Advertise here: Contact FM.

Of course you realize, this means war!
July 7, 2009 10:10 PM Subscribe

Google Chrome OS:Google says it will release a new operating system, built around its Chrome browser, which will be open source and will initially be targeted at netbooks. Shipment is expected second half of 2010. No response yet from Microsoft.

This is the biggest shot across the bow of Microsoft since Navigator and OS/2 and Java. Gates is retired now. I wonder if Ballmer knows how to cope with an existential challenge?

posted by Chocolate Pickle (227 comments total) 18 users marked this as a favorite

About fucking time.
posted by pompomtom at 10:16 PM on July 7 [1 favorite]

I'm not surprised, but I'm stunned.
posted by jragon at 10:21 PM on July 7 [1 favorite]

I think they telegraphed this by calling the browser 'Chrome' in the first place.

"The Browser is the operating system." is a mantra I have heard a lot in the last few years.
posted by rokusan at 10:25 PM on July 7 [1 favorite]

As a Linux user, I hope this goes really far. The more people using the kernel, the more the resources and volunteer man-hours will that will be devoted to keeping the Linux kernel cutting edge. Also everybody will have a strong incentive to make their web content Linux-compatible. This could kill directX/silverlight, if we are lucky.
posted by idiopath at 10:25 PM on July 7 [8 favorites]

I was just looking into Windows 7 today. For the light/cheap consumer version, the full retail package is going to be $199, upgrade version will be $99 (there's a special deal on right now, but I'm not going to take the jump). When my entire netbook cost me $375, the value just isn't there.
posted by falconred at 10:28 PM on July 7 [6 favorites]

HTML5 will kill Flash and Silverlight dead.
posted by signalnine at 10:30 PM on July 7 [3 favorites]

Makes sense to ditch X + ($DE). Most mortals definitely don't need that power of X. The Linux kernel is a natch at this point. A pretty new face over a *nix foundation... lemme see... who else did that? I hope that those who want it can access the power of the command line and make use of userland tools.

Apps that run anywhere and everywhere. This has been promised before... we shall see how this plays out. The thought is a nice possibility.

The multi-platform X86 and ARM support is nice. = ) The latest ARM offerings are looking really swell... with HD video support and super low power consumption. I'm thinking of picking up a Touchbook.

Mmmmmm.... ARM. I loves me some RISC.

:Hugs SunBlade 2000:
posted by PROD_TPSL at 10:30 PM on July 7

I'm not stunned, but I'm surprised.
posted by ryoshu at 10:30 PM on July 7

Will this be as big as Google Wave or Google Calendar?
posted by sien at 10:34 PM on July 7 [3 favorites]

They briefly address the Android/Chrome overlap but I wonder how that'll play out over the long term.
posted by asterisk at 10:35 PM on July 7

Written entirely in javascript.
posted by stavrogin at 10:37 PM on July 7 [6 favorites]

I'm neither stunned nor surprised.
posted by Krrrlson at 10:40 PM on July 7

HTML5 will kill Flash and Silverlight dead.

Yeah, that doesn't seem like it's going to work out, since they gave up and won't have a common format that will work cross-platform and cross-browser, not to mention the lack of IE. For instance, Youtube plans to use HTML5 with H.264 encoding — but Firefox won't support that due to patents. A nightmare.
posted by smackfu at 10:44 PM on July 7 [1 favorite]

Makes sense. Google makes money from advertising (boatloads of money). Like the over-the-air television of the 60s and a few decades beyond, they offer free services to attract clients, offering a base of contacts with which to sell their advertising. The more clients, the more advertising money they make. Think about Android for a moment. They were already offering cell phone services on devices like the iPhone. But they didn't have control. Apple wasn't putting all the hooks in that Google wanted for its services, so they hired one of the key Sidekick developers and created Android. Web browsers didn't run Google's services fast enough so they hired one of the key Mozilla developers and created Chrome.

They are just continuing to do what they have always done.
posted by eye of newt at 10:44 PM on July 7

It will be interesting to find out what the business model is behind this.

This article hints at always connected internet, network storage of files, and of course the OS has to run on something...I'm curious if the OS will be ad driven, subscription based - or if it will be free to the consumer in exchange for ad supported application use.
posted by device55 at 10:45 PM on July 7

I really like the idea, and I would like to say I'll use it, but I probably won't. My two computers are running OS X, one of them with a Windows partition. The Windows partition is used only for games.

Given my setup: if Chrome is somehow able to run PC games -- easily, without a delay in release date, free/very cheaply -- then I would replace Windows in a heartbeat. Since that's not going to happen, Chrome could only serve to replace OS X, which also isn't going to happen.

Still, I guess with netbooks rising in popularity, this is a good time to release a really simple, lightweight OS.
posted by voltairemodern at 10:47 PM on July 7 [1 favorite]

Youtube plans to use HTML5 with H.264 encoding — but Firefox won't support that due to patents

I bet that, within a couple of days of youtube starting to use HTML5 and H.264, there's a FF extension to handle it; and I expect that YouTube is aware of that likelihood too.
posted by nonspecialist at 10:48 PM on July 7 [10 favorites]

Still waiting on that Chrome browser for Mac OS, actually...
posted by hippybear at 10:52 PM on July 7 [9 favorites]

I checked my calendar, no its not April 1. Yes, thats really Google's blog. I shouldn't be this surprised, but I am.
posted by SirOmega at 10:52 PM on July 7

signalnine: "HTML5 will kill Flash and Silverlight dead."

Well, it would have if everyone could have agreed on a goddamn video codec. Google certainly didn't help in that department. [Curses incoherently about H.264]

Unless there's a standard codec that gets universally incorporated into all the major browsers, I think Flash will probably be safe. It has tremendous inertia, and if there's any lesson to be learned from Microsoft, it's that you can make a tremendous pile of money if you get just about everyone locked into your product early. Flash has manged to do that, and dislodging it from its position is going to be a battle as long and vicious as the one to unseat Windows as the standard desktop PC OS, if one that will be a lot more transparent to users.

Anyway, people have been talking about a gOS for years now and I'm pretty surprised they're actually doing it. I figured it was going to be another piece of never-deliverware, like IBM Desktop Linux.

I wonder how much of their codebase, if any, will be shared between the netbook OS project and Android. The netbook and smartphone platforms seem like ones that differ mostly in terms of user interface, but could share a common OS and perhaps even application stack if you designed them carefully to take advantage of the strengths of each platform.
posted by Kadin2048 at 10:53 PM on July 7

A nightmare.

I dunno. HTML 5 stands to benefit everyone who isn't Microsoft or Adobe. I imagine the fine folks at Apple, Mozilla, Google, Opera, IBM, et al will come up with a solution to the common codec problem - even if it means a sweetheart deal where Mozilla bundles a quicktime installer with Firefox.

IE usage has also been dropping steadily year after year, and besides techniques already exist for hosting and displaying HTML 5 video which degrade gracefully: http://camendesign.com/code/video_for_everybody
posted by device55 at 11:05 PM on July 7 [2 favorites]

I hope it's actually lightweight. Lots and lots of people still have the Asus EeePC 900 model (myself included, thanks woot.com), but if it's larger than 4GB, it will be a no go for us lot.

I also wonder when they're going to get around to giving us the option to filter images/ads/flash on the browser, unless that's changed since I took it for a spin. It took me all of 10 seconds to bolt for the uninstall button as soon as I saw my first unfiltered internets.
posted by saturnine at 11:08 PM on July 7 [2 favorites]

My Samsung NC10 just got wet.
posted by wfrgms at 11:09 PM on July 7 [1 favorite]

If Google provides an alternative to running Linux on a netbook that isn't deadeningly slow, I wish them all the luck in the world.

If anything, this will light a fire under Apple's ass to get their OS X tablet PC to market.
posted by Blazecock Pileon at 11:25 PM on July 7

I wonder how much of their codebase, if any, will be shared between the netbook OS project and Android. The netbook and smartphone platforms seem like ones that differ mostly in terms of user interface, but could share a common OS and perhaps even application stack if you designed them carefully to take advantage of the strengths of each platform.

That would be the strategy I'd see them take (as with Apple). With a haptic interface, a netbook-sized device is only a smartphone / iPhone / iPod Touch with a larger screen (and perhaps a tiny physical keyboard).
posted by Blazecock Pileon at 11:28 PM on July 7

If Google provides an alternative to running Linux on a netbook that isn't deadeningly slow, I wish them all the luck in the world.

It's still linux... it just has Google stuff sitting on top.
posted by pompomtom at 11:28 PM on July 7

This is the biggest shot across the bow of Microsoft since Navigator and OS/2 and Java.

Well, hardly, in the case of OS/2, since it was co-developed by Microsoft.
posted by grouse at 11:30 PM on July 7 [1 favorite]

It's still linux... it just has Google stuff sitting on top.

Windows XP and Hackintosh on a netbook are not very usable, due to hardware constraints, and Linux + X11 is not really an OS for consumers. Linux is fine for power users, but it's just not for everyday folks and it's unlikely it will be for another 5-10 years. However, Google might just be the group to successfully put a stripped-down, well-engineered UI onto a kernel right now. Guess we'll see.
posted by Blazecock Pileon at 11:35 PM on July 7

So it's like Moblin, only from a company that gets revenue by monitoring user activities? No thanks.
posted by cmonkey at 11:36 PM on July 7 [8 favorites]

gOS
posted by felch at 11:45 PM on July 7 [3 favorites]

Until ogg is supported by a power-efficient single-chip solution, H.264 will reign.
posted by five fresh fish at 11:46 PM on July 7 [4 favorites]

Isn't Google real revenue-generator data-mining of individual's web browsing, email, chat, and other data habits?

It scares the bejeesus outta me when I consider what my Google searches would reveal about me. Sometimes MeFi threads lead me to look up some pretty … narsty things. Fortunately, there's also my Reddit click-history, which reveals that I'm a whore for a snappy headline, having never quit learned to not click the link.

Oh, to unsee the things I've seen.
posted by five fresh fish at 11:52 PM on July 7 [9 favorites]

I like what I see in that gOS link; computer as oversized smartphone. I can completely see the value in that. For my mom, say.

for me personally, not so much
posted by davejay at 11:54 PM on July 7

It will be interesting to find out what the business model is behind this.

The business model is "cloud-based everything".

Basically, it's been Google's dream forever to get back to old-school client-server computing where we all have dumb terminals, and Google is the gatekeeper in the white lab coat. All your apps and all your data will be in their servers.

Kinda scary to think about, especially at 3am.
posted by rokusan at 11:56 PM on July 7 [4 favorites]

g for glad?
posted by TwelveTwo at 12:02 AM on July 8 [3 favorites]

It's just not right that my OS can also show my butt thong.
posted by twoleftfeet at 12:03 AM on July 8

Will this be as big as Google Wave or Google Calendar?

I'm sure it'll be at least as big as Google Lively.
posted by Pope Guilty at 12:05 AM on July 8

an existential challenge?

Oh, bullshit. Existential challenge my ass.

If you notice, Google almost always fails at everything except search. But even if they were brilliant, impossibly brilliant, why on earth would you think that Google has a chance when neither Linux nor Apple have successfully dislodged Microsoft from the desktop? It's not like, working from scratch, they're going to come even close to what Ubuntu offers, but Ubuntu has maybe a 1% market share. And Apple's software is extremely slick and polished, offers many native apps that Windows doesn't, requires almost no user knowledge, and they've managed to take maybe 12%. Cracks in the foundation of the Windows monopoly maybe, but existential threat?

Hyperbole less. Think more.
posted by Malor at 12:10 AM on July 8 [8 favorites]

Dunno what you're looking up, cmonkey, but having someone know I'm a metafilter tragic and compulsive wikipedia-er seems a small price to pay for a nice distro with a clean, integrated UI, especially if it means I can avoid x11.

People don't have to be doing anything they're ashamed of to not like the idea of everything they do being monitored and archived by an unaccountable corporation. And is there any evidence they rewrote X.Org? It sounds to me like they went the Moblin route and just rewrote the desktop environment using existing toolkits that work with existing X servers.
posted by cmonkey at 12:11 AM on July 8 [2 favorites]

Let me try again: they don't care about Windows. Or OSX. Or Ubuntu.

This is all about being an OS that runs a web browser and nothing else. Turn on device, you're in a web browser. Quit browser, device turns off. That's all it is for.

What percentage of contemporary "computer use" is just the web anyway? We old-fashioned types thinking about running applications on our computers... we're dinosaurs. Very soon.
posted by rokusan at 12:18 AM on July 8 [9 favorites]

I'd hope that Firefox would/will run on this beast. As noted in comment just above (saturnine), I have not and I don't intend to browse the internet with flashing banners and all of the other jive. Google, with their whole 'don't be evil' routine has surely been evil toward Firefox, and the whole open source software community, or so it seems to me. As pointed out above, be wary of people giving you too many things, there's usually a price tag on it somewhere, or one that will show up.

Other than that, I think it's great, potential to really hammer Microsoft in the nose -- WHAP !!! And if it's light -- and if it's aimed at netbooks it'll have to be light -- if it's light it'll keep lots of machines going for lots of years, machines that would otherwise be dumped, for example the old beast compaq celeron desktop I've got sitting idly by.
posted by dancestoblue at 12:27 AM on July 8

I have very mixed feelings about "the cloud is the computer"

On one hand, all devices would be disposable, in that you'd never have to worry about reconfiguring everything every time you get to a new machine. New Internet tablet, plug in your passwords and you are off and ready to go! Storage? no-worries, you have an infinite size of hard drive and buyable raw CPU processing. Having your applications hosted on a cloud also allows easy collaboration and presentation. The cloud allows for an incredible amount of freedom in location and hardware.

The other end of this is that you never "own" your applications, you can't back them up, and unless you host the apps yourself, you are never in control of the how well the service works. And if you have to pay for these apps, it's a subscription - subscriptions for software really rub me the wrong way - whether is ArcGIS, Adobe's Acrobat.com, or any other kind of subscription. I like being able to run old software - Office XP works just fine for me, I don't need all the new crap. I like to feel those performance gains that new hardware gets me when I run old software.

An emotional problem that I have cloud computing really comes down to: I am very dependent on alot of other computers and alot of other people for my everyday computing experience. While it could be argued that I am as dependent on these same faceless people for all my other applications, it doesn't feel that way. I don't get emails from my word processor telling my terms of service are changing and my external hard drive doesn't bill me every year for storage costs. My relationship with my current computer consists of one off transactions, never to be thought of again.

Privacy, backup, application performance [application /OS tweakability], having a fixed cost for computing - these are things that will be way more complicated by moving to a complete cloud driven system. And the cloud is going to built on trust, innumerable TOS's and an arguable fragile network.

Having a pure cloud system also means that I'll never be able to fool myself into thinking that I will only buy this piece of software and that's the only time I'll ever have to buy that hammer again (I buy the good hammers.) The hammer will always be on loan, rented to me - it's not my hammer, it never was. Who wants to pay rent for a freaking hammer?
posted by bigmusic at 12:27 AM on July 8 [15 favorites]

This is a weird press release, and I'm not 100% sure why Google has (a) released it when it did and (b) why they want to go into the OS market at all.

For now then, I'm dubious. The timing of the press release seems to be designed specifically to throw some tacks under the wheels of Microsoft & Windows 7. I can understand that. I believe that MS have started to make up lost ground with Win7, and the latest iteration of development tools are getting good reviews. It's only a matter of time before those good feelings start spilling over into usage of "bing".

The OS itself isn't necessary, and as a lightweight OS would be a retrograde step for most users, it's unlikely to get any traction for anything but consumer devices. I'm not ruling Google out, because it's full of people who are a lot smarter than me, but I can't see this going anywhere.

(Unless, Googles target isn't Microsoft. With the iPhone attacking Android and an OS designed to capture that mac friendly developer crowd, I'm wondering if Google aren't trying to get to second place by slapping Apple around a bit)
posted by seanyboy at 12:48 AM on July 8

Google, with their whole 'don't be evil' routine has surely been evil toward Firefox, and the whole open source software community, or so it seems to me.

Are you insane? Firefox was almost 100% funded by google for a while, and google has released tons of open source code, and tools for distributing and working on open source code.
posted by empath at 12:50 AM on July 8 [11 favorites]

Just so you know the kinds of numbers we're talking about -- google provides close to 90% of the funding for Firefox -- it adds up to 100s of millions of dollars over the past 4 years or so.

Theoretically, it's for 'search royalties', but really, does anybody think that google NEEDS to pay to be firefox's default search engine? If firefox switched to anything else, people would throw a fit.
posted by empath at 12:53 AM on July 8 [1 favorite]

I think this might be relevant. Not the end of ActiveX I don't think, indeed perhaps only the true beginning... as rokusan said, back to old-school client-server.
posted by XMLicious at 12:54 AM on July 8

On the video codec thing.

As far as I know, HTML 4 doesn't specify what image formats need to be supported, and not all image formats are supported by all web browsers. I don't see why video should be any different. Whatever youtube goes with is going to be the winner though.

God I can't wait to never have to deal with flash again.
posted by empath at 1:00 AM on July 8

HTML5 will kill Flash and Silverlight dead.

What makes you think that? Apple are doing their best to kill the multimedia components.
posted by rodgerd at 1:01 AM on July 8

rodgerd -- ridiculous. Youtube is going to support h.264, apple is going to support it, phones and netbooks will support it in hardware, h.264 is going to win, whether firefox likes it or not.

Ogg is never, ever, ever going to be a relevant audio or video format, except for neckbeards who care about open source and never pay for music or videos anyway.
posted by empath at 1:07 AM on July 8 [13 favorites]

Linux is fine for power users, but it's just not for everyday folks and it's unlikely it will be for another 5-10 years.

Heh. I seem to remember sentiments along those lines both 5 and 10 years ago...

I have often thought, myself, that it would be cool to have some device where you press a button to turn it on, and it gives you an instant, fast web browser.

Then I've remembered the actual software I require, for both work and play, that I can't ever see being ported to or usable on a browser platform.

I mean, it may be true that the majority of people spend the majority of their computer time in a browser, but how many people truly do everything in a browser? Even if things like word processors or spreadsheets were possible in a browser (the current examples of these are really nothing more than toys), what about, say, R? ArcGIS? FruityLoops? Lightroom? Cloud computing may be fashionable, but it's got a hell of a long way to come.

I suspect this effort by Google will go much the way of the other Linux netbooks...people like the idea, and the price, then discover it won't run that one piece of software they need.
posted by Jimbob at 1:15 AM on July 8 [3 favorites]

Even if things like word processors or spreadsheets were possible in a browser (the current examples of these are really nothing more than toys), what about, say, R? ArcGIS? FruityLoops? Lightroom?

Jimbob, look at the Google Native Client I linked to above. Doesn't take care of the bandwidth part of the issue but clever caching plus Moore's Law might.
posted by XMLicious at 1:28 AM on July 8

I'd hope that Firefox would/will run on this beast.

I would think that it would behoove Google to make sure that other browsers run on their OS so that they do not run afoul of any more EU or US DoJ antitrust concern. If this new OS is merely Google's flavor or Linux then there should be no reason Firefox would not run.

I'm a bit concerned about Canonical/Ubuntu, who have been gaining new users for their Ubuntu Netbook Remix. Users who might have tried UNR might end up with Google Chrome OS instead. It'll be interesting to see how the Linux community views Google coming into the mix in this manner.
posted by gen at 1:35 AM on July 8

BSD is even more for the power user than linux is, and yet Apple managed to build an incredibly user friendly OS on top of it.

I wouldn't expect google OS to look anything like Ubuntu or Red Hat. It's not using KDE or Gnome...

It'll be interesting to see a 100% cloud computer in action.
posted by empath at 1:42 AM on July 8

From what I have heard, google will not be using X11 for their OS.

If you don't need a remote desktop, and you are only really shipping one application, not using X11 makes perfect sense. X is an old and hairy ball of cruft. The disadvantage to not using it is having a usable set of apps that do not rely on it.

I am sure some enterprising hacker will release a version of X11 and the rest of the standard userland that mainstream Linux apps (such as firefox) rely on. This will most likely be 10x the installed size of what the google OS would ship with by default.
posted by idiopath at 1:59 AM on July 8

I started in the mainframe era, loved it when personal computers came along, and now use Google's web ('cloud') apps extensively.

Mainframes, in the era of personal computers, had several problems that constricted individual users. Time was billed by the CPU-second, storage was often doled out in tiny quantities, and you had to go to the terminal room (what a pleasant name) to work. If you wanted more storage, you could get a tape, but then it would go into the special room with the sysops, who would physically mount it on one of the handful of tape drives available. You might wait 20 minutes before your tape was mounted, and have to finish in 10 to make way for the next user.

Google? Their stuff works, it's free, and the quotas are pretty generous. I don't feel the pain I did in the mainframe days.
posted by zippy at 2:01 AM on July 8 [1 favorite]

Jimbob, look at the Google Native Client I linked to above. Doesn't take care of the bandwidth part of the issue but clever caching plus Moore's Law might.

So now we've got this native client, which will run code off the internet; you just visit a site, it downloads the code, and runs (like a java applet). Developers start writing applications for this client, and in time these applications become increasingly sophisticated, and the cost of that sophistication is file size.

Ok, so we use clever caching (like java web start). So then the applications are storing their files of non-trivial size on the users' harddrives. People's hard drives fill up and they want to be able to prevent applications from downloading in the first place, and also to remove applications.

Ok, so we prompt the user before downloading, and also add a "uninstall web apps" program to remove them from the cache.

From a user experience perspective then, how does this differ from downloading and installing applications the traditional way?

Of course, maybe it's more secure. However, any program with enough permissions to modify files in your 'my documents' folder (like a word processor, or image editor, or spreadsheet, or really any program that does anything useful) has enough permissions to trash perhaps the only irreplaceable files on your computer.
posted by Pyry at 2:06 AM on July 8 [2 favorites]

Even if things like word processors or spreadsheets were possible in a browser (the current examples of these are really nothing more than toys), what about, say, R? ArcGIS? FruityLoops? Lightroom?

If you're running those apps, you're a "power user", not a "consumer".
posted by Blazecock Pileon at 2:18 AM on July 8 [2 favorites]

From a user experience perspective then, how does this differ from downloading and installing applications the traditional way?

I would imagine that it's not going to be that different, perhaps just more automated. Perhaps resembling the Linux package management system more than a Mac or PC - where you simply go down a list and tick the apps you want. Or in the way that some products right now JIT compile a feature or interface of the application the first time you use it maybe it would download and compile on first use.

But I don't think it's really for any practical benefit to the user; the benefit to the user would be primarily pricing. If Google could make it so that the OS and the core apps were free or had a negligible price, and could use the native client as a backup to quickly port (and/or encourage the open source community to port) anything that turns out to be a killer app and as a deus ex machina to provide web applications with features they couldn't normally achieve like direct hardware access - but without any of the security problems inherent in ActiveX, which is what the Native Client is designed to do - they might be able to capture a large segment of both the OS and browser markets. I would think that the object would be to gain control of that stuff as a lost leader (i.e. without necessarily making revenue) first and then use that leverage to make money in other ways.
posted by XMLicious at 2:38 AM on July 8

cut off their air supply
posted by Hello Dad, I'm in Jail at 2:45 AM on July 8

A different way of saying it: what Microsoft has really proved is that first to market very frequently beats a better technical design. I think that Google might be arranging things so that they can beat Microsoft at their own game and in the end have the better technical design too, if the application always resides on their servers (or perhaps within some peer-to-peer network that they control, but in either case providing a much more reliable and cost-effective way of fixing problems in an application after it has been shipped - as opposed to having a user install and uninstall things) and they simultaneously control both the OS and the browser.
posted by XMLicious at 2:50 AM on July 8

Hey, maybe it will work like Chrome does for me and NOT render all the Zalgo mumbo-jumbo. Life would be grand.

/Not a slam against Chrome. It's my favorite browser these days.
posted by educatedslacker at 3:12 AM on July 8

If you notice, Google almost always fails at everything except search.

And mail, and maps. Not to mention IM (how many of your IM contacts are still using, say, MSN or ICQ? Most of mine are on XMPP, typically via Gmail), and keyword advertising, and Google Earth.

There are, of course, a lot of Gogle projects that have fallen by the wayside, but are there more than, say, Yahoo! or Microsoft projects that have failed?
posted by acb at 3:24 AM on July 8 [2 favorites]

Ogg is never, ever, ever going to be a relevant audio or video format, except for neckbeards who care about open source and never pay for music or videos anyway.

Precisely. Ogg is DOA as a mainstream media format precisely because it is an open format. No patents to license = no DRM mandates in the licensing conditions = Big Content gets cold feet, refuses to license its content = no advertising revenue.
posted by acb at 3:28 AM on July 8 [4 favorites]

If this new OS is merely Google's flavor or Linux then there should be no reason Firefox would not run.

Calling it "Google's flavour of Linux" is a bit misleading; it'll basically be a Linux kernel, a framebuffer driver, a window manager of some sort and the Chrome browser engine. Architecturally, it'll have more in common with Palm WebOS than Ubuntu.
posted by acb at 3:30 AM on July 8

acb: theoretically, you could install glibc, GTK, X11, etc. etc. etc. and run firefox, even if it ships with none of these.
posted by idiopath at 3:39 AM on July 8

If you're running those apps, you're a "power user", not a "consumer".

Seriously? Using anything beyond a web browser and a word processor makes one a "power user" these days? By these criteria, I'm trying to think of anyone I know who isn't a total hardcore hax0r.
posted by Jimbob at 3:55 AM on July 8

theoretically, you could install glibc, GTK, X11, etc. etc. etc. and run firefox, even if it ships with none of these.

If my grandmother had ~~balls~~ wheels, she'd be ~~my grandfather~~ a tram.
posted by acb at 4:06 AM on July 8 [1 favorite]

Awesome! Will the new OS stay in beta for 5 years too?
posted by belvidere at 4:13 AM on July 8

I don't think it's going to be anything more than a kernel, a framebuffer driver, and the absolute minimum required to get Chrome running. No window manager necessary.

This is not aimed at people who want to run "other applications" -- what would you want to run?

Mail? That's gmail, via Chrome.
Word processing/spreadsheet? Google Apps, via Chrome.

Everything is already web-based enough that all you need is a browser. And, if it's Chrome, then the G gets to track your usage and target their inline advertising more accurately, thereby improving the attractiveness of their offering and thus their bottom line.

It's taken the back-end development, over time, to get to this point because without Apps, gmail, and everything else going web-based, it'd never work; but now we're there, it's an obvious and dead simple move to make.

This is either Google capitalising well on the state of the game, or evidence of some seriously long-term planning coming to fruition (long-term in Internet Time). I think it's the latter. Well played, Google.
posted by nonspecialist at 4:19 AM on July 8 [1 favorite]

Seriously? Using anything beyond a web browser and a word processor makes one a "power user" these days?

I don't think R and ArcGIS are something most grannies are running on a daily basis.
posted by Blazecock Pileon at 4:32 AM on July 8

Hmmm, it's cool to see all this in discussion about the intricacies of Google's business model, whether or not they've "failed" with all the services they give away for free, etc.

However, I recently read a quote that made it all pretty clear: The more people who use the Web, the better it is for Google.

That's it. There's your business plan.
posted by jeremias at 4:36 AM on July 8

I don't think it's going to be anything more than a kernel, a framebuffer driver, and the absolute minimum required to get Chrome running. No window manager necessary.

So its audience will be limited to the people who are satisfied with one browser window?

If they're aiming this to be an appliance to give to proverbial elderly relatives to look up recipes on and send the odd email, that'd be adequate. But if I understand it correctly, Google are aiming this to be a proof of concept of the web being a full replacement for a traditional OS. In which case, one browser window, even with tabs, ain't gonna cut it.
posted by acb at 4:41 AM on July 8

However, I recently read a quote that made it all pretty clear: The more people who use the Web, the better it is for Google.

Exactly.That's why Google are sinking money into making and giving away things like Android, Chrome and such. Their presence encourages people to use the web in new ways, increasing the value of Google's advertising-backed properties. And their open-standards-based nature poisons any attempt by Microsoft to use their advantage to enclose the web with proprietary standards.
posted by acb at 4:55 AM on July 8 [1 favorite]

My mother uses Audacity to record hypnotherapy sessions and burns them. Don't try to tell me she's a power-user.
posted by Lemurrhea at 4:58 AM on July 8 [2 favorites]

So its audience will be limited to the people who are satisfied with one browser window?

Sure -- although you and I might find it stifling, think of tabbed navigation. You have all your apps that you're running listed right there at the top of your browser window. It's an interface paradigm that's well established, and a lot less confusing for the Average User than raising/lowering or even, gods forbid, multiple workspaces (note: I am not an Average User).

As for the question of interfacing with peripherals, there's a lot that can be done with the plugin/extension model of operation; after all, the kernel provides the device nodes, and if you only have the one application using them (chrome), it makes it all much more simple.

Only time will tell whether it appeals to enough of the "market" to actually begin to make inroads; I suspect a lot of its traction will come from how well Google manages the market for plugins/extensions -- free to download/install from anywhere, written by anyone, or more AppStore like, with certificate signing etc.

I can imagine the posts on boing boing now: "ChromeOS jailbroken!", "Huzzah for the 1337 haxx0rs among us!" and so on.
posted by nonspecialist at 5:07 AM on July 8

idiopath: From what I have heard, google will not be using X11 for their OS.

No, instead, we have AJAX and HTML which is about as clunky as vintage X11 for building user interfaces.

nonspecialist: Mail? That's gmail, via Chrome.
Word processing/spreadsheet? Google Apps, via Chrome.

My antipathy for MSOffice is legendary but Google Apps is little more than a networked rich text editor, lacking key tools that I need on a day-to-day basis.
posted by KirkJobSluder at 5:15 AM on July 8 [2 favorites]

Sure -- although you and I might find it stifling, think of tabbed navigation. You have all your apps that you're running listed right there at the top of your browser window.

Sounds suspiciously like the Apple Dock, the Microsoft Taskbar, or the equivalent UI widgits that have been standard under X11 since Motif and NextStep.
posted by KirkJobSluder at 5:18 AM on July 8

I imagine this will be running on the Crunchpad at some point.
posted by tomierna at 5:25 AM on July 8

This is great. Linux always felt like it's 97.5% there and nobody having enough resources and persistency to bridge the last 2.5%, and I also always thought that X was dragging things down in a huge way and replacing it was too big of a job for anyone.
posted by rainy at 5:27 AM on July 8 [1 favorite]

The biggest potential win for Native Client/NaCl is that it's a straight cross-compilation target for existing C code. No porting to the JVM or CLR, no/little reworking into a 'web' app, little sunk cost if it ends up not being the Platform Of The Future (which it will).

So you've got three directions for apps to be run on a device with no OS to speak of:

Rich web apps. Gmail, gtalk, google docs, facebook, the kind of things Ajaxian creams its jeans over. Realistically this is a good chunk of everyday functionality even for a power user, maybe it's 100% of the functionality for my mother who isn't.
Native apps ported to NaCl. Anything that doesn't quite work as a web app, I'd expect things like Photoshop, GIS software, any kind of professional app to either be ported directly or for incumbents to fill the space with a work-alike. I can't imagine many MMOs won't port clients.
Legacy apps running in a virtual machine. I'd imagine making a hypervisor run under NaCl would be...interesting and probably not all that quick but if you've got enough bandwidth you could point a virtual desktop client of some kind at a virtual machine in a datacenter somewhere. Big corporate shops are already moving in this direction, look at VMware's VDI for example. 3D is coming to virtual machines eventually as well, legacy games have a potential in here too.

This is all dependent on good, low latency bandwidth; I'd expect Google to push even harder for things like whitespace bandwidth and open-access to networks now that they're pushing for mass adoption. An Apple-like App Store might work too.
posted by Skorgu at 5:35 AM on July 8 [2 favorites]

First, there was the rise of OSX.
Then the rise of Ubuntu.
XP did not seem to get fully appreciated until Vista was released...
And now Google is coming in...

Something tells me that MS will be wanting to make an actually operating operating system now... This as I comment using PalmOS...

I smell a great competition.
posted by JoeXIII007 at 5:37 AM on July 8 [1 favorite]

iPhoto, iDVD, iMovie, GarageBand… these are all non-power users apps that basically will suck when implemented as a web app.

Any rich media web-apps will always be painfully bad until we have the bandwidth. And I don't know about you guys but there really hasn't been a widespread bandwidth leap.
posted by schwa at 5:44 AM on July 8 [6 favorites]

When my entire netbook cost me $375, the value just isn't there.

This is why Microsoft invented Windows 7 Starter edition. Details here. "For the first time, we will be making Windows 7 Starter available worldwide on small notebook PCs."
posted by smackfu at 5:48 AM on July 8

However, I recently read a quote that made it all pretty clear: The more people who use the Web, the better it is for Google.

Also -- and not at all incidentally -- it damages Microsoft's ability to leverage their OS profits to complete with Google in the advertising space.
posted by Slothrup at 5:56 AM on July 8

Just to pipe in about Malor's comment re: Ubuntu and MS market share. I get what you're saying -- MS is very entrenched, but really, Ubuntu? I'm a total computer nerd and even equate Ubuntu with Neckbeard. Biased, yes, overlooking a great linux install, yes, but if I have this perception.. how is a lay person supposed to trust it?

Google, on the other hand, is everywhere. "It's Google!" they'll say, and it will have a big G and some pretty art, and they'll load it on their machine... though I guess we're only talking netbooks and the like right now, so this is a little premature.

I just think Google has much better capability to enter the market than Ubuntu or even OS X (Apple? But it doesn't run most of my progrtams right?) did. And even then OS X numbers keep climbing...
posted by cavalier at 6:02 AM on July 8

The timing of the announcement is interesting for something that won't be available for over a year. Scoble claims that there's a "big Microsoft announcement" on Monday. This would be consistent with the announcement of Wave (perfectly timed for the day on which Bing was announced) and Chrome (released the same month as IE8 beta 2).

It's classic business strategy, of course, but when combined with Google's deliberately projected attitude of "aw shucks, we're just folks", it actually scares me a little.
posted by Slothrup at 6:04 AM on July 8 [1 favorite]

Is there going to be filesystem access? The web is terrible at sharing, ironically. If I want to email a (private) hires picture from my Flickr, how do I do that without stepping outside the browser?

The thought of the browser-being-the-OS makes me shudder: it's like System 6 apps that take over the machine, except without a file open/save dialog to let you escape. And no Finder.

As for the apps: there are reasons that people who buy Linux netbooks are dissatisfied, and it's not because they don't like the widgets on Firefox. It's because they can't do their other stuff.
posted by fightorflight at 6:17 AM on July 8

Chrome is an operating system now? The last time an application expanded its feature-set and became an OS, it created a new religion and an associated holy-war.
posted by the cydonian at 6:19 AM on July 8

"Is there going to be filesystem access? The web is terrible at sharing, ironically. If I want to email a (private) hires picture from my Flickr"...

Probably /home access combined with the ability to easily add photos from Picasa. Just like Android, in fact.
posted by jaduncan at 6:21 AM on July 8

Scoble claims that there's a "big Microsoft announcement" on Monday.
From the link:

Josh: part of the Microsoft announcement on Monday runs on Google Chrome (and Firefox for that matter). - Robert Scoble

My take: Silverlight 3 (or Silverlight-for-mobile) will be out. SL 3 offers an out-of-browser mode; so only a part of it runs on browsers. (Then again, Silverlight currently does not run on Google Chrome, so) I was thinking Azure was also another likely option, but a) it wont be "big", and b) no big deal about running across browsers and needing many videos.
posted by the cydonian at 6:30 AM on July 8

I'd really like for HTML 5 to take off in a big way. I'm not sure it's in a position to kill anything, though. I recall hearing a lot of the same buzz about SVG killing flash in the same way, and that never really happened. If the codecs were loadable chunks of code, like protocol handlers or plug-ins, consensus might be easier, but that'd muddle the waters for end-users.

Based on what Google did with Android, I think they've got a good shot at making a good, usable OS for netbooks. People have already tried jamming Android into netbooks with varying degrees of success, so it seems natural for Google to take control of an evolving need.
posted by boo_radley at 6:33 AM on July 8

Here are two additional problems with the whole client/server cloud computing scenario.

First, it's certainly the case with my work that no single software vendor really supplies it all. In my day to day work, I'll use software from six different companies, and I can do so seamlessly after having logged into the employer's file server and mail server. If I need to port data between applications, it's a simple matter of either copy-pasting or opening the file in a new application.

With cloud computing however, the service owns both the software and the data. So if you need to get your data out of the Google universe and into the Zoho, Microsoft, Adobe, or Apple universes, you need some sort of transfer mechanism that likely exposes either your credentials or data.

Second, I think it's profoundly unwise to make assumptions about the computing needs and desires of late adopters and lock them into a limited number of applications in advance. The email chain-letter matron may suddenly discover a passion for blogging or video, or dive into the world of networked gaming at age 65. There are hundreds of hobby-specific applications ranging from ham radio, music, cross stitch, quilting, stamp collecting and genealogy.
posted by KirkJobSluder at 6:34 AM on July 8 [1 favorite]

Hmm,
I'd have liked to see them stick with X. It is getting to be quite OK these days, finally, and the Nomachine remote desktop alone makes X worth using - it is absolutely magical.
posted by krilli at 6:42 AM on July 8

The assumption seems to be that we're all connected to the web, all the time. (Naturally, Google is going to make this assumption -- or not care about people who aren't connected.)

I live in New York City amid tons of options for getting and staying connected. Yet my connectivity is far from perfect. At least once a day, I'm somewhere where I can't get a 3G or WiFi signal for my phone. And my home Internet connection (Cablevision, but the same was true when I was with Verizon) goes down sometimes. Also, web sites go down -- including ones like gmail.

Yet decades of personal-computer use has made me expect instant access to my apps and data any time I want/need them. I can't see switching to a web-based OS until the infrastructure is better. At that point, I'll use whatever is easiest, cheapest, most powerful and most fun.
posted by grumblebee at 6:45 AM on July 8 [1 favorite]

Precisely. Ogg is DOA as a mainstream media format precisely because it is an open format. No patents to license = no DRM mandates in the licensing conditions = Big Content gets cold feet, refuses to license its content = no advertising revenue.

MP3 doesn't support DRM, and it's huge. Nobody has a reason to abandon MP3 for Ogg, because nobody cares about patent issues that don't cost them money, and nobody is going to build a device that doesn't play MP3, but does play Ogg.
posted by mkb at 6:52 AM on July 8 [1 favorite]

Eh, the cloud is a fad. What are we, on the information super highway?

Or more specifically the terminal/mainframe setup has been done in the past and I can't see people trading in their Personal Computers that they can use in anyway shape or form, for what advantages exactly?

I think it's really a flaw in Google's thinking that the net is everywhere and all powerful. Sure if you live in Sweden, but out here in the 'real world' I prefer the power and flexibility of my gaming-upgraded real pc, that run Maya and AutoCAD if I wanted to, has no network issues unless I'm using something online and can do, just so much more.

In the work environment it could be different, but as someone who did half an MSCE on Server 2003, even a noob such as myself could imagine the hundreds of access & security things you would want control over in any work environment that the cloud could easily deny you access to, or make life worse. Heck this OS better play nice with printers, wifi cards, cameras and the billions of other things used in a workplace.
posted by Submiqent at 6:55 AM on July 8

Second, I think it's profoundly unwise to make assumptions about the computing needs and desires of late adopters and lock them into a limited number of applications in advance. The email chain-letter matron may suddenly discover a passion for blogging or video, or dive into the world of networked gaming at age 65. There are hundreds of hobby-specific applications ranging from ham radio, music, cross stitch, quilting, stamp collecting and genealogy.

... most of which require generally more resources than are available on the average netbook; we're talking here about the platform that's one small step above a phone with a full qwerty keyboard, not even a full laptop.

The nice thing about having more OSes to choose from though, is that you can choose the one that suits you. Want to play games and run business software supplied by your employer? Pick Windows. Want more control and stability? Pick a Linux variant. Want something that's useful on the road, does what you need as a traveller, without all the patching/fuss/hassle? Use Chrome OS. Or Android.

Look, I'm not saying Chrome is going to be suitable for everything; I never have. It'll be very interesting to see just how many people "settle" for it as adequate (even good) for the purpose, though, especially if it's a viable way to avoid paying the Microsoft tax.
posted by nonspecialist at 6:59 AM on July 8

Submiqent: since it uses the Linux kernel, it will have the same hardware support: if a hardware vendor co-operates with the kernel team it will be perfect, if they don't it will be spotty and won't become available until the product has been around for a few years.

Regarding access and security, that could be the huge drawback here: if they are aiming at the desktop and putting everything in the browser, I am really afraid that can only mean one thing - browsing the web as root. And that is fucking stupid. We can cross our fingers and hope they use the selinux extensions and have a separate administration application, but if not it will be a throwback to the days of windows 98 network security wise. We may even see the first Linux virus in the wild!
posted by idiopath at 7:03 AM on July 8

Hey, maybe it will work like Chrome does for me and NOT render all the Zalgo mumbo-jumbo. Life would be grand.

T͆ͦ̋̋͂́̀h̋̐a̓ͨͦ̉͑ͤt̊ͫͬ͗ͭͣ ͛ͩͫ̾̍͑͒̇͂i̍̀ͤͬ̓sͬ͊͆̆̏̀ ͤ̆b̓̓ͭ͐̄l̄͛̇ͦͥ̌aͧͬtͩ̑̆̅ͯͤ̒a̿ͭ̎̋n͛ẗ́̔ͫ͊̐͂̎ ͒͌̊̒͑ͣͭ̌ḁ̗̬͕͚̜̃͐̆̒̍̍ͤn̞̣̖͍ͮ͌͗̀͂ͫͣ̆t̤̱͉̫̱̫͙̺̋ͣ̾̈́̽̌̉ͥi̩͍̥̻̯̍͛̾̏̑-̘̪̻̩̭͍̮ͮ͐̑ͮO̻̝͇͒ͥl̮̱̥͖̎ͧͯ͒̌ͦ̇d̲̯͓̳̱̫̪͒̍ͬ̂̀͌̋̅ͅ ̝̠̘̪̹̳ͯͤͫ́̈Õ̮̙̲̉̐́͗̓̏n̜͛͋͐e̯̫̖ͦ̄̇͋̿͂̊ ̗̠͚̳̩ͯ̒ͭͅb̛̜͇͙̙i͉̻̜̲̳͙͎͖͠ͅg̳̠͈͘̕͡o̠̙̗̫̺̯̝͔͞ţ̯̠̘̪̮̯̖ͅŗ̣̺͓̺y͙̯̫̜͟͞a̴̿́̈́̿̑nd̃͐ͫ ̵̈ͮͤͣͦh̾a̷s ͛̏͑͗̂ͭ͆͝n̢o̧̊ͩ p̸̔ͫͪ́͌l̵̉ͯ͛̽ͧ̋̅a͛c̴͂̒è̸ͪ̅ͭ ộ̣̲̯̣̹̯͔̹̹̼̝̱͕ͬ̂̑̒ͦ̐̃̄͛͆͐ͨ̅͆̊n͇̰̖͇̭͍̠̥̞̺̝̪̦ͮ̈̏͌ͥ͆ͩ̈́̈̎͗̑ͩͅ ̣̭̟͍̪̼̟̭̭̖͓̟̹̯͆̈̽ͪ̄̐̅̌͆̀̈́ͣͤ͛ͫͣͧ̚t̬̻̪̰͔̘̱͈͙̯̙ͣ̉̆̋ͮ͒ͪ̓͂ͣ̚h̼̟̯̹̪̣̻͉ͭͦ̈ͫ͗̏ͭͤ̚ͅī̼̲̼̥̯̻̲͍͙͔ͯͬͦ͆ͣ̇s̯̠̠̦̹̞̱̞̟͎̟̠̻̈ͪ̂̓̓̒̍ͧ͊̓ ͉͍̪̣̹̮̺͖͖̙͙͖͋̒̊ͣ͌̅́̔ͦͥ̍ͬ̂̊͗̿̿̚ͅͅṡ̝̲̤̮̱̩̱̟̏ͨ̌̋ͯ̌͊̊ͩ̑͑̇̂͋ị̣̫̟͙̰̦͈͚̘̘͂̅͒͐̉t͖̼͕̖͓͈̥̙̲̪̥̹͛̍̈́̌ͯ̽́͆ͭ͗̌ͅẻ̞̮͖͓͕̪̠̘͔̬͇̟̌̋͂ͮ̈́ͨͣ͌ͭ̽̐̿ͥ̈͑̿̋ͅ.̦͍̦͇̰̰̦̥̬͍͇͎ͤ̏̀̈̃͑̈̊̃ͪ
posted by Pope Guilty at 7:30 AM on July 8

The assumption seems to be that we're all connected to the web, all the time. (Naturally, Google is going to make this assumption -- or not care about people who aren't connected.)

Au contraire! That's what Google Gears is all about. Google Calendar and Google Docs both support offline usage. (GMail does too, but only in the unsupported Labs options.)
posted by smackfu at 7:34 AM on July 8

Windows XP and Hackintosh on a netbook are not very usable, due to hardware constraints

Actually, after I invested $85 in upgrading my $180 900A with a fast 16GB flash drive and 2GB ram, it runs XP quite well. It's fast enough booting up and shutting down that I don't bother with standby. There are some configuration tricks listed on the EEE PC forums that really help. I didn't use nlite, either - standard XP SP2 install CD.
posted by rfs at 8:01 AM on July 8

Google Gears was advertised by WordPress as frickin' essential for use with their service. Trouble is, it killed use with Flickr. To use Flickr, you then needed to uninstall Google Gears. This was a very Microsoft-ey thing to do: Get people to advertise your app as absolutely essential, only to have it not work with half of everything, including widely used ... oh, stuff.

Ubuntu Linux is definitely there, meanwhile. What is not there is any ability to use Microsoft-dependent sites, a la Netflix streaming films (which requires ActiveX), live major network and ESPN broadcasts of sporting events via Silverlight (MS is working with Novell on an open-source version called Moonlight, by ABC and associated cable channels don't accept it). The Netflix failure is particularly ridiculous/egregious, considering that its separately sold streaming movie storage thingie is run on Linx.
posted by raysmj at 8:04 AM on July 8

And Google Docs sucks, except if in re to networking/collaboration. It doesn't even register pages, for gosh sakes, just uploaded a 32 page document of mine as a single page.
posted by raysmj at 8:05 AM on July 8

The assumption seems to be that we're all connected to the web, all the time. (Naturally, Google is going to make this assumption -- or not care about people who aren't connected.)

Google's business is being connected. If you're not connected you can't use Google's products.

This isn't going to put Microsoft out of business, but if could change the landscape of Operating Systems. Some may need Windows 7+ or Mac OS or fullblown Linux and their client-based apps. InDesign, Excel's SQL DB queries, CAD, etc won't be going anywhere for a while.

But many people don't need all the overhead. But in today's business model you're paying for Windows even if you only use to fire up your browser to check your webmail and Facebook. Cycles won't be burned up running a bloated antivirus app.

This is just the kick in the pants Microsoft needs to start innovating again, and not relying on the lasted CPU/GPU and giant hard drives to accomplish it.
posted by birdherder at 8:07 AM on July 8

Google Gears was advertised by WordPress as frickin' essential for use with their service. Trouble is, it killed use with Flickr. To use Flickr, you then needed to uninstall Google Gears.

I had Google Gears installed (on Firefox 3.0 on OSX) and had no problem accessing Flickr with it. Are you sure Gears was at fault?
posted by acb at 8:13 AM on July 8

Flickr in Word Press, a distinction there, for posting Flickr photos on my photo site and blog. Yes, Gears was at fault. I looked to see if other were having the same issue, a Google search showed the answer to be, Absolutely. I uninstalled Gears, and the problem was gone.
posted by raysmj at 8:19 AM on July 8

Heck, I still have gears installed and it hasn't stopped me from using flickr, ever.
posted by boo_radley at 8:20 AM on July 8

With Word Press, not Flickr alone, OK? Move along.
posted by raysmj at 8:23 AM on July 8 [1 favorite]

One of the best things about being an engineer at Google was that we never preannounced products before they were built. It took a lot of pressure off the product teams, we could focus on doing something right and release it when it was ready. But as Google's gotten more ambitious with products like Chrome and Android it's been harder and harder for them to keep it quiet. An open source OS is by necessity a collaborative project, I don't think they have any choice but to have disclosed it. Too bad though, because now all the people working on it have to do so with some public attention and speculation.

I'm looking forward to seeing what Google creates. Chrome is really great and a little laptop that runs a web browser and nothing else will be a nice class of device. There's not really a good lightweight OS for that purpose right now except maybe Ubuntu, and it's hard to call Gnome lightweight.
posted by Nelson at 8:56 AM on July 8

So Google put out Gears... WordPress integrates Gears into their service... WordPress advertises Gears integration with their service... WordPress with Gears breaks WordPress with Flickr... Google is Microsoft.

I don't use WordPress, so maybe there's something I'm missing, but wouldn't it make more sense to blame WordPress?
posted by SAC at 8:57 AM on July 8

I don't use WordPress, so maybe there's something I'm missing, but wouldn't it make more sense to blame WordPress?

I think the point is that "the web" is as fragile a platform as "the desktop" once it needs to support lots of different apps which can interact with each other in ways both obvious and subtle.
posted by Slothrup at 9:04 AM on July 8

Michael Jackson dies. One week later, Google announces a new OS.

Coincidence?
posted by blue_beetle at 9:07 AM on July 8 [1 favorite]

I think this is a pretty stupid idea.

Marco from Tumblr wrote about CrunchPad, which is very similar idea.

It makes sense for Google though. They want you to be in your web browser, available to look at ads, all the damn time.
posted by chunking express at 9:11 AM on July 8 [1 favorite]

I think the point is that "the web" is as fragile a platform as "the desktop" once it needs to support lots of different apps which can interact with each other in ways both obvious and subtle.

That may be your point, but I don't really read that to be raysmj's point.
posted by SAC at 9:13 AM on July 8

Anyone have a source for the claim that they won't be using X? The linked article just says "new windowing system", which is so vague it could mean just another X window manager. Of course, if they're really committed to making it so browser-centric that that's the only app ever running, it could make sense to ditch X. But if it's so browser-centric that admin tasks happen through the browser, they'd be re-creating one of Microsoft's worst ideas ever.

As for Linux on the desktop not being ready... ready for what? My wife is a happy Ubuntu user 'cause her needs are covered by Open Office, Firefox, and GnuCash. So far as I know my other relatives' computer use, that'd cover them, too. I know lots of people tied to Windows at work due to a dependence on some particular app; home users, not so much.
posted by Zed at 9:27 AM on July 8

Well, what flickr plugin are you using? I don't have any problems with mine.
posted by boo_radley at 9:35 AM on July 8

wait, never mind.
posted by boo_radley at 9:39 AM on July 8

Chrome browser for the Mac is running nicely for me as a developer release.
posted by mdoar at 9:43 AM on July 8 [1 favorite]

Re: HTML5 and FF- what's going to have to happen is that Mozilla is going to have to just bite the bullet and use native codecs or find enough cash to license H264. They're not going to release a browser that doesn't support Youtube.
posted by signalnine at 10:03 AM on July 8 [1 favorite]

A couple points.

Another company took a linux-like kernel and spend hundreds of millions of dollars to make it "just work." That company is Apple. And I think by any measure it was not only a brilliant move financially, but technically as well.

With that in mind, Google isn't breaking new ground -- indeed, they're digging a shallower hole. They don't have to support any legacy code, or hardware at all. They just need to make something that boots in two seconds and can run a network stack, Gears, a HTML5 renderer and a JIT Javascript engine. With the lightweight loadout this OS will require, the hundred dollar netbook era will finally arrive. In just two years they'll give them away at cell phone kiosks if you'll sign a two year contract.

Back to Apple. If you haven't looked at Apple's website lately, you might not have noticed that they're now operating as a showroom for a big boatload of 3rd party desktop apps. Including "download a demo" links. Next step for them: the desktop App Store. Which will dovetail with the minibook & tablet to be announced later this year. Again, this device will have 3G capability (along with WIFI) and it will be free on contract, or buy it off contract for five bills. And Apple gets a nice slice of any applications bought OTA, just like the iPhone.

So: as far as I can gaze into my crystal ball, it will be Apple and Google in the 2010s, playing market share one off the other. Apple's advantage will be slick hardware, native applications, a killer app store, and that "Apple polish." Google's advantage will be cheaper hardware (plus form factor variations), the free hosting and storage of your stuff, and, for personal users, a gentle upsell to more storage and output features like printing, and for business users, collaboration, security and customization tools.

Google will eventually win the marketshare because their shit will run on every device for truly everywhere computing, while Apple's functionality-based approach will ensure that anyone wanting to do heavy data processing or computation will have at least one of their machines available to do it (although Google will happily provide remote sessions to access these things from anywhere in the world).

Increasingly, Microsoft will be disenfranchised, because their business model doesn't allow them to "get it." Vista was the nail in the coffin there, as far as the future is concerned. Sure, there will always be Microsoft Windows, and software that runs on it, but no one who has a specific need for their platform will give a shit. When grandma starts using Google Docs to type her Xmas letter, instead of MS Word, the coffin will be ready to lower into the ground.

This all sounds fine to me. I'm very much in favour of Microsoft's demise. For twenty years they've tried too hard to sell upgrades that weren't really ready to be used. And people are really fucking tired of computers that don't work as well as cars, televisions and hammers.

The era of appliance computing is nigh, and friends, I'm excited to be a part of it.
posted by seanmpuckett at 10:10 AM on July 8 [11 favorites]

it'll support youtube with flash still.
posted by empath at 10:12 AM on July 8

The app store in the iphone is so good that i'm really amazed that nobody has done something like that for PCs (except for Steam). If MS tried it, they'd be sued into oblivion, though.
posted by empath at 10:17 AM on July 8

Is there any reason a windowing system couldn't be accomplished through ECMAScript and SVG? One of Google's big pushes with Chrome was to make a damn fast ECMAScript. I should think that they could especially tweak to make the window/OS-component very, very fast. Fast enough for web use, fersure.
posted by five fresh fish at 10:18 AM on July 8

Meebo.com has windows that work amazingly well, they'll even tear out from the website, if you want.
posted by empath at 10:27 AM on July 8

I don't think R and ArcGIS are something most grannies are running on a daily basis.

Sorry. I just had a quick look at my Quicklaunch bar to identify some applications I use every day that aren't ever going to be something you access from the cloud via a web brower.

If it was, say, my father bitching here, he probably would have typed in the software he uses to edit and combine short video clips from his digital camera that he then burns to DVD to post to my elderly aunts.
posted by Jimbob at 10:28 AM on July 8

What seanmpuckett said.

Apple and Google have a co-operative competition going. Google isn't into hardware. Apple isn't into search engines, advertising, or even particularly the web.

Co-operatively, Apple's "lightweight" devices and Google's web services and applications cover a huge chunk of what most people need most of the time. Apple's laptop/desktop OS provides the remainder of what power users need.
posted by five fresh fish at 10:29 AM on July 8

hmmm. Meebo.com implemented as an Opera Unite service would provide all the thrills, and none of the spying-on-users that makes me distrust all of these web services...
posted by five fresh fish at 10:31 AM on July 8

I think this might lead to the re-emergence of the desktop PC.

The thing many people are overlooking is that the netbook will be a person's only computing platform, it won't. People will carry their Google powered netbook with them to do quick things, convenient things, time-killing leisure things.

They will the return to their primary unit, which may be a shared unit, as people can do 50-75% of things on their netbook, to do things beyond the capability of the netbook.

There will be no need to sync, as the desktop will be connected to the same cloud as the netbook.
posted by Mick at 10:36 AM on July 8

What seanmpuckett said.
posted by jragon at 10:45 AM on July 8

When I got my first laptop, I swore I'd never, ever go back to a desktop. Portable access to the web is an amazing, life-changing revolution. Once you've got all the information in the world available at your fingertips while you're sitting on the couch, at the kitchen table, on the toilet, lounging in the back yard, in the car — why, you'll never, ever want to go back to being tied down to a single location.

But I'm now back to looking at desktop machines because, in my employment, I was lucky enough to score a dual-display setup. And let me tell you, once you get two high-resolution screens, so that you've got tons of space for work and reference and multitasking — why, you'll never, ever want to go back to a dinky laptop screen for real work.

If I had an iPhone-sized device for when I'm out and about; and a high-resolution device about the size of a comic book for when I need a slightly bigger, casual working space; and a 4000x2000 22" display for getting real work done, I'd pretty much have it made in spades.

I can see how Google and Apple can co-operatively compete to make this a reality. At each level of device, my needs are very different. The smaller the device, the more online it becomes; the larger the device, the more I need a full OS.
posted by five fresh fish at 10:49 AM on July 8 [3 favorites]

When I got my first laptop, I swore I'd never, ever go back to a desktop. Portable access to the web is an amazing, life-changing revolution ...

I echo that whole post completely. The desktop use to seem like a limiting brick, but now that I have an iPhone, it's the laptop that seems like an awkward compromise. With a Mac Pro back on the desktop doing real work is much more pleasant, and the iPhone has filled that couch-browsing hole far better than I thought it would.
posted by fightorflight at 11:19 AM on July 8

Another company took a linux-like kernel and spend hundreds of millions of dollars to make it "just work."

Some companies have tried but Apple certainly didn't. Apple took NeXTSTEP, which ran the XNU kernel (which is nothing like the Linux kernel in terms of technology, history, or development model), updated NeXTSTEP's 4.3BSD components to code from FreeBSD, and updated the libraries and hardware support. OSX is pretty much NeXTSTEP evolved over the past 15 years. It was clearly a successful move, but the path Apple took to get there is nothing like the path Google is apparently heading down. You simply can't compare the two.
posted by cmonkey at 11:38 AM on July 8

Call when I can run Team Fortress 2 and Photoshop on it.
posted by arcolz at 11:46 AM on July 8

Why the hell would you think it's intended use is for you to play Team Fortress 2 and use Photoshop? Good god.
posted by five fresh fish at 12:04 PM on July 8

OnLive will give you TF2 & Photoshop on an iPhone.
posted by blue_beetle at 12:06 PM on July 8

Call when I can run Team Fortress 2 and Photoshop on it.

probably a few years. You can run Quake in a browser now.
posted by empath at 12:07 PM on July 8 [1 favorite]

You all should check out the HTML 5 demos that are out there:

Photoshop?

Obviously, we're not there yet, but getting there.

More
posted by empath at 12:12 PM on July 8 [1 favorite]

Mick: The thing many people are overlooking is that the netbook will be a person's only computing platform, it won't. People will carry their Google powered netbook with them to do quick things, convenient things, time-killing leisure things.

Except that this runs directly counter to the trends in computing. The Netbook of today has more memory, disk, and computing power than the servers that formatted my first LaTeX document back in the early 90s. The iPhone is more powerful (and certainly has more disk space) than my first home computer. In another few years, the arguments for software-gimping either to get around limited battery and cpu will be moot.

Mick: There will be no need to sync, as the desktop will be connected to the same cloud as the netbook.

I wish people would stop using the term "cloud" to describe what is, fundamentally, a client-server model that's nearing it's fourth decade. It's not a cloud when I use two different machines to connect to the nearest google cluster. It's a cloud when my phone, laptop, work computer and home computer automatically connect to each other peer-to-peer negotiating version control issues to share files.

empath: probably a few years. You can run Quake in a browser now.

Sure, and Photoshop is also built using AIR. Firefox and Thunderbird are built in ECMA-script on top of XULRunner with a few added bells and whistles.

The problem is that the more business logic that you end up transporting from sever to browser, the more your application starts to feel like a java applet, and we all know how popular they were.

Quake is an interesting case, but I find it unlikely that game developers are going to be content with the limitations on the Flash plug-in's 3d support. If you listen to the developer commentary for Portal, they reveal that it took a large quantity of developer time to get the physics and display models for the portals working on their target platforms. I have doubts that such CPU and GPU-intensive algorithms can be implemented in a flavor of ECMAscript.

Your "photoshop" examples are too trivial to bear much more than a passing mention. I'll only point out that the ability to trivially decorate images and doodle on the screen was a showcase of early java applications. Heck, I wrote one in my first computer programming course. The hard part is whether you can make adjustments to large-format raw image files in CYMK and send them to a large-format printer.

So there you get into a mess of issues that I'm not certain that either Firefox or Chrome are able to handle. How do you deal with operations that exceed the physical memory of your computer? How do you translate a file 10,800 pix wide for your printer? How do you optimize processes for 1-8 cores? Do you do the processing on the client or the server? How do you make the editing process usable with data loads that take minutes over 100Mbit/s Ethernet? These are the sort of questions that Photoshop is designed to handle because they occur in real-life production.
posted by KirkJobSluder at 1:23 PM on July 8 [1 favorite]

So there you get into a mess of issues that I'm not certain that either Firefox or Chrome are able to handle.

Yeah, and they're not designed to handle those issues. The exciting thing about netbooks, the iPhone OS, etc is that they're designed for a different class of computing. If you want to run Photoshop then go get a "real computer". If you want to read your email and update MySpace, use a netbook.
posted by Nelson at 1:31 PM on July 8

I'm reading five fresh fish's explanation of things he'll never do without, which seem contradictory at first... and then I understand. I can't help but picture him sitting on the toilet surrounded by three large screens.

I had a six-monitor Macintosh II back in the 80's. That rocked.

No toilet, though.
posted by rokusan at 1:48 PM on July 8

Nelson: Except that those arguments are not that compelling now, much less in the time frame that GoogleOS is expected to come into maturity. Netbooks are "real computers."

Dedicated client-server systems have been around for decades. I don't see what's different here.
posted by KirkJobSluder at 1:54 PM on July 8

probably a few years. You can run Quake in a browser now.
posted by empath at 3:07 PM on July 8

That's nice, because when it came out I ran Quake on a 486-100 with no hardware accelerator. So when you give the browser 20 times the CPU speed, and a hardware GPU that can render Doom 3 at 60fps, it is finally as good as a 486. That's efficient. In a few years, with an 8-core i7 and quad-Nvidia Quadros cooled with a Bose-Einstein condensate, maybe it will be able to run Photoshop 5.

The browser will never do what Photoshop can do. GIMP can't do everything that photoshop two version back can do. Why they hell do you think Adobe can get away with charging $700 for the damn thing?

For the kinds of things a netbook is useful for, an iPhone is better. Browsing, email, media consumption etc. For the kinds of things an iPhone is not good for, like content creation, processor-intensive tasks, a netbook equally sucks.

And Grandma isn't going to buy a netbook with a 10.1" screen. Grandma buys large-print books, remember? They don't sell phones and remotes with oversized buttons because your eyes' ability to resolve detail improves with age. Here's a list of new laptops, not netbooks, but laptops, with 14" or larger screens that grandma can buy, starting at $320, that will run Windows XP.

Everyone thinks Google and Apple are going to carve up the market? I don't know what world everyone lives in, but Google's products other than search and Adsense stink. And all their revenue comes from Adsense. Everything else is a money pit. Gmail after 5 years is only arguably better at some things (and worse in others) than Yahoo Mail is, and yahoo mail hasn't substantively improved 6 years.

All Apple has to do is release a sliding keyboard for the iPhone like some other cell phones have, and no one will every buy a netbook.
posted by Pastabagel at 1:55 PM on July 8

Oops, here's that list of cheap, decently powered laptops. And these will run Photoshop.
posted by Pastabagel at 1:59 PM on July 8

Dedicated client-server systems have been around for decades. I don't see what's different here.

The difference is the Internet is for reals now.
posted by Nelson at 2:01 PM on July 8

Nelson: The difference is the Internet is for reals now.

The internet was for real when Plato shipped. We had it when we used Lotus's communication system on VMS. We had it with X11. We had it with Java applets.

Or let me put it another way, how long will Google OS last on that person's netbook or desktop when Mr. or Ms. Consumer has that particular itch that can't be met via online services?
posted by KirkJobSluder at 2:08 PM on July 8

Calling it "Google's flavour of Linux" is a bit misleading; it'll basically be a Linux kernel, a framebuffer driver, a window manager of some sort and the Chrome browser engine. Architecturally, it'll have more in common with Palm WebOS than Ubuntu.

Running a Linux kernel makes it Google's flavor of Linux.

What you are thinking of is GNU/Linux, the bog standard set of GNU userland applications running on top of a Linux kernel. This is what 99% of Linux distributions are, and you are correct in thinking Chrome OS probably won't be it.
posted by spitefulcrow at 2:09 PM on July 8

ok ok ok. everybody chill out for a second.

I think there's this whole tendency to immediately think of all the things you can't do with something like this, and then go off from there. it's a valid concern up to a point, but only up to a point.

I think it's important to consider that google isn't thinking like that, and that the market is gradually heading away from that kind of thinking, as well. once upon a time, Microsoft was a pretty smart company, and they did a pretty smart thing. It's actually something the tech savvy still criticize them for today, as it started a horrible trend that we still see repercussions from. they built a dumb-ass version of a more capable OS, and they leveraged some considerable marketing and development muscle to support it. Why did they do this? when everybody who understood computers back then would have thought that the techies would have brought the industry to Unix and the rest of the world with them, Microsoft instead brought it to the masses. DOS was basically garbage, and the networking was virtually non existent. It failed to predict so many things (using more than 64k of memory, the evolution of ARPAnet, etc...) and caused the user so much suffering for that lack of foresight down the line, but it made more sense to the layman than Unix did and, comparitively speaking, was easier to use if you weren't a power user. Further, they built "killer" apps that would only run on their OS, and sold them to businesses like crazy as package deals.

Years later, Google would do the same thing with search. Consider hotbot, alta vista, yahoo, etc... they were clunky cluttered interfaces with blinky garishly colored shit all over them. they had options up the ass littered all over their pages and tried to become, on their main sites, these impossible portals to everywhere. then google came along with what is now considered a design revolution and dumbed it down for everyone else. it was so simple! it wasn't necessarily easier to get power use out of it, although they would eventually improve their "advanced search" link to become another powerhouse piece of tech, but fuck everyone could look at google and figure that shit out.

here's google on 4/22/1999
here's hotbot on 04/18/1999

add to that the considerable tech muscle behind the engine, so that it was actually a more useful way to search at the time, and you've got the now legendary start that made them bajillionaires.

but it's not because you and I liked it so much. maybe you and I first hipped some of the cubicle workers to it. but in the end they took each other's advice to use it because one of them would say something like "nah, you just type it in and you got what you're looking for. it's easy, trust me." and off they'd go. it was dumbed down simplicity, and that's pretty much what works.

that doesn't mean there isn't additional material back there for power users to get their hands on. but it's not up front.

so now we've got this dumbed down netbook OS. it's not meant for photoshop users. it's meant for word documents, simple spreadsheets, and the web. you do email through the web, so it's got that, too. it's crazy simple, it's free and it works on the least expensive machines in the world (hopefully) without a hitch. moreover, google can now leverage considerable marketing and development muscle to support it, just like MS did with DOS way back when. hell, they've already built the web apps. if they can manage to sell it to businesses like crazy in package deals, they'll have pulled off something pretty amazing. and maybe they can. if there's anything that gets entrenched businesses to accept change, it's the idea of lowering overhead without having to do too much retraining of their employees.

then afterwards they can make the machines that you or I would want to use. hell, I've got a fucking stacked mac pro desktop at work, with 3 monitors attached, mixing board, studio speakers, fiber channel SAN, 3 digital video decks and more. I'm not replacing it any time soon. but I don't have to. I'm nobody in the grand scheme of things. i'm a niche market. but if they can get some cubicle guys using these things every day in the office, they'll want to take it home with them, too, since they're familiar with it from work and they like what they're familiar with.

games are going digital download, or (as has been mentioned) possibly distributed remote delivery, anyway. if onlive is real, and you can run it in a browser, then a google netbook would be your game machine, too. but that's not what it's for right now. right now it's a simple, stupid, cheap business machine that people wouldn't mind taking home with them, and that's a pretty powerful market if they pull this off.

of course, I have no idea if they will. it could totally fail.
posted by shmegegge at 2:16 PM on July 8 [3 favorites]

probably a few years. You can run Quake in a browser now.
posted by empath at 3:07 PM on July 8

That's nice, because when it came out I ran Quake on a 486-100 with no hardware accelerator.

the quake in a browser is actually quake 3 arena, fyi. significant difference in performance and required hardware.
posted by shmegegge at 2:18 PM on July 8

You can run Quake in a browser now.

You can launch Quake 3 from a browser now. It still runs as compiled code on the user's system, and not as Flash.
posted by dumbland at 2:21 PM on July 8 [1 favorite]

"If I had an iPhone-sized device for when I'm out and about; and a high-resolution device about the size of a comic book for when I need a slightly bigger, casual working space; and a 4000x2000 22' display for getting real work done, I'd pretty much have it made in spades."

One of the remaining unexploited hardware mother loads is virtual displays that we can mount in front of our eyes. The first guy to figure out how to get a perceived 36" 4000X2000 pixel display into a postage stamp sized package will never want for anything again. Not even postage stamped size; I'd wear something like Picard's Borg headset if it meant I could interact with AutoCAD while walking around a site and keeping both hands mostly free.
posted by Mitheral at 2:45 PM on July 8

I have a 1st gen EeePC. It was sort of novelty at first, but when I formatted that shit Linux OS away and installed EeeXP, I had a tremillion times faster and actually useable computer. Really, the Eeenux or whatever the OS is called sucks. Stripped down-XP doesn't. And if I want to run Gmail, hey, still I can! Wow!

Anecdotal point to go with me saying that Google fanboys is nearly as irritating as Linuxxers. Nowhere near Applebois though. (Have yet to meet an MS-fanatic that doesn't work at MS).
posted by mr.marx at 2:46 PM on July 8

schmegegge: I think there's this whole tendency to immediately think of all the things you can't do with something like this, and then go off from there. it's a valid concern up to a point, but only up to a point.

The problem is that such a system stops being popular the instant you find that "must have" feature. For some it's MSWord. As much as I hate MSWord, neither Google Apps nor OpenOffice offer full file compatibility. For some it's going to be that picky feature that doesn't exist in the Outlook Web client. For some, it's going to be plugging in the new widget and be frustrated it doesn't work. For some, it's going to be letting the grandkid play games. For some, it's playing games.

These are the things that have been the hammers that have hit adoption of both MacOS and Linux/BSD. I loved running XFCE, until the point that OpenOffice barfed on a client's document costing me a dozen hours of work.

so now we've got this dumbed down netbook OS. it's not meant for photoshop users. it's meant for word documents, simple spreadsheets, and the web. you do email through the web, so it's got that, too.

Sure, the problem is that very few computer users I know stay within that limited sandbox for long. They discover that there is this great bit of software that supports their current hobby, and they buy it. Everyone I know who picked up a computer just for email and simple word documents ended up using it for genealogy, games for the kids, music composition, a bit of video, or desktop publishing for their clubs.

games are going digital download, or (as has been mentioned) possibly distributed remote delivery, anyway. if onlive is real, and you can run it in a browser, then a google netbook would be your game machine, too.

It won't without DirectX. I'm as much of a fan of digital distribution as the next person (even though a typical title will take 2-3 hours of download time), but will that be possible as a locked-down browser-only environment?

but that's not what it's for right now. right now it's a simple, stupid, cheap business machine that people wouldn't mind taking home with them, and that's a pretty powerful market if they pull this off.

Unless your business is blogging, I don't see how it can be called a "business machine." The alternatives to the software I use for business just are not of the required level of features and file compatibility to be viable.
posted by KirkJobSluder at 2:50 PM on July 8 [2 favorites]

The problem is that such a system stops being popular the instant you find that "must have" feature.

certainly a good point. what if that must have feature is in a google app, though? it seems to me like their whole strategy is to piece by piece take over the individual markets that enhance the allure of an OS. maybe that must have feature will be something google created that other apps don't have? i don't know. like i said, they could also just completely fail.

Sure, the problem is that very few computer users I know stay within that limited sandbox for long. They discover that there is this great bit of software that supports their current hobby, and they buy it.

which says to me that any OS has to stand on the shoulders of its developers. maybe google has something for that. do we know what the developer landscape looks like for the chrome OS? I'd be interested in hearing about it.

It won't without DirectX. I'm as much of a fan of digital distribution as the next person (even though a typical title will take 2-3 hours of download time), but will that be possible as a locked-down browser-only environment?

well, if we're talking about onLive, we're talking about processing that happens entirely remotely. the only thing the netbook or standalone box needs to do would be video processing and downloading the video stream and sending input commands back to the hosting site. but if, again, we're talking about onLive, we're talking about something that could be total vapor. who knows. I like imagining the future is going to go all streaming, all digital content delivery. it looks to me like we're heading that way, but I'd be more than a little presumptuous to assume it'll be smooth sailing all the way.

Unless your business is blogging, I don't see how it can be called a "business machine."

you know, I'm way way way over in the creative end of business. I couldn't really tell you how much your standard office worker does with his machine. so before I step any farther into "talking out of my ass" territory (note: I said farther.) let me just say that I'm really just imagining a very generic business user archetype, that probably stays largely in the arena of office apps and email where their business is concerned. maybe I'm being overly restrictive when I talk about this type of user, but I tend to see them as using things like Excel and Word almost exclusively. and exchange email.
posted by shmegegge at 3:16 PM on July 8

there are no must-have google apps
posted by mr.marx at 3:30 PM on July 8 [1 favorite]

"The problem is that such a system stops being popular the instant you find that 'must have' feature. For some it's MSWord. As much as I hate MSWord, neither Google Apps nor OpenOffice offer full file compatibility. For some it's going to be that picky feature that doesn't exist in the Outlook Web client. For some, it's going to be plugging in the new widget and be frustrated it doesn't work. For some, it's going to be letting the grandkid play games. For some, it's playing games."

To be fair it's not like Vista is all things to all people either. And all those people running XP on notebooks are going to be limited to hyper critical support only once Windows 7 ships IE: essentially none. Many, many people are going to be fully or mostly serviced by this system and it'll come at a much lower cost.
posted by Mitheral at 3:44 PM on July 8

The browser will never do what Photoshop can do

Most people, even some advanced amateur photographers (myself included), will never need all that power, or all of those features.
posted by raysmj at 3:51 PM on July 8

shmegegge: The unfortunate state of things is that with MSWord as a defacto standard, you pretty much need MSWord if you work with people who use MSWord. Both Google Docs and Open Office have the potential to mangle content in destructive ways.

In regards to games, while one segment of the market has embraced web-based delivery, another segment of the market is going to take advantage of terabyte hard drives, quad-core CPUs, and the latest bells and whistles from graphics card makers to deliver increasingly rich environments. Blizzard already prefers to ship its content and updates via bittorent, and we have a number of networked game delivery options including steam. I see few incentives for a developer to continuously render and stream a video feed that could be produced at a higher level of detail on the client's computer.

Mitheral: Sure. I didn't mean to imply that Vista was the perfect operating system. I just don't see the market for such a restrictive OS paradigm off of a phone.
posted by KirkJobSluder at 3:52 PM on July 8

Most people, even some advanced amateur photographers (myself included), will never need all that power, or all of those features.

But most people will want some of those features, and Google is going to leave them hanging while Apple and Microsoft will be there waiting to sell them a real computer to give them back the flexibility they had before for not much more money.
posted by Space Coyote at 3:53 PM on July 8

boo_radley: Flickr Manager, latest version.
posted by raysmj at 3:54 PM on July 8

The unfortunate state of things is that with MSWord as a defacto standard, you pretty much need MSWord if you work with people who use MSWord. Both Google Docs and Open Office have the potential to mangle content in destructive ways.

no doubt. i got nothing.

In regards to games, while one segment of the market has embraced web-based delivery, another segment of the market is going to take advantage of terabyte hard drives, quad-core CPUs, and the latest bells and whistles from graphics card makers to deliver increasingly rich environments.

well, I'm not sure how familiar you are with the state of pc gaming right now, or onLive specifically, but PC gaming is experiencing a pretty heavy slump specifically because of how expensive those bells and whistles are from a hardware standpoint, and console sales have completely eclipsed pc gaming sales specifically because the dedicated machines cost so much less than a gaming pc. onLive specifically is a service that is designed to connect you to huge ultimate gaming rigs remotely that you don't have to own. in other words, your hardware at home just processes the video feed that you're getting from a server center that runs all that hardware for you. you pay a monthly fee (with all the good and bad things that entails in the user's mind) and you never have to worry about pc hardware again from a gaming standpoint. that's the idea at least. it has been demoed running crysis, remotely, on a shitty laptop, as an example.

but this is getting beside my point:

which is that what we're talking about could be just a foothold on a LARGE share of the market, by ignoring the niche users for now. yes, gamers are unfortunately a niche and dwindling part of the pc market. and yes, they will one day have to be addressed if google ever intends to get more than a paltry cultish following for their OS. but I think it's reasonable to assume that google has a plan to address that down the line.

of course, that's an assumption and they could totally fall flat on their face with this.
posted by shmegegge at 4:10 PM on July 8

I think the trend of web services for what could be done better on your dirt-cheap desktop is bizarre, and I'm loath to let Google know even more about me and my data than they already do.

But I'm a crank, and what I think really doesn't mean anything in terms of mass acceptance.

A fast-booting easily portable web-capable device? Make it cheap enough, and I think it could find its niche, no matter how many other things it can't do.
posted by Zed at 4:16 PM on July 8

The thing many people are overlooking is that the netbook will be a person's only computing platform, it won't. People will carry their Google powered netbook with them to do quick things, convenient things, time-killing leisure things.

You're already describing what I do with my iPhone. Quick things, convenient things, time killing leisure things, quick emails, keeping in touch in every way possible. A netbook is better equipped for some things, but it can't compare to something like the iPhone for MANY things. And it doesn't fit in my pocket.

All Apple has to do is release a sliding keyboard for the iPhone like some other cell phones have, and no one will every buy a netbook.
posted by Pastabagel

Ain't going to happen, and it ain't gonna affect apple.

I saw a teensomething kid the other day typing on his iPhone faster than I can type on my macbook. I know, keep him off your lawn.
posted by Dennis Murphy at 4:19 PM on July 8

empath says:
The app store in the iphone is so good that i'm really amazed that nobody has done something like that for PCs (except for Steam). If MS tried it, they'd be sued into oblivion, though.

I have not used the apps store, but isn't this the standard way to install apps on most distributions of Linux, except the software is free? Or do you specifically mean as a software sales platform?
posted by idiopath at 4:58 PM on July 8

Sales.
posted by empath at 5:14 PM on July 8

Is it really? The articles I've pulled up indicate that the industry is slumping across the board, quite possibly due to reduced technology sales during the recession. I personally doubt that hardware costs are a factor. Most games, including Crysis, ship for a wide range of systems, and 3D graphics integrated graphics chips are shipping standard on laptops these days. Throwing money at a gaming system just buys bragging rights and possibly a prettier experience.

Heck, Sims 3 shipped for the iPhone.

But that's just details. The whole concept of hosting a render farm and streaming dozens of hours of video per user and title strikes me as one destined for obsolescence. The gap between low-end and high-end video chipsets is just going to get narrower. In a few years, practically everything will be shipping with some form of hardware acceleration, and there doesn't seem to be an equivalent quantum jump on the horizon.
posted by KirkJobSluder at 5:18 PM on July 8

Google's Vanity OS is Microsoft's dream from the Reg.

Google and MS say: "You have anti-trust problems, we have anti-trust problems - how about we launch online advertising sales and you launch an OS." Sweet!

As far as PC vs console game software sales go look at this recentish Boston globe article with numbers for global game sales for PCs and consoles. Another way to look at the figures is that the PC is still the single largest game platform. The consoles would split sales 5+ ways ( Wii, PS3, Xbox, DS, PS2 ).

The Windows PC has been dying continuously for at least 15 or so years. Who remembers thin client Java machines, Java replacing the OS, Navigator replacing the OS, the year of the Linux desktop (1999-2004)? There must be more.
posted by sien at 6:04 PM on July 8

Google's products other than search and Adsense stink

What? What's wrong with Google Reader, News, Analytics, Calendar? Everyone I know uses Google Maps for driving directions. Google Earth and SketchUp are both fun to play with... I wouldn't say they're the most useful of programs (though you can build Left4Dead maps in SketchUp), but by what standard to they "stink?" Especially since each and every one of those is free (though Earth and SketchUp have paid pro versions). I have no idea whether or not they're "money pits," but I really don't get how you can say they "stink."

Gmail after 5 years is only arguably better at some things (and worse in others) than Yahoo Mail is, and yahoo mail hasn't substantively improved 6 years.

Well now this is clearly a joke. Have you even used GMail? Now I have to list a few of the ways in which GMail destroys Yahoo! Mail. Not "arguably better at some things," head and shoulders better.

1) Conversation threads: This is just brilliant. If I send out an email to 12 people telling them I just spent 15 mins putting together a response to someone insulting GMail, and each of them replies with some variation of "Loser!", why would I want those responses to take up a quarter of my inbox when they can all be neatly stacked into one conversation. It's great!

2) username+awesome@gmail.com: Let's say you really want to join some random blue site, but you don't trust people who ride recumbent bikes. When you sign up for the site, you can give your email address as username+blue@gmail.com and then filter all email with the "+blue" directly to the trash.

3) Labels: Or, instead of trashing it, you could just label it... with as many labels as you want. I can put a green label on all email coming from the people I work with, making it easy to quickly spot them in my inbox. But labels are better than folders because every year, I run a March Madness Pool, and I want to be sure that stuff gets the highest priority. So I just add another filter that puts a bright red label on anything with NCAA in the body so there's no conflicts with where a March Madness-related email from a coworker should go.

4) Free POP/IMAP: Yahoo! charges $20 a year to allow for POP access to their shit email. That's free in GMail. Plus, no graphical ads in GMail.

5) Integrated text, voice, or video chat: No install required. Chatting with someone and suddenly need to go hands-free? Just click the little phone next to their window, and you can start talking.

So there are 5 things that are clearly better about GMail. I'm honestly curious to hear what you think Yahoo! Mail does better.
posted by SAC at 6:15 PM on July 8 [1 favorite]

maybe I'm being overly restrictive when I talk about this type of user, but I tend to see them as using things like Excel and Word almost exclusively. and exchange email.

I may be somewhat ignorant to the way the lowly cubicle monkeys use computer as well, as a scientist...but one thing I'm pretty sure of is that netbooks are pretty irrelevant in this case anyway. Can you see office PCs being replaced by netbooks? I can see OH&S implications in making someone sit in front of a screen and keyboard that size for 8 hours a day, for a start.

I think there's a big irony here. Almost by definition, the kind of people who want some sort of ultra-portable, tiny device to use to access the web and email from anywhere are probably going to be power-users, who (unless they occupy that very, very rarefied position that's probably overstated on places like Metafilter and tech sites: the "professional blogger") will be unsatisfied with only having web and email.

I've thought of another reason my old dad wouldn't want one of these things; he takes his decent-sized laptop on the road with him on caravan/camping trips, so he can plug in his TV tuner dongle and watch TV - far, far from any kind of internet access. Computers are amazing, multipurpose devices these days, and I can see little value in artificially restricting their functionality.
posted by Jimbob at 6:24 PM on July 8

As a self-proclaimed internet super freak
I CAN get by with what GoogleOS is offering
If I need an app - I can grab it from a cloud.

I have had my eye on googleOS
Since Google bought the domain in April of 2004
Many of us have been looking fwd to this
That it will be open source is icing.
posted by will wait 4 tanjents at 7:06 PM on July 8

Most netbooks have video out and probably all of them have USB ports. You can use them with external keyboards and monitors. But, yeah, they're not likely to make inroads as desktop replacements for offices anytime soon.
posted by Zed at 7:23 PM on July 8

Fake Steve sums up the announcement. Worth a read. Especially point 5.
posted by sien at 9:48 PM on July 8 [1 favorite]

A whole lot of you people sound like three years ago you'd have pooh-poohed the idea of anyone wanting and using a small, 4 1/2 x 2 1/2 inch device for web browsing, gaming, and general applications.

Of course, now that you have an iPhone, you can't imagine the idea of not finding it useful!

And now you pooh-pooh the idea of anyone wanting and using a net-based OS. [facepalm]
posted by five fresh fish at 9:54 PM on July 8 [4 favorites]

I do translation, and a big part of that involves using silly proprietary software that only runs on windows, but the other side of that coin is that since I'm working with mostly text, most of the software I use is just some glorified version of notepad with a dictionary attached, so it's very low-power. The operations of translation are pretty simple, mostly involving word-processing plus looking up the occasional word. And I often need to be able to go to my clients, so I need a portable machine. The only place netbooks fail to be my ideal machine is for long stretches of typing, and I have a USB keyboard on hand for that. The cloud idea is also great for translation, because it makes collaborative databases that much easier to maintain and access.

There are definitely places where having a larger machine makes sense, but in certain industries, switching to something like cloud computing and google's OS could really help eliminate a lot of headaches. I have my phone, my netbook, my desktop for dumb CPU tasks, I have proprietary software and licenses spread out across the 3, and it would save me literally half of the time I spend doing my job if I could move all that idiocy to a lightweight, online system. These new low-power laptops and the associated OS's hold a lot of promise for my field, and I'm excited to see what comes next.
posted by saysthis at 11:02 PM on July 8

saythis: there is also the simple fact that if the people who write your translation software find out that a significant percentage of their users have a machine that runs google-os, they will have a huge incentive to port their software to google-os. Lack of apps is completely chicken/egg and if that is the only reason a platform would fail, in the long run it can easily be a non-issue.

Of course there is still the issue of first system syndrome (which is the reason I can't stand anything but Linux, through an odd series of events I learned Linux before I really learned any other system, and to this day nothing but Linux really makes sense to me).
posted by idiopath at 11:13 PM on July 8

Of course, now that you have an iPhone, you can't imagine the idea of not finding it useful!

And now you pooh-pooh the idea of anyone wanting and using a net-based OS. [facepalm]

An iPhone is something that didn't exist before. What a netbook with a "web OS" would be like would be you trying to sell me an old pre-app store iPhone to replace my current app-filled iPhone and telling me it's all I need.
posted by Space Coyote at 1:02 AM on July 9

That's a great point, Space Coyote. Considering the wailing and gnashing that followed when Steve tried to sell the idea that "web apps are all you need" and the explosion that followed an SDK for the iPhone, it's a bit early yet to write off native apps, especially those with a monster framework the size of Cocoa behind them.
posted by fightorflight at 5:23 AM on July 9

A whole lot of you people sound like three years ago you'd have pooh-poohed the idea of anyone wanting and using a small, 4 1/2 x 2 1/2 inch device for web browsing, gaming, and general applications.

You can use this defense for anything. "You mocked how log it took for Half Life 2 to come out and it was great, so who are you to judge any other delayed product?"
posted by smackfu at 5:47 AM on July 9

five fresh fish: Except for the obvious fact that locked-down operating systems that boot directly into an internet client have been around since the 1970s. You can do it now if you like with Linux or MSWin. The fact that those operating systems have existed for years but have never caught on outside of email kiosks and internet cafes should say something. Heck, isn't a java-based web-os Palm's last gasp attempt to avoid getting eaten by RIM, Google, and Apple?

The bottom line is that the proposal for Chrome OS is neither creative or innovative, and I see no reason why we should ignore the fact that these operating systems have 35 years of marginal market presence behind them.

Sony made portable music players ubiquitous in the '80s. Developing a portable player around new miniature hard drives and compressed music files was a no-brainer. Cellphones and smartphones were also experiencing explosive growth when Apple entered the market.

idiopath: The problem isn't just a lack of apps. The problem is a lack of development/deployment models that cuts out a large chunk of the software ecosystem.

SAC: But labels are better than folders...

Please, labels are folders and folders are labels, and it's been that way for a few decades now.
posted by KirkJobSluder at 6:28 AM on July 9

Please, labels are folders and folders are labels, and it's been that way for a few decades now.

Except for the actual example I gave where Yahoo! Mail won't let me put one email into two different folders... Please.
posted by SAC at 7:13 AM on July 9

SAC: Then you should have said that Yahoo's specific implementation of categorization metadata is lacking.
posted by KirkJobSluder at 7:30 AM on July 9

Seeing as how it's clear my comment was pointing out ways in which Google implements E-mail better than Yahoo!, I think I said it fine.
posted by SAC at 7:40 AM on July 9 [1 favorite]

Ted Dzuiba on this whole thing. He is awesome. And he makes fun of TechCrunch, so it's doubly good.
posted by chunking express at 8:06 AM on July 9 [1 favorite]

I'm amazed this conversation is going on so long, folks spending time and energy saying "that will never work". Get excited and make things!

An iPhone is something that didn't exist before.

That's not really true. There were a lot of smart phones with real apps and operating systems before the iPhone. Palm/Handspring devices, Blackberries, Windows Mobile, even Symbian all had serious offerings: some pretty good. Also there's Newton, which while a dead end and never a phone had a lot of UI ideas that came back in the iPhone.

What's different about the iPhone is that Apple executed very, very well. And so it became a game-changing device. For me the key enabler was having a really good web browser on a big enough screen with a good enough wireless network connection to make it useful. For other folks it's the integrated iTunes, or the clean multitouch UI, or the cell phone quality. (Ha, I kid about the cell quality.)

I'm hopeful that Google will create a similarly game-changing netbook OS. I think the field is fertile, that there's enough applications hosted on the web now that a web-only device (augmented with something like Gears) will make sense. And I think Google is innovative enough to make a significantly higher quality product than before, much like how the iPhone is significantly better than Windows Mobile. But heck, I don't know any more than anyone here whether they'll succeed. Can't wait to find out!
posted by Nelson at 8:45 AM on July 9

Nelson: I'm amazed this conversation is going on so long, folks spending time and energy saying "that will never work". Get excited and make things!

It depends on what you mean by "that will never work." Technically, there is not a single fucking thing in Google's proposal that hasn't already been developed and implemented. And in fact, I'm tempted this evening to fire up Virtualbox and make a proof-of-concept Firefox OS.
posted by KirkJobSluder at 8:58 AM on July 9 [1 favorite]

Is it really? The articles I've pulled up indicate that the industry is slumping across the board, quite possibly due to reduced technology sales during the recession.

here's something I found really quickly that shows a trend from 1998-2006:

link

here's a quick google search for pc game sales in 2008.

what's been happening for a while, now, as consoles take advantage of the very trend you pointed out (narrowing graphics hardware gap.), is that consoles are taking over the market, which is commonly attributed to low cost and low cost of ownership.

Most games, including Crysis, ship for a wide range of systems, and 3D graphics integrated graphics chips are shipping standard on laptops these days. Throwing money at a gaming system just buys bragging rights and possibly a prettier experience.

how wide a range do you think that is? you couldn't play crysis on a system less expensive than 1000-1500 dollars (not including monitor, keyboard, mouse, speakers) when it came out. they had to build a system for 500 dollars (which did not include monitor, keyboard, mouse or speakers.) themselves to prove it could be done, and then showed a video of the player fighting only one monster to demonstrate it. meanwhile, costs for the hardware necessary have gone down, so you can get a crysis capable box for significantly less these days, but crysis was also released almost two years ago. further, the laptops you're talking about that can game well still cost more than an xbox360. I just priced out an hp laptop that could theoretically play crysis or far cry 2 at a mewling 30 fps, and it was 1300 dollars. for a machine that, if I ever had to upgrade it for the next generation of games, would have to be entirely replaced at a comparable price point. meanwhile far cry 2 plays on my 360 beautifully.

Heck, Sims 3 shipped for the iPhone.

the iPhone is not a pc. as a gaming system, it's probably classified as a console.

In a few years, practically everything will be shipping with some form of hardware acceleration, and there doesn't seem to be an equivalent quantum jump on the horizon.

well, that's certainly possible. but it's been my experience that for every jump in affordable acceleration, there's a comparable jump in prohibitively expensive high end hardware and games that make use of it. you can say that only the hardcore go for that, and it's true. but the rest of us don't bother making do with informed affordable hardware upgrades. some of us do, but the vast majority of us just get a console and forget about it.

but this is beside the point. the point I'm trying to make is that I think we've departed from the era of the pc that does everything. I think most people buy something that does "the internet and word." i think most people don't worry about games on a pc, anymore. and those are the people that would like a super cheap ultra dumbed down pc that doesn't give them any stress. sure, it would never take off if it didn't eventually take on the rest of the market, too. but I think that's in the eventual plan. maybe it'll flop horribly. it wouldn't totally surprise me, to be honest. but I don't think the idea of building something that can't game and that doesn't necessarily appeal to power users is that terrible of an idea right now.

it might not work right now, either, but I think the philosophy is generally sound. services like streaming netflix, onLive (if it ever happens), pandora, itunes et al. all make me think we're heading in a very specific direction that this google chrome OS also seems to be heading in. maybe they're doing it wrong, and maybe they're doing it too soon. but the idea is sound.
posted by shmegegge at 9:02 AM on July 9

Put yourself in the shoes of your average computer buyer in a store looking at, on one had, a $300 netbook, and next to it a $400 low-end laptop. And he asks the salesman what he can do on the laptop. "Oh, everything you can do right now, and you can run all the software you're used to, and open all your existing files, since it just runs Windows like you're current computer." "And what about this one?" "Well, it just runs a web browser."

Techies don't mind dropping a few hundred dollars on a netbook because it's a secondary PC for them to have because they absolutely can't ever be without the internet no matter where in the house they are. But the majority of the market are households who only own one computer at a time and don't want to take the risk on something that is effectively worse than what they've already got.
posted by Space Coyote at 9:26 AM on July 9

in fact, I'm tempted this evening to fire up Virtualbox and make a proof-of-concept Firefox OS.

That'd be cool, I'd love to see what you come up with. The closest thing I know of so far is Ubuntu Netbook, but I admit I haven't used it enough to really know. It might be fun to try Chromium instead of Firefox. Or if you use Windows as your base OS, you can just use Chrome itself.

Of course, "it's Linux with only Firefox!" is not a very interesting proof of concept. What I'm really hoping Google will do is make something that runs very fast on a $150 three pound device with 12 hours operational battery life. That has you going from zero to reading your email in three seconds. The only reason to even consider these stripped down OSes is the cost and power savings. But really testing that takes hardware.

Technically, there is not a single fucking thing in Google's proposal that hasn't already been developed and implemented.

To be fair, none of us have any idea what "Google's proposal" is. All we've got is the announcement and a FAQ with no technical detail. (Also, why swear? Are you that angry?)

While I'm extending this thread further, here's a really stupid fake Chrome OS screenshot blog that got an embarassing amount of pickup.
posted by Nelson at 9:47 AM on July 9

shmegegge: Of course, you are starting from two false assumptions. The first is that computer gaming is exclusively focused on high-end graphics like Crysis, while I can point you to any number of casual game publishers who are doing brisk business. The second is pointing to the issues in buying a game-specific computer, while on the contrary, I think most people add games on as functionality to a system that's bought for other purposes.

My feeling is that consumers are likely to get pissed when they discover that their locked-down google OS doesn't play the game, or run the software that they want. Games are only one class of application that would be excluded.

the iPhone is not a pc. as a gaming system, it's probably classified as a console.

Except for, ohh, that tricky fact that it's running a specialized OSX and Cocoa with a development environment (built on gcc) that makes it reasonably easy for the game producers to port the Sims 3 over. There is considerable cross-development for XBox 360 and MSWin for the same reasons.

it might not work right now, either, but I think the philosophy is generally sound. services like streaming netflix, onLive (if it ever happens), pandora, itunes et al. all make me think we're heading in a very specific direction that this google chrome OS also seems to be heading in. maybe they're doing it wrong, and maybe they're doing it too soon. but the idea is sound.

Well, how are you going to get streaming netflix if you can't install Silverlight? What happens if onLive demands a custom plugin component to handle the two-way bandwidth? And of course, you won't have iTunes, which is a Rich Internet Application built on Webkit, Cocoa and QuickTime. You can't install it, and it uses a download-once/access-often content model that allows you to copy those files to portable devices or burn them onto CD for road trips. It's the same model that's used by Steam, the Kindle, and DVR's and is the primary innovation behind Google Gears, and HTML5. I'd be really surprised if pandora doesn't cache its interface as well.

Or even just look at the web-based applications that will be the center of this discussion. The whole point of AJAX is to create more responsive user interfaces by minimizing chatter between the client and the server and keeping that chatter to the background. Gmail is a download-once/access-often application. It just runs in a web browser. One of the problems you get into is the more functionality you pack into that AJAX interface, the more it starts to look like an RIA framework.

The idea is certainly sound for some applications. The question is whether it's sound across all applications. And the answer if 35 years of prior experience is any indication appears to be "no."
posted by KirkJobSluder at 10:59 AM on July 9

Exactly. There are shit loads of applications you can't shoe horn into a web page. And even if you can, why would you want to?
posted by chunking express at 11:09 AM on July 9

Nelson: (Also, why swear? Are you that angry?)

Not angry. Just selectively using it to make a point.

My personal axe to grind is that the technology community has very little sense of history. The end result is that we see old ideas rebranded with new buzzphrases and hyped beyond all sense of perspective. I think that bullshit like "Web 2.0" (online community and CSCL/W), "cloud computing" (client-server models), and "Google OS" (a minimalist kernel that launches a web browser) ultimately hurts the technology community as a whole. First, because I think there are lessons to be learned from how those models were implemented in the 70s, 80s, and 90s. And second because of inflated expectations that these old models will suddenly be the Holy Grail computer systems followed by disappointment when they fail to deliver.
posted by KirkJobSluder at 11:29 AM on July 9 [1 favorite]

The first is that computer gaming is exclusively focused on high-end graphics like Crysis, while I can point you to any number of casual game publishers who are doing brisk business.

I assure you, I'm not making that assumption. In truth, I'm having difficulty keeping a close lock on precisely where we're disagreeing, here. If I look at the beginning of our conversation, what seems to me to be the central point of contention is that, thanks to niche and/or high end applications, terminal computing or browser based OSes can't become a viable platform. am I understanding your point correctly? I'm concerned that I'm not, and that the result is us talking around each other.

if I am, then what I'm getting at is that low-end gaming won't be the problem with something like the ChromeOS, if it's heading in the direction I think it is. sure, it's a problem, now. but for now, that's not the target audience. in the future, there would presumably be measures taken to include that audience, possibly including more robust hardware and software support. (but I'm guessing. if I'm wrong, then sure, this is a stupid idea. i think i'm making a reasonable guess, though.) but for now, I think the non-gaming market is large enough to support an OS and netbook style this limited. that's my ultimate point. that, despite the fact that the games market still does and will continue to exist, this OS has a shot at capturing a sizable market, and that that's what google is trying to do. I hope I'm being clear about this. from there, I'd be very surprised if their business roadmap did not then include some form of expanding userbase adoption, including games and desktop workstation users.

until that time, though, I don't think casual and low-end games are going to be a huge stumbling block. (I'm assuming you don't consider world of warcraft a low end game.) consider popcap games, who's arguably the world's leading casual pc games maker. their games run on every device imaginable, including the web. I see a future where that type of game could happily run off of a terminal-like system. It is, to my mind, the least of google's worries.

Well, how are you going to get streaming netflix if you can't install Silverlight? What happens if onLive demands a custom plugin component to handle the two-way bandwidth? And of course, you won't have iTunes, which is a Rich Internet Application built on Webkit, Cocoa and QuickTime. You can't install it, and it uses a download-once/access-often content model that allows you to copy those files to portable devices or burn them onto CD for road trips. It's the same model that's used by Steam, the Kindle, and DVR's and is the primary innovation behind Google Gears, and HTML5. I'd be really surprised if pandora doesn't cache its interface as well.

honestly, the one discussion I don't want to have is getting into the nitty gritty engineering side, because I don't know anything about it. but, because I also don't want to resort to lame cop outs like "i don't want to talk about engineering," let me at least try to describe what possibilities I imagine, with the caveat that I'm describing something imaginary:

first things first: i think we're assuming that google is "just putting chrome on top of [foo.]" and we don't honestly know what the hell they're doing. further, I think we're imagining a dumb terminal, which seems a little restrictive. something like onLive, which I brought up, certainly acts like a dumb terminal (if it works) but it's not the only way. the idea, if I understand the blog correctly, is for googleOS to operate on a netbook, which would certainly allow for partner applications and extensions to be installed and operated, including silverlight (if MS were a partner) or flash (since adobe IS a partner). down the line, chrome could simply operate on a full on laptop, or even a desktop, though at this point I'm obviously speculating. but what's most interesting to me is the possibility that if chromeOS takes off, it will inspire different architectures from current and potential partners. architectures that would take advantage of terminal-like architecture or cloud computing. for cloud computing to actually become viable not just in theory but in practice because of partner support and rigorous support for developers from google. all the while, we could happily run whatever off of hard drives until, MAYBE maybe maybe, they become unnecessary.

but I don't know. does it really sound that undoable? I know it's not possible RIGHT NOW, but you don't think this looks like a step in an interesting direction?
posted by shmegegge at 11:50 AM on July 9

First, because I think there are lessons to be learned from how those models were implemented in the 70s, 80s, and 90s.

I think you're making good points, but I think that you're using a selective idea of "history," here. we could have easily learned a lesson from the old sidecar-expandable computers that extensible hardware is a dead-end. but we took the good parts of the idea, discarded the bad engineering and bad design, and we've got pci cards and usb drives and all sorts of varied and interesting hardware out there that has enhanced our computing. on the other side of the spectrum, history is filled with monoblock computers that taught us not to go THAT route either, but here we are and the imac is a fantastic little machine that many people use and appreciate. it's not really a great idea, to my mind, to reduce all new ideas to "just a variation on [failed earlier thing x]" and then dismiss it. I think the industry in general has a better idea of history than that, and despite its failures along the way tends to separate the wheat from the chaff and use old ideas in new successful ways.
posted by shmegegge at 11:54 AM on July 9

looking at, on one had, a $300 netbook, and next to it a $400 low-end laptop

It's easy to make up straw machines here. What if the netbook is $100, weighs under a pound, and switches on instantly?

Thing is, we don't know anything. We don't know how innovative the software will be, because other than that it'll use a Linux kernel, some form of the Chrome browser, and "a new windowing system" (that a lot of people are assuming will mean not-X), we know zilch. We don't know what the hardware will be, because it doesn't exist yet.

In the early nineties, I heard dinosaurs with an inflated sense of their knowledge of history dismissing the significance of the web, because client-server architectures had been around a long time. Flip sold a zillion cameras as people dismissed it for doing less than almost-as-cheap cameras didn't already do. And high-capacity digital audio players? Who needs to carry all their music all the time?

Things can succeed by doing nothing new, if they do nothing new in the right way at the right time. Fortunes rise and fall on trying to time that. Will the Google OS in combo with one or more hardware makers get it right? Beats me. All I'm saying is that I think it's too early to rule it out, because it's

(Oh, and here's some speculation about the timing.)
posted by Zed at 11:59 AM on July 9

It's a good thing innovation doesn't depend on the like of you folk.

"Oh, Ogg, no one is ever going to need a pair of round discs with an axle between them. You silly fool!"
posted by five fresh fish at 12:29 PM on July 9

shmegegge: I assure you, I'm not making that assumption. In truth, I'm having difficulty keeping a close lock on precisely where we're disagreeing, here. If I look at the beginning of our conversation, what seems to me to be the central point of contention is that, thanks to niche and/or high end applications, terminal computing or browser based OSes can't become a viable platform. am I understanding your point correctly? I'm concerned that I'm not, and that the result is us talking around each other.

Except that I don't see those applications as particularly "niche" or "high-end." MSWord is (unfortunately) a requirement for millions of users. Neither Google Docs nor Zoho are robust enough for academic use. So there you have two huge segments of the market that would not be well served by systems exclusively dependent on web services.

but for now, I think the non-gaming market is large enough to support an OS and netbook style this limited. that's my ultimate point. that, despite the fact that the games market still does and will continue to exist, this OS has a shot at capturing a sizable market, and that that's what google is trying to do.

There certainly is a market for locked-down web-only systems. However it's neither sizable, nor does that market are consumers. I have yet to meet a single late-adopter whose computing needs did not expand in ways that were difficult to predict in advance. The market for locked-down systems is kiosks and internet cafes.

first things first: i think we're assuming that google is "just putting chrome on top of [foo.]" and we don't honestly know what the hell they're doing.

Exactly what they've announced is putting Chrome on top of some, as yet unspecified, minimalist hardware OS. Chrome is not a dumb terminal. onLive can't be a dumb terminal if its going to deliver high-quality video streaming responsive to user input.

for cloud computing to actually become viable not just in theory but in practice because of partner support and rigorous support for developers from google. all the while, we could happily run whatever off of hard drives until, MAYBE maybe maybe, they become unnecessary.

I'm still scratching my head as to how the 35-year-old client/server model embodied by google docs or netflix streaming really justifies the metaphor of "cloud computing." Client/server models have been viable for 35 years, and they are critically necessary for some tasks like surfing the web, and reading your email. The tough sell is arguing that client/server models are sufficient for a broad scope of user tasks for which they are not necessary.

But the question is, why would you want for local storage to become unnecessary when it's a key part of Google Gears and HTML5? Ideally, I shouldn't have to download a specific kilobyte more than once. Furthermore, I'll argue that once I download it, I should have all rights that I'm entitled to for my personal use.

Real cloud computing in my mind would be reliant on cheap and abundant local memory. It would be peer-to-peer and would automatically negotiate version control across the multitude of devices that share my credentials and keys, as well as my secure off-site storage.

It strikes me as pretty silly to argue for dumb(er) clients as part of cloud computing when every vendor is moving in the exact opposite direction. Mozilla has XUL and HTML5 with their own storage mechanisms. Adobe is betting on Flex and AIR. Google has Gears and Android, Palm has their java and ECMAScript webos, Apple has the iPhone SDK and developer tools shipping with XCode. Multiple vendors have internet-based game delivery. RIM has Sun's Java, and Sun has JavaFX.
posted by KirkJobSluder at 12:44 PM on July 9

shmegegge: It's not really a great idea, to my mind, to reduce all new ideas to "just a variation on [failed earlier thing x]" and then dismiss it. I think the industry in general has a better idea of history than that, and despite its failures along the way tends to separate the wheat from the chaff and use old ideas in new successful ways.

Well, two obvious objections:

1) It's incumbent on the people claiming that Chrome OS will revolutionize computing as we understand it to make the case that it will be an improvement on failed earlier attempts. So far you've failed to offer a convincing argument that's the case.

2) As I've said multiple times before. The client/server model is a brilliant success for certain kinds of applications. We are using one of those successes now. The case for chaining all applications to a client/server model is substantially less clear.

I'll also point out that almost all of the applications I use on a daily basis periodically phone home for updates. Does my Word Processor really need to contact its maker on more than a weekly basis, and do I really need a constant contact with a remote server to edit a business letter?

Zed: We don't know what the hardware will be, because it doesn't exist yet.

We do know because they said what the hardware will be. Everything from netbooks to desktop systems.

Zed: All I'm saying is that I think it's too early to rule it out, because it's...

Nonsense. Chrome OS is due out in a year, and its entirely reasonable to take a look at the state of Zoho and Google Docs and say that unless those applications radically change over the course of the next year, they are unlikely to satisfy the intended users.

five fresh fish: It's a good thing innovation doesn't depend on the like of you folk.

Where is the innovation?
posted by KirkJobSluder at 1:04 PM on July 9

It's a good thing innovation doesn't depend on the like of you folk.

Maybe you can explain what's innovative here? It's possible Google is going to put out something exciting, but people are cuming in their pants over a press release. A pretty boring one at that.
posted by chunking express at 1:08 PM on July 9

I don't see any cumming about their press release. I don't see any innovation either, because nothing has been done with it yet.

What I do see is a lot of uninformed claims that this can't possibly be useful or successful.

A year from now we'll see whether Google's come up with a nifty new tool. Personally, I hope it means I can get a hundred dollar comic-book sized touch-screen web interface. A couple of them scattered through the house would be a damn fine thing. With a light OS, they might have a couple days runtime before needing to be charged. That would be sweet.

Or perhaps it'll all fail, like some of Google's other projects.

But for the time being, all we've got here is uninformed, hysterical anger and naysaying. It's pathetic.
posted by five fresh fish at 1:25 PM on July 9

Hysterical? Maybe we're reading different threads.
posted by chunking express at 1:26 PM on July 9

five fresh fish: I don't see any cumming about their press release. I don't see any innovation either, because nothing has been done with it yet.

Which isn't true. The primary user interface (Chrome) has been released. As are the applications that this operating system will point to in the future. What we've not seen yet is how those will be packaged and linked together.

Furthermore, Google's announcement made some very specific claims about what a large number of computer users want from an operating system. We can certainly make informed opinions about those claims based on the behavior of people we know, and the market as it exists.
posted by KirkJobSluder at 1:34 PM on July 9

Except that I don't see those applications as particularly "niche" or "high-end." MSWord is (unfortunately) a requirement for millions of users. Neither Google Docs nor Zoho are robust enough for academic use. So there you have two huge segments of the market that would not be well served by systems exclusively dependent on web services.

which seems to me to say "we would need a web capable Office suite" and assumes that it won't be possible from a 3rd party, at least not to the specifications of an academic community or the kind of user who uses advanced functions, since the 3rd party apps still encounter conversion problems when you export to the MS doc format. am I understanding you correctly?

do you believe it's possible for a 3rd party to eventually overcome those limitations, despite years of not having done so? if not, do you believe it's possible for a 3rd party to take market share from MS with their offering? if not, what do you think about what Zed posted above, this link here speculates that MS is going to offer Office over the web. does that change the forecast for Word users in your estimation?

There certainly is a market for locked-down web-only systems. However it's neither sizable, nor does that market are consumers.

I'm still not sure how "locked-down" we're talking about. I mean, they're saying is got Chrome within a windowing system, yes? shit, that practically describes the relationship between IE and Windows 95, doesn't it? I don't know how locked down or web only we're really talking about, here. here's the quote, though I'm sure you know it:

Google Chrome running within a new windowing system on top of a Linux kernel. For application developers, the web is the platform. All web-based applications will automatically work and new applications can be written using your favorite web technologies.

so yeah, the web is the platform, but that could just mean that the web is the platform the way that flash uses the web as a platform. I don't think that necessarily eliminates the possibility of installed OS level applications. I just think it's a step toward a future of more remotely stored data and applications, but not the total elimination of local storage and apps. does that make sense?

The tough sell is arguing that client/server models are sufficient for a broad scope of user tasks for which they are not necessary.

is it less of a tough sell, though, if it's just a step along a path in that direction? what if we're just naturally heading in that direction. to extrapolate it to its pie-in-the-sky sci-fi extreme, what if we're talking about an age of portable minimal electronic hardware perpetually connected to ubiquitous wireless (though maybe initially wired) remote storage and processing? I don't know if I've been explaining what I'm getting at well enough, but I'm talking about a very distant future, and this being a step along the way. I'm not trying to say it's an entirely new idea, or that Google is blowing minds, here. I think they're taking a predictable step, and one not necessarily destined to fail.

But the question is, why would you want for local storage to become unnecessary when it's a key part of Google Gears and HTML5? Ideally, I shouldn't have to download a specific kilobyte more than once. Furthermore, I'll argue that once I download it, I should have all rights that I'm entitled to for my personal use.

I don't know if I WANT it to, so much as I think it could, in favor of increased functionality down the line.

It strikes me as pretty silly to argue for dumb(er) clients as part of cloud computing when every vendor is moving in the exact opposite direction. Mozilla has XUL and HTML5 with their own storage mechanisms. Adobe is betting on Flex and AIR. Google has Gears and Android, Palm has their java and ECMAScript webos, Apple has the iPhone SDK and developer tools shipping with XCode. Multiple vendors have internet-based game delivery. RIM has Sun's Java, and Sun has JavaFX.

and you don't see these as intermediate steps? from physical data delivery through disks, to digital delivery through download, to remote storage and streamed delivery? the progression seems natural to me.

Does my Word Processor really need to contact its maker on more than a weekly basis, and do I really need a constant contact with a remote server to edit a business letter?

do you need it? no. might you have it in the future, anyway? I think it's possible. I mean, part of what we're talking about, here, is the kind of sharable content that is typified by the architecture behind twitter and other social apps being applied to office apps, complete with the kind of realtime collaboration google has already shown an interest in with Google Wave. I really don't think it's that hard to see this as google's plan. as I said, maybe it won't work, but it seems to be what their strategy is, and now that people are saying they've given MS a reason to put Office on the web, it looks like it's already having repercussions at other companies, as well.
posted by shmegegge at 1:43 PM on July 9

The primary user interface (Chrome) has been released. As are the applications that this operating system will point to in the future. What we've not seen yet is how those will be packaged and linked together.

I think it's worth considering that we haven't yet seen what they're doing to improve or change those applications, though. In a year's time (barring the inevitable release pushback) we could see some radically different applications.
posted by shmegegge at 1:45 PM on July 9

Chrome is open sourced, right? That means you can use their interface to the new windowing system as a reference to write your own app. I give it six months, maybe a year max, before someone ports the openoffice suite to the new windowing system.
posted by idiopath at 2:29 PM on July 9

shmegegge: do you believe it's possible for a 3rd party to eventually overcome those limitations, despite years of not having done so? if not, do you believe it's possible for a 3rd party to take market share from MS with their offering? if not, what do you think about what Zed posted above, this link here speculates that MS is going to offer Office over the web. does that change the forecast for Word users in your estimation?

Until you produce the advanced web-capable office suite, we are talking about something that's even more vaporware than Chrome OS. Is it possible for these web-based offerings to develop in radically different ways over the course of the next year? Yes. Is it likely given what's we've seen regarding prior development? Doubtful.

s it less of a tough sell, though, if it's just a step along a path in that direction? what if we're just naturally heading in that direction. to extrapolate it to its pie-in-the-sky sci-fi extreme, what if we're talking about an age of portable minimal electronic hardware perpetually connected to ubiquitous wireless (though maybe initially wired) remote storage and processing? I don't know if I've been explaining what I'm getting at well enough, but I'm talking about a very distant future, and this being a step along the way. I'm not trying to say it's an entirely new idea, or that Google is blowing minds, here. I think they're taking a predictable step, and one not necessarily destined to fail.

Except that all of the trendlines are pointing in the exact opposite direction.

Smarter devices: My phone is smarter than my first computer. My microwave is smarter than my parent's toaster oven. My laptop is smarter than the entire server farm when I was a university Freshman.

Higher storage density: My iPod stores more data than the tape storage service I had access to as a college freshman.

Smarter clients that leverage local storage: Early web browsers didn't have cache. The next generation did. Now, Gears, Javascript, HTML 5, Flash, and Sliverlight all support local data stores.

The processing and data-storage capabilities at the client end of the cloud are exploding at a rate much faster than Google can slap blades into their server farm. The pipe for physically-connected data processing is going to be at least an order of magnitude higher than what can be achieved over public networks in the near future.

And meanwhile, there are some pretty strong ethical concerns with trusting your data to a service given recent snafus at Live Journal and Amazon.
posted by KirkJobSluder at 2:39 PM on July 9

mkb: "MP3 doesn't support DRM, and it's huge. Nobody has a reason to abandon MP3 for Ogg, because nobody cares about patent issues that don't cost them money, and nobody is going to build a device that doesn't play MP3, but does play Ogg."

The MP3 format doesn't have DRM in it, because it was the thing that caused the demand for widespread DRM in the first place.

It's the last media format that could ever be created without dealing with the DRM issue in some way. All modern formats must either explicitly embrace or reject it, because of how popular MP3 got. I once heard it described as a "naive" format, in the same way that unauthenticated SMTP is sometimes called a naive protocol.*

If the content industries had any idea of what MP3 would bring, there is no doubt in my mind they probably would have paid to have the entire Fraunhofer team assassinated. (If Sony ever develops a time machine, I fully expect that they'll send someone back to 1990, Terminator-style, to do the job.)

* In the same way that it's impossible today for a filmmaker to create film noire without being conscious of the genre, there's no way you could create a new audio file format for mass consumption and not be aware of MP3 and the debate it caused. You might say it was the last unselfconsciously DRM-free format. It didn't have DRM because its designers probably didn't think it was anything that people would be interested in; all new formats have to make a conscious decision of whether to emulate or reject the MP3 model.
posted by Kadin2048 at 2:40 PM on July 9 [2 favorites]

I think the trend is toward using the appropriate tool for the appropriate job -- to make simple things simple and complex things possible.

The idea, to my mind, would be to make it super easy and fast to do the 90% of what you do on a day to day basis through web apps -- browse the web, watch videos/listen to music, email, take notes, write documents... basic spreadsheet/database stuff.

You'd use the cloud for all storage and collaboration (for the sake of convience/backups, etc) and you do 90% of the processing locally.

The advantage is that you have a 'global desktop', so no matter where you are signed in, you have access to your work, etc -- you are never tied to a particular computer, and you have much, much, much less IT infrastructure overhead. Also, simple sharing/web publishing/collaboration, etc..

The difficulty is the 'making complex things possible'. For mass acceptance on 'the desktop', the OS needs to run Windows software. There is just no way around it. Mac's wouldn't sell the way they do today without bootcamp/parallels, etc, and neither will this.
posted by empath at 2:53 PM on July 9

Except that all of the trendlines are pointing in the exact opposite direction.

Smarter devices: My phone is smarter than my first computer. My microwave is smarter than my parent's toaster oven. My laptop is smarter than the entire server farm when I was a university Freshman.

this is an awfully suspect way to describe a trend. my phone is smarter than my first computer, but dumber than my current one in favor of being small and portable. my bluetooth headset is of worse audio quality than my wired one, but is wireless. my phone's address book is smarter than the address book on my first computer, but it syncs wirelessly with apples mobileme remote storage server and syncs over the air with my desktop pc and my work mac which both have more robust calendaring and address book applications, but still rely on mobileme to sync both. my business email, calendaring and address storage is easily even more robust than all of that, because it's my business's exchange server, which is stored remotely. yes, it also stores data locally for offline access, but that doesn't mean it will always have to. my workstation's media storage is stored remotely and accessed via fiber channel instead of on the internal drives. the trend you're describing ignores everything we've seen in networking and storage solutions in favor of comparing a univac to a laptop.

you can present the trends any way you want to, but i think you're being awfully narrow by insisting that the only trend one can reasonably see is toward increased local storage and processing. frankly, I wonder if we'd be disagreeing so much if this were a theoretical discussion outside of the standard google hype. maybe, if we were just talking about the possibility of tech trends heading this way, you'd actually think it were possible instead of outright dismissing it as totally backwards.
posted by shmegegge at 2:55 PM on July 9

oh also:

And meanwhile, there are some pretty strong ethical concerns with trusting your data to a service given recent snafus at Live Journal and Amazon.

yes, this is absolutely true.
posted by shmegegge at 2:56 PM on July 9

shmegegge: my phone's address book is smarter than the address book on my first computer, but it syncs wirelessly with apples mobileme remote storage server and syncs over the air with my desktop pc and my work mac which both have more robust calendaring and address book applications, but still rely on mobileme to sync both.

Yes, but what is actually happening when you sync? The systems in question perform a version-control check and exchange data that is stored within the device. Once you sync, you don't need to query the remote server to look up the phone number of John Doe, Esq.. When you enter "Do" in the search box, it can narrow the choices displayed to John Doe and Jane Donahue. And when you click on the icon for John Doe, you have the number with no latency.

my business email, calendaring and address storage is easily even more robust than all of that, because it's my business's exchange server, which is stored remotely. yes, it also stores data locally for offline access, but that doesn't mean it will always have to.

Stop thinking just in terms of offline access. Think in terms of efficiency. Ideally, you shouldn't have to transmit the same kilobyte more than once. Why do you need to query the net each time you look up John Doe's email address? If it hasn't changed, then fetching it from local storage is quicker, takes less memory, and less power. If you change it, then you can push that change onto the server. If it's changed on the server, it should be able to push that change to you.

Compare a DHTML webmail interface like gmail to a entirely server-driven interface sometime. Even with an ethernet connection, the server-driven interface is a sluggish pain in the ass. The dynamic gmail interface is faster, more responsive, and doesn't have to redraw the entire screen when you change tags, or when it fetches a new message. This is possible because the code to render pages is cached the first time you load it, and reused each time the data is updated.

my workstation's media storage is stored remotely and accessed via fiber channel instead of on the internal drives. the trend you're describing ignores everything we've seen in networking and storage solutions in favor of comparing a univac to a laptop.

The problem here is that networking and storage solutions complement rather than replace local storage. Bandwidth is always going to be a critical limitation on what you can do over a network. One of the tricks of the trade for maximizing bandwidth is to avoid sending redundant information. And that is why you need local storage. Rsync, GIT, mercurial and SVN are smarter and more efficient than FTP. AJAX updates of a DOM are more efficient than reloading the entire DOM. It's this principle that makes Google Docs even possible.

It's not just about enabling "offline storage." It's about maximizing the utility of a limited resource so we can do more interesting things with it.
posted by KirkJobSluder at 3:45 PM on July 9

honestly, this conversation is exhausting me. not that you're exhausting me, but getting really into this WAS fun and is now not, and I'm tired.

so I don't know how much more I have in me, but let me say this: I'm aware that my data is stored locally, and is only synched by mobileme or my exhange server. I'm also aware that this is preferable to remote storage because there's latency involved in remote storage.

but to my mind, that's with current technology, and I see our tech heading further and further outward, rather than inward. eventually, I see remote storage as not only possible but preferable for social and professionally collaborative reasons.

but as you said, the idea is not about enabling remote storage. I think it's become the hangup of this conversation, though, and is part of why this is so exhausting for me. again, I don't think google is trying to enable remote storage for its own sake. I think google is continuing this outward trend, and the possibility of remote storage is part of where that trend could go. systems like onLive could be an example (the game is store remotely, your data is stored remotely.) of that, but it's more of an example of the power of remote servers to provide extremely intensive applications to the home user, given certain technological advances.

but where google is concerned, I don't even think they're trying to make netbooks into client-side boxes that access remote data. how we got hung up on that isn't entirely clear to me, except maybe that I mentioned it as a possibility down the line.

and I think that's part of where we're experiencing a disconnect. you're describing current tech and seem to be saying that there's no reason for it to be headed in the direction I'm specifying because we still have and need local storage. and what I'm trying to say is that, yes, we have and need it now. we have and need many aspects of consumer and professional computing that may function very differently down the line. I think this offering from google could be an indication of continuing that trend. just because my phone stores addresses locally still doesn't mean that that's not a step in the direction I'm describing. and that seems to be part of the disagreement: i'm seeing what we have now as steps toward something very different. and you seem to see what we have as indicators of something far more similar to what we still have. I see itunes storing downloaded music locally as a step toward streaming remote libraries. you seem to see it as a step toward what I assume must be simply larger music libraries stored locally. I see netflix as a step toward remote streaming video libraries. you seem to see it as a step toward more locally stored streaming applications.

I don't know if there's a way to resolve that difference between us, since we're both making predictions. Maybe there isn't one. But if what we're down to is me continuing to try to convince you that I have a reason to believe we're headed in the direction I've described, I just don't know if I have the energy any more. no offense, I'm not saying it's because you're too difficult to talk to or anything. you've been civil and forthright and I'm glad we've talked. but I'm just very very tired. call it old age.

if we continue this, I'll really try to muster some kind of strength to keep going, but don't take it personally if I don't manage to.
posted by shmegegge at 4:03 PM on July 9

Ok, if we are being futurists...

I'm sitting at a cafe somewhere, and there is an interesting TV program in a language I don't understand. I ask my phone, "hey, what's that TV program?" My phone parses my request through its learning algorithms to understand which of the half-dozen televisions with in WiFi range I'm talking about, and does some peer-to-peer chatter with the television to find out. The television, also peer-to-peer, bounces its cached copy of the stream from the beginning to my phone. All of this takes about 30 seconds, and because I'm dining out with a special someone, I record it to my terabyte data store (which is synching to my off-site backup via a smart peer-to-peer algorithm that can use video checksums) and throw the whole process to the background.

My phone has about 8 gigs of compressed and encrypted statistical information that's the basis of a weak AI that scans, evaluates and prioritizes a few hundred information feeds. It's somewhat quirky at times, my tendency to re-read Lukyanenko delivers gossip about the Twlilight movie remake as a false positive, since both have male teen vampires.

On my way home, I witness a car accident and turn on camera conveniently clipped to my glasses. Lacking the bandwith to stream it out at 1080p in real-time, my phone spools and throws a digital signature on each chunk. After the action is over, I shut down the recorder, and the phone bounces it to my backup at about 3/4 speed. When the police officer arrives, I bounce it to her, and the two drivers via peer-to-peer. Then I catch a bus, and watch the show I saved on my HUD.

so I don't know how much more I have in me, but let me say this: I'm aware that my data is stored locally, and is only synched by mobileme or my exhange server. I'm also aware that this is preferable to remote storage because there's latency involved in remote storage.

but to my mind, that's with current technology, and I see our tech heading further and further outward, rather than inward. eventually, I see remote storage as not only possible but preferable for social and professionally collaborative reasons.

I guess I don't see why the two contradict each other. First, I see local storage, caches, and smart-clients as a a necessary part of building a networking infrastructure. Second, I don't think the economies of scale are such that Google, Amazon, and Apple will be able to provide a client/server environment that meets the demand to handle increasingly media-rich content. With consumers buying terabytes and teraflops much faster than companies build serverfarms, I think peer-to-peer is going to be critical for networking.

While you see storage as a hack to get around limitations of bandwidth, I see video streaming as a limitation of older business models. Content providers love streaming because they can enforce a pay-per-view or attach advertising revenue to each view. I'm not convinced that streaming is in the ultimate best interest of consumers.
posted by KirkJobSluder at 4:52 PM on July 9 [1 favorite]

A stream is nothing but a file of unknown length that is still growing. Turning any stream into a normal file is trivial, especially since most of them actually are made into files that are just hidden from you in a tmp directory somewhere.
posted by idiopath at 7:43 PM on July 9 [1 favorite]

I've been amused to see in a couple articles the "tech journalists" talking about how Google has to make absolutely sure that Chrome OS can run Flash and play Youtube videos because this has caused customer returns of other Linux netbooks - without noting that Google owns Youtube and hence will be able to fix problems with at least that application of Flash even after the OS has shipped.

And actually, the Mono project has both the open-source .NET 2 implementation and Moonlight, their open-source version of Silverlight, in pretty advanced development. I wonder if Google will put any .NET components in.
posted by XMLicious at 1:32 AM on July 10

Daring Fireball discusses all of this.
posted by chunking express at 7:25 AM on July 10 [1 favorite]

Wait, people seriously had that much trouble making flash and youtube work on a netbook? I have had troubles with Linux before, but seriously, you go to adobe's site -> the one that firefox sends you to if you click on the "missing plugin" puzzle piece icon, download the default plugin that adobe offers, open the archive you downloaded by clicking on it, then run the program inside. This is IDENTICAL to what you do if you need to use flash on windows (or does windows come with flash preinstalled?).
posted by idiopath at 12:03 PM on July 10

That was what the articles seemed to be implying, that Windows comes with the Flash player pre-installed and hence people who'd never seen a computer without the Flash player didn't realize they could fix the problem themselves, or perhaps that some people were incorrectly installing it.
posted by XMLicious at 12:11 PM on July 10

Wait, people seriously had that much trouble making flash and youtube work on a netbook?

I believe that has more to do with the Atom processor and its limitations than any install / run difficulties users were having. My understanding is that Adobe has been working to optimize Flash to overcome those problems.
posted by hippybear at 12:13 PM on July 10

hippybear: OK, that makes sense, the adobe flash player for linux is a stupid fucking cpu hog, I was able to watch videos but only while running an extremely minimalist windowmanager, no desktop environment whatsoever, and basically shutting down all the demons on my system that were not necessary for network connectivity. Basically I was much better off using downloadhelper and watching the downloaded .flv file in mplayer. Unless google starts offering up raw flv files on youtube by default, or fixes the flash player, they will have the same problem.
posted by idiopath at 12:24 PM on July 10

Flash is equally broken under Mac OS. In fact, This is as good a place as any to mention the Click2Flash plugin, which has made web-surfing much less painful.
posted by hippybear at 12:43 PM on July 10

i wish i could favorite the daring fireball thing 3 times.

I think he's on the right track with the idea of Chrome OS as being a browser that can also run native code.
posted by empath at 12:50 PM on July 10

Anil Dash comments on Google's Microsoft Moment.
posted by chunking express at 1:50 PM on July 10 [1 favorite]

I think he's on the right track with the idea of Chrome OS as being a browser that can also run native code.

Bingo. This is going to be ActiveX controls risen from the dead on zombie-steroids. And some web designers whine about being forced to deal with javascript or PHP. Their heads will explode when faced with c or x86 assembly in the Google Native Client. (But since it will likely be much more cross-platform than anything Microsoft, and especially because I like to see web designers' heads explode, I approve.)

For Flash, there's also the alternative of Gnash, though I don't know how it does performance-wise.
posted by XMLicious at 1:52 PM on July 10

In my experience, gnash is OK performance wise, and atrocious compatibility-wise. It could have gotten better since the last version I tried, though.
posted by idiopath at 12:48 AM on July 11

Microsoft announces web version of Office
posted by XMLicious at 4:18 AM on July 17

Wait, people seriously had that much trouble making flash and youtube work on a netbook?
...
I was able to watch videos but only while running an extremely minimalist windowmanager, no desktop environment whatsoever, and basically shutting down all the demons on my system that were not necessary for network connectivity.

And to think people say Linux is hard to use. Idiopath: Shutting down daemons and killing the desktop environment is not "IDENTICAL" to what you have to do on Windows.
posted by fightorflight at 5:22 AM on July 17

« Older Was Michael Jackson A Pedophile? No! Jackson Was A... | DB Clay, the Portland based ar... Newer »

This thread has been archived and is closed to new comments

Of course you realize, this means war! July 7, 2009 10:10 PM Subscribe

Of course you realize, this means war!
July 7, 2009 10:10 PM Subscribe