Well then, I quit.

October 29, 2003 10:09 AM

Web-based Humor at It's Finest
Words fail me. DeCloak sells (I'm guessing) an HTML templating system that works in tables. But they can't make it work in CSS. The good news is there's no reason to use CSS:
Q: TABLES are for TABULAR DATA and not meant for Web Page Layout . . .
A: Last time I checked, most web sites use a database. And databases are just a bunch of tables in the first place, hence tabular data. [from Zeldman]
posted by yerfatma (54 comments total)

I previewed this a dozen times and left "it's" in there. Administrator please killed me.
posted by yerfatma at 10:10 AM on October 29, 2003

Apostrophe Sinner!
posted by crayfish at 10:16 AM on October 29, 2003

From now on I promise Ill only use it to indicate plurality.
posted by yerfatma at 10:20 AM on October 29, 2003

It's the ambiguity of the title caps with the hyphenated first word that's making me itch, but it's a hilarious article nevertheless. I like "wc3 is useless!" Heh.
posted by DenOfSizer at 10:27 AM on October 29, 2003

That page is a total riot. It reads like the worst apple/pc linux/windows FUD. The person that wrote it obviously did little to no research and doesn't understand how browsers work to assert those things.

The part about how CSS sites don't load faster is the one that kills me. Just random surfing around sites that are CSS based versus those that use tables for long articles should make it apparent how often you have to wait for tables to "fill up" with data before they are displayed.

I should make a version of the front page of this site done up in tables so everyone can see for themselves, I would bet it takes 3-4x longer to load any text.
posted by mathowie at 10:29 AM on October 29, 2003

It's also a deliberate troll that we all fell for. I applaud the author for a job well done.
posted by mathowie at 10:32 AM on October 29, 2003

My own little rant coming on....

While this company's is beign very silly and they are obviously overdefensive, I'm really tired of the strangely high and mighty attitude of the standards zealots.

Those of us building sites in the real world (that is, not lovingly tweaking our own blog) have a hard sell for going the pure CSS route. I've had to abandon or partially implement the CSS approach on the last couple not-for-me projects I've worked on. I'd love to do just CSS, but I can't honestly recommend it for most of my clients at this point.

Why?
-Try this example conversation:
You: "Well, the site isn't going to work in Netscape 4. But that's only 3% of your customer base."
Client: "So hundreds of customers can't see the site?"
You: "We'll give them an upgrade message, but they'll need to download a new browser."
Client: "And we can't do it any other way?"
You: "Well, we could do it with tables and it would work in every current known browser, but it won't be compliant to modern w3c standards".
Client: "Uh?"
... you can go into some details of other benefits, but none of them is nearly as significant as is generally claimed. Rendering time, for example, is very rarely the bottleneck on page load time.

Not to mention the fact that unless the site's designer (99% of the time, the designer and front end developer are different people working together on larger sites) specifically designs with CSS in mind you end up doing just as many silly hacks with CSS as you do with tables. Heck, look up the stuff people need to do to get a flexible three column layout with a nav bar on the bottom. You have to do some horrible hacks to get it to work the same in IE and Mozilla. Oh, right, IE isn't standards compliant either, guess we're just building for Mozilla, Opera, and Safari.

So, yes, these guys sound crazy and/or are trolling and should look into offering a CSS version of their product if their customers want it.

However, I too am a little tired of the overblown arguments from the largely academic standards-first advocates.

I too would love to see a day when all the various http clients out there used the same standards, and those standards included CSS. But right now, a much larger percentage supports table tag layout better than CSS.
posted by malphigian at 10:47 AM on October 29, 2003

Well, how 'bout telling them it's not going to be pixel-perfect in Netscape 4, but it'll be perfectly usable? And now it'll be usable in hand-held devices and more friendly for disabled users. That section of users (though largely unquantifiable, I know) should be growing, unlike Netscape 4 sufferers. I feel your pain, but I think it's a matter of learning how to sell it, which is something developers (myself included) may not be good at. I have a hard time explaining the benefits because I'm so sold I can't imagine the alternative. It's an almost physcial reaction for us here when we have to go work on our old code and see all that nested crap.
posted by yerfatma at 10:54 AM on October 29, 2003

"IF anything, these non-visual browsers need to more intelligent and not dictate to 99.99% of the population on how to surf"

I'm not blind, but I know that overusing tables makes it impossible to understand webpages using audio feedback.

"*THE REASON FOR CSS PURIST'S HOSTILITY
(A lot of these CSS purists don't even have a data-driven site so they have no appreciation of what you can do with a database. Plus their own personal web site typically has a bunch of static pages anyway. They have no idea that after the 10 minute FLASH commercial that people actually want to do something practical besides being entertained.)"

Ad-hominem.

"One should ask these, "purists and elitest" CSS people who advocate FULL CSS, "At what point is abstraction too much?"

Spelling "elitist" incorrectly is almost as bad as spelling "stupid" or "idiot" incorrectly.

"So, what's the savings? Half the bandwidth on 5-10% of the total bandwidth is 2.5 - 5% savings...MAX!!! Wow, lots of savings there!!! That's, TWO-POINT-FIVE to FIVE percent, not 50%...tha't not a misprint."

Any increase in efficiency saves a ton of money if you get a lot of hits.

Mathowie - if it's a troll, it's a damn good one. View the page source - I've never seen so many table tags in my life.
posted by Veritron at 10:58 AM on October 29, 2003

i have to agree with malphigian, only i'd add that in my case it's hard even convincing the dev company i work for: they want pretty graphical sites with traditional tabled sites. i've been trying for well over a year to convince them otherwise but to little fanfare. i finally just gave up and manage to keep up with css in my own time.

on preview: like yerfatma said, i think a lot of it has to do with selling the merchandise. i'm horrible at this, but one of the programmers here is quite knowledgable with css is beginning to use it quite extensively. he's also quite convincing in his arguments and the company is sloowly beginning to see the benefits.
posted by poopy at 11:00 AM on October 29, 2003

oh and what Veritron said. i think this guy is totally serious.
posted by poopy at 11:03 AM on October 29, 2003

thanks to a template i found via metafilter, and some tips from the #mefi gang, i was able to totally retrofit blort. as of october 1, it relys entirely on CSS for positioning - no tables! well "big deal" you might say, but blort was a nightmare of nested tables, 1 pixel gifs for used for alignment - the whole ugly schmear! all of that is gone, the size of the html files it outputs has been quartered, it loads way faster and now looks much better in non-IE browsers. in fact - it looks the same in mozilla as IE, which it never did before. i've been pretty geeked up about it :-)
posted by quonsar at 11:05 AM on October 29, 2003

This may be useful in explaining the benefits of CSS to non-developers.
posted by yerfatma at 11:10 AM on October 29, 2003

This might help as well: The business value of web standards.
posted by fletchmuy at 11:25 AM on October 29, 2003

he's also quite convincing in his arguments and the company is sloowly beginning to see the benefits.

Poopy, is it possible that your bosses give less credence to your arguments because, well, you use a handle like Poopy?
posted by billsaysthis at 11:27 AM on October 29, 2003

I keep trying to follow this link and succeed only crashing IE. Is he using so many tables I don't have the RAM to display them...?
posted by Tubes at 11:31 AM on October 29, 2003

I'm trying to wean myself off tables and on to pure CSS, but there's one thing I still haven't figured out: is it possible to align things relative to the right edge of the window?

ie. window-width - 50px

Using tables, I'd have an empty cell to the right of my data, 50px wide.
posted by krunk at 11:41 AM on October 29, 2003

poopy and malphigian, I used to think standards zealots were also not living in the real world and it wasn't actually until last year that I started giving a crap about standards at all. In the five years before that I thought that coding to standards limited what you could do and also didn't work across browsers, so it wasn't worth it.

After learning all the workarounds, and workaround-arounds, I can now almost code any design in pure CSS. Sometimes it takes a lot of work, and sometimes it feels like a house of cards that can drop at any second, but I have only encountered one or two designs done in tables that I couldn't replicate in CSS exactly (like my personal blog, which uses a table for layout).

But after you get over the hump of learning what works and doesn't, which browsers can do what, etc, you'll get to enjoy all the other benefits of using CSS and the validator for development. This means faster dev time, less code to wade through, less code to deliver to clients, and the validator is a great way to solve display bugs.

So I guess this is just a long-winded way of saying that I was in the same spot you were, and eventually convinced people I contracted for and people I worked for that moving to xhtml/css was worth it in the long run, as long as people didn't mind giving text versions of the site to six year old Netscape 4 browsers.
posted by mathowie at 11:41 AM on October 29, 2003

is it possible to align things relative to the right edge of the window?

Yes, in browsers with decent CSS support (e.g., IE5+, Netscape 6+, etc.). Assuming the item is outside of any container that's not statically positioned (which is a stupid way of saying it's not inside something that's positioned absolutely or relatively), it'd be:

position: absolute;
right: 50px;

I think mathowie's last paragraph is right on: there's an adjustment period during which you will want to hide back under the covers of table-based positioning, but it'll pass. And you can make things look ok in Netscape 4 if you take the time.

Typically I build something so it looks right in Firebird (and watch it in IE6) with a DOCTYPE that throws IE into standards mode. Once I'm happy with it, I break the styles up into two stylesheets, base.css and dom.css. I move all of the stuff I know will cause Netscape 4 to shit itself into the dom.css and then tweak from there until it stops crashing and looks ok. I've started adding a third stylesheet to hide stuff from IE4, but that's just because I'm so anal retentive.
posted by yerfatma at 12:02 PM on October 29, 2003

yerfatma -- cheers! now if only css support was consistent across browsers...

I agree that css saves a lot of time changing stuff once you're happy with the page, but I'd argue that the initial coding time takes longer (well, maybe that's due to standards compliance). I've been doing tables for so many years now that I don't even have to think about it. And I can't say I've ever had to resort to using spacer gif's. How tacky!
posted by krunk at 12:14 PM on October 29, 2003

It's also a deliberate troll that we all fell for. I applaud the author for a job well done. -- mathowie


I was finally able to get to the site and read the manifesto. I'm usually pretty good at spotting irony and trolling, and I don't sense it here...
posted by Tubes at 12:16 PM on October 29, 2003

I'd also like to propose Senior's Law: No matter who the contact person is for a client, they will be running Netscape 4.0 on a monitor with a resolution of 640x480.
posted by krunk at 12:17 PM on October 29, 2003

I'm going to hop right on their insecure order page and punch in my VISA card number! Made a dummy account and followed it up to the end- not a secure transaction. So, while maybe not a troll, it sure isn't a trustworthy company (imho).
posted by internook at 12:28 PM on October 29, 2003

Why?
-Try this example conversation:
You: "Well, the site isn't going to work in Netscape 4. But that's only 3% of your customer base."
Client: "So hundreds of customers can't see the site?"
You: "We'll give them an upgrade message, but they'll need to download a new browser."
Client: "And we can't do it any other way?"
You: "Well, we could do it with tables and it would work in every current known browser, but it won't be compliant to modern w3c standards".
Client: "Uh?"

I approach the argument differently, and almost always win. My points:

Me: "The information will be available and fully functional for 100% of your user base, but those using netscape 4 will not receive many of the visual benefits. That's only 3% of people though."

Them: "So hundreds of customers won't be able to see the site the way it's supposed to look? Can we do it another way?"

Me: "We can use tables and incorrect markup, however I make no guarantee that any browser will display it correctly, at all."
posted by tomorama at 12:34 PM on October 29, 2003

tomorama: That's not really a truthful statement for me (well, unless I wanted to lie), I can make about as good a guarantee of support in either case, since I always test on all target browsers and, obviously, make no claims about future browsers.

And in that case, CSS is no more a gaurentee of proper display on every browser than tables. One just happens to be up to the latest standards.

Anyway, good points all, like I said, I do all my personal stuff nowadays in CSS, I was only trying to make the point that there still are some arguments for legacy development -- let's hope those last few netscape 4 users wise up soon. Or maybe all of you guys convincing companies to go all CSS will frustrate the netscape 4 hold outs into upgrading. Way to take one for the team :).
posted by malphigian at 12:47 PM on October 29, 2003

I make no guarantee on any browser for incorrect markup (because that's what it is, incorrect) since somewhere in some program, an error may occur because of some work-around or proprietary statement, or something, somewhere.
posted by tomorama at 1:27 PM on October 29, 2003

Who are these Netscape 4 people? Surely there are enough Mefites across the planet that we can individually visit all these people and do the upgrade for them.
posted by monju_bosatsu at 1:32 PM on October 29, 2003

position: absolute;
right: 50px;

Or put margin-right: 50px; on your main content div.
posted by hyperizer at 1:43 PM on October 29, 2003

That assumes you want everything to be at least 50px away from the right-hand side.
posted by yerfatma at 2:32 PM on October 29, 2003

Malphigian--
You can tell your clients:
1. Valid sites render faster--based on my nonscientific testing, if a browser doesn't need to go into "quirks mode", it lays the page out noticably faster.
2. CSS will save bandwidth.
3. Netscape 4 can still show the content, it just won't look as pretty, and if you really want to, you can present a dumbed-down stylesheet to it.

(Yerfatma's link covers some of this)

I've had precisely one visitor to my site complain because he's still using Netscape 4. He also uses, I think, a Mac Duo 280.
posted by adamrice at 2:38 PM on October 29, 2003

At my place of work, we still have about 9% Netscape 4 people. They're crotchety, whiny, dysfunctional, passive-agressive....and usually technophobic. "I was forced to take a class on how to use this years ago and I'm not going through that again!" A lot of them are also using old AOL software, which just adds to the fun.
posted by gimonca at 2:39 PM on October 29, 2003

Oh, and even Eric Meyer has said, in so many words, "look, if you need to use a table to position some page elements, I'm not going to get mad at you."

That said, it is pretty straightforward to use CSS to handle 90%+ of what you could do with tables, and plenty more that would be like pulling teeth with tables.
posted by adamrice at 2:46 PM on October 29, 2003

The NS4.x users are all web developers trying to see what the website looks like for the annoying people using NS4.x...
posted by twine42 at 3:22 PM on October 29, 2003

This would be the time to mention CSS Zen Garden -- the most impressive demonstration of the flexibility of CSS so far, and surely required viewing for any CSS denier out there, indeed anyone interested in developing HTML-based sites.

The Garden comes with a large number of style sheets, the best of which completely transform the look of the pages, beyond what you thought was possible with no modifications to the HTML. Try out different style sheets from the column on the right.
posted by gentle at 4:31 PM on October 29, 2003

As a coder I like CSS a lot. I'm not a hardcore html freak and so CSS makes things easier because with building forms i can drop a label in a certain place and know it's going to stay there.

It is a mind set change though - people have been so indoctrinated into using tables for layout that CSS almost feels wrong.
posted by carfilhiot at 4:38 PM on October 29, 2003

On an unrelated to the original post aside, can anyone recommend a site with a good background for using css, especially things like moving from using tables to css. I've played around with it some, but I have to say I'm still stuck on tables for formatting.
posted by piper28 at 7:11 PM on October 29, 2003

is it just ie6 or does drag-over-text selection work kinda wonky for that zen garden site?
posted by juv3nal at 7:32 PM on October 29, 2003

Web-based Humor at Its Finest

(This common mistake bugs me a lot, sorry.)
posted by VeGiTo at 8:47 PM on October 29, 2003

piper28: try glish, and you'll want to have a look at layout-o-matic.
posted by quonsar at 9:48 PM on October 29, 2003

Any web application programmers out there? I've been creating dynamic web content for the last 4 years via database, and rely on tables heavily to produce tabular data. How does a programmer get around tabular information in CSS? Most of the sites I develop require at least some pages of dynamic content. I write ASP code that in turn generates HTML. Tables are very useful here. How fast a page loads is often down to initial design. For example the page you are reading now is very very long. It takes time to load because the initial design behind it decided on displaying every front page post of the week and not just the day (example of design). Not saying that's right or wrong, just an example of how initial design will decide how much content a page has in the first place. Because the pages I design are empty templates without any content in them, I need to know the scale of the content. If I'm looking at 30+ records, I might decided to paginate and not make a 30+ row table.

If everyone's talking about static content primarily here, fair enough. All bets are off. As for dynamic content, tables are very useful (unless you're pulling simple blocks of text from a database in which case CSS or tables can do it easily).
posted by SpaceCadet at 2:56 AM on October 30, 2003

if you been brainwashed for the last 2 years or so on the absolute superiority of CSS (e.g. sort of like the Nazis who thought they were the superior race)

Wow, an anti-CSS rant that Godwins itself. Class.

who says they are going to visit enough pages on your web site anyway to really enjoy the benefits of having that extra large .css file download initially

I can see this going down well with management. "They're going to leave out of boredom after the first page anyway, so screw it!"

[Checks size of own extra-large CSS files for layout and font-styling... CSS file for all browsers: 1,847 bytes. Supplemental 5+ browsers file: 458 bytes.]
posted by rory at 3:45 AM on October 30, 2003

Heh, I like the bit where he calls you all Nazis...
UPDATE:
Seems a lot of people are really UPSET at this article....Well, I guess, if you been brainwashed for the last 2 years or so on the absolute superiority of CSS (e.g. sort of like the Nazis who thought they were the superior race)...
posted by chill at 3:53 AM on October 30, 2003

How does a programmer get around tabular information in CSS?

You don't: tabular data, by definition, goes in tables. It's all the other stuff that you're pulling out of a database and outputting in a loop that doesn't need to be in tables. This site is a great example: each comment on this page is in it's own DIV. If it were one giant table with repeating rows for each comment (and then another row for the "posted by" info), it would take a lot longer for the browser to render. Ignoring the fact the HTML might be 50-100% larger in terms of file size, the browser needs to get the entire table downloaded before it can render it. With CSS it's a good deal faster. When I ripped off this site to learn PHP on my own site, the original design forced me to use tables (because my CSS wasn't as strong as it is now). Switching that design to CSS caused the page to render much, much faster.

I'm kind of a hybrid here at work, doing HTML and applications development. As I teach the pure applications people more and more CSS, they're delighted with how much easier things become. It saves you a lot of logical leaps when you don't have to force everything into tables and it makes all those things the designers wanted to do a long time ago much easier to implement.
posted by yerfatma at 4:58 AM on October 30, 2003

What yerfatma said. I do a lot of dynamically generated sites, both in PHP and ASP. I switched over to using CSS full-on a couple of years ago. It is TONS easier to program around, and the HTML code is much easier to write, easier to read and most importantly, easier to debug, especially when it's being generated by a script.
posted by RylandDotNet at 6:09 AM on October 30, 2003

yerfatma and Ryland, I guess you're not doing much in the way of producing pages that require data in columns then? Almost all my clients require some form of tabular data....maybe it's more to do with the nature of my clients.

...the browser needs to get the entire table downloaded before it can render it

I do know that in ASP you can control this server-side - it will display the table drawn so far with a Response.Flush command (yup, it will display it to the latest row downloaded)....however that's bad practice as it forces two trips to the client. On my own sites, if I'm forcing the client to draw a large table, the design has already failed. For example, this particular site takes forever to load long threads even on broadband (200+ posts can take 30 seconds sometimes). For me that's a design issue - I'd paginate at about 30 posts a page and display the last 30 posts. Can't help but wonder how much strain the Metafilter database takes with these long pages being served up to all it's clients....(don't want to make this a critique of metafilter - there's many ways to make an omellete!).

I'm not anti-CSS at all but I'm certainly not anti-tables. Both have their uses!
posted by SpaceCadet at 6:55 AM on October 30, 2003

As long as we're talking about tables...

1. A while back I translated the comments in the source code for a mini-browser (the kind that might be installed in a PDA). The table-rendering code was by far the hairiest. Tables that didn't explicitly define the width for everything were passed through 4 separate width-guessing algorithms. Imagine how crazy that would get with nested tables. Table rendering creates a huge processor hit.

2. Although you can use CSS with tag-soup HTML, the point is to use it with structurally meaningful HTML--so tabular matter goes into tables. I've actually seen CSS used to render a table as something that didn't look like a table: for, say, a catalog, you'd get a nice layout with the product name at the top of the section, a picture, some facts, etc. Looking at the HTML, all you see is a bone-stock row of cells containing product name, img tag, price, description, etc.

3. Although I haven't experimented with this myself, you can actually use CSS to apply table-like layout to non-table markup. In essence, the CSS says "display this DIV (or whatever) as a table cell here."

4. Check out http://thenoodleincident.com/ for some excellent CSS tips.
posted by adamrice at 7:37 AM on October 30, 2003

Don't want to get too into code here, but here's a simple example of how a table might drawn by pulling the content from a database:-


-----------------------
[table]

[% executequery (select * from databasetable;)
do while rs.eof<>TRUE
%]

[tr]
[td][ %=rs.Fields (name)%][/td]
[td][ %=rs.Fields (email)%][/td]
[/tr]

[% rs.movenext
loop%]

[/table]
------------------------------------

That will loop through and display x amount of rows of names and e-mail addresses depending on how many records it pulls from the database.

(looks weird as I can't use real tags)

Is there an alternative way to do this by CSS?

posted by SpaceCadet at 8:27 AM on October 30, 2003

Sure there is; you could use divs and spans instead of table tags. But that would miss the point. Table tags are perfectly appropriate for presenting tabular data, such as rows of names and e-mail addresses. You can still apply CSS to the table to modify how it looks, but there's nothing 'unstandard' about using table tags if the actual content they contain is a table.
posted by rory at 9:05 AM on October 30, 2003

I guess you're not doing much in the way of producing pages that require data in columns then?

Sure I am. But that's tabular data. So it goes in tables. CSS can clean up your table code too. It looks so strange to me now when I write a table: there's nothing specified in the TABLE tag, 95% of the tds have no attributes set; it's just a nice, clean, unnested table.

I'm clearly not doing the job explaining what I mean by "tabular data". It all comes back to the idea of semantic markup mentioned above. If you think pure-CSS folks are zealots, wait 'til you get hit semantic markup. I have ugly arguments with other developers here who want to know where I get off saying it's not good enough to use DIV and P and whatever else.

If I can't get across what's tabular data (items with a number of pieces of related information) and what's not (everything else), I'm not going to be able to explain why you should look into ADDRESS or CITE or whatever. If you imagine a web constructed of perfectly coded XHTML pages (CSS + what I'm calling semantic markup), can you see how those pages are just one enormous (XML) database waiting for you to query them as you see fit?

Is there an alternative way to do this by CSS?

Here's something I did recently for a list of groups that have a control to delete them floating at an arbitrary point on the right-hand side, where you'd normally have a two-column tbale so everything's vertically aligned properly but the second column is also aligned:

div.privilege {
width: 400px;
}
div.privilege .label {float: left; width: 250px;}
posted by yerfatma at 9:09 AM on October 30, 2003

> At my place of work, we still have about 9% Netscape 4
> people. They're crotchety, whiny, dysfunctional,
> passive-agressive....and usually technophobic.

I qualify for all except technophobic. I maintain a fondness, though, for old as well as new tech. I still have Netscape 2.02 and 3.04 installed (also Mosaic 3.0) and though I mainly use Mozilla 1.x I have the useragent override string set to report Netscape 4.75 on Windows 95. If a website tells me "sorry, you can't come here" I don't go there. I just adore showing people their websites in Netscape 2.

I also dig stinky cheese.


> "I was forced to take a class on how to use this years ago
> and I'm not going through that again!" A lot of them are
> also using old AOL software, which just adds to the fun.

I bought a car years ago (1978, specifically) and I would be very much annoyed if they altered the interstates so that cars older than 2001 can't drive on them.

Consider the recording engineer. He might like to mix exclusively for $20000 speakers. He might like to tell the customers "if you want to play our disks, you have to upgrade your playback equipment to our specs."

But he can't do this--not if he wants to get work, anyway. He has to find a mix that sounds good on high-end equipment and car stereos and boom boxes and joggers' earphones.

Thus also with web designers. You can prune your requirements so as to be widely compatible, the way the client wants, or you can tell your client "I only design for niche markets where the user understands XHTML 1.1 and XML 1.0 and CSS level 3 and DOM level 2 and ECMAScript 2.62 and MathML 2.0 and SVG 1.0." Or you can design whatever you like, as a hobby.
posted by jfuller at 9:24 AM on October 30, 2003

I disagree with the basis of your argument. If you're arguing "No site should lock out Netscape 4 users (unless it's a secured site that requires encryption NS4 can't provide)," that's fine. You can visit CSS sites without a problem. They just don't look like they would in a newer browser. Highway road surfaces get upgraded all the time. There's no reason I shouldn't try to build a better road just because some old cars can't take full advantage of it.

I don't think anyone advocating full CSS wants to lock out old browsers (except in extreme cases like ESPN where Netscape 4 users cannot see advertisements and are getting what the company considers a free ride).
posted by yerfatma at 9:45 AM on October 30, 2003

I just adore showing people their websites in Netscape 2...

...watching MTV on an old black-and-white TV... listening to electronica on a mono turntable... playing Grand Theft Auto in 16 colours...
posted by rory at 9:58 AM on October 30, 2003

jfuller:

The problem with Netscape 4 isn't that it's OLD, the problem is that it's BROKEN. Brokent, busted, failable, it fucks your shit up.


You can, using totaly standard CSS and HTML make a page that will display correctly on every web browser made other then netscape four. And that's the problem. NS 2, NS3 great. NS4? No, go away upgrade or down grade or something. But stop using that peice of shit.
posted by delmoi at 10:10 AM on October 30, 2003

The alternate text: "Nested Templates".
posted by delmoi at 10:14 AM on October 30, 2003