What comes after “open source”

April 2, 2019 8:55 AM

Steve Klabnik muses about a conflict within open source, and what comes next.
posted by a snickering nuthatch (27 comments total)

Uhh ... what the fuck is this even implying?

I’m not sure exactly how it happened. I think the lazy answer is “GitHub!!!!”. I do think GitHub played a role, but I think the answer is more complex than that. I personally think that gender plays a huge role. But that’s a different essay. Regardless of why it happened, something did happen.

Emphasis mine, but: a billion red flags flying into the sky here about whether this is a person we should be boosting.
posted by tocts at 9:10 AM on April 2, 2019 [12 favorites]


Apache License 2.0
BSD 3-Clause "New" or "Revised" license
BSD 2-Clause "Simplified" or "FreeBSD" license
GNU General Public License (GPL)
GNU Library or "Lesser" General Public License (LGPL)
MIT license
Mozilla Public License 2.0
Common Development and Distribution License
Eclipse Public License


Choose any two.
posted by sammyo at 9:10 AM on April 2, 2019

Steve Klabnik is an anarchist or maybe a kind of communist, not some kind of alt right guy. My best guess is he meant that with open source being overwhelmingly male, participants are relatively less willing to do the emotional labor you need to sustain a social movement or ideology.
posted by vogon_poet at 9:16 AM on April 2, 2019 [6 favorites]

Yeah, the part about "this is the fault of the WOMEN-folk, but that's a rant for another day," followed immediately by protracted whinging about developers these days, and how they're not worshipping properly at the Altar of Stallman, makes me think this guy can go fuck off into the sun.
posted by Mayor West at 9:16 AM on April 2, 2019 [4 favorites]

Given his larger public persona and track record on trying to make open source projects more inclusive, I really am pretty sure that’s the opposite of what he meant.
posted by vogon_poet at 9:18 AM on April 2, 2019 [30 favorites]

Meanwhile, real programmers are getting real work done with a mix of open source and proprietary software.
posted by Nelson at 9:21 AM on April 2, 2019

Given his larger public persona and track record on trying to make open source projects more inclusive, I really am pretty sure that’s the opposite of what he meant.

OK, I have done some minimal research, and have found evidence that supports your perspective over mine. I retract my wish for him to fuck off into the sun. That said: he specifically called out gender as an explanatory factor of change he doesn't like, while talking about how the programming landscape is different than it was 20 years ago. I'm a developer, his ostensible audience, who doesn't know enough about this guy to know whether he's a horrible misogynist or a standup guy who's trying to be more inclusive in technology. Given the pulse of the industry, I'm more likely to suspect the former than the latter. Mea culpa for jumping to conclusions, but that is a stunningly tonedeaf thing to say in an article about the evolution of open-source.
posted by Mayor West at 9:25 AM on April 2, 2019 [14 favorites]

My read matches vogon_poet's, knowing little about this guy. I think when he said "gender" he probably meant "culture", i.e. the notorious toxic masculinity of many OSS projects is limiting their growth and ability to compete. And it should probably be seen as telling rather than damning that he conflated these two factors.

Ultimately as I understand it there isn't even much of a conversation to be had about gender in the OSS community because the lion's share of non-male devs stopped trying to engage in any significant way long ago. So I'm not sure what else he could mean.

I mean, if I'm wrong I'll stand corrected, but that's my read.
posted by potrzebie at 9:26 AM on April 2, 2019 [2 favorites]

There's been a long running feud between the FSF crowd and the MIT/BSD/etc. crowd for decades. I'm not sure why he thinks this is recent.
posted by Candleman at 9:54 AM on April 2, 2019 [7 favorites]

It seems that he teases his arguments without actually making them. I bet this will be an interesting article when he finishes it.

Edit: oh, second link. Duh.
posted by sjswitzer at 10:00 AM on April 2, 2019 [1 favorite]

(I think I screwed the post up by not signaling strongly enough that the second link is the main link...)
posted by a snickering nuthatch at 10:08 AM on April 2, 2019

Bit of a rant here, I apologize, but it is on topic.

This year marks 20 years since I took my first job with a Linux/open source related startup. I've been through phases of being passionately pro-Linux, pro-open source, and through a trough of despair and back again. When I was young I thought open source could change the world. It made a tiny dent, I suppose, but mostly the world changed it. Same way the world co-opted the hippies and other counter-culture movements. The powers that be figured out how to harness the parts they liked for profit and scuttled the rest.

Part of the problem is that users and non-developers rarely enter the discussion. One might think that users should factor into the discussion, as should those who support open source projects and products in ways that aren't related to code - but they rarely do.

Free software folks like to talk about the "ethics" of software licensing, but they might as well be shouting at the moon for all normal users care. And I lost any and all interest in that position when I read RMS' response to a blind user about how they should just wait for free software tools to be available to do their job, and it was unethical for them to use proprietary tools - it would be better, according to Richard Stallman, for them to be unable to function as a blind user / programmer than to use proprietary tools. Fuuuuck that.

The problems with open source and free software vs circumvention of those licenses' spirit and intent is not going to be solved as long as we live in a capitalist society and it costs money and time to produce software. Full stop. The problems facing free software and open source are the same problems that everything else has - and if we can't as a society decide we're going to take climate change seriously, or find it in ourselves to oust Trump from the White House or cancel Brexit, then why do we think open source is going to be a happy little exception to capitalism and human nature?

It costs money to develop good software and all the trappings that surround it, particularly at a pace and quality suitable for normal humans who aren't developers. As long as money is a factor it will be distorted by business interests and any and all licenses and movements will be corrupted to suit business interests. I would be happy, of course, to be proved wrong - but watching 20 years of projects being derailed by competing business interests, greed, personality conflicts, and other assorted failures of people to get along towards a common goal, I am not optimistic. (That's not exclusive to open source, but in particular...)
posted by jzb at 10:11 AM on April 2, 2019 [31 favorites]

I think another problem with the corporatization of open source is how GitHub has become the new resume for programmer types. If you want to get a job as a programmer or developer, many employers want to see that you have a history of writing code and putting it out into the world. And that history needs to be ongoing...

Say you contribute some code to a couple projects on GitHub, have a small hobby project, and leverage that into a paid gig as a junior developer. If your full-time development job doesn't give you time to keep contributing to open source projects, or even your own open source hobby project, when it comes time to look for a better job, you're up a creek. This narrows the range of people who can actively contribute long-term to open source projects, and even a narrower subset of those people will have the wherewithal and ability to contribute and manage open source as a concept.
posted by SansPoint at 10:16 AM on April 2, 2019 [10 favorites]

I'm not sure why he thinks this is recent.

This is different. Take a look at this Hacker news post. The context is that Cloudflare has taken Wireguard, and forked it into their own product. I'll quote:

tw04 1 day ago [-]

> Forking an upstream project to implement decisions without upstream’s consent is a tried and true open source software process, implemented by thousands of projects over the years. Claiming that they don’t support standards, solely because they don’t support another implementation of those standards, is incorrect and inflammatory.

If upstream is doing something you don't like and refusing to work with you, sure.

When upstream actively petitions you to not fork, asks you politely to work together, and you refuse to work with them, that is far, far from a "tried and true open source software process". That creates a fissure in the community and it generally ends up poorly for everyone involved.

My comment is far from inflammatory, it's a statement of fact, and something cloudflare has refused to acknowledge or respond to. Which just further drives the point home that they aren't acting in good faith.

This is about community, not code or licensing. This is also the result of the power imbalance between actual smart development communities. When one side is sitting on IRC, and the other is a multi-billion corporation like Google, the ant and the elephant can both dance with each other the same, but one will get squashed. The ascent of the fast smart giant company over the dumb company means that they can use your tools, and even open source some code, but nothing important, while take full advantage of the bounty of code. Google will open source data formats for storage like Protobuf, but the algorithm that determines what you see when you type something in the search bar will never see the light of day. The keys to the kingdom are code that is run on computers we don't own.
posted by zabuni at 10:24 AM on April 2, 2019 [12 favorites]

Meanwhile, the open source hardware world already has a certification process and mark. I don't understand why FSF/OSI hasn't been sitting up and paying attention to what's been going on in the hardware world for the past decade. The whole software licensing argument just feels so ossified, like it's round two of vi vs. emacs or something.
posted by phooky at 10:36 AM on April 2, 2019

Mea culpa for jumping to conclusions, but that is a stunningly tonedeaf thing to say in an article about the evolution of open-source

It's a post on his personal blog, where he can reasonably expect that most readers are familiar with him.
posted by ripley_ at 11:14 AM on April 2, 2019 [5 favorites]

Akin to SansPoint's point, there's the $WORK where you just can't get a clear sign-off on being able to contribute code due to lawyers and such. Nobody would ever give me the OK to publish that cool module I wrote because I couldn't find an existing module that did what I needed. So nothing made for $WORK no matter how generic made it out into the open source world. :(

zabuni, that WireGuard vs CloudFlare thing seems to not be all that bad. CloudFlare is happy to contribute their code back upstream. It seems they're just throwing programmers at it with the intent to move fast and can't be slowed down by being a sub-project and in WireGuard's workflow. I'm pretty happy that the CloudFlare VPN thing is WireGuard, WireGuard rocks and I hope to benefit from this parallel development.
posted by zengargoyle at 11:39 AM on April 2, 2019 [2 favorites]

It seems they're just throwing programmers at it with the intent to move fast and can't be slowed down by being a sub-project and in WireGuard's workflow.

See above my point about the failure of open source due to people being unable to work together on common goals as motivated by greed, etc.

"Thanks for this neat-o project, but we can't be bothered to work with you, so we're going to just go over here and do our own thing - and if it happens to be duplication of effort or even sabotage what you're doing, well. We can't be inconvenienced."
posted by jzb at 11:49 AM on April 2, 2019 [2 favorites]

→ Part of the problem is that users and non-developers rarely enter the discussion.

If the only avenues for discussion are managed by developers for developers, and user forums have devolved into tone policing about structuring requests to conform to esr's dreaded "smart questions" diktat¹, it's no wonder that user voice is missing from open source. It would take a forum for users, moderated by users, to break that cycle.

¹: Paul Stoffregen's How To Get Tech Help From Strangers On The Internet is way better anyway.
posted by scruss at 11:59 AM on April 2, 2019 [4 favorites]

Emphasis mine, but: a billion red flags flying into the sky here about whether this is a person we should be boosting.

I can see why someone not familiar with Steve's politics might misread this, but it is a misreading. He expanded a bit in a comment on HN:

What I will say is that the speculation of this as being about "brogrammers" is incorrect. I don't want to super get into it, because my thoughts aren't fully formed, and this is a hot-button topic. What I will say is that, in the last 20 years, our discipline has grown, and women are returning to it. If you take the large public figures in both movements, like RMS and esr, both have attitudes and beliefs that are... not great, on topics like sex, gender, and other social issues. In general, the social stuff around these movements is very masculine, and often socially regressive. This makes it hard for women to care about these movements. Obviously, these are mostly broad generalizations, and there have been some women who are involved here. I know several women who are very passionate about both Open Source and Free Software. But I've also spoken to many who say "never", and the numbers speak for themselves.

posted by jacobian at 2:23 PM on April 2, 2019 [9 favorites]

a conflict within open source, and what comes next.

This was a good read, but the what comes next part remains very un-answered, as he dismisses both of his own proposals at the end. Sigh.
posted by aspersioncast at 4:25 PM on April 2, 2019

If folks need more context, consider perusing geek feminism's wiki. Stallman has a variety of creepy, public opinions, and regressive humor. I won't recap the litany here for brevity.

ESR... wrote a guide to picking up women for nerds, and his blog subheading is literally "Sex, software, politics, and firearms. Life's simple pleasures…" Searching it for feminism yields gems like 'Why Hackers Must Eject the SJWs'.
posted by pwnguin at 8:25 PM on April 2, 2019 [3 favorites]

My comment on the Morozov/O'Reilly piece six bloody years ago, drawing from Geert Lovink writing in 1999, still applies. In fact, it applies even more when you consider what the keys to the kingdom now are.

The volume of capital which is circulating at the higher level of applications and e-services which build on top of the Net will gently push aside old software configurations. Roots are fading away, getting irrelevant (sorry, Kittler). Capital, with all its weight is about to smash the Open Source movement. Not with repression. Not in an ignorant way. There is a growing respect, with bits of appropriation here and there. But life goes on. [...]

Linux people won’t starve, that much was clear at WOS. Economics is not even an issue for them. They already got jobs. Some are even being paid to write free software. For the coding class, hobbyists, midnight hackers, economics is not a real topic.

As zabuni and jzb note, this is about community in the context of capital. On the one hand, certain kinds of capital now directly sustain large-scale open source projects because of the positive externalities. But you don't need to be a student of Marx to see the fault line that opens up when a particular type of capital enters the equation at a particular order of magnitude with the expectation of a particular type of return. On that side, it's the unseen code that dumps out its negative externalities like toxic waste into a river.
posted by holgate at 9:08 PM on April 2, 2019 [2 favorites]

One thing that doesn't get enough attention: software licenses operate at the granularity of programs, but often it's collections of programs that have value. Both MacOS and Android have cores (kernals + system libraries and applications) that are open source or free software, but then have proprietary layers on top. It's a neat hack by the controlling companies, they get a lot of the benefits of open source while also keeping control of the overall combination.
posted by nnethercote at 11:43 PM on April 2, 2019 [1 favorite]

Is it true that developers don't even know the FSF definition of open source any more? Or less than they used to? I'm not sure about that. But a big part of what happened to its momentum seems pretty straightforward - big software companies embraced "openness" in a way that doesn't buy into the more radical ideological commitments behind free software, but does cater to the basic desire of developers to have nice tools and be able to poke around in stuff. And meanwhile the free software folks never did a fantastic idea catering to users who aren't developers or technical hobbyists, so the external demand for their products that aren't operating systems for a server has remained pretty static.
posted by atoxyl at 12:38 AM on April 3, 2019 [1 favorite]

Strange to not see him address the issue of how free software and open source licenses work in relation to SaaS. Talk about forgetting your history.

As someone who has been around the free software and open source worlds since Linux 0.11, and participated in some significant projects since that time (but not very actively in almost a decade), I’m really failing to see the new conflict or new solutions here.
posted by jimw at 12:46 AM on April 3, 2019 [6 favorites]

Meh, I feel like I read this exact same argument about 15-20 years ago during the big BSD/Public Domain vs. GNU licensing arguments. A lot of those arguments are being revisited with the issue of public access to research results today. I remember "Open Source" as a term for marketing non-copyleft licenses, which were partially adopted by people who wanted to make their work available but didn't want the burden of enforcement implied by GNU copyleft.
posted by GenderNullPointerException at 6:33 AM on April 3, 2019 [1 favorite]