We can only have true collaboration in FOSS if we stop treating it as a competition (a.k.a. "this project is shit and my project is so much better for those irrelevant technical reasons that are in fact just opinions but they matter really much because I am the best coder on this world and decided so")

@karolherbst this is why i hate modern linux sm, everyone is just like this all the fucking time whereas all my projects are built with the understanding that people with different opinions will use them

@karolherbst don't forget the profit motive... the reason RedHat Directory Server exists is because we wouldn't sell OpenLDAP to them, so they bought Netscape instead, and advertise it as the only enterprise grade LDAP software, even though their code was so decrepit they finally had to adopt OpenLDAP's networking and database libraries.

Our project *is* better. But we refused to be owned by them.

@hyc @karolherbst damn, wonder how many of those Red Hat has pulled

@ozamidas @karolherbst Statements like this are unprovable unless actually legally confidential information held by everyone involved are released to the public. I personally would take it with a grain of salt.

@karolherbst This is tone policing, aka the language of the oppressor.

Who's going to decide what technical reasons are relevant? Who's going to tell the difference between legitimate criticism and useless badmouthing?

We can only have true collaboration in FOSS if we can all operate as peers and discuss software quality in good faith, without any attempt at stifling criticism and appointing ourselves as guardians of what is allowed to be in the conversation.

@ozamidas @neal @karolherbst … says the person who used to work for Dead Rat, Inc. (according to the profile)

@neal @ozamidas @karolherbst it was 21 years ago, there's nothing privileged about the information now. https://www.cnet.com/tech/tech-industry/red-hat-acquires-aols-netscape-server-software/

@neal @ozamidas @karolherbst the landscape was different then, all of the commercial Unices were still major players. We were focused on cross platform compatibility for Solaris, HPUX, AIX, z/OS, etc. I was concerned that being owned by any particular OS vendor would damage our support for the other platforms. We suggested a partnership, much like what we had with HP, but RedHat insisted on ownership so we walked away.

@neal @ozamidas @karolherbst anyway, back to the original topic, some projects are objectively better than others, better code written by better coders. Some projects are moribund legacy code on life support, cash cows whose only value is in the support contracts with their existing customers. Others are actively developed, pushing the envelope in efficiency, scalability, etc. Some projects are just people punching in for payday, others are about passion for perfection.

@karolherbst i don't understand this post but maybe I am not into linux politics enough and this doesn't usually apply in my circles

@mirabilos @ozamidas @karolherbst So? I didn't work for Red Hat during the time this happened. I was barely a teenager then. I don't know anything confidential about that time either. The time I worked at Red Hat didn't really expose me to anyone from that time either.

@neal @ozamidas @karolherbst idk man, you sound very corpo there

@neal @ozamidas @karolherbst Former redhat directory server engineer here. It's not complete or fully true.

@mirabilos @ozamidas @karolherbst I'm a engineering professional who has been a party to M&A deals before. I know what goes on in those things. It's rarely so simple.

@ska @karolherbst

Thank you, Laurent, for expressing this better than I could have.

@star it's about tech elitism where people claim that specific projects are trash (and forced upon everybody, if they are also into conspiracies) based on the existence or lack of characteristics they claim are universally good/bad.

Like imagine somebody hates bananas for being a bit slimey, but it's not enough they don't like it, they say it's a terrible fruit and nobody should continue to eat it and it's just forced on everybody through big fruit, because it's clearly bad.

@karolherbst @star got it, so you're talking about systemd

@SRAZKVT @star I'm not talking about anything specifically it happens with all sorts of projects and I'm seeing those kind of discussions where one just opens with "project X is shit, because $arbitrary reason. Can we just do this thing that makes sense in my niche use case but it's terrible in the general case, but I still say it's superior for every use case, because I know things" and then just make it sound like it's a hard fact they talk about where it's "simply" an opinion.

@SRAZKVT @star anyway, point is, this sort of tech elitism needs to stop.

Like we can choose to collaborate on projects instead of making up some fake war just because somebody doesn't like the color.

If you want to create a competition, sure go ahead, I think that's great. But badmouthing the think you don't like based on arbitrary reasons is going too far.

@karolherbst @star a lot of those ive seen aren't arbitrary reasons, but actual bad design decisions that gets shoved aside as unimportant. but maybe we see different things

@SRAZKVT @star "bad design" is also arbitrary to be honest.

Like who decide what's bad design? It's all made up. Not to say it's not real, or there isn't experience to know what works or what not.

But I think we treat those "technical reasons" too much of a hard facts where it's "simply" just opinions on what worked best in your own experience, but that shouldn't mean we should treat our own experience as universal laws of programming.

@karolherbst I wish more FLOSS projects embraced end user research and accessibility. Debates between developers with no end user voices in the mix have been the standard for decades and I am tired.

@akareilly oh for sure. And I think that's ones reason why we are in the state we are where too many devs go like "it's obvious it has to be this way, because I'm a smart engineer".

While doing the opposite of what software engineering should be about, that is, figuring out what are the constraints in the using environment of your software, which for software engineering sometimes includes users that have no idea about anything computers :)

@SRAZKVT @star or maybe to put it into different words, we should treat programming more like art, where everybody has their own style and techniques due to experience, but we shouldn't just go around and tell others to do the thing we are doing, just because it worked out for us or because we think it creates good results.

Like sure, you can have opinions on those things, but it goes too far to start random wars and complain why everybody use the obviously sucking solution.

@karolherbst @star but software is made to be used, not just looked at, some design decisions will just hurt practicality, and those who pay for it at the end is end users

@SRAZKVT @star sure. But here is the thing, we figure out what's bad not by declaring it violates technical reasons, but because we make end user studies and they tell us what's bad and what's good.

@karolherbst @star but end users don't always know about what's best for internals, those are only really useful for ux. but bad internals can still hurt users, for example, complex data formats would hurt reimplementations, and shall the original turn bad in some way, that's just going to hurt users

@SRAZKVT @star well sure, but we don't know that until we get to it.

And we can always collectively change things if we decided to do it. Yeah, sometimes it's more work than it has to be, but often you have to do the bad thing first to learn why it's bad and which mistakes to not repeat.

We can't know those things prior trying it out.

And complexity is often necessary and a result of experience with the problem space understanding of all the pitfalls and niche use cases.

@karolherbst it’s not as clear cut as you paint it to be. GNOME and KDE cooperate all the time and it gives us nice things, but sometimes some things are shit and the way forward is to acknowledge it and compete with them. Without competition, we’d still be using sysvinit, e-s-d, CORBA…

@afranke I think the problem is, that we often have to do the thing to learn that something wasn't great and the environment and users change over time.

Which might be bad today, might have been a good enough solution in the past.

Of course we have to acknowledge that things change and we need to be open for adaption and exploration of better solutions.

I do agree that some cool projects are created out of spite, but I think I have a problem with spreading negativity around.

@karolherbst yes, it’s all in measured response.

Part of the problem is also the opposite direction though. Creators of new things get a lot of hate comments, so it’s hard to blame them when they respond aggressively.

@afranke yeah.. we are all social beings, and we all have our limits on how much negativity we can deal with before we have to vent.

It's totally fine to explore things, and it's fine if something I don't like gets adopted, because I do trust people to make an educated decision, because they probably have different priorities than I do and come to different conclusions based on their needs and their experience working in a different usage environment.

Unless his account disappeared overnight, I think Karol blocked me 🤔

I'm sure disagreeing with him makes me a destroyer of FOSS, or a Nazi sympathizer, or something.

Be wary of anyone in a position of power who cannot accept dissent and frame it as your enemy. They are only in favor of collaboration as long as it serves their own interests, not necessarily yours. It is obvious in the way e.g. Google approaches Open Source; but unfortunately this mindset does not seem limited to big corporations.

@ska clearly, not wanting to centralise control of software means you're a nazi prepper freak

@SRAZKVT Disliking what systemd has done to Linux and being vocal about it makes me somebody who should be silenced, no matter how technical I want to be about it (and especially since I want to be technical about it, because it is a little harder to ignore)

(Edit: typo)

@ska well clearly those technical points are opinions and don't matter, programming is art and you can't make bad art right ?

@SRAZKVT They've never used the art argument. So far, Lennart has just dodged every technical discussion, and Karol's position is that technical quality doesn't matter, systemd has won and we should all rally behind it. 🤦‍♂️

@ska karol used the art argument in a reply to me

@SRAZKVT omg

@ska https://chaos.social/@karolherbst/115835958264682794 here

> we should treat programming more like art, where everybody has their own style and techniques due to experience, but we shouldn't just go around and tell others to do the thing we are doing, just because it worked out for us or because we think it creates good results.

@SRAZKVT Ugh. So, yeah, there is no such thing as good engineering practice, or experience, or science. It's all just subjective.

Okay then, don't shove your shitty art down my throat, I don't want it in the software that runs my fucking car.

Talk about never wanting to take any responsibility ever.

@SRAZKVT @ska using llm is art too. /s

@pinskia @SRAZKVT I understand the sarcasm, but my point isn't that programming isn't an art (it definitely is), it's that it's also a science.

@SRAZKVT @pinskia @ska it’s all three. 50% craftmanship, 20-40% art, 30-10% science.

@karolherbst @star it’s not like every food recipe suddenly got a dependency on bananas and you have to reorganise all your kitchen shelves because the banana packages don’t recognise the difference between the small and the large bin

@ska heh I boosted this simply because I have had people do this toxic "I'm the best coder ever" bullshit trying to "compete" with projects I maintain like pkgconf.

it is toxic behavior and we shouldn't take charlatans who behave that way seriously. that of course does not mean immediately rolling over when people present bad engineering for the community to accept.

@ariadne Yeah, I know people like that too, trying to make a name for themselves through competing with other people's software.

I find them relatively easy to expose, because everyone in the community knows I'm open to, and crave, technical discussions. So to anyone interested in comparing merits and coding ability, I have a simple answer:

Bring it on.

@karolherbst @SRAZKVT ​​ ​​ ​​

@karolherbst The opening of Embedded Recipes (2024?) mentioned "coopetition" while showing logos of a whole bunch of their sponsors.

I think *healthy* 'coopetition' is good, because then you have more options for which projects you choose to use rather than just one unique project. Healthy means that they all cooperate in a way, but still kind of gently compete with each other (not the kind of thing where one says "oh this project is garbage mine is much better!!", but just existing alongside, with different feature sets and such, and users get to pick what they want without one saying "oh this project is garbage you should pick this other project instead")

I'd love to see this kind of thing more

@fun competition in e.g. sports exists because there is a market for it, not because the athlete want it, most of them would be happy either way or do it without the audience.

But a lot do it for the entertainment aspect, not because it's necessary.

We _could_ do it for entertainment in FOSS as well, but so far it looks more like it's imposed by annoying users and most of the devs don't really want this kind of attention.