Post by @stefano@bsd.cafe

Stefano Marinelli
@stefano @bsd.cafe

Over the past few days, I've spent quite a lot of time revisiting some tools I wrote for Linux many years ago. Some of them have been running in production for well over a decade. They work well, I've kept improving them over the years, and I think they could be useful to other people too.

So I started cleaning up the comments (in Italian, mostly), reviewing a few things, thinking about making them public. And then I stopped.

I'm not sure I want to release them anymore.

It's not about the work involved. I enjoy writing things, and I enjoy sharing them. I've been doing both for almost thirty years. But lately, with some of my projects, I've found myself spending more time explaining why I chose to do things in a certain way than discussing the software itself.

Why didn't I use this? Why didn't I follow that convention? Why isn't the development history publicly visible? Why not simply do what everyone else does?

And, to be clear, I have absolutely nothing against these questions. Asking questions is perfectly legitimate. What bothers me is the assumption that, if someone does things differently, there must be something wrong with their approach. Sometimes even something suspicious.

Last Saturday, during my talk at Linux Day, I mentioned something that I consider fundamental: Open Source is about freedom. Not about replacing one set of imposed conventions with another.

Thirty years ago, people asked me why I didn't just use Windows. Today, when something doesn't compile on a BSD, someone inevitably asks why I don't just use Linux. Which is rather funny, considering that I use Linux extensively and have done so since 1996.

But that's not the point, is it?

I don't dislike conventions. I don't dislike Git, public development or any of the practices that have become common over the years. I use many of them myself. I simply don't think they should become requirements for being considered a legitimate member of the Open Source community.

I have always disliked monopolies, including monopolies of thought. And I find it rather sad to see the same pressure to conform appearing in a world that attracted me precisely because it offered an alternative.

Perhaps I'm just getting a little tired. Not of developing things, but of having to justify my choices every time they differ from what people expect.

So, for now, I think those tools will remain where they've been for all these years: running on my servers, doing their job. Maybe I'll change my mind. I hope so, actually.

But right now, I don't feel like going through all of that again.

And that's a pity, because I still believe that sharing what we create is one of the best things about Open Source.

#OpenSource #BeFree

106 likes
Refresh interactions
36 boosts
Replying to @stefano@bsd.cafe
MaddieM4
@MaddieM4 @raphus.social

@stefano I think that's entirely valid.

These days, I put a lot of work into explaining why I do things, not really to silence critics, but because I want people to be able to easily evaluate if my software fits how someone wants their tools to work, not just "does it do a task." I also use my software as case studies of patterns I think are healthier than the mainstream, or experiments worthy of exploring.

Both of these are *also* reactions to thought monoculture as a problem. And they're a higher effort than a lot of people have the spoons to bother with, because thought monoculture is exhausting and oppressive.

Stefano Marinelli
@stefano @bsd.cafe

@MaddieM4 And I think that's an important distinction. I actually enjoy explaining why I do things differently. Sometimes the reasoning behind a particular choice is more interesting than the software itself, and discussing different approaches is one of the things I've always loved about Open Source.

What I find exhausting is having to *defend* those choices, as if doing something differently were a mistake that needs to be justified.

I'm perfectly happy when someone questions my decisions. I've learned a lot from those discussions over the years. But there's a difference between curiosity and the expectation that everyone should eventually arrive at the same conclusions.

And yes, that takes energy. Sometimes more than writing the software in the first place.

Replying to @stefano@bsd.cafe
Tomáš Odehnal
@tomas @odehnal.online

@stefano Thanks for sharing the thoughts, Stefano. You know, that's actually the reason why I have notifications on for your profile, to not miss any of your posts. Because you do things differently.

Replying to @tomas@odehnal.online
Stefano Marinelli
@stefano @bsd.cafe

@tomas thank you, Tomáš!

Replying to @stefano@bsd.cafe
François 🇺🇦
@FrankauLux @mastodon.social

@stefano
Bien faire et laisser dire... :-)

Replying to @stefano@bsd.cafe
vi ⁂
@vgarzareyna @mstdn.mx

@stefano
> What bothers me is the assumption that, if someone does things differently, there must be something wrong with their approach.

I like to think that if someone did something different, they probably knew something I didn't

Replying to @vgarzareyna@mstdn.mx
Stefano Marinelli
@stefano @bsd.cafe

@vgarzareyna sometimes it’s just a way to avoid monoculture

Replying to @stefano@bsd.cafe
Andrew Bartlett
@abartlet @mastodon.nzoss.nz

@stefano Yeah, there is a longstanding issue with the parasocial relationship between users and maintainers.

The demands of users are ever increasing, even before the current strong feelings on AI or LLM use, we had security requirements and "best practices".

They days of users accepting the code "as is, with no warranty" have long gone.

Replying to @stefano@bsd.cafe
Raster
@raster @retro-gaiden.com

@stefano La portabilità è quella cosa che si è sempre sottovalutata, ma poi si piange quando manca.
Una volta era per Linux che non aveva ports di roba comune su windows, ora che Linux è travolto dalle LLM si piange che i BSD non sembrano allo stesso livello, peccato che anche tanto software per Linux è diventato troppo integrato con estensioni uniche che portarli è un lavoro complicato.
E' una lezione che non si va mai ad imparare, e poi ci cadono sempre.

Replying to @stefano@bsd.cafe
argv[-1]
@argv_minus_one @mastodon.sdf.org

@stefano

Today, when you write code with your own hands and keyboard, someone inevitably wonders why you aren't using AI. And those who wish to avoid using AI-written code are finding that there are suddenly very few programs to choose from.

So yeah, following the herd is not always best.

Replying to @stefano@bsd.cafe

@stefano@bsd.cafe Just to add another angle. Sharing via public git repos will be subject to AI scrapping.

These things soak up this information, and spit it out. Well, from what I have seen, the standard from AI is not that great, often wrong, and copious in nature. It doesn't encourage me at all, and it leads me to believe that these so-called programmers who published this stuff in the first place (and that AI is subsequently feeding from) don't actually know what they are doing themselves.

A lot of the work I have done can't be published either because it is company confidential, or quite literally - a state secret! So, I've never got into the habit. Even with the code written now I often don't bother, because of a deep mistrust of large corporations.

Why should they have my ideas, my work - sweat and tears, aching fingers hands, without any credit or recompense?

Adding AI to the equation is like adding salt in my wounds, or my face in the dirt.

So, NO!

I have the council of the few people I actually trust - over 40 years worth!

Their guidance and criticism is enough for me.

Replying to @stefano@bsd.cafe
josemanuel
@josemanuel @social.sdfeu.org

@stefano Sometimes I ask those kinds of questions, too. But I'm not trying to be judgemental, just genuinely curious. Why did someone do things this way? Do they think it's better? Does it make them feel more comfortable? What can I learn from this? Is there a better way I haven't thought about?

Stefano Marinelli
@stefano @bsd.cafe

@josemanuel Asking is perfectly fine. It's a sign of interest, and it makes me happy.
A little less interesting is when people start with the "why aren't you...? It's suspicious?"

Suspicious of...what? You have the code. If you can't read it if you don't see all the commits, it's your problem, not mine 😆

Replying to @stefano@bsd.cafe
benz
@bentsukun @mastodon.bsd.cafe

@stefano The breathless demands of the purity police demanding proof that no LLMs were involved in the creation of your just-released software 🫠