Notices by Lars Wirzenius (liw@toot.liw.fi)

@djm While I agree that the Postel / RFC robustness principle, especially as widely understood today, is largely the wrong thing to do in protocols in 2023, things were different in the early Internet and early web. Then the problem was not "how do I do this networking thing really well", but "how do I and others do this at all so that it gets wide use?". The principle mattered more back then. Without it, we might not have the Internet at all today. We might only have closed gardens.

@davidrevoy Related. https://www.youtube.com/watch?v=AbSehcT19u0

"OpenPGP for Application Developers" is now public. I did some proof reading on that, so I'm biased, but I think it's a neat and clear intro to OpenPGP . It's more on the conceptual level than nitty-gritty bit level formats or details of the cryptographic algorithms, but it should help developers using OpenPGP understand what's happening. The site is not quite complete, but a very good start and you can help make it better.

https://openpgp.dev/
https://fosstodon.org/@hko/111573842808340744

@joeyh Thank you for the recommendation. I hadn't heard of Phanby before, but I've been trying it today, and it's overall a much nicer user experience than the native UI of Mastodon. I've not had any big problems so far, but it'll take a while to find how to do all the things I'm used to doing with Mastodon.

Overall, with half a day's experience, five out of five stars. It's easy enough to try, for anyone curious.

https://phanpy.social/

If you found a piece of paper, what writing on it would convince you it was dropped by a time traveler?

I blogged about why #Debian is the way it is.

https://blog.liw.fi/posts/2023/debian-reasons/

I've started a new job earlier this month. I now work on and for #Radicle , a distributed git hosting system and peer-to-peer code collaboration system. I'll be working on continuous integration support.

It's open source and written in Rust, and is making git be #distributed again.

https://radicle.xyz/

@neil @pwaring I notice that for many people, an important reason to have a blog is to earn money. Medium and Substack are based on this idea, and at least they are not surveillance based advertising.

I can sympathize with people wanting to earn some or all of their living by writing on the Internet.

As a read, I do not like the experience of having to click away popups to be able to read.

@n8 @mlinksva Of course, it is true that universities have people who study software engineering, so in theory they have this knowledge in-house.

@n8 @mlinksva I heard many times over the years that research software tends to "not up to professional software engineering standards". Not having a requirements document is part of that.

In my experience, in the software industry, very little software of any kind has a requirements doc, at least not a useful one, never mind anything else that one might expect of quality software.

I'd like to sit down with researchers, specifically, to discuses how to improve this, some day.

Happy birthday, #Debian

On this day in 1993, that's 30 years ago, Ian Murdock announced the imminent release of a new Linux distribution.

https://wiki.debian.org/DebianHistory?action=AttachFile&do=get&target=Debian-announcement-1993.txt

@jesse As individual tracks or as ISO images so that you can preserve inter-track pauses?

Just asking.

On this day, in 1991, Linus Torvalds first announced Linux to the world, in the comp.os.minix Usenet group. I had written a tiny bit of code in that code: the string formatting parts of the kernel logging function printk.

Today, that code (with changes and improvements by other people) runs on billions of devices, on all continents, on all oceans, in orbit, and on Mars. Possibly other places.

https://www.cs.cmu.edu/~awb/linux.history.html

@neil The first Linux installation on my computer was the first Linux installation, but I didn't do it myself.

Later, the first installation I did myself was with the boot/root floppy that LInus used to make.

After that, MCC, SLS, Slackware, and Debian. I'm still on Debian.

https://lwn.net/Articles/928581/ has more of the story.

"Open source" has an actual definition. The definition is important. If you relax or ignore the definition, the term becomes meaningless.

A lot of people call things "open source" that aren't. Some are just lazy. When corporations do it, especially very large tech corporations like Meta, it's an attempt to make the term "open source" meaningless.

@clacke @joeyh @interpeer Yup. https://joeyh.name/blog/entry/become_ungoogleable/

@joeyh the commenters on the orange site really don't like you combating Google, do they?

@bagder Some of us learn slowly.

This realization was triggered a "workflow tool" for a programming language where running a program builds it before running it, but doesn't actually write the built binary to a file. Building the program does that.

Foot gun: if one has built the binary earlier and has a binary lying around, running the program with the workflow tool doesn't update the binary. Result: head-scratching when running the binary directly gives different results.

It turns out that after programming for 39 years, I've come to the conclusion that programming tools that contain foot guns are not fun.