Yoan
Andrew NesbittWhat happens when a large open source project dies?
LisPi@andrewnez > You could measure it: look at the fork and dependency graphs of dead projects over time, count how many new projects cite a dead dependency, compare the half-life of a whale fall in npm versus crates versus rubygems.
That's more a measurement of neglectful maintainers than anything else though.
That's more a measurement of neglectful maintainers than anything else though.
I Quit Smoking@andrewnez good topic, great article
Troed Sångberg@andrewnez Wonderful writeup, many thanks! Really enjoyed the read.
Torsday@andrewnez Indeed a very good read. I took a long break from keeping up with technology news and focused on family and a more semi-digital career until recently, when I started keeping up with the FOSS evolution and the Fediverse. This resonates with me, I feel like I am diving into a combination of evolution, archeology and scavengehunting.
Thomas Arildsen@andrewnez very compelling framing of the life cycle of open source projects. Thanks for this perspective. This is an interesting read.
Sassinake!the 2010 were the years I went back to school in IT, and by 2019 I realized something was going terribly wrong... and soon after burned out.
we made the world or work less physically demanding, but made it complex beyond our understanding, until every project breaks apart like the Tower of Babel.
Kynx@Sassinake @andrewnez whale fall is a great analogy. But what of the ones that somehow stay afloat? From the PHP world I’m thinking of projects like drupal - don’t use it, but it’s been around for yonks and last time I checked the code looked very well cared for. Or the small ones that are essential and still mostly one person’s work. Is there any common denominator?
this article is about open source projects that die.
@Sassinake @andrewnez um yes. I did read it thanks.
sTiKyt 
@andrewnez
Interesting, let's hope same happens to MinIO, since the project got archived and it was the best option
Interesting, let's hope same happens to MinIO, since the project got archived and it was the best option
Jennifer Moore 😷Interesting, thanks!
John Ulrik@andrewnez Excellent analogy, excellent writing. Thank you.
PS. Same with trees: Left alone, it takes a forest ecosystem hundreds of years to fully decompose and recycle an old fallen tree, in many stages.
Paco Hope@andrewnez This is really insightful. I love the metaphor.
One thing that has changed, since some of the examples you listed, is a bit of a diversification of code repositories. Even though GitHub still has an overwhelming dominance, significant projects are moving to self hosted or codeberg, etc.
I can’t decide whether this is worth acknowledging in the metaphor. Like the repository services are oceans and a whale fall in one might not nourish projects in another. Or a different way to think of it is that how easy it is to fork/collaborate and submit PRs has an impact on how easy it is for the ecosystem to feed on the carcass.
The more I think about it the less important it seems. I think maybe maybe my idea is an irrelevant detail. The fact that someone can pick up the dead project and host its code anywhere is probably sufficient.
This is how I can tell it’s such a good metaphor, though. It provokes lines of thinking and reasoning by exploring the metaphor and mapping back to the real world.
I think this blog post is going to resonate with me a while. Thanks!
clew“The file formats we commit live after us; Runtime optimizations go to the grave.”
The conclusion made me wonder how many other industries went through similar periods of diversification and consolidation. You need interoperability to be important for the equivalent of an API … Bolts? Recording media? Bicycles? Canning jars?
Vlad-Stefan Harbuz@clew @andrewnez Industrial epistemology studies just what the body of knowledge of engineering is, and what processes were used to reach it. Engineering standards are in many ways similar to Open Source standards/protocols/etc.
(1) https://www.tandfonline.com/doi/full/10.1080/19378629.2022.2124025#d1e75
(2) https://link.springer.com/chapter/10.1007/978-3-319-16172-3_9
Not just the proper name but *footnotes*! thank you.
Aral Balkan@andrewnez Interesting read, thank you.
I wonder what things would look like if we had universal basic income and actually taxed corporations to the point where a trillion-dollar corporation was an impossibility. (Hey, a girl can dream, can’t she?)