Programming

21903 readers

392 users here now

Welcome to the main community in programming.dev! Feel free to post anything relating to programming here!

Cross posting is strongly encouraged in the instance. If you feel your post or another person's post makes sense in another community cross post into it.

Hope you enjoy the instance!

Rules

Follow the programming.dev instance rules
Keep content related to programming in some way
If you're posting long videos try to add in some form of tldr for those who don't want to watch videos

Wormhole

Follow the wormhole through a path of communities !webdev@programming.dev

founded 2 years ago

MODERATORS

snowe@programming.dev

Ategon@programming.dev

MaungaHikoi@lemmy.nz

UlrikHD@programming.dev

Old Code Gets Younger Every Year. The threat of decaying technology looms, by Marianne Bellotti (medium.com)

submitted 2 days ago* (last edited 2 days ago) by HaraldvonBlauzahn@feddit.org to c/programming@programming.dev

4 comments fedilink hide all child comments

Archive link: https://archive.ph/A7LI4

Marianne Belotti has worked at large institutions with modernizing decades-old code bases. She is author of the book "Kill it with Fire" [review].

From that book's author bio:

Marianne Bellotti has worked as a software engineer for over 15 years. She built data infrastructure for the United Nations to help humanitarian organizations share crisis data worldwide and tackled some of the oldest and most complicated computer systems in the world as part of United States Digital Service. At Auth0 she ran Platform Services, a portfolio that included shared services, untrusted code execution, and developer tools. Currently she runs Identity and Access Control at Rebellion Defense. She can be found on most social networks under the handle bellmar.

top 4 comments

sorted by: hot top controversial new old

[–] HaraldvonBlauzahn@feddit.org 12 points 2 days ago* (last edited 2 days ago) (1 children)

And this is also a potentially very dangerous difference between AI generated code and legacy code written by humans:

Legacy code might look messy and incomprehensible at first, but any part of it very probably was written for real reasons and in logical steps, and therefore is intelligible (though with effort).

On the other hands, AI generated code will look plausible, especially to the person who wrote the prompt, but in fact it was never understood by anyone.

Yet to the engineering manager or laymen, both look the same.

And this difference is going to bite much harder if the code runs for some longer time and is used in critical infrastructure systems.

[–] Solumbran@lemmy.world 2 points 2 days ago

Messy code by humans generally isn't logical because it went through multiple modifications/deletions that are not visible anymore.

I'm not sure that bad devs are much better than AIs, and they're the ones using them to code now. I think the problem stems from not giving a shit about code making sense, and AIs just increase the problem.

I mean, I saw code that was essentially saying things like "if True == False" and when asking about what the purpose of the block was to the original dev (who wasn't even such a bad dev), they just had no idea. Humans are chaotic and nonsensical and messy code comes from that, not from a weird logic, and you can't comprehend it most of the times because it's a mix of confusion, lack of planning, mistakes and laziness.

[–] HaraldvonBlauzahn@feddit.org 4 points 2 days ago* (last edited 2 days ago)

I can resonate with her view on excessive, deep dependencies and breaking transitions like Python 2 -> 3 as a larger threath than old COBOL systems. As e.g. Konrad Hinsen wrote, such breaking transitions are a problem for long-running institutions and endeavours such as science. And this includes computing infrastructure itself - for example, The Python Wiki is a MoinMoin Wiki , and this wiki software is still implemented in Python 2.7 since its authors did not have the resources to port it!

I note that such breakages affect my own choice of tools. For example, a while ago I started a personal project, a kind of a mini-database which is designed to quickly capture input, and allows to export and analyze data with graphs. The first version was in Python, but because I anticipate to use it for many years, I rewrote the second version in Guile.

[–] HaraldvonBlauzahn@feddit.org 1 points 2 days ago

Oh, and in my opinoin, the huge reliance on libraries and large dependency chains is also a potential large problem for infrastructure written in Rust.