I consider writing software a kind of craftsmanship.
Uncategorized
8
Posts
5
Posters
0
Views
-
I consider writing software a kind of craftsmanship. It's important to me to think through what I build, to consciously make design decisions based on knowledge and the existing problem space. To understand why things work (or not!) and be able to solve problems as they come up. To keep things readable, understandable, maintainable.
In the end, I want to be proud of what *I* build.
-
I consider writing software a kind of craftsmanship. It's important to me to think through what I build, to consciously make design decisions based on knowledge and the existing problem space. To understand why things work (or not!) and be able to solve problems as they come up. To keep things readable, understandable, maintainable.
In the end, I want to be proud of what *I* build.
@foosel
The amount I say "Just because it works doesn't mean it's right."Functionality, maintainability, efficiency, elegance, scalability, portability, development cost, development time.
Looks like the MBA crowd has noticed they can use tools to affect one or two variables, and since they don't understand the others, those must not matter.
We'll be cleaning this mess up for decades.
-
@foosel
The amount I say "Just because it works doesn't mean it's right."Functionality, maintainability, efficiency, elegance, scalability, portability, development cost, development time.
Looks like the MBA crowd has noticed they can use tools to affect one or two variables, and since they don't understand the others, those must not matter.
We'll be cleaning this mess up for decades.
@PhilSalkie That's my worry too. And while I'm pretty sure I have (and will keep) the necessary skillset for that cleanup, I DO NOT WANT TO CLEAN UP THIS MESS 🤮
-
undefined valhalla@social.gl-como.it shared this topic on
-
@foosel
The amount I say "Just because it works doesn't mean it's right."Functionality, maintainability, efficiency, elegance, scalability, portability, development cost, development time.
Looks like the MBA crowd has noticed they can use tools to affect one or two variables, and since they don't understand the others, those must not matter.
We'll be cleaning this mess up for decades.
@PhilSalkie @foosel and part of maintainability is managing of complexity (the accidental type, where we as devs have control), and with gen AI this is hard (if not impossible) to control. This defers maintenance costs, which as we all know (right?) increases them, and also makes the whole system harder (sometimes impossible) to change.
This model only makes sense if the software being shipped is short-lived/throw-away, a model fit for VC-backed companies designed to be acquired.
-
@PhilSalkie @foosel and part of maintainability is managing of complexity (the accidental type, where we as devs have control), and with gen AI this is hard (if not impossible) to control. This defers maintenance costs, which as we all know (right?) increases them, and also makes the whole system harder (sometimes impossible) to change.
This model only makes sense if the software being shipped is short-lived/throw-away, a model fit for VC-backed companies designed to be acquired.
I'm sure we've all seen demos which were shoved out into production, then somebody realized the model wouldn't scale, or had all sorts of vulnerabilities and brittleness.
Welcome to the future
-
@PhilSalkie @foosel thanks, I hate it
-
I consider writing software a kind of craftsmanship. It's important to me to think through what I build, to consciously make design decisions based on knowledge and the existing problem space. To understand why things work (or not!) and be able to solve problems as they come up. To keep things readable, understandable, maintainable.
In the end, I want to be proud of what *I* build.
@foosel as a long time user of Octoprint, I’d say you’ve wildly succeeded.
-
I consider writing software a kind of craftsmanship. It's important to me to think through what I build, to consciously make design decisions based on knowledge and the existing problem space. To understand why things work (or not!) and be able to solve problems as they come up. To keep things readable, understandable, maintainable.
In the end, I want to be proud of what *I* build.
@foosel There's some of us still out here!
-
undefined oblomov@sociale.network shared this topic on
Gli ultimi otto messaggi ricevuti dalla Federazione
-
@brooke SPOILERS!!!
-
ok i'm almost at the end of fellowship
boromir just died
then off to flight sim
-
Bonjour @Neko0001 et bonne journée.
-
@aeva planet earth. all my constituent particles are loaners and shall be recycled eventually
-
@aeva My cat, most likely
-
hmm i should reboot the ol' PC into Windows and fire up Flight Simulator, it's been a while!
been busy ripping DVDs/Blu-rays lol
-
@aeva@mastodon.gamedev.place my many, many creditors.
-
RE: https://infosec.exchange/@sophieschmieg/116218879218610494
Yay test vectors!
I will write properly about this, but we are going pretty far to test ML-DSA *and make it easy to test,* so I am hopeful ML-DSA bugs will be rare compared to classical [EC|Ed]DSA bugs.
These test gaps were identified by writing multiple alternative ML-DSA implementations and mutation testing *those* to find missing vectors to then bring back to the Go implementation, and share on Wycheproof.
