Des nouvelles de présentation de Master 2 sur l'IA générative

One dev thought it would be a good idea to put everything into the cloud without estimating the bill :-D

Pain in C++

• tools and compiler/platform differences
• ergonomics and (thread) safety
• community

Why Rust?

• fun & cool: better for a hobby project. It is a need.
• Great tooling
• Options are easier to use than C++'s pointers
• use system over #include
• simpler dependency management
• killer feature is Send and Sync, statically enforcing rules around threading.

Ship of Theseus strategy for the rewrite: component by component. The fish app should work exactly the same.

So if you are trying to draw any conclusions from this, consider the context: A group of people working on a thing in their free time, diverting some effort to work on something else, and deciding that after the work is finished it actually isn’t.

Gripes with Rust

• portability between OS: it allows to miss systems and ignoring version differences
• string translation and localization: format! are checked at compile-time
• building to other targets: "it is often better to use if cfg!(...) instead of #[cfg(...)] because code behind the latter is eliminated very early, so it may be entirely wrong and only shows up when building on the affected system.

They also report mistakes they made.

There are good with quick-wins of the port to Rust. There is also some sad ones: CMake is not removed yet. Cargo is missing some features to install third dependencies (.fish scripts, 130 pages of documentation, the web-config tool and the man page generator).
Cygwin is not a supported platform.

Un retour d'expérience sur la réécriture d'un projet de Java en Rust. Les mesures montrent de meilleures performances partout, et une difficulté supplémentaire par rapport à l'écriture du code en Java.

Il a fallu remettre certains systèmes d'équerre pour que les futures mises à jour soient plus rapides.

"Faire bien" du premier coup, c'est mission impossible avant d'avoir fait le tour du métier assez longtemps.

C'était soit ça, soit construire sur des sables mouvants.

Le besoin de simplicité pour prendre des billets de train n'a jamais été aussi grand.

Someone proudly posted in Astro Discord how they figured out how to feed Astro docs to their LLM, then proceeded to show its instructions to "get started with Astro"... which were not just worse than our recommended, but WRONG. Like, you'd FAIL. And I saw it in a heart beat.

Never been happier to be an old woman, "just working on docs", who will never be taken seriously in this industry anyway so I can yell that your LLMs are shit, and I don't have to care what the sensitive tech bros think.

The cultural heritage sector needs as many small, focused tools as it can produce. It needs them in the long-term to finally reach the goal of a common infrastructure that can be employed sector-wide. It needs them in the short-term to develop the skill and the practice required to make those tools successful. We need to learn how to scope the purpose of and our expectations of any single tool so that we can be generous of, and learn from, the inevitable missteps and false starts that will occur along the way.

As of this writing, these four tools might be more SFO Museum specific than not.

(in the context of big tech companies)

the priority of a project is to ship!

But it’s really important that one person on the project has an end-to-end understanding of the whole thing: how it hangs together technically, and what product or business purpose it serves.

You only know you’ve shipped when your company’s leadership acknowledge you’ve shipped.

you have to get clear on what the company is looking to get out of the project. [...] Align your work and communication accordingly!

Second, no matter the project goal, your leadership team will always have basically zero technical context about the project. They will rely on you for estimates, to answer technical questions, and to anticipate technical problems. Maintaining that trust should be your top priority.

How?

• track record of having shipped in the past.
• project confidence
• project competence
• communicate professionally and concisely. Share updates.

Then getting to production! Often a key detail is missing. Sometimes the user documents are stored in memcached and are MB large, or the data stored are unexpectedly sensitive legally sensible.

Can we ship right now?

Bring up the feature to as many eyes as possible!

If you want to ship, you need to do the exact opposite: you need to deploy as much as you can as early as possible, and you need to do the scariest changes as early as you can possibly do them.

J'ouvre l'espace "dépôt de documents" sur le site de ma mutuelle, et je vois un Thumbs.db.
Purin encore un machin géré sous Windows. Non seulement géré sous Windows, mais accédé via cette saleté d'explorateur de fichiers Windows :facepalm:
J'imagine les employés qui viennent double-cliquer sur les fichiers déposés par les adhérents😬
Les probables problèmes de droits entre répertoires, les fausses manipulation (entre adhérents) trop vites arrivées quand on manipule de fichiers à la souris, etc

There are now a whole new class of potential issues, design considerations, and challenges that Rust introduces to web developers which I think is often under-estimated.

Fixing one ownership issue often just shifts the problem to another place, and without a strong understanding of this mental model, it can be very difficult to identify which path from any given node is actually the "right" one, which gets closer to the real solution.

The documentation is awesome, but some crates are under-documented.

Editing Rust is not so much supported in VSCode than Typescript.

The feedback comes from a typescript developer.

Human feedbacks

Never mind that you can also harvest code from any of your shelved projects. I mean why rewrite the Fisher–Yates shuffle algorithm if you have it already in a shelved game? Code for switching the monitor depth (on those early Macs) I moved from game to game… Polygon-point collision code, a sine lookup-table for quick trig functions, a dot-product routine, cross-product routines…

was the whole exercise of my guerilla programming technique a wash? Maybe. But it always served me well

There were times too when a coworker might have said, “You should have used a Bloom Filter” and I was able to come back with, “Yeah, already tried that but the typical data we are seeing is so small that the performance gains were negligible and added unnecessary complexity to the code base so I tossed it.” Boom!

A.B.I Always Be Iterating.

How the author evolved about getting news and updates.
First from social media, then from a curated list on Reddit.
To multiple RSS readers (Feeder, Miniflux) to Yarr.

They share their feeds in the post too :)

A feeback about starting a small business. It is full of insights and experiences.