throwaway032023

MONTROSE has become less eclectic and fun. When Mary’s mural went away is approximately the down fall. Time marches on.

I remember when pypy was only 25x slower than c python.

Pypy

“that's declined so precipitously”

That sounds hyperbolic and unserious.

https://medium.com/flutter/whats-new-in-flutter-3-32-40c1086...

Flutter is doing well on Desktop. Is it going to be the same experience as writing c# on windows or swift on OSX? No. Will it work? Yes.

Note the ffi/jni gen announcement. This has been in works for a long time. FFI has been stable for awhile. The generators are new.

Why not Zappa?

https://github.com/zappa/Zappa

What is Canada’s transportation infrastructure like? USA is blessed with waterways for the middle/eastern side. Canada is massive. How can shipping not be wildly expensive?

Openresty distribution of nginx with lua scripting is game changer vs. Apache.

On your tip toes and falling down with a thud is also a qi gong exercise.

Moribund. Developer went full time with offline replication vendor. Interesting project. Devil in details.

Feels is the key to that statement. :-) It’s these “feels” that got USA to its 2024 election outcome.

> It's clear just from using the language that it was designed as a more "serious" Javascript replacement

What is your usage of Dart?

JavaScript is a target platform for Dart. The Dart interop story is the point. It’s a cohesive glue language. First class interop with JavaScript, Java/Kotlin, C++ and Swift/ObjC. Community interop supports go and rust.

Maybe I am confused. We are comparing Dart to what again? Yes, Flutter is what Dart is known for and it’s, imho, important for their to be primary customer to drive evolution of language.

I don’t do JavaScript development but anytime I bump into it it’s a cluster with regards to tooling (maybe this has changed in the past 3 years). Dart comes ootb with tooling that is a cohesive experience. The technology is impressive and I think it’s filling a niche.

Dart reminds me of Python 20 years ago. Python shipped with IDLE, glued platform (win32com), corba, java, etc features together.

Dart is doing similar stuff but first class interop from the supplier (Google/dart team) for the platforms (js/wasm, mobile, desktop). It is pretty damn amazing and ambitious (maybe too ambitious?). And pub/package system is nice - dart is statically compiled and Google is one of the main innovators in software supply chain security - I think there is real possibility of it being orders of magnitude more secure than python/javascript. (This is not the case currently)

I guess I see it as a “I can stay in this world and do most of what I need”. Are there sharp edges? Yes. But the tooling and the community can resolve this fairly quickly. I don’t need to jump between abuncha languages with tooling disparaties.

How does it compare with pottery? https://pypi.org/project/pottery/

Yeah this isn’t drive-by contributions where you shit out a PR that works for you but makes the test infra fail and you get upset. Ppl need to think big picture.

One big problem, I think, it may not be visible all of the test infrastructure if your outside google. Don’t know. Gerritt is its own world and googlers live there not in GH (it appears).

Absolutely true. It’s so much easier to upgrade dependencies and if it works / most likely works. We had a small UI regression in a very large app and that hadn’t been touched in 1 year.

This sounds bitter and misinformed. Check automotive grade Linux and check GitHub for flutter embedders. Flutter being deployed all over the place. It’s impressive. The tooling and DX is excellent.

There is a comment above that shows the founding former didn’t want to fix tests in his PR. Flutter is massive and it’s critical infrastructure. The tests are critical to up-keep. The tests are probably a majority of the work. Which makes sense for something that has such a large surface area.

I think this is an emotional reaction from someone has never managed a large project. Merging PRs and releasing them isn’t the problem. I hope they understand that.

Flutter is amazing. Web, desktop and mobile that actually work is nearly magical. Yes there are priorities. Yes the desktop isn’t as high priority of web. It does work. Hell. Collect money and fund Hixie or others to prioritize patches community feel should go in. That is constructive and methodical/reproducible.

Forking a massive code base, applying PR and not conforming to the source repo standards (tests whether that is golden or not) - just seems not well thought out.

I think community could fund a well known developer to fix desktop issues and have PR adhere to Flutter standards. It’s a ton of work. But seems like it’s a money/dev/process issue. Not a PR merging exercise.

Words are cheap.

Seems like:

- guy wants to add PR more aggressively to a code base than Flutter currently is

- isn’t clear if they have a concept of handling compatibility between fork and original

- why not be a labor organizing entity and work with the flutter team?

Some of the PR that I am aware of that are stale and could be good candidates - are difficult design decisions and stakes need to be put in the group.

It seems ill advised. This is huge amount of work with limited upside and huge downside (adding uncertainty to outsiders looking in) and tons of effort (just in infrastructure maintenance).

Edit: if I were conspiracy oriented I would think it’s to harm community. I would pay this no mind.

I think Dart is underappreciated. Almost the entire compiler is written in Dart. It has a VM, AOT compiler, FFI/Native, and first class Javascript interop with ability to compile to javascript. It has first class WASM support. Null sound safety. They have experimental support for macros. Pretty impressive.

What is a “protected class?”