Implementing the Tastenet ( ) mockup in Flutter. Flutter seems kind of good. There are going to be lots of native desktop apps made in it. Canonical has embraced it.

I'm not confident that the average flutter app will have fewer memory leaks than the average web app, I've come across an API that seems to encourage leaks already ( but the project probably has a lot more agency to address the issue than web browser maintainers do.

· · Web · 3 · 0 · 0

Why are most of the values in the default Theme null. Oh well. I'll just use my own theme provider.

@cblgh Regarding Web of Trust stuff, I think I've settled on using {length of shortest path to endorsed blocker < length of shortest endorsement path} as a blocking criterion, and {length of shortest endorsement path < limit} as an inclusion criterion (the shortness of the path to a tagger will sometimes be used to sort content, as a heuristic for quality, but it is not a very accurate heuristic, serious users will use their clique's preference aggregation thingies instead, for quality rankings)

@cblgh I think the spread activation algorithm trustnet uses makes more sense as like, a real model of the epistemology of transitive endoresement, but it seems like it'll always be harder to compute/reason about and I can't see why it would need that level of accuracy.

@faun glad that yr exploring alternate terrain! the raised eyebrows were at the *very* similar name to trustnet fwiw :^)

@cblgh Yeah, sorry about that. I'm looking for another name, but it's just... so fitting... lol. Taste is all.

Maybe I should call it tasteweb. Yeah, why not.

👋 also settling on Flutter as the main tech for an expected long-term project over here.

The only questionmark is the web target support, which seems to suffer from perfomance problems. Even though we won't be targetting the web mainly, there's some parts that will need to be web-exposed and it'd be a real shame to have to introduce JS etc to the stack just for those. It does seem like that's the way it'll shake out though :/

I've tried a bunch of production web-apps flutter link to in their various online docs, and all of them have just a hint of uncanny lag on a desktop PC :(

@s_ol If I even provide a web build there will be a banner at the top that says "get the fucking native version you fool"

In the latest Flutter Engage™ yesterday they talked about making the web target much faster btw. You should look into that.

Oh yeah I checked that, but it's basically just the old CanvasKit backend coming out of beta unfortunately :(

In my case, I would like to use it for the kind of use-case where you get a link via email or some other comm channel, where you don't necessarily know what devices it's gonna pop up on, and then you just have to "land" there and save or accept some content. So it's not like you'd stay there using the app anyway, but it should feel very very lightweight too

the CanvasKit thing is "fast" because webassembly, but it also means you're pulling in a complete binary for the 2d rendering lib Skia on page load, which isn't really ideal. also it's kind of a Frankenstein thing because all current browsers already use Skia under the hood, but you can't punch through to that from JS or WASM so it's basically a slower reimplementation of the browser itself.

also since the CanvasKit thing still struggles with relatively simple views at times from what I've seen so far, I think that maybe the actual graphics rendering isn't the bottleneck, and it's just that having a whole DOM-ish tree and all event-handling in Dart is not quite fast enough. Have absolutely no data on this though, so take this with at least a pinch of salt

Sign in to participate in the conversation

Merveilles is a community project aimed at the establishment of new ways of speaking, seeing and organizing information — A culture that seeks augmentation through the arts of engineering and design. A warm welcome to any like-minded people who feel these ideals resonate with them.