@downey @dcharles525 In their messaging, I don't see Purism standing against that idea. They just want to express it differently.

I'm not saying their idea will succeed, but I won't villify them for exploring it.

To summarize: WebKit uses some Perl scripts (Perl seemed well suited to this task, btw) to compile a custom header file format, with help from C++ templates into code which converts the C++ datamodel into JavaScriptCore objects and back. To handle the prototype properties/methods, it constructs the hashmaps JavascriptCore indexes into at compile time.

I was a bit distracted getting through that, with more important work.

@ekaitz_zarraga I believe you've probably heard of my Odysseus web browser, so I'll just say what I'm doing next:

I'm preparing to package it for Debian, Arch, and maybe Guix. And then I'm looking to harvest unvisited links to present as personalized recommendations.

Hopefully that alongside the webfeed discovery support I'm releasing now will help people rely less on central silos! And I've got more planned that way.

@Shamar @ekaitz_zarraga Cool!

How about you write an EBNF grammar and more sample programmers for it, and I'd be interested in discussing it. Plus you'd be most of the way towards writing an interpreter.

@ekaitz_zarraga @Wolf480pl @Shamar @grainloom As for Servo, it looks like it has pretty much caught up already.

The core is quite intertangled (mostly for the sake of DOM APIs), and has proven harder than I first thought to disintangle. But I also originally missed dependencies of Servo's that I could reuse.

Always knew of WebRender though, which'll be very useful.

@ekaitz_zarraga @Wolf480pl @Shamar @grainloom From reading the code NetSurf looks pretty solid. Though I have suspicions that layout/rendering is too simplistic (1 code file each), and it's got a half-hearted JS implementation (can't ask for more there) that I don't think has to complicate the code *quite* as much as it does.

If they gave up on the JS I don't think it'd take too much to catch up, though I'm not clear how much.

And he's got a list of things he wants to be able to do without JS. discuss?

"1. Why can't we have a standard search element that filters a list on the client side (similar to how ng-repeat | filter: worked on Angular)?
2. Wouldn't a standard HTML element for drag-and-drop sorting be awesome?
3. More advanced validation functionality, like comparing equality of two different form fields.
4. The ability to do the modal/checkbox trick above without it feeling like a hack and writing weird CSS."

Another thing that comes to mind when I think of an app sandbox like that is maybe it's bytecode (having a compiler can simplify things) should be based on a purely-functional Haskell-style "GMachine".

That can be very efficient to JIT, and (given the I/O is performed using Functional-Reactive Programming) allows the order of execution to be more in control of the interpretor rather than the sandboxed program. Maybe randomising the order of execution can help thwart vulnerabilities?

I'm not personally interested in using it, but if you do want a cross-platform app framework I've got a few ideas:

The key would be to make the sandbox itself as simple as possible, in order to best secure it against any Spectre-style vulnerabilities. Which btw Google doesn't think ever can be truly fixed. That is if it can be a library running within the sandbox, or if it can be compiletime developer tool, it most definitely should be.

A couple toots got me thinking:

I really think the biggest offront overcomplicating software today is the web browser. Ultimately I think the fix is to remove JavaScript and give the browser more flexibility to present webpages however works best for it's UI devices.

But I do hear from people who would miss having The Web as a cross-platform app platform. Personally I think the web fills that role very poorly, and the only it does is because of corporate politics I think are ruining The Web.

Interesting project I've been working on in the hopes that someone I know will pay for it: https://git.nzoss.org.nz/openwork/libdggs

It's an efficient and accurate GIS library, based on new notations "DGGS" from the OGC standards body.

@fosslinux If it helps I've heard a story of someone sneaking Libre Office into school and teachers not noticing the difference.

Oh great, this again!

'An earlier protocol called DNS over TLS was already available but since it runs on port 853 and “does not look like HTTPS”, network operators that dislike DoT can easily block it. Most corporate networks will in fact do this by default.'
Hence why DoH.

https://blog.powerdns.com/2019/02/07/the-big-dns-privacy-debate-at-fosdem/

@yabirgb @ekaitz_zarraga You can do the same for WebKit and Blink if you want. In those browsers (I don't know about Gecko) the HTML is modeled exactly the way you see it from JavaScript, so no surprises there.

I am implementing my own browser engine too, but I haven't implemented it yet as many don't yet exist in Memex. And when I do I'll probably implement it a bit differently, in order to bring interaction into the layout tree rather than the document tree.

Let me lay out my what motivates me to develop software: I do it to argue for what I perceive as improvements in the tools we use. And to see if those arguments hold up.

So I absolutely do care about having others use my software, and for that to benefit them. Because I want to see if my ideas benefit you.

Finally getting around to watching https://m.youtube.com/watch?feature=youtu.be&v=3kEfedtQVOY

It's beginning with start: while your computer must be Turing complete, the communication protocols it uses should not be.

@albin From Stallman's artical:

"There are so many ways to use data to hurt people that the only safe database is the one that was never collected. Thus, instead of the EU’s approach of mainly regulating how personal data may be used (in its General Data Protection Regulation or GDPR), I propose a law to stop systems from collecting personal data."

@albin It's mentioned in this talk, I think https://idlewords.com/talks/website_obesity.htm

But definitely these links:
https://ar.al/2018/11/29/gdmr-this-one-simple-regulation-could-end-surveillance-capitalism-in-the-eu/

https://www.theguardian.com/commentisfree/2018/apr/03/facebook-abusing-data-law-privacy-big-tech-surveillance

@albin Well, Stallman, Maciej Cegłowski, later @aral , etc have talked about the concept of a General Data Minimisation Regulation (Aral's term).

I do think we should be campaigning for that. I don't know what else.

Yay! I'm one step closer to being able to support Content Blockers like @aral 's Better in Odysseus!

I just got notified that the issue to expose that functionality to browser UI developers like me has been marked fixed!

So when will this update land in the Debian repos?