News for hackers. Stuff that matters.
Re:Firefox is on a slippery slope by bobajeff on Saturday December 16, 2017 @04:16PM (#15942165)
This isn't just a problem with Web standards. Think of how few complete implementations of C, C++, Python, Java there are.

So yeah the web is complex but so are most popular runtimes.

Re:Firefox is on a slippery slope by corndoge on Saturday December 16, 2017 @04:23PM (#15942202)
The difference is that the web was originally designed to show documents, not build applications. The failings of the web are entirely a result of turning the browser into a VM for Javascript applivations.
Re:Firefox is on a slippery slope by bobajeff on Saturday December 16, 2017 @04:58PM (#15942397)
I like the idea of having the web just being a static document format with hyperlinks and embeddable objects. No DOM just a tag for some narrowly defined supported formats (eg. .form, .game, .social, .store)
Re:Firefox is on a slippery slope by userbinator on Sunday December 17, 2017 @12:23AM (#15944298)
Think of how few complete implementations of C, C++, Python, Java there are.

C++ is at least an order of magnitude more complex, but there have been plenty of C implementations, some even entirely the work of a single person[1]. "I wrote my own C (subset) compiler" seems to be a reasonably common thing on HN too.[2][3][4][5][6][7]

On the other hand, I haven't see very many "truly independent" webpage rendering engine implementations (e.g. HTML4 or HTML5-subset, CSS2.x), so if anyone wants to give it a go (or Go, if you like...), they are more than welcome to, if only to increase the diversity of available implementations --- something that could probably handle HTML4/CSS2 might not be all that difficult, and especially so if you don't care for 100% identical results to the mainstream browsers (which often differ slightly too.)

[1] https://en.wikipedia.org/wiki/Tiny_C_Compiler

[2] https://news.ycombinator.com/item?id=13914137

[3] https://news.ycombinator.com/item?id=15463738

[4] https://news.ycombinator.com/item?id=13778353

[5] https://news.ycombinator.com/item?id=8558822

[6] https://news.ycombinator.com/item?id=11903674

[7] https://news.ycombinator.com/item?id=9125912

Re:Firefox is on a slippery slope by pcwalton on Monday December 18, 2017 @10:38PM (#15957536)
> something that could probably handle HTML4/CSS2 might not be all that difficult

As someone who did a lot of the work to implement a from-scratch HTML4/CSS2 engine, I struggle to come up with words that would adequately express to you just how much you are underestimating the difficulty of this.

Re:Firefox is on a slippery slope by mattgreenrocks on Saturday December 16, 2017 @05:07PM (#15942448)
Size is not a requirement of a programming language: look at Lisp, Smalltalk, Lua, etc to see simple systems that can be fully comprehended and are quite powerful.

If anything, the monstrosities produced by committees are less powerful and less beautiful. (That we put up with them says more about us.)

Re:Firefox is on a slippery slope by wbl on Saturday December 16, 2017 @08:01PM (#15943280)
Have you seen the Common Lisp specification?
Re:Firefox is on a slippery slope by flavio81 on Sunday December 17, 2017 @12:04AM (#15944243)
>Have you seen the Common Lisp specification?

Yes, it is huge and gigantic, but worth it's weight in gold. The stuff "just works" and has almost no limits. I'd say that spec is a masterpiece.

Re:Firefox is on a slippery slope by erikj on Saturday December 16, 2017 @09:08PM (#15943551)
Common Lisp is big because it's a bridge between MacLisp and InterLisp communities of the 80s. There are much smaller (and just as powerful) Lisps like Scheme or ISLISP.
Re:Firefox is on a slippery slope by kazinator on Sunday December 17, 2017 @11:33AM (#15946368)
Scheme is poorly expressive for daily programming and it's because of its small size. The documentation quantity for a useful Scheme like Racket is much larger.

Also, do we count all the pages expended on SRFI's? Or not?