Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Gosub – An open-source browser engine

github.com

Post reply on HN
Print view View original

161–170 of 173 posts

Re: Gosub – An open-source browser engine

#161

Post by PaulDavisThe1st »

xigoi wrote: post #157
PaulDavisThe1st wrote: post #110

Earlier quoted context omitted.

Try replacing "should" with "wouldn't it make more of a positive change in the world if" and see how that leaves you feeling.

Wouldn't it make more of a positive change in the world if you went and did something useful instead of posting snarky comments on Hacker News?

It isn't snark.

I've been writing an open source DAW for 25 or more years. Every time I see somebody else working on their own new open source DAW or DAW-adjacent project, I find myself wondering whether it would be a better or worse thing if they put their efforts into our already-existing project.

Re: Gosub – An open-source browser engine

#162

Post by Arcuru »

Zardoz84 wrote: post #91
dmz73 wrote: post #81

Earlier quoted context omitted.

Isn't this how generative models are supposed to be used? Instead of generating a sub-par (no pun intended) logo myself, I get a model to build one for me. It is the same as using any computer software to achieve the goal - human is providing the inputs and hardware and software are generating the output.

Instead of generating a sub-par (no pun intended) logo myself, you should PAY to someone to make a logo, instead of stealing work from other people.

I generally agree with that..for a for profit project.

Gosub, despite have 2.6k Github stars, has a grand total of 0 sponsors. They literally take in no money. This is someone's unpaid side project.

I expect a designer would be welcome to contribute a design, just like coders can contribute code.

Re: Gosub – An open-source browser engine

#163

Post by shiomiru »

rixed wrote: post #127
sharktheone wrote: post #87

Earlier quoted context omitted.

Yes, the HTML parser, probably the component that is the most spec compliant. I think the CSS parser also is. For the initial phase, I think it is more important to get something working. For a later implementation, the spec should dictate what we do and what not. But with a small team, it is just not feasible.

Isn't one of the difficultlies of writing a browser that your parser must accept lots of invalid inputs and render it like other browsers would?

HTML5 parsing is relatively easy, because behavior for tag soup is 100% specified. That is, in theory, all conformant HTML5 parsers should output the same DOM for any arbitrary byte sequence.

Layout is much harder because the CSS spec is full of UB, so you will have to do some reverse engineering there.

Re: Gosub – An open-source browser engine

#164

Post by exe34 »

PaulDavisThe1st wrote: post #161
xigoi wrote: post #157

Earlier quoted context omitted.

Wouldn't it make more of a positive change in the world if you went and did something useful instead of posting snarky comments on Hacker News?

It isn't snark. I've been writing an open source DAW for 25 or more years. Every time I see somebody else working on their own new open source DAW or DAW-adjacent project, I find myself wondering whether it would be a better or worse thing if they put their efforts into our already-existing project.

from your point of view, they're wasting their time and effort. from their point of view, they'd rather fail than do unpaid work for you.

Re: Gosub – An open-source browser engine

#165

Post by legends2k »

mxuribe wrote: post #66
sharktheone wrote: post #38

Earlier quoted context omitted.

lol GOSUB as a BASIC command, yes I can see the reference. It actually referes to the BASIC command.

Thanks for this! I didn't really use BASIC much back in the day, but I am from - ahem - an older time...and the name seemed so familiar, but quite couldn't place my finger on it. so, thanks for clarifying this! :-)

Logo? Lisp? Punched cards?

Re: Gosub – An open-source browser engine

#166

Post by IshKebab »

PaulDavisThe1st wrote: post #159
jancsika wrote: post #125

Earlier quoted context omitted.

> There are long lasting bridges made of stone, concrete and in the right climates, even wood. Not to mention rope when it suits the purpose. I'd be fine with someone building a new browser out of stone, concrete, wood, or even rope. Just don't build it out of C/C++ because those languages aren't memory safe.

Rust is a risk-reduction strategy. It is not a risk-elimination strategy. There are risk-reduction strategies one can follow with C++, which offer some fraction of the reduction that Rust does (opinions vary on the value of the fraction). Rust enforces risk-reduction strategies, which in some contexts may be of value all by itself. Language choice does not eliminate risks in software.

> opinions vary on the value of the fraction

They don't really. It's been shown by many studies that 2/3 of security bugs are memory safety errors. That's the minimum that Rust can eliminate.

It actually should eliminate more because it also has the strong type system and tree-ownership style that help reduce the chance of logic bugs unrelated to memory safety too (similar to Haskell and other very strongly typed languages).

This is probably the biggest data set:

https://security.googleblog.com/2024/09/eliminating-memory-s...

Unfortunately they don't break out non-memory safety vulnerabilities, but they've almost eliminated memory safety vulnerabilities by writing new code in Rust.

Re: Gosub – An open-source browser engine

#167

Post by IshKebab »

PaulDavisThe1st wrote: post #160
IshKebab wrote: post #129

Earlier quoted context omitted.

Nobody is saying that. But to continue your analogy, do you not think that in a world where all the bridges are made of wood it would be notable to say that a new bridge is made of steel? Of course there are impressive projects made with C. But we generally don't build large bridges out of wood anymore do we.

> do you not think that in a world where all the bridges are made of wood it would be notable to say that a new bridge is made of steel? Notable, sure. But it could turn out to be Tacoma Narrows bridge. We never did build large bridges out of wood, I think. For the right context, wood is still a preferred material for some bridges.

> We never did build large bridges out of wood, I think.

Yes we did. They just haven't survived. For example London Bridge was "various wooden bridges, AD 50 – 1176".

Re: Gosub – An open-source browser engine

#168

Post by kibibu »

qzw wrote: post #29
codetrotter wrote: post #20

Earlier quoted context omitted.

> more of an intellectual curiosity than anything which seeks to become useful I am reminded of this: “I’m doing a (free) operating system (just a hobby, won’t be big and professional like gnu) for 386(486) AT clones.” – Linus Torvalds, announcing the software that became what we now know as Linux

One can only hope we can get a browser the same way we got Linux, but as RMS was fond of pointing out, Linux’s success had a lot to do with the existence of a full suite of GNU software that could run on top of a new kernel. If only we had something like GNU with browser engines, I’d be more optimistic about the chances for success of many new entrants.

> Linux’s success had a lot to do with the existence of a full suite of GNU software that could run on top of a new kernel

I've heard this a number of times; but circa 1991 I'm not convinced. The critical GNU components to scaffold Linux adoption were:

- GCC + C runtime

- A shell (plus a bunch of small utilities)

You could make a reasonable argument that Linux was and remains a much bigger achievement, namely because GNU has never actually managed to make a usable kernel (Hurd still has no USB support, for example. I don't know if this is for technical or political reasons). On the flip side, there were plenty of compilers and shells kicking around that Linux could have bundled.

RMS makes all sorts of wild claims about how much of a distro is "GNU software" (a criterion that is never clearly defined - is any non-kernel GPL software part of GNU?) by comparing lines of code; but at the time of these claims distros tended to bundle anything and everything that you might possibly want.

Contemporaneously to all this, 386BSD had created a full freely-distributable Unix with no GNU code; and its descendants might have captured the mindshare were it not for an extended copyright battle from AT&T.

Re: Gosub – An open-source browser engine

#169

Post by PaulDavisThe1st »

IshKebab wrote: post #167
PaulDavisThe1st wrote: post #160

Earlier quoted context omitted.

> do you not think that in a world where all the bridges are made of wood it would be notable to say that a new bridge is made of steel? Notable, sure. But it could turn out to be Tacoma Narrows bridge. We never did build large bridges out of wood, I think. For the right context, wood is still a preferred material for some bridges.

> We never did build large bridges out of wood, I think. Yes we did. They just haven't survived. For example London Bridge was "various wooden bridges, AD 50 – 1176".

I guess "large" has a bit of wiggle room built in.

Re: Gosub – An open-source browser engine

#170

Post by lioeters »

p0w3n3d wrote: post #144
dang wrote: post #63

Related: I'm fed up with it, so I'm writing a browser - https://news.ycombinator.com/item?id=37608580 - Sept 2023 (477 comments)

how to introduce progress into civilisation? Just make some devoted engineer angry

So true. It's how we got rms and the GNU project.

> In 1980, Stallman and some other hackers at the AI Lab were refused access to the source code for the software of a newly installed laser printer, the Xerox 9700. ..This experience convinced Stallman of people's need to be able to freely modify the software they use.

And Linux.

> Frustrated by the licensing of Minix, which at the time limited it to educational use only, he began to work on his operating system kernel, which eventually became the Linux kernel.

Post reply on HN