[HN Gopher] Understanding Complexity Like an Engineer - The Case...
___________________________________________________________________
Understanding Complexity Like an Engineer - The Case of the
Ladybird Browser
Author : akling
Score : 145 points
Date : 2024-02-12 08:58 UTC (14 hours ago)
(HTM) web link (t-shaped.nl)
(TXT) w3m dump (t-shaped.nl)
| yogorenapan wrote:
| Looking at the screenshots, it's crazy good. I only frequent 5-6
| sites on a daily basis; will actually daily drive LadyBird if at
| least 3 work
| ofrzeta wrote:
| Related (chat with Andreas Kling about SerenityOS and Ladybird):
|
| https://news.ycombinator.com/item?id=36620450
|
| However I don't understand why many people are longing for
| Ladybird as another free browser. There are already some and
| Firefox is losing marketshare every day. So please do use it.
|
| Ok, I guess Mozilla Foundation's running the Firefox project is
| not to everyones liking, so that would be a valid reason. But not
| to throw out the baby with the bathwater.
|
| Firefox is a great browser and the only reason I sometimes use
| Chrome is that more and more sites require Chrome (e.g. Teams).
| apex_sloth wrote:
| Side note: Teams works for me in firefox on linux if I change
| the UserAgent with this add-on:
|
| https://mybrowseraddon.com/custom-useragent-string.html
|
| For these urls
|
| https://teams.microsoft.com/
| https://statics.teams.cdn.office.net/
|
| To Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36
| (KHTML, like Gecko) Chrome/85.0.4183.102 Safari/537.36
| Edg/85.0.564.51
| jwells89 wrote:
| Changing the user agent seems to make things work more often
| than not in the case of sites/apps like this. My inclination
| is that most of them don't actually have hard browser
| requirements and just haven't been tested against anything
| but Chrome and the devs (or those managing them) would prefer
| to not change that.
| jcul wrote:
| It seems to work for me on FF / linux out of the box, without
| any user agent change.
| timw4mail wrote:
| I use Firefox, but I still want more browser engine diversity.
| Blink and Webkit have a significant shared ancestry, and the
| only other major one is Gecko.
| jwells89 wrote:
| Some of the discontent comes from Firefox not being very
| interesting compared to other browsers.
|
| The various Chrome-clones are doing interesting things like
| adding toggleable vertical tabs built in (Vivaldi, Edge) or
| even vertical-only (Arc) and Arc is playing with using native
| OS widgets for UI across platforms for example.
|
| There's also functionality advantages. All the chrome-clones
| have better out of the box profile support, as does Safari as
| of the last major release or so. Safari also IMO has a better
| stock tab groups implementation than any of the other major
| browsers.
|
| Firefox can be fixed up to support these things, but you'll
| need multiple extensions and if you want to make it all look
| good (e.g. hiding the ugly sidebar header, extending sidebar
| into toolbar area, configuring Sidebery/TST/etc), a substantial
| amount of tinkering is needed. More of this should be rolled
| into the base browser and "just work".
| TheRealPomax wrote:
| Firefox isn't made, run, or owned by the Mozilla Foundation,
| though. You're thinking of the Mozilla _corporation_ , which is
| a for-profit company working on the Firefox product.
|
| The Mozilla _Foundation_ is a separate advocacy and campaigning
| non-profit (which is also why donating to the foundation means
| your money goes towards advocacy and campaigns, not Firefox)
| DiggyJohnson wrote:
| It's disingenuous to not explain the connection between these
| organizations. You're making it sound like they both happen
| to be named "Mozilla", when the reality is much more nuanced
| and interdependent, but in terms of their current funding and
| leadership, as well as their history.
| TheRealPomax wrote:
| No, it's not. Having worked at the foundation for a decade:
| to the vast majority of folks, the difference that matters
| is that "one makes Firefox, the other does
| campaigning/advocacy/outreach".
|
| The history behind _why_ there are two different entities
| is fascinating, but also doesn 't change the fact that they
| are two completely different companies with completely
| different org charts and completely different focuses,
| united under the Mozilla manifesto.
|
| (And even in the tech crowd, far too many people have no
| idea that there are two different things called Mozilla,
| and that their donations to the Foundation go to "the
| things the Foundation does", not Firefox)
| DiggyJohnson wrote:
| Honestly I didn't realize that you worked personally with
| the Mozilla Foundation, so I'll defer to you (and color
| myself surprised).
|
| My apologies, and thanks for the information.
| fifteen1506 wrote:
| Which one fired most Rust developers?
|
| The foundation or the corporation?
| TheRealPomax wrote:
| Corp. The Foundation doesn't make products.
| wheybags wrote:
| I'm aware and I hate it. I have zero interest in most of
| the crap the foundation gets up to, but I keep sending
| them money because it's the only way to "support
| Firefox". I want to support Firefox but I resent every
| cent I send them because I know they're wasting it on
| irrelevant crap.
|
| I would really love to see a mass resignation of Firefox
| engineers, who then set up a new nonprofit + ff fork, and
| I could just switch my attention and my donations over
| there.
| TheRealPomax wrote:
| You can't fight for an open web just by making a browser.
| As dire as the Firefox market share is today, it'd be so
| much worse if the Foundation didn't spend its time and
| effort making sure the folks who don't give a shit, but
| should (e.g. politicians, law makers, lobbyists, news
| outlets, etc. etc.) learn why they should give a shit
| about the open web.
|
| It sucks that you can't donate to Firefox directly, but
| calling what the foundation does "irrelevant" feels like
| you're not looking at the very real big picture of
| everyone wanting to lock down the web, all the time,
| everywhere.
| WesolyKubeczek wrote:
| Yes, but sometimes, just sometimes, I. Want. To. Support.
| The. Freaking. Browser.
|
| I realize that there's some creative bookkeeping and
| bureaucratic barriers set up so that the only entity that
| can meaningfully support Firefox financially is Google in
| the end, but I hate the fact.
|
| I'll probably donate to Andreas instead.
| TheRealPomax wrote:
| I'd much rather have the option to sponsor specific
| bugzilla issues, to be honest, I don't want to give
| Mozilla any money if it means it just pads out the CEO
| salary a bit (here's hoping whoever the new CEO will be
| takes the job in the understanding that it's not a
| million dollar position, with a sensible salary instead).
| NoboruWataya wrote:
| But the Corporation is wholly owned by the Foundation,
| isn't it? In that context, I don't see how it makes sense
| to see them as completely separate. The Corporation is
| answerable to the Foundation, the Corporation's profits
| go to the Foundation and I imagine it is similarly
| possible for the Foundation to fund the Corporation
| (unless its charter prohibits it or something).
| rgbrenner wrote:
| Yes, claiming they're separate is like saying Walmart and
| Walmart.com are separate. Except in that case, Walmart
| only owns 88% of the .com... so thats actually more
| independent than the Mozilla Corp.
| asadotzler wrote:
| Corporation profits don't go to the Foundation. That's
| not allowed. The Foundation owns the Corporation and you
| are correct that MoCo ultimately answers to MoFo which is
| their full owner, but tax law doesn't let a non-profit
| create a for profit to generate funnel commercial
| activity cash back to the non-profit as that defeats the
| purpose of having non-profits and giving them tax
| exemptions, so the Mozilla Foundation gets its money from
| fundraising while the Mozilla Corporation gets its money
| from commercial agreements. Those pools of money are
| almost entirely separate. Your overall point stands. MoCo
| ultimately answers to MoFo in the big picture.
| asadotzler wrote:
| Now you're misrepresenting things. The Foundation fully
| and wholly owns the Corporation and if the Corporation
| ever strayed from the Foundation's intentions, the
| Foundation could literally shut it down or sell it off.
| Leaving off that critical detail -- and I was an original
| MoFo and original MoCo employee that predates anyone else
| here, suggests you've got some agenda here other than
| informing folks about the corporate structures.
| TheRealPomax wrote:
| It may feel like an important detail, but not one that's
| realistically relevant? In no real world would the
| foundation ever sell the corporation off? O_o
|
| And no one at the foundation has any say over what
| happens to Firefox (certainly not during my tenure) which
| was the original point: the comment said "I guess Mozilla
| Foundation's running the Firefox project is not to
| everyones liking", which perpetuates the mistaken belief
| that the foundation has anything to do with Firefox.
| joshmarinacci wrote:
| Adding another free browser changes nothing. People don't
| choose free browsers because people don't choose. The owner of
| the platform chooses (with the notable exception of Windows
| where Google has spent billions of dollars to advertise it).
| metaxy2 wrote:
| Chrome is also very big on macOS even though it's not
| preinstalled, although yeah, that also has to do with the
| marketing. It seems hard to find good stats on this but it's
| definitely in the web developer zeitgeist. On mobile, though,
| people do pretty much just use whatever comes installed.
| DiggyJohnson wrote:
| I'm not sure if longing is the right word, and I think you may
| be confusing excitement and buzz around Serenity and Ladybird
| as longing for another free browser, when it is more accurately
| excitement and curiosity about a technically very impressive
| project.
| NoboruWataya wrote:
| I think it's more excitement about the _idea that it is
| possible_ to write a web browser from scratch, than about this
| specific browser. This is just one of the most promising
| implementations of that idea.
|
| I am very much team Firefox and think it is important to
| protect and promote it as much as possible. But I don't think
| the attention that people are giving to Ladybird would
| otherwise be dedicated to Firefox. Maybe when Ladybird gets
| more functional and stable they might become competitors but
| not right now.
|
| IMO it is exactly because building a fully featured browser
| from scratch is considered basically impossible today, that we
| so desperately need Firefox to succeed.
| andrewstuart wrote:
| >> I think it's more excitement about the idea that it is
| possible to write a web browser from scratch
|
| Depends on your definition of "a web browser".
|
| It's not possible to clone Firefox or Chrome without huge
| huge resources and time and effort.
|
| It is probably possible to make something like a web browser
| from 20 years ago.
| _heimdall wrote:
| That might be true, but if that's where we've found
| ourselves it seems like the best course of action is to
| start over with a fresh web spec focused on staying as
| simple as possible.
|
| Browsers have never had a sustainable revenue model. If
| we're at a point where building one is so complex that it
| requires a massive resource investment we're setup to
| eventually be left with only one or two highly centralized
| and controlled browsers. That defeats the whole purpose of
| the web.
| asadotzler wrote:
| maybe, but it absolutely won't happen, so why bother
| suggesting it?
| zamadatix wrote:
| I think Ladybird has already surpassed that mark. Maybe not
| in 100% of attributes, technically modern Chrome/Firefox
| don't even meet that mark, but it's also already gone well
| beyond in most of them. This including that they are doing
| everything in house from the image decoding to the font
| rendering (well, on Serenity - for other platforms like
| Linux/macOS they have QT bridge their implementation to
| existing stacks).
|
| Still a long ways to go to be a Chrome/Firefox for sure but
| that it's already beyond what you suppose might be possible
| is just the point of why it's its own kind of excitement.
| squarefoot wrote:
| > However I don't understand why many people are longing for
| Ladybird as another free browser. There are already some and
| Firefox is losing marketshare every day. So please do use it.
|
| Having more options in the FOSS world is always a good thing.
| Firefox unfortunately lost market share because the Mozilla
| Foundation does absolutely nothing to promote it, and that's
| not going to change until they find a better source of revenue
| than Google.
| intelVISA wrote:
| It's time to sunset Firefox and usher in The Next Age of libre
| browsers.
|
| Mozilla has betrayed FOSS for Google monies, at last the
| chickens come home...
| arcanemachiner wrote:
| Maybe we should wait for a viable alternative before
| sunsetting things and ushering in ages.
| dale_glass wrote:
| > However I don't understand why many people are longing for
| Ladybird as another free browser.
|
| I'm longing for a free web browser *engine*.
|
| Neither Firefox nor Chrome like being used as a component of
| something else. Chrome is used by Qt, but only by force through
| a pile of hacks.
|
| I'm really hoping a new project drops that nonsense and just
| allows rendering HTML to a texture without needing a big
| corporation to maintain a patch set to make it actually
| possible.
| jcul wrote:
| That's strange, I sometimes (once every few months) use teams
| with Firefox on Linux and it seems to work fine.
| epgui wrote:
| Very cool project, but this article isn't really about
| complexity, and complexity makes a background appearance at best.
| misterdata wrote:
| Author here, I guess I could have chosen a better title. This
| article is about diving into complexity that seems
| insurmountable at first, with an engineer's mindset.
| fifilura wrote:
| He was lucky that the fix was to disable the functionality.
|
| More likely would have been that it was non-compliant but chrome
| renders it anyway.
|
| And from there starts the work of figuring out how exactly to
| make it bug-compliant with Chrome.
| andrewstuart wrote:
| I put it to you that the web browser and the operating system are
| the two most sophisticated applications ever built.
|
| No doubt it's possible to build a simple web browser that gets
| some 50% of the job done and could serviceabley display some
| websites.
|
| However there would a very very long tail of detail and nuance
| and edge cases that would be very very hard to catch up.
|
| This is why you should use Firefox and we should never lose
| Firefox.
|
| If you don't follow it closely, the pace of new feature
| development in web browsers is stunning. There is a huge amount
| of new stuff going in constantly.
|
| I'm not knocking this project..... developers can build whatever
| they like. I'm just observing that the web browser is already the
| Pyramids of Giza or some other such gigantic human endeavor.
| kerrie wrote:
| But is it really necesary to use the such a bloated piece of
| software as a web browser? i think it would be better to
| discard the whole web and replace it with the gemini protocol.
|
| It can do 99.99% of what you use the web for with less resource
| usage. you can implement your own gemini client in 100 lines of
| code.
|
| For the remainder 0.01% of stuff, use a dedicated application.
| Dont trust it? Use it with docker or similar.
|
| As far as Im concerned all the browser bloat is worse than
| useless. Much of it is just to spy on you. Im sure these
| dedicated spying apps, you call web browsers, are deliberately
| full of security holes so they can upload your data to their
| servers
| andrewstuart wrote:
| The web browser is one of humanities greatest achievements
| and I don't mean that ironically.
|
| The more you understand about the modern web browser and how
| stunningly powerful it is, the more you should be amazed.
|
| And far from bloated, the modern web browser is trim and fast
| given the unbelievable feature range.
|
| And if you don't trust something you're free not to use it.
|
| When I see Chrome or Firefox and I deeply gladdened.
|
| Go back 25 years and any time you wanted to do something at
| the user interface in any context it was hard and glitchy and
| maybe couldn't be done at all. Want to do something in a user
| interface today? Chances are the browser can do it.
|
| Truly breathtakingly beautiful and powerful software and I
| love it.
| samatman wrote:
| Ladybird is doing an excellent job of demonstrating that the
| putative unreproducible complexity of the modern web browser is
| just FUD.
|
| I'm glad they didn't listen to all the people repeating the
| conventional wisdom that writing a modern web engine from
| scratch is impossible. Success is the best proof.
| asadotzler wrote:
| Use it for a week. It's not really a complete web browser,
| IMO, kind of like trying to navigate a massive Microsoft
| Office document repository with Open Office in 2004.
| samatman wrote:
| It isn't finished, that's true. But my bet is that at some
| point it will be.
| asadotzler wrote:
| It's been a while since I had a need to know, but Firefox is
| certainly over 20 million lines of code by now. The web
| platform has a lot of features to support, both standardized
| and proprietary, and it needs to also work when web authors
| mis-write their sites and apps. (Forgot to close that tag,
| Firefox still works, that kind of stuff.)
|
| What are the chances that a group of people come together to
| build another 20 million line browser? I'm with you in thinking
| not great. I'm also with you in supporting Firefox. That
| shouldn't be surprising as I'm one of the guys that started
| Firefox back in 2002 at Netscape when I was a member of
| staff@mozilla.org
| tantalor wrote:
| Engineers don't do stuff for fun, they do it for money.
| Retr0id wrote:
| no true scotsman
| robador wrote:
| > As an example, the recently launched website for Dialogic
| requires about 700 JavaScript-modules, and has about 19.000 lines
| of code of its own. There is however a staggering amount of
| complexity hiding beneath: just this website alone depends on
| (potentially) another 1.3 million (!) lines of code in modules it
| uses.
|
| This is fine.
|
| Seriously though, that's insane.
| AlienRobot wrote:
| The irony is that I clicked on the website and it looks like
| you could make it with wordpress...
| misterdata wrote:
| Author here - this figure is insane but also a bit of an
| exaggeration. I calculated it by running 'cloc' on the source
| code directory, not excluding node_modules. Many modules in
| there are actually unused or only used at build time. The built
| output (from Nuxt) that is actually run is much more compact
| (though partly minified so line counts are not that
| interesting). Also, Nuxt specifically is a bit overkill here,
| but greatly improved my productivity developing this website
| (and was chosen because it aligns with our other web
| applications, so less 'innovation tokens' [1]).
|
| [1] https://boringtechnology.club
___________________________________________________________________
(page generated 2024-02-12 23:01 UTC)