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

Mozilla's open source AI strategy

blog.mozilla.org

Post reply on HN
Print view View original

201–210 of 213 posts

Re: Mozilla's open source AI strategy

#201

Post by 7bit »

Vinnl wrote: post #121
giancarlostoro wrote: post #51

Earlier quoted context omitted.

I think the problem many of us have is it feels like Mozilla invests more effort into everything other than its browser, we see the things they do outside of Mozilla, I mean it was so bad Thunderbird had to become its own foundation due to lack of funding! I would love for there to be a world where Mozilla maintains Firefox and can make for product projects that provide higher value. I also have a pipe dream of one d…

At the same time, Firefox last year gained tab groups, vertical tabs, a user-friendly profile switcher. Split view and tab notes are under development. It sometimes feels like it's moving faster than ever, and that's disregarding all AI features. (Disclosure: I work at Mozilla, but not on Firefox.)

You must be joking! The user profile switcher is NOT user friendly. It's lacking in every aspect of UX compared to Chromes profile switcher. And it has ads in it. Of all places, why the heck did they put ads into the user profile switcher.

Re: Mozilla's open source AI strategy

#203

Post by _flux »

capitainenemo wrote: post #132
mkroman wrote: post #129

Earlier quoted context omitted.

Ah yeah, there's this: https://github.com/mozilla/TTS I can't seem to find anything that mentions a Firefox integration though?

Click Reader mode on a web page, then the read aloud option in the sidebar. Note that how well it works on Linux will depend on your distro and default settings, as is common for Linux world. They do try to provide setup instructions if your linux distro has issues. ... now whether that model is integrated by default, no idea. I imagine that depends on size. Oh, and mozilla's off-line translate for private translatio…

I tried it, and I'm sure it's not a based on any language model. Sounds like espeak-ng or its ilk, so it probably uses whatever's available?

The TTS repo was updated 5 years ago and links to installation instructions for Ubuntu are broken :/. Also the docker repo is archived, but only recently, so maybe it works.

All in all, doesn't look like the project is alive at all.

Re: Mozilla's open source AI strategy

#204

Post by capitainenemo »

_flux wrote: post #203
capitainenemo wrote: post #132

Earlier quoted context omitted.

Click Reader mode on a web page, then the read aloud option in the sidebar. Note that how well it works on Linux will depend on your distro and default settings, as is common for Linux world. They do try to provide setup instructions if your linux distro has issues. ... now whether that model is integrated by default, no idea. I imagine that depends on size. Oh, and mozilla's off-line translate for private translatio…

I tried it, and I'm sure it's not a based on any language model. Sounds like espeak-ng or its ilk, so it probably uses whatever's available? The TTS repo was updated 5 years ago and links to installation instructions for Ubuntu are broken :/. Also the docker repo is archived, but only recently, so maybe it works. All in all, doesn't look like the project is alive at all.

It definitely uses whatever is available, but on review of the models (and there are quite a few) the ones I checked included instructions for integrating into linux speech engines.

What your distro uses by default might be espeak-ng, but you could use mozilla TTS or festival or piper or anything else. It makes sense to not bundle an entire system that repeats things that are available on all desktop environments and mobile these days, but instead call out to those from the browser functionality - especially given the enormous size of high quality speech models - not to mention the localisation issue. I think the TTS project was just to increase availability of high quality FOSS alternatives. Getting those into your distro is more of a distro packager thing.

Re: Mozilla's open source AI strategy

#205

Post by glenstein »

blibble wrote: post #187
glenstein wrote: post #182

Earlier quoted context omitted.

They already spend more on developing the browser now than at any point in the entire history of Firefox, and that's after adjusting for inflation. They ship millions of lines of new code every year and apply thousands of patches. It's probably one of the biggest and most active open source projects in the entire world. And just so we're clear, are you suggesting they shouldn't have an endowment at all, or that they…

> And just so we're clear, are you suggesting they shouldn't have an endowment at all did you read my comments in this chain? > I would go so far as to say I think the VC fund is the single best idea Mozilla has ever had for long term financial independence. do you look at Mozilla and see it as an example of an organisation that selected good investments over its existence? would have it done better putting that mone…

You started this thread by falsely claiming Mozilla had no endowment. You were fully possessed by the certainty that this was simply more proof of their bad management. Can you acknowledge you were wrong about that instead of just shifting to new accusations?

Re: Mozilla's open source AI strategy

#206

Post by glenstein »

criticalfault wrote: post #194
glenstein wrote: post #170

Earlier quoted context omitted.

Look at what happened to Opera. They fell apart, abandoned their Presto engine for Chromium and sold to an outside investment group and now they serve ads based on user data. There was, in my opinion, no better browser company past or present than Opera in the 2000s and 2010s (sorry Mozilla). But their example exposes the fallacy of assuming that building out great features guarantees market share gains.

opera had a different business model. I don't think they had millions upon miliona that Mozilla gets from Google for antitrust reasons. opera had to earn their money. this aside, Mozilla just now implemented tab grouping. does that mean they are going to, because of the added features, follow opera's path? my point was that someone said how they increased development speed. and I'm saying they are breaking record in…

What are you talking about? Opera and Mozilla are both in the business of trying to deliver a good browser, which means good features. Different financing doesn't change the mandate to deliver good browser features, and Opera most definitely did rely on search licensing as their primary income stream anyway, despite attempts to diversify (for years they relied on Google and Yandex and got more money from that than other forms of financing).

As I said, my opinionated hot take is that Opera was probably the best ever at delivering features and performance beloved by users, but that wasn't enough to move the needle on market share, which is why Opera perfectly exposes the fallacy of assuming better features = better market share. In this context appealing to "business structure" is a deflection.

Also, this tab grouping argument is mistaken both on its own terms but more broadly as a stand in for the argument that the Mozilla team has supposedly done nothing. Firefox had native tab grouping years before Chrome ever had it, had arguably the best tab grouping extension of any browser due to an intentional choice to invest in an extension ecosystem that made that functionality possible, and for the most part, Firefox has never not had tab grouping. What's new is that it's back as a baked-in default rather than merely present as a best in class extension.

The idea that Firefox has done nothing is an unfortunate impression that comes from looking at a serious of unfortunately critical tech headlines and losing sight of nuts and bolts development. I don't have the patience to recite everything here, but every year they push millions of lines of new code, thousands of patches, and deliver measurable improvements to major browser components like webGPU, javascript rendering, shipping production quality rust code, and more for a browser with 30 million lines of code.

Re: Mozilla's open source AI strategy

#207

Post by Vinnl »

7bit wrote: post #201
Vinnl wrote: post #121

Earlier quoted context omitted.

At the same time, Firefox last year gained tab groups, vertical tabs, a user-friendly profile switcher. Split view and tab notes are under development. It sometimes feels like it's moving faster than ever, and that's disregarding all AI features. (Disclosure: I work at Mozilla, but not on Firefox.)

You must be joking! The user profile switcher is NOT user friendly. It's lacking in every aspect of UX compared to Chromes profile switcher. And it has ads in it. Of all places, why the heck did they put ads into the user profile switcher.

Are you talking about the new one, or the old one? The old one was indeed horrible, but the new one is pretty nice IMO. I have no idea what you're talking about when it comes to ads though, I don't think either has that.

Re: Mozilla's open source AI strategy

#208

Post by zamadatix »

WorldMaker wrote: post #108
dandellion wrote: post #96

Earlier quoted context omitted.

I don't know why people left Firefox, but I know why *I* did. And it was three or four years ago (after using it for 15 years) because I got annoyed at them for removing many features that I used over the years, and because I tried a Crommium based browser and it just had better performance and better ad-blocking. That's just one anecdote, but feel free to correlate it with other anecdotes to find some patterns and r…

My father, who is very non-technical has never left Firefox and stuck with it for decades, even against Microsoft and Chrome's tactics to try to claim default browser and constantly install them into his face. My father particularly hates Chrome because he never understands how it keeps reinstalling itself despite his best efforts. His taskbar is often a mess of all three browsers because he can't figure out how to k…

These anecdotes say the people still using Firefox don't like Chrome/Edge and that few cared when Edge switched from Spartan to Chromium. I don't think anyone disagrees with any of that. It's the anecdotes about the different reasons people actually stopped using Firefox for under debate, not the reasons a few have still stayed anyways.

I.e. IE (couldn't resist :)) can be said to have used the exact same shenanigans, as mentioned above, but there were other reasons droves of people still decided to install and use Firefox back then anyways. People no longer make the same decision to install and use Firefox, so if the shenanigans haven't changed... then what did? This is where the common refrains that Chrome managed to be a better browser (particular on mobile) for that decade or that Firefox managed to regress in certain ways come from. Sure, Chrome absolutely got its growth blasted forward by marketing and bundling, but people decided to stick with it and stop using Firefox for reasons unrelated to that. Sometimes niche reasons, sometimes general reasons, but the story was never something like "Chrome invented marketing and bundling, which resulted in Firefox losing its easily gained massive market share of the time".

Re: Mozilla's open source AI strategy

#209

Post by glenstein »

jcranmer wrote: post #113
devsda wrote: post #75

Earlier quoted context omitted.

End users are easily influenced but they could have targeted developers. I think they should have pushed for a gecko based electron alternative. End user dont really care if their favourite markdown editor or notes software is based on electron or gecko but it would have made sure that developers do not target, develop and test for only chromium based browsers.

> I think they should have pushed for a gecko based electron alternative. They did! At least three different versions of it!

Is anyone else seeing this pattern? "Mozilla should have an endowment". They do! "Well they should have invested the endowment!" They do. "They should have done a gecko based electron alternative". They did. "They should have tab grouping". They've never not had it, between native support and extensions. "They should be spending on the browser." They literally spend more now than ever in their history.

It's vibes and drive by cheap shots, all the way down. I get that dabbling in adtech is not great, I get that they've cycled through side bets recently without committing to them (unlike Google?!), but it's an ounce of truth with every pound of nonsense. Mozilla Derangement Syndrome.

Re: Mozilla's open source AI strategy

#210

Post by nusl »

Yoric wrote: post #99
nusl wrote: post #12

I guess replies on this thread are evident that Mozilla has lost much of the trust and goodwill it once enjoyed. Admittedly I am also very skeptical that Mozilla has the ability or genuine interest to make this work.

I feel that, at some point around the Brendan Eich-gate, the Internet decided that Mozilla was always wrong. Change the shape of tabs? We received rape threats. Change it back? Bomb threats. Bringing in new APIs for add-ons that make Firefox faster, more secure, more stable and doesn't break all the time? No, we want addon $X, we don't care about security. I'm not going to claim that everything Mozilla has done is ri…

I get that. Somehow this feels different. Mozilla recently changing the privacy policy I think caused a bunch of it, and then leaning heavily into AI initiatives at a time where I think people just don't want it, contributes too.

It feels deeper somehow than just raging at UI changes or something, since the FAQ change basically betrays what Mozilla has been touting all along, so the trust in their future promises is hurt a lot too. Willingness to change this indicates some kind of internal change that doesn't bode well.

Sure, it's probable that their intent was not the way it was interpreted but we only have what they say to go on. We don't read into things like lawyers or give companies the benefit of the doubt; almost every time this happens it hurts the user.

Post reply on HN