The future of Mozilla Hubs | Hacker News Reader

The future of Mozilla Hubs

hubs.mozilla.com

84 pointsKarrot_Kream63 comments

Before people pile on the typical negativity I tend to see surrounding Mozilla, can we step back and admire this type of product shutdown announcement?

The code is open source, and the announcement emphasizes the ways in which the community can own it if they so desire.

If every SaaS that shutdown went this way, I think it would be really pleasant.

> If every SaaS that shutdown went this way, I think it would be really pleasant.

Indeed... just think about Google's regular service shutdown announcements

What was Mozilla Hubs? Never heard of it until now.

The community took Mozilla's announcement of shutting down parts of the org very negatively, but your comment agrees with my feelings: they're shutting down the stuff nobody uses.

Hubs is/was a private 3D world for conference and teaching, it's essentially metaverse: https://techcrunch.com/2022/11/29/for-20-a-month-you-can-hos...

But I've never seen anyone actually use it. Good on Mozilla for making the hard decisions.

I've considered using it at work but it wasn't really great. There was only a web client.

I think it was a weird tangent for Mozilla, but the product itself is not bad at the core. Probably some more specialised org will take over management of it like what happened to Mozilla's other abandoned products.

Yes, let's hope! The strategy has worked out sometimes - Google shut down 'Google Refine' 10 years ago, it got turned into 'Open Refine', last release five days ago. https://github.com/OpenRefine/OpenRefine

It's a hugely useful tool if you're working with messy Excel-scale data, i.e., most biologists or social scientists.

I think of the weird things that Mozilla does, this one made sense. It's a web-based environment. At the very least it's useful to push Firefox to do 3D better.

If you're looking for something desktop based, by the way, I work on such a thing.

> I think of the weird things that Mozilla does, this one made sense. It's a web-based environment. At the very least it's useful to push Firefox to do 3D better.

Fair enough, good point.

> If you're looking for something desktop based, by the way, I work on such a thing.

Oh I've tried several. Spatial (which is now more general metaverse and less businessy, sadly), Arthur. Immersed. And now we're using MS Mesh. Mesh kinda sucks (it works, but you can't do anything and even the meeting host has no control over the instancing). But it's free with O365.

Mesh is basically a rebadged AltSpaceVR by the way, but with most of the features removed, curiously.

I'm kinda curious which one you're working on though :) I know there's more than the above but I haven't tried them. The biggest problem of all of them though is the usecase. The why. Even though Mesh is free now, I have a really hard time justifying why to actually deploy it.

The whole convention thing is not really worth it IMO, and not many sites bother with WebXR anymore. Personally I'm a huge VR enthusiast and just hanging out is a perfectly good reason for me but I have bosses that are a lot more critical :)

> I'm kinda curious which one you're working on though :) I know there's more than the above but I haven't tried them.

I'm on Overte, an open source fork of the dead commercial High Fidelity.

We don't have WebXR right now, but we could grab that functionality off another fork quite easily. The main reason we didn't is that we're desktop VR focused and that the readily available code is slow and buggy. But we do have a pathway there should there be enough interest.

> The biggest problem of all of them though is the usecase. The why. Even though Mesh is free now, I have a really hard time justifying why to actually deploy it.

Our deployment process is extremely simple. Both the client and server can be compiled trivially, there's no complex dependencies.

Our usecase is doing whatever you want. Unlike most everyone else we've got a distributed model where we don't control what you do with it. It's a bit like a 3D webserver in principle.

The software is very suitable for massive environments though.

The team that build Hubs was also part of the team that built AltspaceVR.

Everything is a web client. Just because they wrap it in Electron and peddle it as a “native client” doesn’t usually make it so.

I don't think most other "metaverse"/VR-chat type stuff is electron. Mozilla wanted this to be a web client presumably to be a good demo/killer app for webXR and related tech.

That's how it started, but then Mozilla laid off their WebXR team. I never understood why they did that but kept Hubs. And there were weird things, too, like Hubs would work on the Quest in Meta's WebXR-compliant fork of Chromium, but not on desktop in Google Chrome with all the same necessary WebXR features. Truly bizarre priorities.

We gave it a try during covid and while it was fun, it gave some people vertigo so we had to drop it as a work tool. It was fun though.

It's Mozilla's metaverse platform

That’s basically it. I used it to visit a virtual art opening during the pandemic times. It was a little wonky but did work. I was able to talk to people and roam around the space that was set up.

I'm with you. This is probably the first time I heard about this service. The usage numbers they give in the announcement would be great for any of my customers but not particularly important for Mozilla IMHO. Much smaller than Firefox.

neither have I which I think is the core of the issue

it's a hardly known hardly used product

and with the community edition another company could pick it up if it's just anywhere close to profitable

For anyone else who missed the Feb 13 announcement:

Mozilla downsizes as it refocuses on Firefox and AI (176 comments) https://news.ycombinator.com/item?id=39362481

> refocuses on Firefox

About damn time; they let their most popular product languish.

How so? What browser capability lacks in Firefox?

Innovation.

Reliability, speed, power user features

chrome is only faster if you account for all the tricks, like ultra-aggressive pre-loading of links and using their own DNS instead. With those two things off it is slower today.

Also, most google properties were caught red handed slowing down on firefox, like meet and youtube.

I use Firefox because they don't threaten to shut down power user extensions every few years. I find it faster and more reliable because it doesn't chew all of my RAM and lead to swap / early-oom-kill.

I'm quite confused by folks' opinion of Firefox in this thread because my only problem with it is that maybe once per year (or less) I find an online checkout that is subtly broken in it (sometimes it's just my adblocker). But I have occasionally found sites that are inverse, only working in Firefox.

They allowed google to push them around on some key things:

- no addons on mobile for more than a decade! for nothing other than google-told-them-not-to.

- moving to webextensions because "google have a better standard for cross browser extensions". Broke all extensions for a long time and google did not honor their promise from day one (e.g. try to install flutter debug extension on firefox)

- reverting features on devtools for no good reason or explanation (e.g. try to change code or update a variable value after hitting a break point on the debugger.)

- No decent search on bookmarks. they keep adding and removing the keyword field. etc.

- marketing. Like it or not, it is the reason badly informed people here, right next to this comment, claim it is slower. lol. marketing lies but it matters. q.e.d.

(I worked at Mozilla for 9 years, including a couple in mobile engineering)

> no addons on mobile for more than a decade!

Huh? What are you talking about?

> for nothing other than google-told-them-not-to.

Google never dictated any product decisions.

> > no addons on mobile for more than a decade!

> Huh? What are you talking about?

Oh, right, there was some five addons that were "allowed". How on earth could I have forgotten that for one decade we were allowed five addons! shame on me. All hail the benevolent mozilla overlords.

> Google never dictated any product decisions.

well, it was the official mozilla excuse (we have to prevent random code. we have to submit which addons will run for review on the store. etc). So it was either google dictating product decision or mozilla having other reasons and blaming google. One or the other.

The bigger issue is that Firefox hasn’t moved beyond what the browser market is right now.

Another way to say that is what browser capability does Firefox have that Chrome lacks.

I remember that pre-Chrome Firefox was leaps and bounds ahead of nearly every browser in the market (especially when factoring in extensions) other than maybe Opera.

But Firefox has lagged in many ways since then. About the only really great thing that has come out of Mozilla/Firefox in the last decade or so is possibly MDN (yeah, their documentation website, which is frankly awesome to the point that MS shut down their own and paid Mozilla to improve MDN).

Even setting aside functionality and features, there’s basic housekeeping stuff that Mozilla can do so much better at. For example, have you ever tried contributing to Firefox code? It’s awful. They have their own, largely unmaintained, bug tracking system that no one uses. And as much as I love mercurial and constantly whine about how it is superior to git and should have won out, it didn’t win out. Git won. And yet Firefox is still a multi gig Mercurial repo. And it’s a mercurial repo hosted on Mozilla’s own hnmaintained forge that no one is familiar with and knows how to use. Even if you don’t want to hop onto someone else’s proprietary forge, which is completely understandable and laudable even, then why not pick up something more modern like Gitea, and run your own instance and contribute back making Gitea or any other Gothub/Gitlab alternative better and making it easier for folks to contribute to Firefox code? Why is Firefox still 1 massive repo instead of breaking it down into multiple smaller components so, for example, web developers who can very well contribute to the Firefox user facing features which are largely written in HTML/JS don’t have to go mucking around with C builds?

There’s so much Firefox needs to do that appears not to have been focused on at all.

The web moves fast, but my experience with Passkeys on Firefox mobile (Android) has been subpar so far.

Can't even use it to authenticate to GitHub, etc

However I can somehow excuse it as Android support for Passkeys isn't that great either. I had to factory reset my phone yesterday and when Google asked me to authenticate my Google account I had to plug in my Yubikey (enrolled as a Passkey) and as soon as I plugged it in (couldn't do that one over NFC for some reason) it asked for the PIN, but since the Yubikey was plugged and detected as a keyboard it would hide the on-screen keyboard, effectively making it impossible to type in the PIN. Someone at Google has to test the UX once in a while..

Oh, I don't know, what about the ability to run WebXR apps without a shim through the pre-standardization WebVR API? You know, like you would need for a metaverse app.

Owning a browser is really a gateway to deliver other mbetter margin products imo.

No one's paying to use a browser these days but there can be a really good angle to approaching a browser as a gateway to using ai baked in and other tools for productivity.

From one bandwagon to the other it seems... Doesn't sound like they learned their lesson. Unless they're pivoting to embedding multi-modal shitpost generators. That might actually be popular.

I also remember the last time they sacked a bunch of people and claimed they were refocusing on Firefox.

Now remove fucking Pocket and all the other stuff no one wanted and just maintain your damn browser.

The funnier part to me about Pocket is Mozilla could have just re-implemented it as their own service.

I'm still sour about them having a password vault, and then not maintaining it for iOS any longer (and maybe android?) when they were my default password manager for over a decade. I have opted to use BitWarden, but what I preferred about using Firefox's password manager was that it was built-in.

Honestly if Mozilla just focused on keeping Firefox up to date that could be a winning move. Whenever MV2 is phased out of Chrome, people will start seeing more ads and noticing their adblockers don't work so well anymore and start browser shopping. They don't need pocket, they don't need Mr. Robot, they don't need the metaverse or crypto. They just need to suck the least.

I have used firefox for I guess 15 years now and only have a dim awareness of what pocket is. I definitely see the hostility towards it but I don't really understand the depth of it.

If you don't like a feature you could just ignore it, right?

The future? I didn't know it had a past! I ignore everything Mozilla except Firefox and Thunderbird, and I'm slowly quitting Thunderbird.

Going to try to move to Seamonkey for everything soon.

That's wild, that SeaMonkey is seeing updates.

Netscape had some good ideas about Internet clients early on.

Do you think SeaMonkey has the right developer resources to wrangle that code base for security?

I pray that developers leave the project alone as much as possible. Too much love is what is bloating Firefox. At least FF is ripping out their Rust remnants.

I thought Thunderbird was doing better as of late? They had some interesting presentations at FOSDEM, and apparently are now on their own.

Yeah, Thunderbird is finally getting better.

They’re gonna ruffle some feathers as they ditch unmainrained, and unmentionable functionality that has been languishing in the code base for over a decade and currently serve little purpose other than being a security nightmare, but they’ve been improving Thunderbird significantly, especially in terms of speed and performance.

> They’re gonna ruffle some feathers as they ditch unmainrained, and unmentionable functionality that has been languishing in the code base for over a decade and currently serve little purpose other than being a security nightmare

Like what?

That's why I'm leaving - I use Usenet.

I'm just beginning to use Thunderbird, and I really like it. It's lighter and less noisy than Outlook, and far less annoying than having a gmail tab open at all times.

I used it for my 1:1’s with a coworker who wasn’t too comfortable with such talks. We strolling around through a virtual world, me as a dancing woman in a pyjama and he as a fox made things more more relaxed for him.

I hope a new community forms to support hubs, it's such a cool platform!

Right above you (at the moment) is someone saying they couldn't convince others what's cool about it.

What is cool about it?

It's the only comparable platform that can run on anything from a phone to a PC to a VR headset. It's fully open and based on open standards.

It's flawed, but I've seen it used to great effect for community, performance and arts events.

Genuine question.

I went past the Mozilla office in SF the other day on the Embarcadero. Looks like they closed up shop there. Did they move or focus on being all remote?

It's disheartening to learn of this development. Although my personal usage of Mozilla Hubs is limited, I frequently utilize NTT DOOR [1], which is built upon Mozilla Hubs and enhanced with additional features. Mozilla Hubs has gained some traction in various sectors across Japan, particularly among those eager to join the "Metaverse" trend. However, with the potential decline in upstream support, we may witness a reduction in its adoption, as maintaining and securing it could become challenging. On the other hand, NTT DOOR is overseen by a team that appears quite capable, especially given their ability to create extensions. Nevertheless, there's a lingering concern about its long-term viability as well.

[1]: https://door.ntt

I always wanted to use Hubs, but was never able to convince others what was cool about it.

Its as cool as VR powerpoint...

I suspect you're trying to be snarky but you're also inadvertently right.

It's as cool as the content you put in it. I've been to several excellent events in Hubs.

I call it efficiency and think it's good. If the resources freed from this project are utilized in improving the firefox browser itself (need of the hour), it's win-win for all. The browser's performance and long outstanding bugs, especially on the android/ios platforms, is the reason people are still sticking to chrome and clones. I really hope they will do it.

People are sticking to Chrome and clones as they got an advertising budget that is higher than the whole Mozilla foundation cash flow and are installed as default browser on the OS.

Combine that with aggressive approaches that try to persuade users from switching when trying to change the default browser on an OS.

The features most people on HN are complaining about are very gimmicky and not even widely used in the developer market.

Interesting to see Mozilla dropping this while others are all in, like:

"Croquet for Unity: Live, network-transparent 3D gaming... but it's so much more":

https://news.ycombinator.com/item?id=38769416

https://croquet.io/

Reply on news.ycombinator.com