Mozilla lays off 250 employees while it refocuses on commercial products

For me, the feeling of getting kicked in the shins by a diva designer every-other update has risen dramatically in the past few years, as has the prominence of (at least the feeling of) 'closed wontfix dontagree' issues for common and longstanding gripes on the bug tracker and GitHub. The unfortunate nature of a bad feeling is that it will outweigh a positive feeling from another change of equal consequence.

I would not be surprised if it was the same for other users. It results in implicitly giving less benefit of the doubt when another potential controversy comes up.

Other application developers are held to a lower standard because they have already come out the other side - people already simply assume the worst about them. The paradoxical anger comes from the fact that they don't want to do the same with Mozilla, but feel more and more that they'll have to.

> I would not be surprised if it was the same for other users.

Yep, you can count me in this group as well. The Firefox team goes out of it's way to make so many changes that just seem useless or annoying; it's baffling to me. It really feels like a team with too many devs and designers sitting around needing to create work. I very much doubt that's actually the case, but that just means it's a widespread management issue.

That said, it doesn't make me want to stop using Firefox because the only other option is Chrome which has bigger issues.

The 'closed wontfix dontagree' attitude, or letting important requests sit there open for over a decade -- some with tens of thousands of comments -- is what killed Firefox more than anything.

In the enterprise world, Firefox lacked a few, small, but critical features:

1) MSI Installers

2) Group Policy Administrative Templates

3) Proxy configuration from Windows

4) Enterprise PKI integration

Some of these are supported now, but for about a decade there was at least one person in Mozilla with a philosophical opposition to doing anything that is seen as helping an enterprise Windows network deployment.

I'm pretty certain that Firefox still doesn't work properly in a large corporate environment. At any rate, I've given up trying, as have millions of other administrators. We installed Chrome, which "just worked", and moved on.

The result of this is that enterprise web applications were written for Chrome, not IE or Firefox. Chrome became mandated and automatically pushed to every machine. It has become the new IE6, for better or worse.

Firefox missed that boat.

> MSI Installers

Why was that important? I was under the impression that exe and msi installer had no real difference between them. Obviously I am incorrect but I am wondering why.

True, there's a spectrum. There's the "download wizard" stub installers, then "interactive only" installers, then the ones with unattended command-line flags, and then there's native support for the operating package management format.

In Windows there are further nuances, such as installing per-user, per-machine, or both. Similarly, MSI support often implies support for transforms (MST files) and patches (MSP files) also, which is important on large networks. Back in the days of constrained bandwidths, MSPs were great for rolling out updates without killing the WAN, but few vendors would provide them.

Firefox tended to prefer the interactive install wizard installers and hence deploying it at scale was an enormous pain in the arse.

For example, the Enterprise CA thing actually interacted with the packaging. You had to crack open the Firefox files, download some obscure NSS command-line tool that they regularly moved around on their website to spite admins, and inject your corporate certificates into the Firefox-specific Root CA file. After this, everything had to be put back together in some way for deployment, typically by repackaging the files into an MSI.

Similarly, instead of ADM Templates that allow settings to be pushed out via Group Policy, you had to do hideous things to JavaScript files. These also changed regularly and had all sorts of limitations.

There was just no way anyone in their right mind would do this every few weeks to keep up with the Firefox release schedule. IT admins have other things to do, not just babysitting Firefox, the one special and unique flower that refuses to play nice with Windows.

The only other obstinately anti-admin products I can think of that were this bad were the Java Runtime and the Adobe suite of products. Even Adobe provided an ADM template at least, even though they published it as a PDF.

The last update of Firefox mobile really made wonder how it passed QA.

The navigation bar on the bottom is an interesting expiriment but still I quickly moved it back up.

The new tab layout is worse. Each tab need much more space so you can display about half the number of tabs on the same space.

The top sites when you opened a new tab also disappeared. It seems you can put bookmark or something to replace them but the top site feature was really good. I'm likely to want to visit sites I visit the most and that used to be a touch away

> The last update of Firefox mobile really made wonder how it passed QA

They had it as a separate app, as it was (seemingly) a rewrite that uses GeckoView. So having less features and unstable features was somewhat expected for users that went out of their way to install it.

Naturally, a lot of the old features weren't supported, and some were likely feeling like unmaintainable legacy code in the old version (Fennec) to begin with. So a rewrite may have made sense, for where the products were at and where they wanted to go. At least, in my opinion.

But for whatever reason, they decided that they had reached an acceptable level of feature and usability parity as to replace the old version with the new one (Fenix) in the play store. Somewhat forcibly moving all of the current users onto the new browser, silently for auto-updaters and update-all-ers. The downgrade path for those who want or need the old version is... well, I didn't see any documentation when I googled, so presumably it's "install the old APK and hope things work".

It's a cynical and somewhat egotistical approach to software development, disappointing to see from Mozilla and yet another entry on the list.

At this point, I would not be surprised if they find the idea of the extension whitelist - rather than a more open platform as is the desktop and was Fennec - far too appealing to move away from in the future.

That's somewhat doom-and-glooming, but many months ago I had thought they wouldn't end up trying to shove Fennec users onto Fenix in this way, and yet... here we are.

There is now, and has been for some time, an advocate of Firefox in the enterprise and last I checked that is almost entirely what he worked on. His name is Mike Kaply.

But you’re right: too little, too late.

> It really feels like a team with too many devs and designers sitting around needing to create work.

This is how a lack of a vision manifests - if there was a vision, there would be meaningful work for everyone, instead of people inventing unwanted features to keep themselves busy.

And of course, vision needs user & customer research, it's not a thing a 'leader' could hallucinate with no external inputs.

Also they basically removed the tablet mode - here I am with a 10 inch tablet and no tab bar, no back button on address bar & no keyboard shortcuts, perfect! :P

More things the dropped in lates "stable" Firefox for Android: - save as PDF - print support (!!) - downloade manager - full URL display - about:config (!!!) - extensions (other than 9 extensions they carefully cherry-picked)

On several important Windows servers on a customer's LAN that are accessed via Remote Desktop, Firefox is able to tell it wants to auto-update... and it fails to do so, since an unprivileged passer-by cannot (and should not) update important software, and no actual administrator bothers with browser updates after initial setup and certification.

The same reason MySQL is one or two orders of magnitude more popular than PostgreSQL.

Thinking it is a good idea to leave Windows users behind. You don't hurt Microsoft by doing that, you hurt yourself.

MSI installers can be automated with PowerShell. They can be pushed system wide in a Domain.

I am sure there are other reasons, but these two are the important ones.