No? First of all, I'm saying that allowing sideloading is actually hostile to the majority of users, and I'm also saying this threatened the very survival of Mozilla, so this needs to be part of the consideration.
You must see that your ideas are not working. Have you been able to attract new users, or even prevent existing users from leaving? No? So then how can you be so convinced that you are right?
Firefox is software, not an experience. Winning over people that don't care is not an accomplishment, even if you could do it. Trying to be better than Google at nannying your lower functioning users from hurting themselves is futile, and punishes everyone else.
You're feverishly stripping away everything that made Firefox superior. Driving away developers who have made unique extensions that cannot be replicated elsewhere.
And for what? What is your end game? Have you seen the Chrome store? It's awful! The majority of extensions are very low quality and often misleading. And they are unresponsive and have to employ strange hacks to do things that Firefox extensions did very simply (like taking a screenshot, or interacting with other extensions).
And what is your ideal outcome? To have Firefox be considered a good browser by Google's standards? How about what the users standards?
You are trying to slowly turn emacs into an cheaper Microsoft Word. And patting yourself on the backs for doing it.
You are saying it, but not making any argument or offering any evidence for what you're saying.
Not to mention this "telemetry" process is entirely opaque, and used to justify decisions like this in a very Ex Cathedra sorta way, where you either accept the conclusions they have made from data you can't see, or be treated like your opinion is irrelevant.
Its a very hostile way to run a supposedly "Open Source" project.
It's under a GPL-compatible license. Fork it and do what you like.
This is a pretty FUD-y way to FUD your FUD.
Then why is that not itself an obvious solution? Use that data to provide a default-on blacklist of known-malicious addons. Then you have a blacklist rather than a whitelist, so it can justifiably be more difficult for the user to override it and the user has no motivation to do so for a malicious addon they didn't intentionally install to begin with.
It also gives you the opportunity to allow the user to specify a blacklist provider, so that if the original vendor ever gets compromised because they're forced to operate in an oppressive jurisdiction, they can farm out the low-resource task of hosting the blacklist to someone who isn't compromised, and the user (or the distribution packaging the browser) can make that determination for themselves.
Blacklisting depends on a limited supply of whatever you're trying to control. UUIDs are not limited.
More generally, if you come up with an obviously superior solution to a problem that someone else has claimed is important and difficult, and has spent considerable resources addressing, perhaps it would be more constructive to investigate or ask questions to test your understanding of the situation before assuming and asserting that the other people are doing it all wrong?
I'm not saying mozilla has the best possible solution here. I don't know what that would be. I do know (I work for mozilla though not in this area) that the step was taken to address real, and very active, threats to security, privacy, and stability.
Which is a good reason not to use the extension ID as the sole method to enforce the blacklist, and is why anti-malware software generally uses a signature-based approach.
> More generally, if you come up with an obviously superior solution to a problem that someone else has claimed is important and difficult, and has spent considerable resources addressing, perhaps it would be more constructive to investigate or ask questions to test your understanding of the situation before assuming and asserting that the other people are doing it all wrong?
You're implying this is a novel problem that hasn't been widely studied by everyone and that they're not actually doing it all wrong.
> I do know (I work for mozilla though not in this area) that the step was taken to address real, and very active, threats to security, privacy, and stability.
There are many ways to solve a problem by trading it for a set of different problems. Centralized authoritarian control is exactly that. And those things can be popular in the same way anything that externalizes costs and internalizes benefits can be -- because that type of thing seems attractive to the people not directly paying the cost of it. Until the victims (in this case the developers and users whose apps you prohibit) devise a way to protect themselves. Then you end up in an antagonistic relationship with your users and addon developers. What's the cost of that?
Meanwhile there are other solutions that don't do that.
"It's not gonna happen. Firefox would be dead by now if it still allowed users to customize their webbrowser and make Firefox unusable (as in slow, unstable, and btw insecure), because users wouldn't keep up with that for too long. They'd just use another browser. This may be okay with you for now because you know how to protect yourself against misconfiguration, but in the long run Mozilla needs market share to stay relevant and be able to compete with the richest companies on this planet."
Substitute in any positive feature that puts the user in control of their web browsing experience, power over something means the power to break it too, using that as a reason to remove functionality is stupid.
Doubly so when you pile on the "omg if they do this mozilla is doomed" hysterics.
If the problem in your point of view is that I didn't /prove/ how big of a problem sideloading was then yes, I didn't even attempt to do that. There's a separate subthread on that question.
That's why your logic is flawed. If a construct can be used to "prove" various known-false statements then it has no value in proving anything.
> If the problem in your point of view is that I didn't /prove/ how big of a problem sideloading was then yes, I didn't even attempt to do that. There's a separate subthread on that question.
There is no proof of it there either. The most popular platforms (Android on mobile and Windows on desktop) allow the user to load their own applications, showing not only that it can't be a major cause of failure in the market but that it seems to be a characteristic of the player with the most share.
Yes, if... But it can't. That criticism of my statement didn't make sense.
> There is no proof of it there either. The most popular platforms [...]
I meant this: https://news.ycombinator.com/item?id=18583257
It applies directly. Anything that gives the user a choice can make the user experience worse for a user who makes the wrong one. But it also makes it better for a user who makes the right one in a way different from what the developer would have had to use as a default -- because sometimes something is right for 70% of the users, so it should be the default, but the other 30% are better off with something else. Taking away the choice makes the 30% worse off to benefit the 10% of the 70% that would have chosen incorrectly for themselves. That is not a relative advantage.
> I meant this: https://news.ycombinator.com/item?id=18583257
I'll reply there, but note that you haven't addressed my point -- other platforms survive and indeed have the largest share without prohibiting users from installing software, even when competing directly against others that do.