Hacker News Reader

Comment by zerocrates | Hacker News Reader

Just another casualty of telemetry.

I'm sure it doesn't help that Chrome killed its own encoding menu years ago.

> Just another casualty of telemetry.

Sure, but you're missing the bigger picture, which is that each feature has all sorts of other costs (maintenance, security risk, codebase complexity, etc). As mentioned in the bug, this feature got in the way of implementing another feature.

Telemetry helps guide the balance between those costs and the value the feature brings, and you can disagree with how that balance in each decision, but it still is a fact that codebase health is important and if you don't prune, you'll very quickly end up with a huge messy project, no matter how careful you are along the way.

Of course, the other option would've been to rewrite/refactor the feature, but again, is an engineers time better spent working on something that helps 50 people or 50K people?

Exactly. With infinite engineering resources, you can afford to keep every feature around and maintained forever.

But nobody has infinite engineering resources. Maintaining everything comes at a cost, and eventually you have to make some hard decisions about which features to abandon so you can reallocate engineers to features people actually use.

Frustrating for the small number of people who use that specific feature, but you can't please everyone.

> But nobody has infinite engineering resources. Maintaining everything comes at a cost

Irrelevant in this case. There are far more effective ways of paying down technical debt and enhancing future maintainability than removing a small, self-contained feature that occasionally addresses a critical user need. This is Firefox we're talking about, they've got a humongous legacy codebase to play with.

They still have the feature, but not the menu.

I bet the maintenance burden of the menu is extremely low.

I would be prepared to bet that most people that use the more advanced functionality disable telemetry.

And then they wonder why the developers think nobody uses said niche feature

Given that it's known that power users disable telemetry, developers should not rely on telemetry alone when deciding to remove a feature. The opposite is spiting power users, and you can see how that's going for Mozilla.

My feeling is that telemetry as used in most places as a gut check to ensure that what you are doing isn't a terrible idea. It isn't necessarily going to drive you to invest further into something (you generally have better ways of getting that kind of feedback).

> developers should not rely on telemetry alone when deciding to remove a feature

What better data should they use?

> you can see how that's going for Mozilla

HN commenters (not necesarily readers) react this way to most changes in everything. Mozilla doesn't stand out IME.

> What better data should they use?

For one, they could take comments on their bug tracker to heart instead of locking the discussion as soon as there are enough users unhappy with a change.

Should I have to sit around defensively pressing buttons I like to keep around but don't have use for every day just to be sure the devs don't think they are redundant? Seems silly.

Alright 15 minutes have passed, here we go again. Back, reload, view page source, bookmarks,...

Really at this point users should band together to steer Firefox back into better graces. Really all that’s needed is a bunch of servers running Firefox with something like Playwrightjs or Selenium and just juke the stats on features since the marketers and developers are ONLY going to make decisions based off that data.

No, you should use it as you normally do, as will others, and then they'll know what is actually used.

And then the vendor wonders why privacy conscious users don't use their browser (a core demographic, judging by their product line and social media.)

They should just remove Firefox since telemetry shows not a lot of people use it.

Their telemetry is coming from Firefox installs. Kind of impossible to escape from that cave.

Chrome's position was "use an extension for this", but then users have to risk privacy by installing a third-party extension. What harm does having the option tucked somewhere in advanced settings do?

Every release they probably have to test that it still works.

Not that I would ever justify removing something like this - it's clear that for a percentage of the userbase, this is necessary for them to browse. But I can see a product manager trying to cut "fat" if they see that 0.01% of people use it.

Isn't this issue fixed by websites fixing their encoding? If no browser is able to properly display the content, I imagine they will be fixed pretty quickly if they are at all maintained.

Yes, but, they’re not. Huge swaths of the web are effectively abandoned, and will not get fixed.

Then the best option would be to copy out the useful content to new maintained sites and let the old ones die. Rather than keeping browsers tied to supporting long long dead features. UTF-8 has been standards for over a decade now. I think there has been plenty of migration time. More than flash had.

Reply on news.ycombinator.com