3 ms·
No, don't. They're not that hard to code, but both Google and Mozilla keep making changes and then bitching at you through their app stores to change your thin
by Animats 4y ago
No, don't.
They're not that hard to code, but both Google and Mozilla keep making changes and then bitching at you through their app stores to change your thing. I recently dropped Chrome support on an add-on because 90% of the users were on Firefox. Since it blocked some ads, I figured Google would continue to whine, and it wasn't worth the hassle.
- satvikpendem 4y agoInteresting, based on browser share I'd have expected 90% to be on Chrome instead. What's the extension (or rather, why are most on Firefox)?
- Animats 4y agoBecause Firefox users install more extensions than Chrome users do. Or at least they used to. I don't really care; it's an old project I keep going for existing users.
- boltzmann-brain 4y agowhat is it?
- gnicholas 4y agoI've had an extension on Firefox and Chrome for almost a decade. Chrome is 90% of our users. It's possible that FF users are more likely to install, as mentioned above, but that doesn't outweigh the sheer number of people on Chrome. With the manifest v3 update, we're changing our freemium structure, and updating our docs. But since FF hasn't figured out all of their manifest v3 stuff, we're not able to update for that 10% of users. It's going to be confusing because our marketing material will reference the new freemium model. But it's no contest as to what we should primarily reference, since there are so few people on FF by comparison to Chrome.
- lvass 4y agoUse tampermonkey if possible. Or better yet, extend software that doesn't break, like Emacs.
- alanbernstein 4y agoHeh, I agree with this viewpoint, but "software that doesn't break, like Emacs" sounds extremely generous to me.
- crtified 4y agoThe lack of platform control creates a paradox where the more useful the extension you make, the more you're beholden to dance to somebody else's tune, ad infinitum, in order to retain the advantage you've created. Unpredictability is at the heart of it. You can invest x $/time, and not know if what you've created is going to stand for 1 year, or 10. On the gripping hand, when the platforms in question are the dominant platforms of, err, all computing in this era, then the above described issue tends to be a minor one, in context, for many applications.