4 ms·
I know and I feel bad about it, but I have to make some compromises due to limited resources. Supporting all major operating systems is already a huge undertaki
by jarek-foksa 7y ago
I know and I feel bad about it, but I have to make some compromises due to limited resources. Supporting all major operating systems is already a huge undertaking even if you use Electron.
Besides, everything seems to suggest that Firefox will be out of the market in the next couple of years. At least unless Mozilla adopts Chromium engine like Microsoft and Opera did.
- terragon 7y agoI understand that you have to focus your resources onto Chromium; but don't count Firefox out yet. The way Google is behaving nowadays coupled with the progress that Firefox keeps making, there will be a sizable group of us Firefox users for the foreseeable future. And, we're a vocal bunch, too!
- dkersten 7y agoI’ve also noticed a small uptick in my circle of friends abandoning chrome for firefox lately. I personally also switched back to firefox this year after years of using only chrome.
- Moru 7y agoI would say chrome is the IE of old, breaking things and moving their own way. They can still do something that drives away the users too.
- lol768 7y ago>Besides, everything seems to suggest that Firefox will be out of the market in the next couple of years. At least unless Mozilla adopts Chromium engine like Microsoft and Opera did. A browser rendering engine monoculture really isn't a good thing that we should wish for. Chrome is very much the new IE at this point.
- michaelmrose 7y agoIs there any particular reason to believe this?
- lol768 7y agoThe latter point? Yeah, there's a few things I can point to: - PNaCl, which was essentially Chrome's ActiveX. Thankfully going away in the near future, after nobody else implemented it. See https://www.cnet.com/news/mozilla-coder-chrome-violates-googles-own-blink-principles/ https://www.cnet.com/news/mozilla-coder-chrome-violates-goog... - The severe limits applied to the webRequest extension API, breaking extensions such as uBlock Origin, without cross-browser consensus: https://blog.mozilla.org/addons/2019/09/03/mozillas-manifest-v3-faq/ https://blog.mozilla.org/addons/2019/09/03/mozillas-manifest... - General implementation of new technologies in the browser, without a proper spec/consensus. See <portal>, which has introduced tangible security vulnerabilities into the browser: https://research.securitum.com/security-analysis-of-portal-element/ https://research.securitum.com/security-analysis-of-portal-e... - Signed HTTP Exchanges and AMP. The controversy around these two has already been well-documented. - Poorly thought out plans to roll-out default SameSite behaviour whilst a glaring bug was present in Safari which would be exacerbated by the behaviour change (requiring SameSite=None). I'm all for preventing CSRF by default, but these sort of changes should be planned and implemented considerately. See https://news.ycombinator.com/item?id=20905396 https://news.ycombinator.com/item?id=20905396 - Poor implementation of CT reporting which sent spurious reports to sysadmins even after the policy was supposed to have expired. See https://crbug.com/786563 https://crbug.com/786563
- michaelmrose 7y agoI mean any reason to believe Mozilla will be out of the running in a few years.
- lol768 7y agoAh, then you should have replied to the parent! But no, I personally don't think so - provided web developers do their job and stop designing for one browser. I've been impressed with their recent work - Quantum, developer edition, WebRender etc. Don't care so much for the sync/VPN/other external services, though Firefox Send is quite neat.
- liziwizi 7y ago"Besides, everything seems to suggest that Firefox will be out of the market in the next couple of years" yeah, that's what they've been saying about Linux for over 15 years now. i wouldn't bet on it. we, the FOSS community, will always need a non-commercial browser that embraces the open source core values.