3 ms·
I worked at Mozilla on Firefox OS until shortly before the project was killed. I wouldn't say that's the case at all. Andreas Gal was considerably more influent
by geoelectric 8y ago
I worked at Mozilla on Firefox OS until shortly before the project was killed. I wouldn't say that's the case at all. Andreas Gal was considerably more influential on Boot2Gecko and FxOS than Brendan, at least once it came out of Labs.
Mozilla really just got caught in between being an OSS organization and working with proprietary phone companies, carriers, baseband developers, etc. We were used to working transparently on wiki, Github and IRC, and suddenly had NDAs and confidential code that partners insisted on. There was a large culture clash, and the effort both took a bunch of momentum from the browser side and ended up putting unique technical constraints on Gecko and the Firefox Browser, both in terms of necessary functionality and on the development cycle.
That last could be a killer because you couldn't do anything quickly as everything had to be tested at length for interoperability, etc. FxOS was usually pinned to something older than the browser development (consider the security bug implications there) for extended periods of time, and teams had to shuttle fixes back and forth. Given that some of the changes were basically hacks to get the phone going and had no benefit to the browser, I'm sure that caused a lot of friction.
Partners also wanted to see formal processes guaranteeing their stake, and that was at odds with the more "steer don't plan" way Mozilla tended to run. The people that were brought in that were more familiar with formal process and could make partners happy weren't as familiar with Mozilla's community or culture, and that also churned things a bit.
Basically, it was a perfect storm of issues to deal with--Brendan leaving was just one of them, and possibly the least important. Morale wasn't awesome, and the window of opportunity for a new smartphone OS was closing--if it was even open in the first place.
While there was some talk around shifting to feature phones as a focus/market gap (basically what Kai has done), it was very late in the game. After a point, I think a decision was simply made to cut bait and get back behind the browser again. Quantum was the (IMO very good) result of that focus.
Honestly, Mozilla did as well as I'd expect any organization of its type and size--which is not as well as I'd expect a greenfield organization tightly concentrated on the goal to do. KaiOS is that latter thing.
- cmroanirgo 8y ago> ...suddenly had NDAs and confidential code that partners insisted on. There was a large culture clash... I was wondering why it failed when it had so much going for it. Tragic. It makes you wonder the secrets that Google withhold...
- fabrice_d 8y agoNDAs and confidential code were mostly to let chipset vendors ship closed source version of the telephony components, so they would not have to disclose their extensions to the RIL standard packets, the integration with a-gps, etc. However there was an open source implementation (MozRil) that worked well enough. FirefoxOS failed commercially because we could not get on board the content partner(s) that make or break a mass market Mobile OS.
- javagram 8y agoThanks for your comment! Really appreciate the insight.
- BrendanEich 8y agoFirefox OS did not come out of "Labs" -- Andreas and Chris worked in Client Engineering, led at the time by Damon Sicore. I took all of engineering including server side pieces formerly under Labs led by Mark Mayo in January 2013. Firefox OS, neé B2G, was a skunkworks project for sure, which needed shaver and me as exec sponsors to avoid being killed in the cradle. KaiOS's success could be read as us being too early (but we always thought we were late, starting in 2011); or as Mozilla not sticking it out and working on the low end to make deals such as the JioPhone one and others which I think were under way with Verizon. Whatever the case, KaiOS pulled off the vision and used code and engineers from FirefoxOS, which counts as a win for the concept. Edit: so if the "window was closing", did it re-open? I think rather the window was always open, some variable amount, and awaited Reliance Jio's amazing growth via free-for-N-months 4G data. Android never fit in the low end, so the window was always there at some width of openness.
- andreasgal 8y agoWe had a concrete business opportunity with several high volume customers lined up. The big driver was 4G transition. To convert legacy 3G bands to 4G carriers have to move feature phones onto an LTE capable platform. KaiOS managed to capture the same opportunity. The main reason that Firefox OS was shut down was that management wanted to focus on western markets and desktop, which was and is the only revenue driver. I left since I didn't see any growth or opportunity or interesting problems to solve in that Venn diagram.
- geoelectric 8y agoSorry about the mistake re: Labs. I had thought B2G started there, but I should have gone back to check myself.