Not saying the comp package was justified. But it was a job doomed to fail.
New Edge was released in January 2020 with Chrome around 70% of the browser market. By May 2022 new Edge was the second most popular at like 10%. Safari gained and took second in 2023, Edge reached record high browser share 11% last month, with Safari also growing leading second place. Chrome is down to 65%. Yes Edge hasn't crushed Chrome, but it's going toe-to-toe with Apple, and is far ahead of any other contender (FireFox, Opera, Brave, and the rest).
Microsoft has started aggressive tactics like opening links in Edge even when it's not the default browser [I'm observing, not defending that] and it results in things like this on Reddit[1]: "When did everyone switch to Microsoft Edge, and why? I work in cybersecurity for a software vendor and over the last 3-6 months have noticed Edge has completely dominated my customers' web browsing choices. [...] the last six or so months it's been nearly 100% Edge."
Chrome grew by being better, faster, and signing in with your Google account to sync state between devices, and with ChromeBooks. Today, Apple is knocking it out of the park and Microsoft can underhandedly force many businesses over to Edge and is jumping on the hype with "your AI powered browser enhanced with GPT-4" and "prioritises your privacy with adblock and tracking prevention". What's Google doing which could push Chrome back up or stop it sliding down by tempting people away from Apple/Microsoft?
[1] https://old.reddit.com/r/sysadmin/comments/1ah5yv6/when_did_...
Instead they freaked out about overall browser usage going to mobile, and threw shit against the wall trying to make some kind of mobile strategy stick no matter how late they were to the table. But I personally still use desktop universally at work, and sometimes browse that way at home too. Maybe Moz could have maintained significant market share in just that space and still be more relevant today.
To be fair, that was more Gary Kovacs and immediate successors as CEOs than Mitchell. But Mitchell had a lot of influence on direction even then, and could have decided to pivot back to core competencies once she did take the reins directly. Servo looked pretty promising to me. At least we got Rust’s success out of it. Maybe she could have leveraged that more.
Fabrice seems pretty convinced that embracing feature phones would have made the difference, and KaiOS shows that the FxOS platform certainly is more well-suited for that than for smartphones. She was involved in the decision-making around killing FxOS right as Mozilla started exploring pivoting to feature. Whether that would’ve helped drive Mozilla’s higher-level goals of browser synergy and making sure the web stayed open, I don’t know, but you’d at least still know their name as a thriving organization.
Or if data privacy is the way to go, maybe pivoting to that focus years ago would have been successful. I guess we’ll see how much potential it has, but I bet it would be even more if Mozilla and Firefox still commanded the respect they used to.
I just think Mozilla had way more routes to success than the routes they picked. Almost everything they tried either struck me as too late, too unlikely, or something nobody really wanted. That’s what I meant by questionable execution.
The OS folks tried to spin it into a few different things. Shipped a Firefox TV OS but that only lasted for a year or so. In parallel there was the IoT plans that never really took off.
Don’t take my words as being too critical of the Android version FWIW, at least the Gecko one. I think that’s been one of the few mobile successes. But what I personally perceived as discounting desktop as a viable strategy was a gross mistake IMO.
Yeah, I agree FxOS was largely a dead end for Moz by the time it was killed and there wasn’t much choice there. My take was partners pretty much lost faith when it became obvious we’d never have performance parity with Android even if feature parity were reached.
But I’m not sure what they—or Moz—expected. FxOS was running core apps on multiple layers of browser stack. Even a Dalvik-based app was going to be faster than that for having a JIT optimized to that purpose, much less native core system utilities. WebAssembly could possibly have mitigated some of that, but partners were universally putting FxOS on the lowest of low end smartphones since cheap was the selling point for them.
I also think the split personality it put on development of Gecko was a huge detriment to Mozilla’s momentum. The browser-focused devs plainly didn’t appreciate the pressures FxOS put on the stack. Then there was the bit where a zero-day in the browser would have been unfixable without potentially compromising all the phones that wouldn’t be able to update for weeks or months later across a disjointed ecosystem.
I left largely over my dissatisfaction with how Mozilla handled that whole period of time and particularly the FxOS effort. It really damaged my impression of Moz as an effective org, primarily due to the leadership and execution issues and what struck me as extreme tunnel vision.
I hear you that after being on that path, Moz was pretty much forced to fold when partners walked away, but maybe some unrealistic expectations were set in the first place. And maybe there wasn’t enough consideration for how that half-pivot would divide the org and drag on Gecko.
I do think the opportunity for feature phones could have panned out technically, though the way reorgs played out may still have gutted it. But when I said I don’t know if it would have advanced browser synergy or the open web (i.e. pushed market share back to full-featured Firefox) I’m saying I’m deeply skeptical it would have been anything other than a purely business success.
KaiOS isn’t a mission-based product and just needs to succeed in being adopted. Moz would have needed to make it mean something. Plus I’m just not sure the audience for feature phones and the audience for web-standards-requiring websites are the same. That’s why I think brand recognition and cash influx were the only real potential gains there.
I think IOT played out well after I was gone, but another comment on here suggests it was never a serious effort so much as a tactical positioning.
Re: TV OS, if you mean Panasonic I’m not sure exactly how much of Mozilla’s code actually survived into that product. I’m pretty sure I’d heard that there was a lot of work done on under the hood on Panasonic’s part to make it viable, while keeping the look and feel plus the FxOS brand on the label.
Maybe that was exaggerated or Moz took a more active role in it after I left, but I didn’t get the impression we had many people in the org dedicated to it when the first implementation happened.