Nokia and Open Source — a Trial By Fire 205
An anonymous reader writes "The H has a damning piece on Nokia's open source smart phone projects, Maemo and MeeGo, and why they failed. 'They did dumb stuff like re-writing the whole networking stack, duplicating as they went. So instead of re-using NetworkManager and improving it, and getting to market fast – they re-wrote, got something that still doesn't work well, failed to push Linux forward, and failed. Repeat that for every technology pick and you get the idea,' said Andrew Wafaa. 'The N900 was a great product. Immediately [after] it was launched it was announced that it was a dead product, ISV-wise. They announced a Qt re-write/project re-set. Then they merged Maemo into MeeGo, giving another project re-set. Then, when they were coming up to release in September 2010, there was another project reset to switch to a different Qt technology (even the Qt groups in-fight in Nokia). In consequence they have no shipping product.' At the same time, 'both Nokia and Intel were working on separate handset UIs using Qt, the former proprietary, the latter open-source. A better worked example of squandering your leadership role and wrestling yourself to the ground is hard to see. Nokia deserve their trial by fire – and I hope the people who truly screwed up the amazing Linux opportunity that was the N900 get shut down in the process.'"
Writing was on the wall for N900 before it began (Score:4, Interesting)
Iterative development FTW (Score:4, Interesting)
It's weird how many engineers fall into the trap of trying too much when settling for good enough would be the right solution. You can always improve stuff in the next version, even if part of the code is ugly. I think the Hurd project has shown how well it works when you insist on getting it "just right".
There are exceptions (Blizzard for example), but often Good Enough is just what you need. Especially with OSS, where the user base doubles as QA and a feedback channel for new ideas.
Re:And they ignored the North American Market. (Score:5, Interesting)
That was Steve Jobs great brilliance. Nokia wasn't going to to play the crappy network game, and basically gave up on the north american carriers as worthless, incompetent, and not worth dealing with.
So Steve Jobs comes along, releases a device that, at launch, was inferior to Nokia's offerings, and was saddled by an outdated network. But suddenly people could see the potential in their phones, if only they had a decent network, and a decent OS. Nokia had (for the time) a decent OS, but no connection to the network, and by the time the network was getting fixed Apple had used off the money they generated to actually build a decent OS. Now you have RIM, Google and Apple all devouring marketspace that in the rest of the world was basically owned by Nokia, because they didn't catch up on innovation.
Don't get me wrong, I'm not a huge fan of Steve Jobs or a lot of the nonsense he pulls, but credit where credit is due, he forced the antiquated network providers in the US and Canada to start pulling their heads out of their asses. That should have been done by Jim Balsillie or Mike Lazaridis of RIM, but they didn't get it.
And now we have phones that are basically computers that can make phone calls. Nokia understood the phones that can do other stuff model, but it doesn't get computers that can make phone calls, and RIM is in the same boat. MS, Apple and Google all get it, it's a matter of how well they can execute and any number of other factors for them.
Re:Who is laughing now? (Score:4, Interesting)
I strongly suspect this was the Symbian side of the company trying to shut the obvious Linux path down
I always assumed it was the other way around. Symbian, in particularly EXA2, was a solid kernel, with a good capabilities model, mostly-usespace device drivers, a great design for power management, and a large base of existing software. It even had a working POSIX implementation, but someone at Nokia saw Linux as more buzzwordy than Symbian, so decided to throw out a working stack and replace it with something horribly experimental (I own a Nokia 770 - power management is atrocious and the Linux OOM killer is pathetic).
Rather than take their existing, mature platform and evolve it slightly, they took a stack that was designed for the desktop and tried to wedge it into a mobile phone. The result was a mess, but all of the focus for smartphones was on Linux, not Symbian. I'd love to have a modern Symbian phone, with a working POSIX layer shipped as standard, but Nokia didn't want to ship anything like that.
Re:And they ignored the North American Market. (Score:4, Interesting)
It's almost like Nokia is the Commodore of the phone market. The Amiga took off in Europe but not so much in North America, where they kept pushing the Commodore 64-derived stuff. The Amiga had its fanboys here, but it never established a huge presence. Meanwhile, the Mac, which was technically inferior in nearly every way to the Amiga for quite a long time, took off. By the early 90s, I saw plenty of Macs, but only ever saw one Amiga in person. Its owner even showed off how it could emulate a Mac and run faster than a native Mac at the same clock rate would run. Technical superiority didn't matter, though.
Yes, it's not a perfect parallel, and Nokia's specific problems are certainly different than Commodore's. But lack of focus behind the cutting edge platform dogged them both.
I remember the joke at the time that if Commodore tried to sell sushi, they'd market it as "cold, dead fish." Kinda reminds me of how Nokia pushed Maemo.
Re:Who is laughing now? (Score:4, Interesting)
" ... Worse, Apple had been rumored to be designing a mobile phone as early as late 2002. For the industry (Nokia et al) to not have made any plans to circumvent this (shut them out with some exclusive contracts, start development of a touch screen phone themselves, etc.) was another example of "falling asleep at the switch."
Although you are probably right, I find the reasons why they felt they could safely ignore an Apple smartphone somewhat interesting.
There is a leaked memo from an executive meeting at RIM where they sat down and had an iPhone in hand to evaluate for the first time, just after launch. I'd give a link, but it wasn't widely reported and I'm not sure where it's at (so feel free to call it bull, but I did read about it somewhere I no longer recall a few months ago; I use a Blackberry).
Their hardware engineers had been telling them that the User Interface and enhanced features (beyond simply making a call) would be limited and easy to add to existing RIM products because there was no way you could put the necessary processing power into the phone and maintain battery life while still having a usable compact form factor. They had estimated battery life to be on the order of 2 to 3 hours if you actually used a feature that didn't involve a simple cell call.
So, prior to it's actual launch they saw it as lightweight competition and of no real threat. Combine that with RIM's belief (and Palm's) that a smartphone had to have a keyboard, although that doesn't really apply to Nokia, perhaps, and there's a recipe for complacency.
On the teardown report discussed in the meeting, the RIM engineers admitted they were taken by surprise at the level of miniaturization and compact layout of the PCB and components, which allowed Apple to stuff a huge battery inside; one much larger than RIM believed could fit prior to the launch. So there was a scramble on two fronts (hardware and software), not just one as they had somewhat expected. One is said to describe the first unit they were able to get a look at as "it's all battery". One can imagine a similar meeting at Nokia.
Some have suggested the delay in releasing a non-GSM version of the iPhone was essentially due to the difficulty of reducing the component footprint to allow for sufficient battery size.