4 ms·
This guy is so ex-Nokia, it's sad. He rates the latest Maemo/whatever as a higher chance of market presence than Firefox OS. OK, Nordsman... we understand you a
by contingencies 13y ago
This guy is so ex-Nokia, it's sad. He rates the latest Maemo/whatever as a higher chance of market presence than Firefox OS. OK, Nordsman... we understand you are grieving for your lost mobile device business... but you can do it in private.
Linux, via Android, commodified smartphone platforms by giving away the OS and thus commodifying the position of mobile device hardware manufacturers. This was coming anyway, but was a brilliantly timed commercial play by Google to head off Apple's strength in devices. Google's allies were the carriers, who were unhappy with Apple owning (via signup-time credit card) who they perceived as their customers.
FirefoxOS and that new Jolla thing the Nords are working on basically use exactly the same hardware and OS platform (ie. Linux). There's no huge investment there, the investment is in higher layers of the software stack, and in FirefoxOS' case in marketing to different segments (the developing world, which still has 'growth', and necessitates less competitive component acquisition for manufacturing batches).
To say a small team can't do something similar is plain wrong. I'd say the effort required is similar to launching a new Linux distribution. As long as there is a niche you can cater to, there is no reason it won't survive and thrive. CyanogenMod is only the beginning: cross-pollination is a good thing.
Start with preferably https://github.com/mozilla-b2g/ https://github.com/mozilla-b2g/ (or otherwise http://source.android.com/source/downloading.html http://source.android.com/source/downloading.html) and begin thy mission to conquer portable computing!
(Edit: Added content after the initial paragraph after it was downvoted. Fair enough, it was a bit snarky :)
- davb 13y agoFirst paragraph snark aside, I agree. To say that any tech market is completely closed to new entrants seems slightly ridiculous. Sure, the barriers to entry have increased - if you're chasing mass-market appeal, a strong app platform is as important as excellent hardware. However there will always be some niche to serve or an opportunity to tackle bloated or less open, established competitors. I think Josh's option is very biased and colored by operating in a certain type of environment. Not that it's any less valuable or valid than any other, but I disagree with him.
- sharpneli 13y agoThe approach by Jolla and FirefoxOS guys is actually quite brilliant. How to get into markets where it's relatively cheap to make a new android phone but anything else requires very expensive contracts with the manufacturers? Solution: Just use the android drivers directly. In my Jolla phone there is the good 'ol android /system/lib/. As an example the libGLESv2.so (the OpenGL ES 2.0 library) in /usr/lib/ simply opens the counterpart in /system/lib with android_dlsym and calls the equivalent functions (and in case of floating point arguments moves them to int registers). And what's the overhead of this? 9 instructions and 2 loads (for the function pointer). So it's essentially completely free. Same approach applies to rest of the system. They just take the core of the android and run on top of that with just the bionic -> glibc layer on top. Brilliant. The actual lib can be found at: https://github.com/libhybris/libhybris https://github.com/libhybris/libhybris That means they can basically just walk to some SOC vendor (like they did to Qualcomm) and say "I want to buy a mid range Android SOC" and they have everything ready.