It's amazing that a group of volunteers can do a much better job at improving the stock AOSP experience than huge companies like Samsung or LG. The official software on some of the Android flagships out there is seriously awful. Best of luck to them.
The biggest problem is that Samsung and LG think they need to customize & brand the phone to be recognizable as their own product. They're spitting in the face of OSS when they go and half-ass their own versions of standard components. Things are even worse when you get carriers into the mix trying to provide their own unique user experience.
A great example of how this can blow up in their face is a recent T-Mobile visual voicemail app. Not only is it slow, buggy & constantly crashing but they had the audacity to place advertisements in it. For some reason, customers didn't like being forced to use their time & metered bandwidth to view shitty ads (the "Click here to upgrade your memory!" sort) while using a service that they had already paid for. If you look at the Play Store, you'll see brazillions of 1-star reviews complaining about this. Fortunately, T-Mo listened & removed the ads. The app is still slow & buggy. I still can't uninstall the Slacker app or remove the Fox News bookmark from the browser but at least I don't have to stare at advertisements while waiting for my voicemail to crash.
Manufacturers & Carriers need to recognize that the value they bring to the table with Android is by providing good hardware, low prices & good service, not crapping up an awesome phone with reminders that Samsung & Verizon are the Best Thing Ever.
Manufacturers do that, even in the face of evidence that they're terrible at it, because they really want to avoid ending up like PC manufacturers. Consumers treat PC makers as interchangeable suppliers of largely similar products, forcing them to compete on price, so they have very small margins.
Samsung is very keen for you to think of your phone as a Samsung phone, not as an Android phone made by Samsung. So to them it makes sense to reinvent standard components, even if it doesn't make the user experience any better. And Samsung does seem to be managing this, though I don't think any of the other Android OEMs have.
Cyanogen may be better for the end user, especially the technically savvy end user, but the metric of success for manufacturers' Android variants are a mixture of end user experience and other factors like brand differentiation. It's possible that they're achieving exactly what they're trying to, which basically boils down to being able to produce an ad that shows off the features that their phones have that their competitors' Android phones don't. Once the sale is made it doesn't matter too much how awesome the user experience is or isn't.
No company that wants to stay in business can afford to completely ignore user experience. Samsung, for example, is every bit as concerned as Apple about this and every recent study I've shown puts their phones ahead of Apple in user satisfaction. Samsung, HTC, LG, Sony etc very much want to do everything they can to make sure that people that buy from them today think of them first when it comes time to upgrade.
Maybe the very bottom barrel discount generic Android device makers don't care about this but you get what you pay for.
There's quite a few features in both AOSP and OEM skinned devices that were inspired by community ones. The most obvious one I can think of offhand is quick controls in the notification area to toggle wifi and other things (especially the way Samsung did it is identical [though uglier] to how it is in Cyanogen).
Brand differentiation is important, but there are more ways to achieve that than, say, "shovel a crappy skin over ASOP"; "timely upgrades to new versions of Android" would be a good one (most vendors fail", minimising crapware (one point I give Sony), and "form factors and hardware features I like" (with two young kids, the waterproof Sony phones have a leg up here for me).
Actually, I don't find it surprising at all. If you knew anything about the decision making structure and culture at companies like Samsung and LG, you'd assume that it's a total camel factory (camel: a horse designed by committee). These guys on the other hand are very very close to the users and know what people actually will actually find valuable and use. They aren't just putting features in their OS because research shows that those features market well and will help sell handsets.
Why do you think that? Would you find it amazing that a group of volunteers could do a better job of developing <app or site> than huge companies like Microsoft or Hewlett-Packard?
Samsung has a different agenda with their software, so while it may suck to you and to the tech savy crowd it may bring Samsung new customers, or make them money in a different way.
If anyone from Cyanogen is reading this, I'd love to know if you guys have any plans to make CM the most secure OS on the market. I hate to make this into another NSA thread, but it would be awesome if CM focused on making a phone that I can trust to do my bidding instead of the bidding of some other master that didn't pay for the device in my pocket.
Moxie Marlinspike is working with them on their implementation of a more secure SMS app[1]. It's based on his previous secure SMS app I believe[2]. While I'm not sure I would use their CM Account¤ (also optional), the source[3] for the website mentioned in the article is up on their Github for anyone to critique. Site repo only appears to be the frontend though.
This is the first thing I started thinking when reading about the news. I have loved CM mainly because it was a way for me to dump the cruft on my device provided by the manufactures.
I had high hopes watching things like redphone emerge for the android space.
I hope that CM takes this in a direction to implement technologies that may liberate people from the fears of privacy loss.
Used Cyanogen since my old G1 ran it. It was a better experience than the stock android. Several android phones later, it started to be hit or miss. Really depended on what phone you had, and how supported it was.
My question is this, is Cyanogen, or any ROM really going to be relevant moving forward?
With all the changes to Android, moving towards one single API choke point (with Google Play). Aren't OS updates going to be less important? I'm thinking daily or more frequent ROM updates will get you very little in the future.
We're in an interesting spot, because typically GSF is
licensed to OEMs, not software vendors (us). But becoming
a legitimate business entity and partnering with an OEM
are the first steps to licensing GSF. Most of the
technical hurdles have already been overcome (passing
CTS).
Tom Moss, who is on our board, is the ex-head of Business
Developments and Partnerships at Google. He basically
drafted all the agreements to license GSF, anti
fragmentation clauses, etc. He'll be very helpful as well
on this front moving forward.
In the past I was always buying an Android phone first and check if CyanogenMod runs on it second. Several Android phones later and after holding back on older 4.x.x versions by countless manufacturers, I have learned that if an Android phone doesn't run CyanogenMod it's kind of worthless (from a hacker point of view).
These days I always check if an Android phone runs the latest version of CyanogenMod first and buy it second. My current phone is a Galaxy S2 (i9100) which according to CyanogenMod stats[1] is the second most popular phone. It runs Android 4.3 and you can update it nightly right from inside the settings. (Settings -> About -> CyanogenMod updates)
My first Android was a cheapy which I knew I could install CM on and get a stock Android feeling. I like a few of the additional touches that CM added though. Same as you, several phones later and I just rooted, unlocked and installed a JellyBean ROM, tried several variants but since my phone was no longer supported I opted for a custom ROM.
It was too hit or miss as you say, some deviating too far from stock Android for me. I've decided to go back to ICs when my device was supported by CM and a stock Android feeling...
They'd have to seriously impress me to go fully CM and deviate away from stock Android.
This makes me worried - what I really loved more than anything else about this project was that Cyanogenmod really tried to embrace the spirit of FOSS. Now, in reality, it doesn't really work that way because of proprietary drivers or google services, but the spirit was there...
Now I have my doubts about whether they'll keep true to that philosophy and spirit. I mean, I guess Red Hat has been pretty awesome for Linux, so maybe they'll go that route and calm my nerves a bit.
I saw someone ask whether they would remain open source in the AMA, and they gave kind of a murky answer which involved saying that certain parts would become closed and proprietary - really not a good start to this whole thing in my book. The whole reason I switched to CM in the first place was to get away from crappy closed software that I had little control over.
As a satisfied user of the latest CyanogenMod nightlies on a 2 year old Galaxy S2, this is great news! CM has breathed new life into my outdated hardware.
However, I have to echo the concerns in the post's comments about Gapps[1] (Gmail, Maps, Play Store). I wouldn't say the dependency is as strong as something like Zynga/Facebook, but if Google ever chose to lock Gapps down, I don't know that I'd stick with CM.
You can roll your own Gapps quite easy (before flashing a custom ROM) if at some point Google starts going after anyone and everyone distributing it (which is what they would potentially target if they did). I've done it before when compiling the AOSP source and there's nothing out yet because it's a new AOSP version.
Google currently has no incentive to lock down those apps. The one place that they have the least incentive to lock down is the Play Store, as CM users will still be sending money their way.
right but they do lock them down. As it says on the parent url, "Due to licensing restrictions, these [GApps] apps cannot come pre-installed with CyanogenMod and must be installed separately." Google heavily licenses GApps.
The incentive for them is to stop folks like Amazon installing Google Apps on their Android fork, and to encourage Chinese/3rd party/etc indie OEMs (the kind no-name tablets you see on eBay for $100) to establish a relationship with Google in return for then being able to supply GApps on their devices.
Holding the keys to GApps is what allows Google to give away the rest of the castle (Android) and still maintain control.
Except, much like Apple's walled garden approach, they want to ensure that what is in their app store works. Because if I pay for a product at your store, and it doesn't work, I complain loudly, demand things, etc. That's a headache to deal with, for Google, for the app developer and for someone using a stock phone looking at an app with terrible reviews.
So in a way, they do have an incentive to make sure that unknown builds can't access their app store.
It is a pretty easy distinction for users to make. Google can easily add a disclaimer that Google won't support or guarantee performance on non-Android devices. If they want to keep CM users happy about that they can offer trial versions so that users can test, or maybe even offer a money-back guarantee for an app.
Google does lock down the Google Play services, and their incentive for doing is the license fee$$$ they can charge to the compliant device manufacturers.
First of all, I didn't know you could just build an operating system startup (edit - in 2013). Second of all, they seem to offer some pretty compelling features, but more importantly, solve the pain point of supporting older phones that don't get the new Android OSs.
Question - they alluded to revenue generation in the article, stating that "If you’re the default OS on a device and you have 50 million users, there are a lot of ways to make money,". What are some of these ways?
The current friction point is loading the CM-OS onto your existing device (rooting, boot loaders, ./adb push, etc).
Offering a self-service tool that does it - either on the phone or via USB is something many will pay $20-$50 for I would guess.
Once you've got the OS on the phone, you can offer your own App Store (potentially along side the Google Play store), auto-install apps and perhaps set defaults like search engine to be google-competitors in return for revenue.
They might also sell direct to OEMS, especially smaller handset OEMS, so the OEM doesn't have to worry about participating in the Android ecosystem directly.
"CM is already in use on millions of handsets, and with the simplified installer that the company is announcing simultaneously with this financing, that number is sure to grow quickly. We believe that CM is poised to become one of the largest mobile operating systems in the world."
I am curious what the business model is for a company like this. Google and Microsoft use their mobile OSes to sell services while Apple and Blackberry use their mobile OSes to sell hardware. What does Cyanogen use its mobile OS to sell?
They use their mobile OS to sell... their mobile OS. I'm sure there are a lot of OEMs that would love a cheap, precompiled Android to ship with their new devices. Less work for them, and Cyanogen gets to deal with updates and porting.
Wasn't that kind of supposed to be the point of Android though? Granted some work is involved with getting Android onto various hardware, but one of the reasons cyanogen exists is because the OEMs started going way to far with their android customization.
Most companies are terrible at putting together Android support and waste money on engineering teams that don't know what they are doing. If they could buy something that's 90% ready to rock + support contract I'd think it'd be pretty compelling.
And with open source they wouldn't be locked into a single vendor.
I hope their business model is going to be something like BYOD (bring your own device) and "pay us a few bucks a month to protect your privacy". Or, alternatively, buy your phone/OS from us, we'll support it and provide security updates for the next nn months/until the next major release or until End of Life for the device.
Cyanogen could potentially sell this to device manufacturers if they can make it truly Android-compatible (including Google Play compatibility) without requiring license fees paid to Google, and that would reduce cost per device, thus give a competitive edge on device price.
custom apps. access to beta's. with 50 mill users, just charging $1/mon for beta access alone would be plenty of revenue. Beyond that, there could be plenty of cyanogen official apps that provide valuable services that users would pay for.
the biggest challenge will be divergence from google. Presumable google apps will not longer be supported (then again, they are available on cyanogen today but you can flash that on). Question is, how many android users will move to cyanogen as opposed to finding another developer who can someone clean up and build android for devices.
CyanogenMod's quality has gone down in the past few months. I'm running CM 10.1 on a Nexus 4. The home screen crashes occasionally and, after the most recent update, my ringer doesn't ring. Community QA is a tough problem.
Does anybody remember what the spat with that one Cyanogenmod contributor was all about a few weeks ago? I'm ashamed to say that I don't remember what it was, and now I can't find anything on it.
He was talking about about some kind of scandal as they were about to monetize the project.
CyanogenMod, the company’s free open-source replacement firmware, has more than 8 million users, CEO Kirt McMaster says. But that counts only users who have elected to share data with Cyanogen, he says, estimating that the true number is two to three times that amount. "There’s always been lot of talk around who’s going to be the third dominant mobile computing platform," says McMaster, who previously co-founded Boost Mobile. "Windows Phone would probably be number three now. If you look at what our actual user base is, we might be equal to or greater than that."
Right now GApps, which includes things like Play Store, Chromium, Calendar, GMail, etc., doesn't ship with cyanogen because some of the apps are closed source. You can install it separately after the install if desired, and for most folks I assume it makes that phone a lot more useful.
I wonder if cyanogen started to make enough inroads on android's market share if Google would try to make it more difficult to install GApps.
Cyanogenmod thrives based on a couple of very grey legal areas:
- first, as you mention, the ability to install the "Google Apps" binaries, which include basic functionality (Play Store, google service apps, etc). If Cyanogenmod is viewed as any sort of threat, Google can immediately crack down on this distribution. CM could still be useful without Google services, but it would be decidedly less useful.
- second, the ability to acquire and redistribute proprietary binaries which ship with devices and make them actually work. A majority of CM-running devices rely on binary blobs extracted from the vendor's software for hardware support; CM rips them from the factory install and redistributes them.
So CM in its present form is dependent on 1) Google's good graces in allowing gapps to be redistributed and 2) manufacturers looking the other way on the redistribution of their proprietary blobs.
The only conclusion I can come to is that CM must position itself as a non-threat to the business models of both manufacturers and Google. That will be an interesting tightrope to walk.
