CyanogenMod definitely doesn't belong in PlayStore. Having an app like that on Play Store enables many non techies to try it out without understanding repercussions. Who do you think these people will call?
Here is the order:
1. Carrier (Walk in Store)
2. Play Store Support Number (Surprise Surprise Google actually provides phone support)
3. ????
The best I can tell, this is like all of those teenagers installing Napster, Kazaa, or whatever on grandma's PC. Feds show up and "OMG, fines!? I didn't know! I'm computer dumb. I paid $10/month for all nsync!"
I love what CM is doing, but if lay persons and even "experts" are getting burnt, who is responsible? Who gets blamed? Google? Because that's the first place consumers are going to go or to distributors or manufacturers who are going to also going to defer to Google.
In fact, the ability to install something outside of the Play store framework seems like one good test for whether someone is ready to make the jump.
I have no idea what the percentage of people having problems is, but it certainly happened to more than one person...
When problems creep up on a product like this, the user will get shafted badly in the end. Google will refuse to help (cue complaints about Google allowing potentially destructive apps in Play store) as they don't support either the software or the phone, the manufacturer will also wash their hands off it, CM does not provide support for this from what I understand.
I have been using Android phones since 2010 (rarely on a stock ROM) and even now flashing anything is done with a prayer on my lips. There are just too many things that can go wrong with the way it is done right now.
I concur with justin66. The necessary skills to deal with something going awry would seem to be good criteria for searching out Cyanogenmod for themselves and making the jump rather than casually clicking on "install" in the play store.
One downside of CM raising money is that they will have to go for high impact initiatives as a must-have than as a welcome side-effect of the stellar work that they have already been doing. Guess they will eventually go the route of CM phones or phones that come with CM pre-installed. The money they raised does not make sense for building a purely software offering.
An app that bricks even <1% of the devices just doesn't belong in the app store where it gets exposure to lots of non technical users.
Why doesn't CyanogenMod replace or repair bricked devices, when their software caused that problem in the first place?
Why should Google or the device manufacturer pay for it?
Users installing it outside the market though as it was always installed have no more right to Cyanogen giving them a new device than Debian or the Linux Foundation would if you destroy your PC installing Debian Linux.
https://github.com/CyanogenMod/android_packages_apps_CMAccou...
Aaaaand this is why making root and roms easy access is a terrible idea. I miss the days where knowledge of at least adb and fastboot were required.
It's not politically correct to say, but it's true.
I was excited when CM first announced their plans but I had assumed that, if they were going to release a product aimed at the average consumer, they would first ensure it was suitable for use by the average consumer. If those two Reddit threads are anything to go by, they didn't. That is a massive and potentially fatal mistake on CM's part.
Before they launched the new installer, the traditional CM user needed to have a minimum level of technical chops to even attempt to get CM onto their device. That meant that, if something went wrong, they had a chance of saving themselves. By contrast, the average consumer is non-technical and if your product is aimed at non-technical users, you need to make fucking sure that it's literally idiot-proof because Murphy's Law applies.
Code doesn't exist in a vacuum. It forms just one part of a broader system that can include other code, hardware, network connectivity and people, any of which can behave in unexpected and irrational ways.
At the risk of generalising slightly, developers tend to (a) think that things are going to work the way they expect them to, (b) expect people to behave rationally, and (c) assume that they don't have to worry about dependencies that are outside their control.
Reality is that (a) things often don't work as you expect them to, (b) people (especially non-technical ones) don't always behave rationally, and, (c) if you are relying on something outside your control to work properly, you better make sure that the code you write is capable of handling errors, failures and exceptions.
I constantly hear developers using the words "should" when talking about issues. The user should make sure that the battery is fully charged before starting the install. $3rd_party_dependency should work as advertised. The user shouldn't pull out the USB cable in the middle of an install.
Assume that things aren't going to work as they should, even if you can't imagine why they wouldn't. Why would the USB connection be interrupted in the middle of an install? Who the fuck cares why?! The user might be an idiot. They might clumsily knock the phone off the table. They might have a two-year old child who wants to play with daddy's phone. It doesn't fucking matter why or how it could happen, just assume that it might and figure out how to deal with it if it does!
In fact, assume that your product is destined to be used by someone who is going to do their darnedest to fuck it up. I used to answser the tech support line for an ISP back in the '90s and the one key thing I learnt from that experience is that users will do the most stupid things imaginable.
Any time you use the word "should", it's a fuck-up waiting to happen! Exception-handling, exception-handling, exception-handling! Not just in code but in the processes and procedures that surround the product that contains your code. Test it to death and then test it some more. Get your non-technical friends to try using it and watch them. Any time they need help or prompting, take note and change your product accordingly.
This is why empathy is such an important quality in product managers. Empathy is what gives one the ability to put oneself in another's shoes. Imagine I'm a housewife whose technical prowess extends to putting on the dishwasher. Imagine I'm a college student who fancies himself as the next Mark Zuckerberg and thinks that coding is writing HTML and CSS. Imagine I'm an impatient high-school senior who wants to differentiate himself somehow from all his friends who also have Android phones. If you're releasing a product via the Play Store, it needs to cater to all those potential users and more.
CM need to get on top of this. There's a massive difference between an obscure piece of software aimed at geeks or hobbyists, and an Installer aimed at the mass market. Even the rumour that the installer is bricking phones could do serious damage to CM as a brand (not to mention put them at risk of litigation - how much do you want to bet that some shark isn't already thinking about a class action lawsuit?). Just look at how much focus and effort Elon Musk puts into rebutting any suggestion that Tesla cars are prone to fires. He knows that confidence is crucial when you're marketing to consumers.
CM need to adopt the same mindset.
What are they doing with those statistics? What could they do with them in the future? What could they be forced to do with them?
If it's so harmless, why isn't it opt-in?
> Every single website in the world does it
Obviously not. Websites don't have access to hardware the way CM does.
If you would quit trolling for one moment, you could easily find your answer[1][2]. It's no more likely they will get away with sneaking something in than any other Linux Distro would as there are many eyes, including myself, looking at their source as a hobby. There are also those that dislike Cyanogen and would love to see them slip that monitor their source constantly. That is a good thing, and serves a purpose, unless it's a false alarm that "cries wolf" and gets repeated so much, everyone just ignores it when there's finally a real issue. Think about that please when you keep dragging up this current non-issue.
They also give a preview of what they are collecting when you go to the "opt out" screen on your device and the source is on Github. If you're super paranoid, compile your Cyanogen yourself minus the stats instead of relying on what they give prebuilt.
http://www.cyanogenmod.org/blog/cmstats-what-it-is-and-why-y...
And now, the question nobody will answer: Why isn't it opt-in?
E.g. Rom managers such as this one. https://play.google.com/store/apps/details?id=com.koushikdut...
> I don't know WTF the cyanogenmod team was thinking, non-technical people messing with CM will just lead to negative press.
I know atleast three people who tried to install Linux and formatted their Windows partition in the process. So you mean Linux CDs and DVDs should never have been bundled with magazines?
I don't think there's any danger of grandma seeking out CM and going through the process to install it. CM has over 8M installs, having a handful of bricking incidents is nothing at that scale.
Hopefully this has nothing to do with CM making its own handsets http://www.rethink-wireless.com/2013/09/20/cyanogen-aims-cre...
Firstly, a PC != mobile phone.
This isn't a bloody OpenMoko phone - this is a working consumer device that's sold to you as a "dumb" device. It's like buying a television or DVD player, flashing some weird third-party firmware, then taking it back and saying, gee, it just stopped working...
Or flashing your car's ECU chip.
The manufacturer is certainly not expected to fix your mistake. They might for goodwill, but they have no obligation to.
Yes, there is certainly a danger in the proverbial grandma flashing her phone (why do we always pick on grandma's? I mean, geez, Hopper Grace was a grandma wasn't she? Lol).
I've seen countless phones soft-bricked by bad flashes. It's like upgrading your BIOS - if you get unlucky, or the power goes, you're SOL.
Google's just being responsible. The whole point of Android is that you can sideload ADB's easily.
Think of it as a entry test - if you can't load a non-PlayStore ADB onto your handset, you have absolutely no business installing a custom ROM.
Put it this way - you ask me how to flash your car's ECU.
I say, well, firstly, you need to pop open the bonnet and read me some numbers. And then you said to me, "Gee, how do I do that? I've never popped a bonnet open".
My next thought would probably be, "Err, are you really sure you want to flash your car's ECU?".
Cheers, Victor
That particular ROM Manager requires having root and an unlocked bootloader prior to use while the CM installer does not. That alone is a big difference, since it implies someone should be aware slightly of what they're doing if they have gotten past the unlocking/rooting stage.
And those users are surely not expecting that they are going to pay themselves, since they installed the software from the official Google Play store.
What CM does (doesn't matter if intentionally or not) is externalizing the cost for support to Google and the device manufacturers, since they themselves are not willing to replace or repair devices that have been bricked by their own software.
Also the process wasn't foolproof even for people capable of dealing with technical issues[2].
It isn't polished enough to go in the play store IMO. I don't know WTF the cyanogenmod team was thinking, non-technical people messing with CM will just lead to negative press.
1: http://www.reddit.com/r/nexus4/comments/1qkfaw/cannot_uninst...
2: http://www.reddit.com/r/nexus4/comments/1qi5lw/official_cyan...