Hacker News new | comments | show | ask | jobs | submit login

Yes, it's true the things listed are bullshit. Yes, I guess we can take it or leave it.

Sometimes the "take it or leave it" position ends up being presented in a way that it is framed that those of us who choose not to partake in the rape of our liberties are backwards luddites (non-facebook member here, also got their domains blocked to kill their insidious tracking). I hope that criticism of those who say "no" is not where this article series (if that's what it is) is headed.

They're all bullshit except the 'Android is open' thing. That's just a plain fucking fact.

Despite the incessant barrage of snarky 'open' jokes from professional Apple frother John Gruber (typically redeemed somewhat by being funny) and M.G. "Mini-Gruber" Siegler (not so much), you can go download the fucking source code:


Sure, tons of manufacturers will sell you junky locked down Android-based phones (which they can do because... tada... it's open), you can also just go buy the unlocked phone that Google offers, and install whatever the hell you want on it.

And if you really want to root your phone and install your own build of the OS, it's basically as easy as ↑↑↓↓←→←→BA and boom-you're-done.

For any reasonable value of open, Android is. Claiming otherwise is just some weird kind of (apparently contagious) gobbledygook nonsense making the rounds among people emotionally invested in iOS. (Which I do use btw; although Android is open, it is still not good enough for me yet.)

Except for a really long time, ie between Android 2.x and 4, you couldn't download the source code.

Edit: why is that being downvoted? It's a fact. HN can be a strange beast at times.

"A really long time"?

Seriously--I know it's been a busy year--but the nine months (from February 2011 to November 2011) during which Android was being used in certain tablets without being open sourced hardly qualifies as "a really long time", especially in comparison to operating systems which have never been open.

I'm not sure how your'e able to say nine months isn't a long time (especially in this industry!) with a straight face.

Also, I subscribe to a pretty liberal interpretation of "open source", and "source will be downloadable at an undetermined date in the future, if we feel like it" really doesn't qualify.

There is, of course, the colloquial expectation of open - the implication that Google has never shied away from is that openness means choice. Of course, even with the source code, the vast majority of Android headsets are locked down - the openness does nobody any good except people who build custom distros (that normal users can't even install themselves!).

The "open" argument, while arguably, technically correct (the source is available... kind of), fails to meet the bar for what techies consider to be essential to the concept of openness, nor does it meet the bar for what laymen would consider to be openness.

For me an open platform is one that can be used, maintained and extended in ways the original platform owner actively disapproves of. Android fits that description, because you can in fact fork it and do what you want, even if its anti-google (like the kindle fire).

I consider android similar to x.org when it was called xfree86, and oo.o before it became libreoffice. It's definitely open source, but it remains to be seen whether the project will develop an open culture. If not, a hostile fork will happen at some point.

Six months passed between the announcement of the iPhone and its release. More than a year passes between each iPhone generation. 9 months, even in the mobile world, is just not "a really long time", and we're talking about a version of Android that only ran on tablets (which was, incidentally, why its source wasn't released on schedule).

Calling Android "open" is not just "technically" correct. It's really true: you can download the source to Android and build your own versions. Cyanogen does this. Amazon does this. Multiple handset manufacturers do this. Anyone can do this, and many do.

You don't address the parents argument or mine at all. Motorola Xoom owners could not download the source code for their OS for most of its existence. Neither could Cyanogen. for example, get source for ICS beta releases, which means the many obvious UI and UX issues in the initial release could not be fixed by anyone but a small portion of people inside Google.

That's true. But still, Arment posted this today. Yesterday, daringfireball.net linked to a baseless Siegler snarkophany in the same vein. Etc etc etc.

(Edit: No idea why your factual comment's getting downvoted, though. It'll hopefully get rectified as more people roll through this thread.)

It was being downvoted because it simply wasn't factual. Android was not closed "from 2.x to 4". Honeycomb (3.x) was first released on devices in February 2011 and its source was opened in November 2011. 9 months is not a "really long time".

"Android is open" is BS for the plain and simple fact that (time delayed) source code availability is of no consequence to the vast majority of Android users. They say it as though it means something, it means nothing, rather like water vendors claiming "100% fat free" on a bottle of water.

Marco didn't say the claim was factually incorrect, he said it was BS, it is.

Although I'd love to have a few bottles of your fat-free water, it's a bogus analogy. It's more like claiming "100% trans fat free" on a package of cookies. That the vast majority of eaters don't care about or even know what trans fats are doesn't make the statement less true, or less relevant to those who do know and care.

Android's openness is consequential to those who care about openness. And almost every Android user I know (yes, all fairly technical) likes that aspect and indeed takes advantage of it.

To such people, the open aspects -- both the fact that the OS is open source, and that they can install whatever apps they want without having to find a buffer overflow bug in the OS somewhere to exploit first -- are attractive.

The "vast majority of users" aren't the arbiter of whether or not something is consequential to me, or to the other people who generally prefer their systems to be more open.

Not really. Android's "openness" doesn't include the ability to actually modify the software running on your device, which means that on its own it's useless to everyone except hardware manufacturers. It's only on the rare occasions when those manufacturers actually choose to do more than is required on them and open up their hardware to modifications that "openness" is of any value at all.

> doesn't include the ability to actually modify the software running on your device

OK, I am uncomfortable with my weird position in this thread as the Android Avenger (especially seeing as how I don't even use Android except to fuck around), so I apologize in advance for the breakdown in my civility function.





I don't really get what you're saying here. Some Android devices have locked bootloaders, but that doesn't affect the openness of AOSP. If anything, you're making an argument for GPLv3 which would require Android OEM's to allow for open bootloaders and (S-OFF (htc)).

But I can assure you, my Galaxy Nexus is running a CM9 alpha that I built from source last night.

Replace "some Android devices" with "the vast majority of Android devices". That is Marco's point.

All HTC devices since September, most devices from <=2010 (and many, many more with community support); all Samsung/LG devices; some early Moto devices (and all if you don't count the kernel via the community) can have custom software installed.

And if that's Marco's point, it's pretty damn poor. How does what Motorola does with the bootloader of their phone have to do with the code that Google puts out exactly?

As has been explained many times in the past, Android's relative openness compared to iOS at the source code level does not translate to measurable benefits for the majority of users because of carrier/manufacturer lockdown.

Rooting is a security exploit, and I don't consider it relevant to whether the OS is practically open.

Now you're just shifting around your position. I thought it was about the unlocked nature of bootloaders? Yes, in my last post I included some that require "root" but, #1, not all of those root methods are through user-facing security exploits, and #2, most of them allow custom software out of the box from the manufacturer (specifically without rooting or really any work at all), I noted the ones that require "community" support. Your characterization is disingenuous.

Now it's about the openness at the source code level? I think you should look up the CM statistics. There is a measurable percentage (or may tenth of a percent) of Android users that run custom software, so I benefit from that. I and others will only buy devices that will have CM support.

Further, the open nature of Android absolutely trickles down to the users. There are a dozen different (and with different physical form factor) Android models that are free on contract to buy. There is only one WP7 that I'm aware of that that is true for. Lower development costs are passed on to users, I've discussed this further somewhere else on this thread.

First, by that standard anyone's claim of openness is BS, because most users of any open source project don't actually use the source themselves.

Second, Android is "open" in ways that have nothing to do with source availability. Even with a closed-source Honeycomb device, you can run apps from arbitrary sources, access the filesystem, replace OS components like the keyboard and launcher, and develop your own apps without asking (or paying) for permission.

Really? An operating system that is cheaper to OEM manufacturers (and thus cheaper per-device costs) as well as a vastly expanded variety of physical hardware isn't a benefit?

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact