Ubuntu Phone will include a Terminal application
wiki.ubuntu.com
wiki.ubuntu.com
Coupled with a powerful processor and plenty of spare solid-state memory, Ubuntu Phone will allow me to walk around and travel with my entire desktop environment in my pocket -- including round-the-clock access to everything available in official and third-party Ubuntu repositories.
(Have you ever needed access to an editor, language, database backend, etc. when you happen to be without a laptop? No problem, Ubuntu Phone will allow you to install and run all these things on your phone.)
I run Raspbian, another Debian-based distro, on my Raspberry Pi (ARMv6 CPU) and haven't run into any missing packages yet. Can't imagine there are many projects out there these days that are still architecture-specific, outside of compilers and JITs and so forth, many of which (e.g. LuaJIT, V8, LLVM) already support ARM.
It is really promising for the current software written against glibc and friends to have all these arm devices running homebrew ARM builds without many hitches.
Especially compared to a platform like, say, Windows, where even though there is an ARM version (Windows RT) which is extremely stripped down and awful on all fronts, there is no software for it because of API compatibility issues and the fact that without open source recompiling most of the Windows software catalog is impossible.
Almost the entirety of Linux userland software is very portable because the server space needs it to run on x86/sparc/power/itanic/whatever. There are exceptions but they tend to be rare and temporary, because hardware vendors don't want customers switching to x86 just because foo app doesn't run on their architecture, and for open source apps the hardware vendors can fix it themselves.
Meanwhile Linux is about the only sensible thing you can currently install on the vast expanse of old PowerPC Macs that can't run current versions of MacOS but can run current versions of Ubuntu or Debian, so any that get recycled into a personal server or a DVR box or whatever will have users pointing out any problems and requesting they be fixed.
And for the most part portability is portability: If you find x86 assembly somewhere and replace it with portable C and an ifdef to use the asm only on x86, or fix an endian issue, you haven't just fixed it for PowerPC or SPARC, you've fixed it for everyone.
This, incidentally, is why having a single architecture is so unhealthy: It promotes everyone forgetting about portability entirely, which prevents new, better architectures from taking hold merely because it's so much more work to port the existing installed base of software.
For a less pretty view (and one that will only make much sense if you already understand how the Ubuntu archives are organized (poorly)), check out http://ports.ubuntu.com/
really looking forward to it.
If I don't know what's available at my destination then I take the extra bag.
(although I use a 3G iPad for emergencies - it is an extra bag but only a small one)
(Also, who cares about speed? It's not like you are going to be rendering or building on it, so even the most modest of smartphones today will be more than enough. We no longer need the latest and greatest to be productive.)
I'd feel a little unwelcome if Hacker News was reduced to make sarcastic posts poking fun at people who want interesting tech on their smartphones.
Honestly? No. I can't imagine the agony of trying to type out a SQL query on my phone. I have a Macbook Air, which is light enough for me to carry around with me the vast majority of the time- it's great.
If I wanted something smaller I would perhaps get a tablet with a keyboard (or a netbook, if they weren't all terrible)... but doing any meaningful work on my phone? No thanks. I can barely bring myself to type out a reasonably-sized e-mail on mine.
However, the rest of the phone sucks, so I can't recommend it.
I use my phone right now for sshing into my home machines and getting analytics, benchmarks, error logs, making spot changes to files, editing users and groups, etc. If you have a smartphone with a foldout keyboard, you're set. The possibility of me being able to do this with urxvt and emacs running on the phone is just awesome. Interacting with postgres on the cli with my phone is a dream as well. Obviously writing out huge statements is a pain, but you're generally just running one liners through to find some info.
unless you meant an actual bluetooth keyboard, then its just a matter of foldable vs fixed and what you like more. My dad uses this:
http://www.quill.com/android-keyboards/cbs/283296.html?cm_mm...
for his nexus 7. Works out pretty well.
I used to perform most of these tasks on Nokia phone (yes, right!) - N900. Too bad, it didn't pick up on the market.
Wrong pick on the OS name, I guess :P.
You know where a market for disruption exists? The millions of iOS devices that are now being relegated to the trash bin of tech. Very soon everything from iPhone 3GS and back will be obsolete. You will not be able to update apps or OS. I have a nice pile of iPod Touch units that I can't develop for and will not move past their last update (was it iOS 4.2-something?).
You don't have to build any hardware. It's already there. And it's nice too.
Provide a path for a Linux (or whatever) phone to be loaded into these devices and you instantly create a market for probably hundreds of millions of devices that will either end-up in the trash bin or forever forgotten in a desk drawer.
If I were Microsoft I'd throw money at making W8 Phone run as a viable replacement for iOS on these devices. And I'd make the software 100% free of charge. Instant access to millions upon millions of customers who will be facing a very real choice of having to spend hundreds of dollars to get a new iOS device or, instead, try W8 Phone for iDevices for free. Yes, apps will be important. For some, this will not work. For others it would be a no-brainer.
The same would be true of Google/Android.
If I could load something else onto my half-dozen now-obsolete iPod's I'd do it. Yes, I know you can Whited00r up to a somewhat crippled iOS5, but that's not really a solution, certainly not for a hundred million devices.
iOS nothing has been released so you have to reverse engineer everything.
Never, ever, ever going to happen. Apple would provide the access required to do that, so you'd need to reverse engineer the device in order to install a custom OS on it- something that Apple would likely take legal action against.
Add to that the perception that WP is a dead-end OS for crappy old devices, and you're ruined your brand too.
And Microsoft would never, ever ever do something like that. I'm pretty sure they would face legal troubles if they did.
EDIT: Apparently saying something that goes against the current on HN will start a shit storm (downvotes, obviously).
Your logic is broken there somewhere.
In fact he specifically said "I fail to see how this will make the Ubuntu Phone more appealing to the average user." shortly followed by "my point was that this will remain a niche phone."
-Edit to remove somewhat unnecessary snark-
I don't think 'your logic is broken there somewhere' is all that snarky in the grand scheme of things, in fact it's a pretty clear statement of fact. The two things "terminal apps are not appealing to average users" and "this phone will remain niche" are absolutely not logically connected.
Since you can assume that the terminal will be never or very rarely used by average users you can make the advanced functionality available as command line applications and leave the most basic/common options in the GUI.
A naive user is far more likely to check a box and click apply by accident than they are to type "rm -rf /".
A good example is a modern version of Windows, the control panel is full of options,tabs,checkboxes etc that can do weird stuff to your system and consequently make it harder to use because (until recently with powershell) windows has never had a useful commandline.
IMO, to build a successful smartphone that is appealing to the average user, that smartphone will require a base of desirable applications.
I see basically two routes to attempt to achieve this:
1) Build a phone that is so desirable to average users that app developers will find it profitable to develop for it (eg, iPhone).
2) Build a phone that is so desirable to developers that many will pick it up regardless of the profitability of the market. Then hope that these apps are of sufficient quality/desirability that they become part of making the phone desirable to average users.
#1 is incredibly difficult to achieve right now due to the maturity of the market, with two major players. Even Microsoft is having a tremendous amount of difficulty with this route. For Canonical, I believe it would be effectively impossible.
With #2, getting to step 1 (modest developer adoption) is easy. Where it often breaks down, is in getting the developers to build apps that aren't just appealing to them (eg, more terminal apps, more ssh apps, more special purpose tools that appeal to geeks, etc). But modest developer adoption is better than no adoption at all, perhaps?
I honestly don't see Canonical having much success with this at all, but I don't think avoiding apps that are appealing to coders will help in their case.
Step 1: Get coders to use your device.
Step 2: Coders want their own apps to run on their own device, so they port them.
Result: Now your device has lots of apps and you've removed one of the most significant barriers to normal people using it.
If these really just acted like little linux boxes wouldn't that be a huge plus for corporate support? Being able to SSH into these phones to install VPN or other default applications instead of forcing the users/ui to see them? What about being able to 'flash' a corporate phone to a default install or manage permissions?
shrug just a thought
So they should really prioritize that part. Leave the developer/geeky stuff for later.
I would rather they concentrated on making a decent standards-compliant terminal emulator and didn't have to hack up the shell to accommodate it.
It might still be appropriate to have some terminal/shell distinction, but emulating an old text-only terminal imposes UI issues. For instance, my shell sometimes prints "you have new mail" to tell me about it. But my computer already has a notification system built-in - I don't need random text strings in my work area. I think the ideas in this mockup are at least moving in the right direction, which is using all of the capabilities of the computer.
Between Android, iOs, the remnants of blackberry and the Microsoft offering there will not be a whole lot of room for a new entrant that is not compatible with any of the above. A hackable phone that is not tied to one of the three largest 500 lbs+ gorillas sounds like a really neat thing to have.
Here's to hoping battery life will be acceptable and that they won't spoil it by tying it to Ubuntu services all over the place. Ideal would be NBR + phone capability, possibility to hook up an external display and hold the marketing.
It seems exceedingly likely that if more than a few dozen people use this, someone (if not Canonical itself) will produce an Android compatibility layer.
What I wish Canonical would do is to do it right. The naive implementation is to take the existing Android source and hack it into Ubuntu like WINE, which is almost certainly a lot easier when the source is already available. What Canonical ought to do is superset Android natively: Make all of the APIs available the same as they are on Android, so that you can take the source to an Android app and just run it on Ubuntu, but you can also start from there and make relatively few changes in order to comply with any differences in Ubuntu's usability standards and have your existing Android app running properly and natively on Ubuntu in very short order.
That would give Ubuntu all of the Android apps right away, and then give Android developers an easy way to fix any compatibility-related issues with running those apps on Ubuntu without making significant changes to their existing code.
Meanwhile now you have Android running next to POSIX and developers have the temptation to use all the traditional gnu stuff plus whatever Ubuntu adds to make it appropriate to the form factor, which has the developers pushing Google to put that stuff in Android.
There is a foreseeable future in which Android and Ubuntu end up as different distributions of the same basic OS the same as RedHat and Ubuntu today.
If what you want is to run Android apps, stick with an Android phone.
Where do you get zero control? Android is licensed under GPL and Apache. The only way Google "controls" it at all is by funding its development. If Google goes in a direction Canonical doesn't like they can fork it at any time with the only cost that they have to fund all future development themselves, which is apparently what you want them to do from the start.
>If what you want is to run Android apps, stick with an Android phone.
What if I want to run Android apps and gnu apps at the same time?
The market already well served by iOS, Android, etc., is people who want to terminal OUT OF their phone and onto a real computer someplace.
What are the real-world tasks you expect to do with the former that you couldn't do, or couldn't do as well, with the latter?
I'm also secretly hoping Ubuntu is paying attention to security with all the mobile spying going on and not just concentrating on making the latest supercoleslawesome twitter integration while they develop this.
No, this is not Richard Stallman speaking. I just want to keep my data on my own server, or at least have the choice to host it at a company in my own country.
That's why I count on free software to close the gap. Free as in: free of corporate interests, no need to make money, no need to provide access to intelligence agencies. I pay for the hardware, and I pay to transmit data. But the data stays mine.
http://www.jonobacon.org/2013/01/23/community-driven-ubuntu-...
It's a subtle distinction I wanted to point out.
Not speaking on behalf of Canonical, just in my own capacity.
"Someone would like to have a terminal on the Ubuntu phone" would be more accurate.
There are plenty of "official" blueprints that never made it into the product.
And there are still those awful rumors of Canonical writing their own windowing server and compositor. They spread themselves so thin sometimes that you rip the bread apart.
What I'm more worried about than the choice of html engine, is if they are going to use chrome or firefox webapp manifests, because they are incompatible and one of them is necessary for first class web applications.