Re: "Does this mean Linux gets a real chance on mobile?"
Most Smartphones run Linux, I know mine does.
Canonical founder Mark Shuttleworth is no stranger to exploring rarefied territory. The man has, after all, been to space. His interest in new frontiers means Ubuntu, the Linux distro he created, is also poised to make a great leap - to go where no Linux has gone before. Shuttleworth's plan to take Linux beyond the desktop …
Linux kernel and tools maybe, but not the Linux UI.
The only reason Android succeeded was its open-ness and because they didn't try to use KDE or Gnome as part of the interface. It's clunky enough on a desktop, it would suck massively on a phone.
Yes, I know it's an open debate with arguments either way but really.
The Android kernel is still significantly different from the Linux kernel in terms of architecture changes.
To even just run on most devices you need OEM crap or even some port just for x86.
Linux kernel <> Linux operating system (GNU/Linux).
Android is not a Linux distro just as much as any custom or embedded mod is not a distro. Next thing people will say OS X is a FreeBSD distribution....
So yes the article is right to ask ""Does this mean Linux gets a real chance on mobile?"
>The Android kernel is still significantly different from the
>Linux kernel in terms of architecture changes.
It's not. The fact that a vanilla Linux userland will run on top of an "Android" kernel proves that... oh and the Android patches are being/have been merged into the mainline kernel.
Openness gets you something among those who are followers of license politics.
It was an os that could take on iOS and also mop the floor with Blackberry and the antediluvian Windows phone os. At first it was at minimal to no cost and then it cost a little bit (with the check paid to Microsoft). So, manufacturers used it. Google had a revenue stream that increased with usage and did not depend on license fees, so it can subsidize the carriers and ease their ability to say yes when the manufacturers pitched the Android phones to them.
It also had to work reasonably well and Google had to show that it was going to work on its sub-optimal parts. Hardware getting faster so the cost of the vm and memory management effectively disappppearing for the user also helped.
There were other mobile oses that were arguably more open and they went nowhere. I think Meego would be a good example.
So, no, openness was not the only reason. I'd argue it wasn't even the most significant reason.
Linux doesn't have a UI, unless you count printk.
The rest is actually completely unrelated to Linux, and comprises a boot manager, an init system and userland tools, which together with Linux you could call an "operating system". Although all of these components, including Linux, are interchangeable with alternatives. For example, Linux can be replaced with Mach, Hurd, Darwin and others, while the userland tools can be either coreutils, busybox, toolbox or a variety of alternatives, and choices of init systems range from OpenRC and systemd to upstart and initng.
Everything after that is purely optional, again interchangeable with alternatives, and also has nothing to do with Linux, but typically comprises a display server, a login manager, a window manager, a file manager and up to about a million applications and games, in various formats, including ELF, Dalvik and others.
The bit that you call the "UI" is probably the window manager (or sometimes a monolithic "DE" that combines several of the aforementioned components), which again has many alternatives, including Gnome Shell, KDE Plasma, Mate, Cinnamon, Sense, TouchWiz, Trebuchet, (stock Android) Launcher, Unity, Openbox ... the list goes on, and on.
Referring to all that as "Linux", then claiming it only exists (or only matters) on a limited and rapidly dwindling range of hardware dubbed the "desktop", is utterly devoid of any meaning. All that really matters is that it's Free Software, and thus can run on any architecture and any particular combination of Free Software operating system components, whether it be GNU and Darwin on PPC, Android and Linux on ARM, Cygwin and ntoskrnl on x86 or any other combination.
Provided the necessary dependencies are included, Free Software can run anywhere. It may interest you to know that there's an X server for Android, for example, which opens up a whole new range of possibilities.
Or as someone else put it, far more succinctly ... what Linux UI?
> Linux doesn't have a UI, unless you count printk.
> The rest is actually completely unrelated to Linux,
Yes. But all of that stuff can be built for Android just as the Mac desktop could be built for PhoneOS. There's no reason that a phone couldn't run a conventional desktop if the right inputs are attached. If the device is fully in your control, that becomes a lot easier.
People already jailbreak their devices or install SSH servers on them in order to treat them more like old school Unix servers.
Running conventional desktop software is just the next logical step.
An X server is perhaps the wrong thing for an Android device. However, X libraries would allow any other X server to connect to your Android phone and login to it graphically from across the network.
Surely, the *correct* place to put the X-Server is on the Display Device, which might be your Phone/Tablet when used *non-docked* but might also be your TV when *docked*, then your compute device (phone/tablet) can truly work independantly of the visual interface. Given the cost of the little *Chrome Cast* device or any number of other HDMI ARM Smart TV/USB adapters, it would add ~ £30 to the cost of the display unit.
Cool ;-) we just re-invented the "X-Terminal" the original "Thin Client" that worked perfectly well right up until the industry got all hot under the collar about "Thin Clients" - all good things come around, and around, and around.
"Linux is nothing but windows or mac with a different paint job"
how much more retarded can clueless nonsense get...?
Linux is a fundamentally different operating system with its desktop purposely made to look similar to windows and mac to make it less alien to potential new users.
So, what are differences that would actually matter to OS users?
- A file system that autonomously checks the health of your data (rather than windows never knowing if anything is wrong, till you actually try to read the file.)
- no corporate shackles or taxes, no hidden backdoors for NSA or corporate buddies.
The rouble is that it is possible that convergence will occur on other platforms long before Canonical delivers a usable product. Despite Shuttleworth's evangelism, they neither have the marketing power or a cohesive developer base to deliver the product in time, while others will simply borrow some ideology and claim it as their own. The interface-that-used-to-be-called-metro is already on PCs, XBoxes and Phones
Just to be clear and get back to basics, this isn't about products. Nobody has any desire to deliver products. At this moment in time, everyone wants to be in the service business so it's all about "_aaS" not products. On one side you have trinkets* that are usually called products purchased by the consumer and these trinkets are designed to lock the consumer into particular services. The other purpose of the trinket is data collection and intelligence distillation. This intelligence is then provided as a service to business.
So what you have is a trinket that delivers consumption as a service to the consumer and consumers as a service to businesses. We might get actual products when we've decided we've had our fill of iFartFacebuchTwaddle and are going to get back to being productive. Unfortunately it looks like that time isn't coming soon given that it seems folks feel they do enough working at a job they hate and need their "me time" to be spent in a semi-vegetative state rather than attempting to improve their situation and get a job they enjoy. Granted, it's probably hard to get paid to spend time in a semi-vegetative state and I suppose it's healthier than the drugs and alcohol that was the standard bearer of vegetative states for the last 50 years.
*trinkets because I don't really care if it's an iPhone, Chromebook, slablet, gaming console or whatever else is currently in or coming down the
sewer pipe and it seems appropriate.
Another interesting thing, which happened in 2012, was that Mint and Mageia bypassed Ubuntu as the most used Linux desktop distros, and Ubuntu became the only Linux distro a large part of the community actively dislikes.
The thing is that just like the interface previously known as Metro, Ubuntu's Unity interface sucks gigantic donkey balls on desktop, and for the exact same reasons: they're dumbed-down touch-centric interfaces which expect you to search for everything, and designed for cellphones and tablets, not to be easy to use with mouse and keyboard.
Both companies are intentionally throwing the desktop under the bus, motivated by the common wisdom that the PC is dying, and the future is cellphones and tablets.
Or I should say WERE intentionally throwing the desktop under the bus, Microsoft met its Waterloo with that strategy, currently they're trying to undo the damage and figure out a new way forward, and heads are rolling: not just Metro chief architect Sinofsky but even major shareholder and grand fromage Steve 'Monkeyboy' Ballmer himself are history.
The moral of this story is that it's not a good idea to flip off the core business which feeds you. And I'm not sure Canonical has really got this memo yet.
> But unlike Microsoft, Linux desktop users may be a dying breed.
The key issue is if desktop users of ALL OS's are dying breeds. Microsoft developed Win8 because they think the desktop is dying and wanted to accustom their users and developers to a mobile interface while there's still time.
The reason the industry thinks the desktop environment is dying is clear enough: PC sales are flat, mobile device sales increase exponentially, and technology has progressed to a point where stationary PCs are ridiculously overpowered for the vast majority of users needs (email, surfing, watching video).
Personally I think the analysis is partly wrong. PC sales are flagging not because they're being replaced by cellphones and tablets, but because PC is now a mature technology similar to a fridge or TV: because they're so overpowered, there is no need to replace them before they break. A five year old PC can do all you require of it, even play the newest games, so why replace it? Cellphones and tablets are immature technologies and still evolve rapidly, and so tend to be replaced yearly.
Unlike Canonical and Microsoft I don't think it's an either-or situation, I think in the future people will have BOTH a mobile device and a PC. You can't play demanding games on mobile devices, and it's a pain to create content (e.g. write texts, use spreadsheets) on mobile devices.
On an even more personal note I also think that the "PC flatlining" dynamic will be upset quite a bit by next-gen games and 3D headsets due to be released the next year, all of which can only be used on new PC:s.
Much sense there, I can also see the 3D headsets or some other developing technology propping up the desktop model for a while.
My uses tend to favour the desktop but I'm keen to move away from WinTel due to the obstructive licensing, energy inefficiency and just plane 90's thinking.
I want to use software locally, on a very lower power APACE (apparent physical interface) but I need a screen that large enough that I probably don't want to hold it.
Tablets are OK but (for me) fall down when trying to concentrate on what is being created beyond the interface and gui. A screen hovering in front of my vision with a neural entry system would be nice. Maybe the PC with morph into a server in the house that feeds physical or implanted headsets so we can see the creativity not the gui. Right now I'd like to move towards Arm type processor efficiency but have something like a PC lurking in the house to offload those heavy jobs when required, I certainly don't want to be web browsing on a PC that burns 150W insists on rebooting at the worst possible time with a hundred and thirty year old keyboard system.
A nice observation. I know I have phone+NAS+PC+tablet. Ironically, I got NAS+tablet about the same time, so perhaps this is simply the market maturing. i.e. something to "serve" media...
But something I acquired from the lab (and others may have from gaming) is the idea of big desktop with multiple monitors. It took a while but I think that is the point of KDE activities. You can have "workspaces" with different setups, that you can chip away at problems... It is more refined than normal Linux desktops.
For those who run windows, this like having multiple PC's with specialised content for each, but each accessible from the main interface.
I would hazard a guess that the sheer diversity of components has probably allowed consumers to find products to serve their niche? E.g. If you have an Iphone, you are highly likely to have a Mac, and perhaps an Ipad. That is what Apple would like. But in reality phones come on contract, your home PC probably doesn't change for 5 years, and tablets really are a refinement of "browsing" but not "creating" type of activity.
Didn't the article mention docks as a possible way for mobile devices to bridge the gap and become PCs (complete with mice and keyboards) one they become powerful enough (almost there IMO--more and more full-fat retail games are appearing on Android).
> The reason the industry thinks the desktop environment is dying is clear enough:
That is a straw man argument, 'the industry' doesn't think that "the desktop environment is dying", it thinks that the desktop _market_ is declining (for the reasons that you point out).
> Microsoft developed Win8 because they think the desktop is dying
No they didn't. They see a decline in _sales_, not a decline in usage. The reason they put Metro onto Win8 is because Windows Phone was not selling, plus they were working towards tablets with WP interface. Consultants told them the lack of WP sales were bacause of unfamiliarity with the UI. Bu putting this UI in Win8 they were going to make it "the most familiar user interface" (see MS announcements) and then users would _demand_ the same UI on their phones and tablets.
""" Myerson expects the coming release of Windows 8 for PCs to give the software giant's mobile platform a huge consumer awareness boost.
"The most familiar user interface for people worldwide will be Metro very soon," Myerson said. "And that's just great for the phone business."
Microsoft needs to get that message out to consumers so they understand why Windows Phone is a better choice for them, Myerson said. """
Microsoft always intended that users would have (buy) all 3 types of devices: desktop, tablet, phone; and that all would have the same UI. They thought this would be a sales point against Apple and Android.
And thats it in a nutshell, PC Hardware is more than sufficient for 90% of all users and has been for some years now - the SW (OS / Apps) can't fully exploit fast Multi Cores CPUs. MS did too good a job with Windows XP, an OS that started off as being good enough at launch, then through various Service Packs became too good (and stalled MS's 3/5 yearly OS Upgrade cycle that was enjoyed from Win286/386/3.0/3.1/95/98/ME). In previous years, a household might have 1 PC that the family would *timeshare* on. Today, everybody has their own device (Smart Phone / Tablet) maybe even a couple of devices each. MS have recognised this need to be in the mobile market, but they messed up. The Win8/Metro is sufficiently different from the Win95 derived interface that has lasted right up to 2013 with Win7 that users are now thinking "why should I spend all that $$$ to learn something completely alien when I only need a cheap Tablet / Smart Phone to visit eBay / Facebook / YouTube. In fact the interface on that new-fangled Tablet (which costs about the same as a laptop) looks like the one I already use on my phone, so I don't need to learn new stuff - so to hell with MS and Win8".
So it's like 0.00003% of total numbers of desktops used worldwide comparing to 0.000025% ?
You know that anything less than 5% is statistically insignificant, don't you ?
To be crystal clear about my point - no one in the world give a slightest damn about Linux desktops, let alone their sandpit wars. Get over it and buy a Mac if you can't stand Windows.
Well that comes across as a bit of a troll...?
It is difficult to get accurate statistics as PCs are sold with Windows pre-installed. Everyone knows that Microsoft would hate it if you saw the price as PC $500, PC+OS $600. Once you get a choice, you may not choose Microsoft. Statistical significance depends upon what you are trying to observe, and things do change. There was a time when android was %0....
The trojan horse here is Android. As someone pointed out , it is under the hood a but not the gui.
Its just a matter of time before "something" gets some momentum as a gui.
I think the difficulty many of us have is the way in which MS of Ubuntu seems toxic to collaboration.
Many of us think that the technology on the average Linux desktop just works just fine, and that is what we want. The ability to add new features gracefully.
"You know that anything less than 5% is statistically insignificant, don't you ?"
Is this some new use of the word statistics that I am unaware of ? Even if desktop Linux was 1% that's still 10Million+ users that got off their arses and installed it - THAT is very significant
"Is this some new use of the word statistics that I am unaware of ? "
Nope, its the well regarded 2 standard deviations from the mean contains 95.45% of all observations in a normal set - so strictly he should have said 4.55%
"Nope, its the well regarded 2 standard deviations from the mean contains 95.45% of all observations in a normal set "
That's assuming a normal distribution.
In any case the observations outside 2 standard deviations aren't necessarily insignificant esp. if they are real which in the case of Linux desktop use they are. No, the OP was quite wrong in his use of statistically insignificant. Much better to use a small minority.
> That's assuming a normal distribution
His error is much more fundamental than that: he's confusing measured amount of change (e.g. "the poll showed democrat support is up 1.5%...") with confidence interval ("...this change was statistically significant...") or possibly with margin of error ("...and well above the margin of error of 1%").
(And if someone wants to argue about the difference between confidence interval and statistical significance in my example, please insert an icicle in your orifice of choice.)
Here endeth the OT.
"To be crystal clear about my point - no one in the world give a slightest damn about Linux desktops, let alone their sandpit wars. Get over it and buy a Mac if you can't stand Windows."
And if I hate Windows AND Mac AND routinely work in graphics so need a GUI no matter what?
Why do you hate Windows and Mac though? OS's are there to run programs - they mostly all do this in the same way, especially if you use WIn8 in Desktop mode or stick with Win7.
I am not a Mac expert at all but it is certainly seems a reliable system with most programs available. I also imagine there are almost no functions possible that Windows cannot perform reasonably well with some program or another.
So, it either comes down to some extremely specific application that you simply must have that will only run on some Linux distro - which seems highly unlikely - or you just *hate* these other OS's for irrational reasons. I will go with the latter reason - when was the last time you used any of them for an extended period?
I, for one, have used Linux extensively. It is a very stable desktop, hardly ever needing a reboot. It is, however, not an easy system to use - one has to continually try to find out how to do things, find things using searches, ask people for help, and use the command line rather a lot.
The graphical interfaces available for some command line programs are obviously basic, not intuitive and barely make it easier than the command line apps and a long, long read of the help system to ascertain how to use the program you need.
Whilst Windows does have very complex command line programs, they are rarely needed, the graphical interface is the norm, looking at the performance apps or the event viewer for instance allows for a huge amount of information and detail, accessed easily and graphically.
I am well aware that Windows requires updates regularly to remain fully secure - this is a feature of its popularity rather than any inherent insecurity or unreliability. Updates are nearly always painless and a monthly reboot of a system that takes 30s to come online again is hardly an issue.
I am often a professional user of Ubuntu and often found it got in the way of producing output. Since compiler work is often offloaded to servers etc. I am perfectly happy running emacs on Windows, accessing files elsewhere and performing makes etc. on the server (often a Linux server - most of my jobs involve embedded work and Gnu).
Still, Linux does save some money at the outset - immediately lost of course if I have even a one hour delay because of it. It is excellent for servers, controlled by people who live and breathe the system, without a GUI to speak of and where Virtual machines can be used to further increase reliability. Mind you, I hear Windows Server can do this too and make it easier to run Windows client software.
There is one thing better than free and that is Easy.
"You know that anything less than 5% is statistically insignificant, don't you ?"
I suspect you are confusing the concept of a significance level, essentially a probability, with a simple proportion of the population.
To return to the main subject: I suspect Canonical are 'going for it' with mobile devices in the expectation of more powerful processors becoming available. I really like the idea of a smartphone that I can dock and get a desktop with Office package and printer drivers on. I shall consider purchase when I can see, hold and try a working product!
Desktop: Typing this on RHEL7 Beta with the EPEL kernel as I need ath5k wifi drivers on this old thinkpad. Not bad actually, especially with the Hide Top Bar Gnome Shell extension installed.
> Get over it and buy a Mac if you can't stand Windows.
I like free software, especially in the libre sense. And open source. Especially bearing in mind the NSA's subversion of closed-source security products. Neither Windows nor OS X are an option.
Linux, (Free|Open|Net)BSD? Now we're talking.
Neither have to rule the world and if they always stay below 1% market desktop share, well, that's not ideal but good enough for me.
Also, this bit about below 5% of the desktop market being "statistically insignificant", what is that even supposed to mean? If someone sells a piece of software and reaches 1% of the desktop market, let alone 5%, they are doing pretty well. Linux is doing just fine over here and on millions of other people's machines and that's what matters to us.
You clearly don't "compute" for a living.
There's around 500 computers in our company, not counting thin clients.
Lots of people have laptops and desktops (generally running Windows 7), but by far the largest share of computers are in the compute cluster, which exclusively runs ("Ta-da"): Linux (except for the thin client servers running Solaris).
The rest of you use computers as glorified typewriters to send electronic memos to each other, and to search for inane clips of One Direction, the Beiber (whatever that is), celebrity upskirts (not a pretty sight if you accidentally get Brian Blessed I am told), obscure train model kits, and to argue about trivia on Wikipedia. So basically any old operating system with a web browser will do for you, even crappy ones like Windows and MacOS.
The fact is the vast majority of people don't use computers to "compute", and therefore their use of so called computers is misnamed and largely irrelevant to those who actually "compute".
Okay you don't like Unity. I wish you well on whatever interface you do like. But have you used Unity? It's absolutely wrong to suggest it's not easy to use with a mouse and keyboard. Hold down the windows / super key and you'll see an impressive number of genuinely useful keyboard shortcuts.
Windows and Unity are excellent for use with mouse and keyboard. They both have a dock with all your favourite programs, and each lets you can type the name of a program instead of going through a menu system designed back in 1995. Unity takes it further by letting you type the command you want in your application instead of remembering special keys or using the menu bar. Just because you can use a finger, doesn't mean you have to.
Windows laptops are going touch-screen so the Modern interface is more suitable to the hardware now. When I switch from my Transformer (Android) to my old Windows laptop I find myself instinctively prodding the screen.
"... and each lets you can type the name of a program instead of going through a menu system designed back in 1995."
Do you mean like typing "emacs" or "vi" to launch your text editor…? Because I'm pretty sure that I was doing that WELL before '95. If you're going to give examples of how your OS/UI of choice is so much more modern and enlightened than others, you might want to suggest something it does that WASN'T in use before 1982.
Please don't call me a troll, but I suspect both Microsoft and Canonical think from very different point of view than we, Linux desktop users, have.
Average Linux desktop "power" user (especially regular ElReg reader) is a person smart enough to find, - at last, - distro that suits him, pick his desktop of choice, throw several customizations in - and finally, everything works! Same applies to MS desktop, really, when a professional customizes it for productive work it is VERY different from what do we have in first place, after installing a stock Win flavour.
But both MS and Canonical aim at newcomers, at now-children-then-teenagers. What is the first computing device in child's life? A personal computer with a keyboard and mouse? A laptop? A tablet???? NOOO. It's, suprisingly, A PHONE ! Every child at school age has it's own PHONE!!! And a laptop is SECOND device in life of a child, not first! *This* is the real reason of dumbed-down, touch oriented, "converging" GUIs such as Gnome3, Unity and TIFKAM (didn't seen Win8 yet, really, but Unity + Gnome did piss me off more than enough in late 2010).
The little 6-8 year old guys & girls that first got familar with swipe-and-smear paradigm on daddy's tablet, than on their own phone, and only then on a PC - THAT is a customer both MS and Canonical are aiming at.
Think of a market that grows with several millions of newcomers ever year - a young ones who don't 1) have any stereotypes regarding WIMP paradigm yet and 2) get familiar with touch interface many years before they will be using software for work. And you will understand where all money and effort go.
"The little 6-8 year old guys & girls that first got familar with swipe-and-smear paradigm on daddy's tablet, than on their own phone, and only then on a PC - THAT is a customer both MS and Canonical are aiming at."
Balls. My ten year old was using a laptop at the age of three (on his own, very well), got his own laptop at 4, tablet at 8 and phone at 10.
Actually I find this view of "smart users don't want Unity / Gnome3 / whatever" to be held by people who think they are smart because they don't want Unity / Gnome3 / whatever.
Neither Unity nor (as I understand it) Gnome3 are that great if you want to spend all your time tweaking the micro behaviour of your window manager; getting your window translucency just so; making sure that your animations of windows opening and closing exactly fit whatever obsessive criteria you have set; and that you have programmed in and memorised 30 different key combinations that do such pointless things as minimising all windows whose name matches the regular expression "/.*o.*c.*d/".
On the other hand they are great if you actually have difficult and demanding work to do that actually requires you to spend minimal time actually farting around with your window manager, but getting on with doing some insanely difficult signal processing algorithm, stats, simulation problem or something.
That is they are excellent as immediately useful and productive systems for smart people, who don't want to spend all day fiddling with the settings.
Having said that, I don't like the design choices in Unity that require click to follow focus (I prefer focus follows mouse), but for the rest of it I prefer having a sensible UI that makes tweaking for usability obsolete.
If you don't like these UIs or agree with their choices, then fine, but don't call a whole load of smart people stupid for preferring to use a system that does not require them to spend hours of potentially useful time fiddling with settings and customisations that hold no interest to them just in order to be able to start working...
Personally, I think the reason PC sales have flat-lined is because the requirements of your average home user hasn't changed in years. Surfing, social networking, youtubing, etc... all fine on a 4+ year Laptop. If you are not too bothered about HD content, you can go back even further. The one I use is 2 years old and I have never had it using more than 50% CPU. (Granted it is using a quadcore i7)
Using Ubuntu Desktop, by the way.
I've just replaced my desktop system, or at least I'm in the midst of trying to.
The only driver for replacing a 10 year old PC was that many modern websites were becoming too slow on the old box. Everything else I want to do it does fine. It's still fine for write training manuals (I'm forced to use Word these days), presentations (PowerPoint) doing the accounts (Excel, this could just as easily be LibreOffice), nothing else that I use the machine for needs more power. But the replacement is a refurb unit which has the advantage of coming with W7, and saves a few bob too, replaced the hdd with an ssd which for most PC tasks make far more difference that any change of CPU.
The biggest hurdles as I can see are:
1) Having to replace all the SW, I'd much rather be running the versions of Office I know, but can't so I've had to spend almost as much on the SW as on the HW. A lot of the SW won't be replaced
2) Moving all the old stuff over is just such a hassle.
Actually most of the stuff just won't be moved over. All the SW I just need to know is there, just in case or that I use occasionally is just going to end up running as a VM on my laptop. Otherwise the SW costs would make the replacement impossible to justify.
As a hardcore Linux guy that's been using it on the desktop since 0.99pl13, I'd install Windows 8 on someone's machine just as quickly as Ubuntu nowadays. I've gotten tired of "'tudes" and Shuttleworth's in particular. If I want to hear manager idiocy, it's not much different than "developers! developers! developers!"
Personally I'd be quite happy for the fanboys (of all types) to completely move away from desktops to almost anything else. Then perhaps those of us that do actual work on our machines won't be bothered by all this 'latest-and-greatest' crap and all the OSs & desktops will slim down to a maximum ease of use form.
Are you having a laugh.
Never in a million years.
I've been a Linux convert since Slakware 1.1 came out. I use a number of dofferent Unix systems before that but personally I have totally failed to see what Ubuntu and its ilk would do for me. Sure for some people it can be an alternative to the mess that is TIFKAM but IMHO Unity is not much better.
I installed Mir on my desktop yesterday. It's worked well for me so far. Ubuntu Touch is a work in progress but it's not too far off. A powerful yet flexible computer in your pocket is an awesome vision, and I hope Ubuntu Touch itself takes off.
Personally, I can't recall a time since the early 90s in which there has been this much disruption and innovation in user interfaces. No doubt some paths will be the road less travelled, but hasn't it always been thus?
I'm not sure why I'd want to run Blender, Luxrender or my own, personal, build of Luxrender (for doing vision science, different colour spaces etc.) on my phone. I understand why I run them on the machines I have that run Linux (can program, build stuff and so on). I understand why the machine I do my email, writing and stuff like that runs Windows XP - it still does the job. My phone is from Apple from a long time ago (3GS) - still just works. Linux on a phone for Linux sake is silly unless it is better specifically for being for a phone - which I doubt.
"The "desktop" version of Ubuntu in 2018 will be your Ubuntu Touch device docked to an 8K ultra-HD monitor. Who doesn't want that?"
It may not take Ubuntu until 2018, but if this is what you want right now then there already are Windows products which do this. When using the tablet you have the "Modern" touch-optimised interface, when docked you have the traditional desktop with monitor, speakers, keyboard, mouse and ethernet. There are also devices which dock into a laptop keyboard, but they tend to be larger tablets.
8" Toshiba Windows 8 tablet in a Belkin dock with a full desktop setup (36:43 ish):
Google has now got a single person in charge of Chrome OS and Android, so they might decide to integrate a Chrome desktop into Android devices (yes, Chrome DOES cater for offline apps!). And when Apple decides to "invent" the paradigm everyone will realise it's amazing and start to actually use it.
It's a bit of a stretch, but Asus might have the basis of the ideal solution with Padphone - phone docks into tablet for a better touch experience when you're not "mobile", tablet docks into a keyboard for a work experience when you are mobile, then it docks into a desktop for maximum productivity when you're in the office - all off a single processor. I think they even had a bluetooth-enabled stylus-come-headset at one point, which docked in the phone!
I switched back to a small distro Ubuntu LTS variant because the Debian variant I was running had not installed the right graphics driver and suffered UI crashes to blank a screen. The Ubuntu LTS did the drivers right. I"m guessing less restricted driver options under Ubuntu made the difference? It forced me to re-evaluate a some things.
I think at some point people using alternatives to Ubuntu that are still Ubuntu under the hood becomes less a protest but instead more of an evolution to where people need to go with their desktop UI paradigm and that is fine. It no longer is an issue if Shuttleworth creates an alternate UI when people can and already do roll-their-own. I think Mint and others like it are proving that to the degree that there is no longer any point in expecting Shuttleworth to provide us the UI we want for us.
Ubuntu is the dog chasing any car that comes by its house. First it was desktop, then it it was the T.V. (remember that) now its the phone (Oh lets not forget the server space, and the cloud). Judging by what they did in the desktop space which was pretty much Linux with ugly wall paper and ugly icons and a window manager that makes Windows 8 look great, I don't expect to much from them in the mobile space.
One thing this article and may others like that hail the end of the desktop forget, is what desktop computers do - they produce the digital content be it from a word processor or a 3D CAD program. Do you think demand for such content is going to diminish ?
The Dells, HP's, Lenovas should stop letting Microsoft drive them to oblivion. Why do they not pull their resources together and get behind Ubuntu, or some facsimile, and keep their profitable desktops relevant. What they are loosing in sales from one business quarter would pay for all the development needed to make polish Linux desktop
Whatever "lack of polish" you see in the Linux desktop isn't the problem. It never was. The market was won by the crudest command line interface available that subjected you to manual memory management.
"Quality" has nothing to do with this.
It's all about the apps and legacy interfaces that date back decades.
Something like Steam or Oracle or AutoCAD is far more important than KDE vs GNOME.
We think we need a full-power Linux phone. Some of us already saw it coming, in the Maemo project. That had everything. You could do devops on a Maemo phone. You could use a Maemo phone for monitoring (or cracking) WiFi networks. Then it became too visible and was killed by tragic mismanagement.
Modern phones may be weak compared to modern desktops, but modern phones are very powerful compared to desktops 10 years ago. Remember when Windows XP was still new? Those desktops could certainly run desktop operating systems. So, smartphones can, too.
But I think that we're moving into a different paradigm. When PCs took over from mainframes, we were not running mainframe programs on PCs. We were running vastly weaker systems with no concept of memory safety and a whole lot of security problems. When PCs grew to many multiples of mainframes' original performance, we didn't then run mainframe programs. We finally got multitasking and virtualization and so on, but they were tacked onto existing PC systems, and it has been a struggle to make those safe.
So, I think that the iPhone paradigm will dominate. I'm not sure of the exact drivers of that: The mainframe systems had the disadvantage that the major mainframe OS vendors (IBM) wouldn't license them to run on PC. Desktop Linux is free. I suspect that cell phone companies like phones to be devices that depend on being connected, so customers don't dare run the thing without a cell contract. Also, people don't like change, unless there's a status improvement or drastic usability improvement that can come from it. I don't think desktop Linux brings the types of improvement that inspire change for most people, compared to Android.
When I hear that you can't work on a mobile device, I'm reminded that PCs were said to be unable to do work. All that naysaying was interpreted as a challenge. People discovered that they could indeed do work on a PC. So it will be with these horribly limited mobile platforms.
Both PC's and mainframes needed keyboards and tv screens to get something done. The PC added a mouse.
If you can add a tv and a keyboard then you can do the same things in a smartphone as you did on a PC, and a mainframe.
But just as the PC was never capable of running mainframe software, the smartphone will not do it either. The OS maker needs to create an OS that can seamlessly support an app that works on small touch screens and on big keyboard-enabled tv screens. But that is actually quite difficult. So an OS makes who only does touch is quicker and will outcompete the integrated OS maker. Then, by the time the integrated OS is ready everybody will be using the touch OS and nobody is going to switch.
Then the third party software creators. The smaller a device is, the less money people are willing to pay for software. Orders of magnitude less. Which means that a business selling PC software will make orders of magnitude less if they have a product that runs on a smartphone too. They are not going to do that.
New entrants in the software business might be able to create such software. But in the beginning, they cannot compete with the encumbents for a particular software niche. While OTOH, there is money to be made on smartphone-specific apps.
> But just as the PC was never capable of running mainframe software,
That is not true. The IBM PC lineup included the IBM PC XT/370 which ran mainframe software as well as PC-DOS. There was also the IBM 3270 PC but that was just a mainframe terminal that was also a PC-DOS PC. see Byte Fall 1984.
I would like to see a Linux OS take off and get real market share and mind share in the broader device market, but as a power user, I would never personally want that Linux flavor to be Ubuntu. Of late their party line is that they know what the users want better than the users do. And *maybe* they do for users in general, or for new users that (somehow) find themselves using a device running Linux. But they sure as hell don't know what I like, because I don't like a lot of high-profile things they've done in recent Ubuntu releases.
That attitude of knowing what I want/need more than I do myself has been classic Apple for ages now, and it always kept me away from them. That attitude was recently adopted to much angst of late by Microsoft with TIFKAM, and it kept me away from Windows 8. And it keeps me away from things Canonical the same way.
You can be pioneering without immediately throwing old paradigms under the bus. You can be progressive without remaking *everything* from scratch without a smooth transition. Does it take longer? Most likely. Does it cost more? Possibly. Will it give you a happy base of users who you help through the transition? I think it would.
It's great to be looking at the long road where today's 6-year-olds will be the device consumers of the future, but the fact is that right now we've got a ton of users bridging the desktop and phone/tablet paradigms. Making *them* want to use your product would get you significant market share which those 6-year-olds would grow up seeing. I don't understand the strategies that seem to decide those transitional people are irrelevant and/or can be made to change their opinions en masse. Given how poorly it seems to have gone in general, I think I'm probably not alone in feeling that way.
If you paid attention to the latest findings from psychology, and the common practices of the best marketers, you would already know that users don't really know what they want. Not even you.
When the iPhone and iPad came out, they were revolutionary in their category. Yet, millions of people who are familiar with the old paradigms bought them. There's a major disconnect between your point of view and reality.
I'm certainly not an MS Shill but I really feel that Canonical have lost the plot. They seem to want to do everything but really only have limited resources. I see lots of promises but not much production quality software being delivered.
I was once a fan of Ubuntu but around the 10.4 release I said this is not good enough. I've gone for stability of CentOS and SUSE rather than new features on my Linux systems and dabble with Fedora in a VM when the mood takes me.
I don't know what all the faux controversy is about, if you don't like it, you can always change it, after all it is open source.
"Like many, I don't use Ubuntu on the desktop"
I've been Linux only on the desktop, at home, for a number of years, and haven't noticed any lack of usability ..
What separates Linux from Windows isn't a coat of paint, its the API that determines which programs run on Windows and not on Linux.
And while new interface ideas are welcome, and more powerful low-power chips are useful, I think plugging a keyboard and a mouse into a smartphone with Linux on it would be even more important than plugging a big monitor into it.
So the article, in a couple of places, seems to have gotten ahold of the wrong end of things.
Not necessarily. A dock that can connect a phone to an HDTV can just as easily hook you up to a mouse and keyboard. That's how laptop docks work, and with a little polish, the same can be done with smartphones. The main obstacle is a standardized way of doing this breakout, as most phones only have the one USB port, and AFAIK you can't do MHL, USB OTG, AND recharging simultaneously on the one port, so an alternative is needed. Perhaps using wireless display casting and Bluetooth for the mouse and keyboard, leaving the port free for charging (perhaps use that as a triggering mechanism), and this is just one idea.
I have a nexus 4 and I have it permanently underclocked to 700mhz (Because it is plenty speedy enough for what I use my phone for), and I use Ubuntu on my desktop so if I can literally rock up to work with my phone in my pocket and have a fully working desktop ready to go then I can definitely see the benefit in that. Only thing would be storage but as most new phones have USB on the go I guess they could all be connected up with a hub for all that? In the words of Jasper Beardly - "What a time to be alive"
Unfortunate name. I called my last boat Mir because it was a metal box that cost sackloads of money, was a permanent work in progress, and at one point I couldn't close the hatch because of the cables running through it (look it up).
Nominative determinism isn't compulsory, but I really wouldn't want to name anything I wanted to be a commercial success Mir, because.
Back to the topic.
My worry about Ubuntu mobile is that it will have exactly the same problems as BlackBerry OS 10, and for the same reasons:
Insufficient development resources
Not enough money spent
bribing marketing to US carriers to get critical mass.
Jolla are already there. Grandiose promises for sure, but they've delivered and without the backing of a billionare like Shuttleworth.
Full linux, innovative UI using standard cross-platform toolkit (Qt/QML). A fully functional IDE in QtCreator.
Support for Android apps.
Oh and now I've turned off the NFC, battery is at 64% after a day and a half, with admittedly light use, but WiFi has been on all the time.
That's not even covering the future possibilities with their real USP, "The Other Half"
The only reasons Android succeeded were :
It was free, and good enough to gain a foothold.. It remains free.
Samsung et al could then slim down their programming effort as there was no longer an operating system to write. (modify Android/sort out drivers/compile and test)
This saves them a shed load of cash. Hardware all the high end phones have a decent camera, touch screen, accelerometer, more than enough memory, quick processor (predominately ARM based)
The differences between them all are packaging paper cup or ceramic mug and flavour latte or cappuccino.
To the El Reg spell checker.. Flavour is spelled with a U between the O and the R. W are not all American.
"Samsung et al could then slim down their programming effort as there was no longer an operating system to write. (modify Android/sort out drivers/compile and test)"
They still sell quite a few Bada (?) phones, and they seem to be expending some effort on Tizen (?) which may be an alternative to Android in the long run.
Samsung are trying to play the long game, and get people attached to their brand. Android is a necessary evil at this point, as provably few people actually care what OS is running on their phone, as long as it does what it says on the time and meets their needs.
Samsung will reach a moment in time where changing Android for Tizen (or Jolla hehehe) at some point. Whether they jump will be interesting.
Granted, I completely support a Linux Mobe/ Desktop interface where much of my work runs of servers in my datacenter, but I can load all the lovely tools of nix-ness that I want using a docking station as the thin client. My larger concern is how many Sysadmins will go postal from all the users looking to replace lost phones, etc.
And, I just can't see business wanting to supply smart phones to say thier marketing intern.