We know what this means
Just as they used the 8.04 LTS as their testing ground for Pulseaudio, so will they use 14.04 as testing ground for Mir. Brilliant!
The new touchscreen-friendly user-interface system promised for next desktop version of Ubuntu has been PULLED just weeks before its release. Mir and XMir – the Linux distro's replacement for the age-old X Windows graphics stack – will now NOT be enabled by default in the 13.10 desktop build - due out on 17 October. Mir is …
The new gubbins is all still in there, just you get plain Xorg by default. Basically a one mouse click change (and a restart presumably, at least of the graphical layer).
Personally, I congratulate Canonical's developers on going with a sensible default. Perhaps they might get used to doing that if we encourage them (hence the beer).
"That's a slight pain in the backside for Canonical's programmers, who will have to maintain their own unofficial, XMir-supporting version of the xf86-video-intel driver, rather than rely on Intel to maintain support for the compatibility layer."
But isn't that the Linux way? That is the beauty of Linux: bundles of similarly-named but slightly incompatible chunks of source code that some people bundle together and call 'distributions'.
There can now be yet-another-version of Ubuntu: Mirubuntu (nicknamed 'Mirror') for the Mir-version of Ubuntu. There can't be an Xubuntu as that is already taken. I suppose there can be a Wubuntu for Wayland.
This coldly explains why there will never be a 'year of the Linux desktop'. There is endless reinvention of things that don't really need reinventing.
UNIX (and Linux) on the desktop is not really broken for me. I am told that 'X11 is broken'. Yet I don't recall having X11 crash this year and I use UNIX and Linux on my desktop all day long in a technical environment. But here comes a new sort-of-X11 which will have its own bugs. And we will get another notepad app, or a new packaging system with ever more convoluted commands. Or a new font from Ubuntu!
Canonical have not chosen to go with Mir as replacement for X for technical reasons - they want to have their cake and eat it with the development and commecialisation of Mir. On the one had Canonical want community contributions to help develop the software and on the other they want to cash in on Ubuntu/Mir in the mobile/hand-held/embedded markets.
To this end Canonical have imposed a Contributor License Agreement that gives them the right to relicense any and all code as they see fit. This gets around the inconvenient fact that, as they are using community code, they are not the sole copyright holder and thus effectively allows them to block other developers from modifying (or even supporting) Ubuntu/Mir in the markets they seek to exploit.
For a full and more eloquent explanation see Matthew Garrett's posting.
It has nothing to do with crashes, it has to do with some of the crap design decisions in X11.
I've been using X11 as a desktop since Debian "Woody", and I can categorically state that the X11 cut'n'paste "design" is a horribly broken abortion. There needs to be just one buffer/selection/method/whatever you want to call it, and that's it. This is the major spot that Windows has a leg up... cut'n'paste usually just works as expected on Windows.
Granted it was designed long before people knew their ass from a hole in the ground, GUI-wise, but there needs to be a patch fix things and to return the same damn thing no matter what the (broken) app asks for.
Ctrl-V & middle-click should return the same damn thing when pasting.
This will save my co-workers from "PASTE YOU MOTHERFUCKING PIECE OF SHIT" and similar howls of pain and agony.
I won't even get into the stupidity with which X11 handles multiple monitors and how most of the window managers manage to make all the wrong decisions all in a row.
Ugh. I have basically the opposite experience from what Gene Cash here mentions. I just *loved* the simplicity and non-verboseness of X clipboard hanging ever since I first successfully installed a RedHat 4 (1996), and through my progression via different distros (even *BSDs) until I settled in Debian. And, after many years, I don't only know there are different clipboards, but use them for different things. Of course, I'm quite a well-trained user by now — But the system does what I want. (Or I think the way the system wants me to? Hmmm...)
Now, as for the "stupidity with which X11 handles multiple monitors"... Most of the time, at my office, I work in a dual-monitor setup. My right monitor is rotated 90° (portrait, 1440x900), my left monitor is at "regular" 1280x1024. I hated multi-monitor setup back in the XFree days, but today, with a single xrandr command, I get everything I need — And I get it always reliably. In my laptop, switching between a regular, dual-screen and mirrored-screen monitor is also straightforward; when I see my colleagues fiddle with Fn+F5 until they get the right projector+screen layout, I cannot but laugh.
So, yes, X11 gave me quite a bit of headaches back in the day. Nowadays, it's one of the most unobtrusive, reliable pieces of software I use. And, yes, I'm one of those obsessed with no-binary-blobs software.
This coldly explains why there will never be a 'year of the Linux desktop'. There is endless reinvention of things that don't really need reinventing.
You know, you have a very appropriate name..
The reasons Canonical gave for developing Mir are fairly straightforward, and stated in the article. Perhaps you'd like to read it? X works just dandy on the desktop, but desktops are dwindling in market share, and X across devices is a ball-ache, hence Mir.
There will never be a "year of the Linux desktop" (a phrase I've honestly never heard actual Linux users give a shit about) because desktops are ten minutes away from not mattering to most folk.
The list of reasons given by Canonical was a lot bigger until it was systematically taken apart by Wayland developers that took it badly to Canonical spreading FUD about stuff where Wayland's progress has been "slow and steady" (especially where it has to wait for X to come to Wayland), while the supporting libraries they had built for having Wayland work on Android graphic drivers were merged into Mir for Ubuntu Phone without a word of attribution or acknowledgement, which was BAD FORM.
Smart Scopes also seem to have become a little smarter. Search results were a bit more relevant than when I have tested this feature in the past, though there is still plenty of junk. Why Britney Spears albums come up on a search for "Thailand" is something only Canonical knows. Maybe.
There is a very subtle bug in the heart of Linux - certainly the versions since 8.04 (7.10 was rock solid). It's to do with the mouse ... on a very few configurations of hardware, the mouse focus gets trapped in an invisible window, and the buttons stop working. Only resolution I (not being a Linux developer) found was to restart the X server.
Bug was still there at 13.04. I know this, because the thread I found on launchpad for it is still there, and still unfixed. https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/41301
Now this bug caused me to completely drop Ubuntu. I had to since (a) there was no guarantee a new mouse would fix the problem and (b) if you have to buy new hardware to run an operating system you may as well have Windows.
No matter what you may think of Microsoft, it's inconceivable they would let such a bug fester for 6 weeks let alone 6 months.
Fix the foundations before you paint the house.
That is the second time in as many weeks that I have heard the complaint that existing bugs aren't being addressed by Canonical.
It really puts some of us off the product.
Is the chap's name Reis or Ries? The article's author uses both.
Checks the blog entry...
Ah it looks like Ries.
Well based on my mouse experiences I simply can't advise anyone to use Ubuntu. I stopped using it on my home PC, and reverted to Windows 7 (does what I want). I kept it on my home "server" since I always accessed that remotely. However I had to move from Ubuntu when I installed the 12.04 upgrade, as they managed to completely bork the security mechanism which means remote users are treated differently to console users. I had spend days having to research and change them to work for me*, and they were undone at a stroke.
Switched to Debian, and never been happier.
Sorry, I have used desktop PCs since 1981. Anyone who thinks serious businesses are going to wake up one day and say "Let's Ubuntu" is living in cloud cuckoo land. And the reason I say this is nothing to do with the pros and cons of Ubuntu, and everything to do with the attitude of the Ubuntu community. Witness the downvote my OP got, and (doubtless) this one will get too.
*Which I painstakingly detailed and gave back to the community. As far as I know they are still the only guide to configuring PolicyKit anywhere in the real world.
And yet on every single machine I have installed Ubuntu on, there have been no mouse problems whatsoever. So I don't hesitate to recommend Ubuntu.
That's eeePc, 2 different Acer desktops, home made desktops from Duron 1600 upwards, quad core Dell monster thing at work, Father's Acer Revo. Now thats all sorts of HW. I think the mouse bug you are seeing must be pretty rare, as I never even heard in mentioned, even here.
I would suspect that you have an X Windows client program that is warping the cursor back into it's window every time that that it detects that it is going outside. As for the buttons, it could either be changing the button mappings, or re-parenting the root window to trap the button events. These would not be problems with the X11 server, more like a poorly written client using the server incorrectly.
I actually wonder whether you are running an older ATI radeon card (something like a Thinkpad T30 or T43). If that is the case, it could actually be that you are suffering from a problem with the display driver not being reset correctly after suspend/resume. If that is the case, then disabling KMS may be the solution (google nomodeset or modeset=0). It's a pain, I admit, but the upstream developers regard R100/RV200 and R200 as being obsolete nowadays, so are not interested in fixing the problems. The adoption of Unity and now MIR (especially if they completely ditch X11 compatibility) will be close on a deal-breaker for me and Ubuntu. The shiny is just not worth losing X11 for. At leat Wayland has an X11 compatibility component.
I've got 12.04 running on my T30, and get strange things happening, but I am working with the pm-suspend tweaks to make the problem less intrusive.
Thank you for a positive contribution to the discussion of this issue. It sounds plausible. It wasn't a laptop but a desktop. Interestingly enough it *did* have a Radeon card though. However other reports - albeit scarce - demonstrated there was no single common hardware element. Different motherboards, chipsets, video cards and mice. At one point I suspected it could be a wireless mouse issue. But switching to a wired mouse didn't help, and others reported the issue happening with their wired mice.
As I said previously, I am convinced the "bug" originated after 7.10 I started using Ubuntu at 6.10 and went through 7.04 and 7.10 with no problem.
Now for all their faults, Microsoft take bugs - even subtle ones - *very* seriously. I know this because many years ago I hit an incredibly subtle bug somewhere in the TCP/IP<->DCOM architecture. I would instansiate a COM object on a server, and despite it being sent and responding to keepalive pings, it would have disappeared when it was called. It was impossible to work with, as testing for the existence of the object returned a true. So you'd test for existence. Get a yes. Go to call .... crash. This happened every so often, but without rhyme nor reason.
After weeks of liaising with MS Windows support, they sent two experts out to our office. They installed the source code of the appropriate DLL on the server, and captured a shed load of debug. They went away, and delivered a customised DLL to fix the issue. At this point I changed jobs so don't know what they did or if it was an underlying Windows issue subsequently fixed in a service patch.
Anyway, the point is *that* is taking support seriously. And at no point was the MS answer "change server hardware" or "change client hardware" or (seeing as it was a network issue) "change routers". Yet in my dealing with the Ubuntu community it was suggested that to "fix" the problem I should buy a new mouse. Or video card. Or (and this was quite a common one) just get a new/different PC.
Now call me old-fashioned, but with attitudes like that, Linux - certainly the Ubuntu flavour - ain't going anywhere near a corporate desktop. Downvote me as much as you like, but it won't change the facts.
Oh, and my contribution to getting Linux on the desktop is to sneak 6 copies into my work, when I was tasked to setup a training environment for a Windows VM. I put 6 copies of 12.04 (with Unity !) onto some old PCs, and got them to autoboot into the VM. Not one of the 30 users over 2 weeks had any issues, and they all thought they were running Windows Server 2008 !
"Is this important? In terms of distinction between kernel and OS, not really, no, though there are plenty of people who would vehemently disagree with me on that one. But in terms of remembering there are plenty of other Linux distributions out there (which may or may not have the same bug, depending on where it originated), yes it is. A bit. Relatively speaking, of course."
Well, for the average Joe it is quite important. How's anyone supposed to know which distro is good and proper, and which ones to avoid? And are the recommended ones popular because they appeal to a certain type of geek, or because they are actually well thought out for the average end user. It's especially difficult when the answers to these question seems to change every year or so. And as for desktop consistent experiences - ha!
I'm not sure either about the 3-D gestures bit. I can't imagine why you couldn't make 3-D gestures work, seems like a bullshit reason.
Intel is backing Wayland for obvious reasons, however I have yet to see anything official from Nvidia support, although it really isn't needed. Last I checked with Wayland was in December of 2012 and Firefox and Google both dropped support. However, I never read a definitive reason why.
You can draw a few things about the future of which desktop will be "next" for Linux...
1. Intel directly supports it.
2. Firefox and Google may have dropped it, but haven't denied future support.
3. Nvidia does have a developers section on their official site for Wayland, and it's active.
4. Valve is supporting it.
The above list has a 1 hardware vendor, and 4 very potential software supporters, all of which that are major players in their respective fields.
1. Ubuntu WANTS to support it.
The above list is complete, with the exception that there is at least 1 hardware vendor that will NOT support it. Basically, it's like VHS vs. Blu-Ray.
As Mayor Guliani once said with his best Tarkin impersonation to a firefighter asking what to do about the burning WTC 5: "PULL IT" (because why did we spend the past 2 months setting the demolition charges if we don't use them now?)
[...okay what he actually said was "PULL OUT" (because a majorly large part of the support structure is missing), but that's not nearly as fun.]
"[Wayland] which was designed in 2008 by Linux kernel engineer Kristian Høgsberg who is now a software engineer at Intel."
I guess he didn't want to work on someone else's code. I can't say I blame him. If you know your own code, why bother with someone else's that aims to do the same thing?
if Ubuntu themselves decided to yank it. After they released Unity and the uproar that caused I would of thought they probably had no quality standards. Then the privacy stuff with search etc.
So if they found enough reason to yank it, it's got to have some serious bug infestation.
Myself I am still on 10.04 LTS for my desktop even though the desktop components are no longer "supported" (I still seem to get other updates like kernel and stuff though - not that I care).
At some point I need to make a decision I figure not until I have a new hardware platform. I don't feel my Linux desktops or laptops are in any sort of real security threat.
I really have adjusted well to my basic GNOME 2 + brightside (GNOME edge flipping for virtual desktops) over the years and have no interest in changing it. I have a very little experience with the "legacy GNOME" mode of GNOME 3 on Debian 7. I got to hating it within about 4 seconds (about as much time I needed to realize right click on desktop did not work by default). I've only used Ubuntu for the hardware support in recent years on desktops and laptops, before that (and on all my personal servers) I run Debian stable(since 1998).
I have read/heard about other projects like Mint etc which sound promising. Or maybe I should just go old school and use Afterstep again............that might be too much of a shock, as much as I miss the awesome native virtual desktop capabilities of afterstep.
Valve, apparently. Given that SteamOS may be an Ubuntu respin and that the official client only officially supports Ubuntu.
At least, from the look of it. Why do they only have a DEB repo and why does the package rely on Ubuntu proprietary packages like Jockey (meaning the package can't be installed upstream in Debian without hacking)? Sure, they are unofficial RPM and .tar.xz packages for OpenSuSE and Arch respectively, but users of those gets no support from Valve.
The chip giant, which backs the rival display engine Wayland, gave little reason for the snub. The engineer who kicked Canonical's XMir code out of the xf86-video-intel driver simply said: “We do not condone or support Canonical in the course of action they have chosen, and will not carry XMir patches upstream.”
What's that about "[giving] little reason"? I see quite enough reason in that engineer's quote:
"It was your idea, mate, not mine. You deal with it, I have my own stuff to take care of."
No talk of being "committed to customers", "feedback", "community" or any of that horse dung. Just a simple and concrete reason, for once.
Ah, if only they'd pay engineers extra to also do the work of PR drones!
Biting the hand that feeds IT © 1998–2019