I can see his point. But maybe a good mechanic *would* clean his workshop whilst listening to music? Some of what a programmer does is just "housekeeping" and not equivalent to a mechanic at work on a vehicle. I like to switch music on or off depending on whether I need stimulation, relaxation or concentration. For mechanical stuff, or when I'm in "the zone", I think music can help propel me through greater productivity. When I really need to think something through in detail I usually switch it off.
I think the principle of requiring proof is good. However, I suspect a lot of managers would not consider being treated as one of the programmers to be the compliment that the programmers intend it as. If he's managing the programmers and perceives that as "I'm in control of the programmers" rather than "I'm here to help this team get its job done" he might not appreciate being approached to justify a management decision as though he were a peer. Not that your suggestions are a bad idea in all cases, it just requires a bit of care to determine what you can get away with.
I'd think for any given action it would be very helpful if the programmers could act as a group, to minimise the risk that an individual is perceived as "the awkward one", even if he is trying to represent the group's interests.
Well, even they are not concentrating 100% on the thing that keeps them alive - driving the car safely and efficiently. They'll also be managing their sleep cycle, planning fuel stop strategies, reporting back to the pits on the condition of the track and what tyres they need, analysing the car's behaviour to predict potential failures. Also, I would bet real money that many of them thought, at some point or other, "It'll be good when I've won the race" or "It'll be nice to sleep" during the driving.
They do demonstrate that people are capable of extraordinary levels of alertness and concentration for (relatively) short periods of time but even that isn't really 100%, the brain is a very parallel device so it would be surprising if it was only doing one thing!
Whether racing drivers can or cannot concentrate 100% though, I think we'd probably all agree that it's not reasonable to expect workers in a, say, 9am-5:30pm, 5 days a week job to be able to concentrate 100% on something for all their working hours.
There's a dude here: http://lwn.net/Articles/366266/ who suggests that the blob Nouveau currently downloads to the card is reasonably well understood and should get rewritten at some stage.
I'm not wanting to contradict you, by the way, just wanted to note what the options are for people wanting to use Nvidia / Nouveau but who are wary of the blobs.
I'd have said the hardest part is reverse engineering how to talk to the hardware. If new hardware comes out they'll have to reverse engineer the new features / changes in that but they shouldn't have to start again from scratch.
This is good stuff, I think. They're not going to shove the whole Nouveau device driver into the kernel, it'll follow the modern X.org / Linux model of having kernel modesetting and a DRM driver in the kernel and a whole load of other stuff living in userspace.
Kernel modesetting (KMS) means that one entity, the kernel, always controls what graphics mode the video card is in. That's useful because pre-KMS, X.org might have changed the mode *without the kernel knowing*. That's one reason Linux can't easily have a Blue Screen Of Death - the kernel doesn't know what it can send to the graphics card to display it. BSOD isn't a feature you want to *see* but if you have a kernel panic, it'd be a lot more useful to actually see it, rather than it being hidden by your frozen X server! I'm not aware of graphical kernel panics currently being supported but at least it could be done in principle now, AIUI. KMS also reduces unnecessary modeswitching "to make sure" that you otherwise get, so switching between console and X should be quicker, as should switching between X sessions (fast user switching). KMS is also what's used by the new bootsplashes, like RedHat's Plymouth (which other distros, e.g. Mandriva) are also moving towards. DRM, in this context, is the Direct Rendering Manager and is how GL apps get direct rendering access to the graphics card, in a controlled way. I don't know so much about that though;-)
The Nvidia open source driver "nv" doesn't support KMS or any 3D. The Nvidia proprietary driver doesn't support KMS but does support 3D (with good performance). Many distros have tended to use nv by default, some do ship nvidia though. Either way, you don't get the nice boot splash and neater terminal switching that a KMS driver would get you. The Nvidia proprietary driver is good performance-wise but it also tends to lag the open source drivers in terms of features a little; I think Nouveau (at one point? may not still be true) was aiming to support Xrandr features that Nvidia's did not. I've also heard that Nvidia's driver has issues with suspend.
Because of all this, expanding Nouveau support is a good thing. Nouveau are also in the process of reverse engineering for 3D support but they have some way to go. However, I've had the impression that it's getting towards being better than the 2D-only nv driver. So at *least* it will mean that when installing on your system you can expect a decent boot experience and correctly-working basic 2D graphics, with suspend/resume behaving sensible, etc. So it could be making life better for users *soon*. But as 3D support improves, things should get better still.
The Linux kernel devs generally take a stance these days that all kernel code ought to be merged into Linus's tree as soon as possible. It's really rather impressive to see this process working and the kernel devs (mostly) really following through on this.
You want "make localmodconfig", which I think was also added recently, possibly to 2.6.32 actually. This builds a kernel using a local.config file, except that it only compiles modules that show up in lsmod. So if you boot off your vendor kernel with a squillion modules, let it load the modules you actually *use* then do make localmodconfig, you can make a kernel that only contains those modules. I don't know what it does if module names etc change, maybe you'd need manual fixup then - should still be less work than you currently are doing though.
Doing it in software for purely virtual hardware is useful. I know it's been used to sync disks across the network on Xen hosts, the idea being that if the local and remote copies of the disk are kept in close sync, you can migrate a virtual machine with very low latency. Should be able to do similar tricks with other Linuxy VMMs. Having software available to do this stuff makes it easy to configure this sort of thing quickly, especially if you're budget-constrained, hardware-wise.
I think your comment about asking the engineers to solve the right problem is very insightful.
But I'm curious - did Windows have more fine-grained locking than a single kernel lock at the time Linux introduced SMP support with 2.0? I can imagine Windows may well have been better re locking scalability back then. Both Linux and Windows have been using increasingly fine granularity locking over the years, which is nice. It's somewhat frustrating that the Big Kernel Lock is still hanging around but at least it's not on most / any important critical paths now. And one day hopefully it will go away properly:-)
I object to that, sir - in the UK we have the finest stream contrivances money can buy. We've been building "railways" for a number of years and in London they even have horseless carriages! Simply spiffing. I'm typing this on one of our latest teletype machines, which is almost entirely automated. It only requires an office of 50 clerks or so to handle the Slashdot Javascript rendering. Actually, the clerks have something of an attitude time, insisting on collating and compiling the reports I requested on a "Just In Time" basis. I'm off to have a word with them now, toodle pip!
That's the normal way of getting road legality for an unusual vehicle. There might be a less conventional method: I once read that there is a nuclear-powered car somewhere in the UK which, for some bizarre reason, was granted a royal dispensation to allow it on the roads - thereby bypassing the normal regulations. That probably also makes it a bit awkward to revoke; I imagine if it's true we just rely on the person who's got it being a good sport and keeping it in the garage.
" I profoundly believe that. And when you steal from us, you steal from them." That's neat. It's literally true in a sense. But I don't like the implied viewpoint that - because EA pays the artists a cut - the artists and EA's interests are perfectly aligned, cutting EA's profits is bad for artists, EA is a purely good force for the artists. Sounds somewhat similar to the music industry's "But you're stealing from the artists!" line. Some of these people may even believe this stuff but we all know it's a bit of a shaky argument.
I don't usually do this but really the parent has an excellent point. It's bad enough that they have increasingly large amounts of DLC you can't resell or buy on a second hand market. Treating the second hand market as basically piracy is a) bad for consumers and b) stupid. When they find a way to make me pay full price for all games by eliminating second hand sales *I will buy fewer games* and I therefore won't buy the DLC for them. The people who didn't like the games and traded them in so they can afford new games will also be able to afford less. They'll be shooting themselves in the foot but, like the music industry, they'll be doing it in a way which makes themselves *think* they're winning. And if they do that, it'll take ages for sanity to break out, if at all:-(
Maybe I'm being a bit dire but this is the way I imagine stuff going.
Agreed about not using clicking; I'd only really want the tabbing if I could switch using a keyboard shortcut.
It does seem maybe a bit redundant to having both virtual desktops and window tabbing but I think I'd use the extra level of "categorisation" it gives - in some circumstances anyhow.
Focus-follows-mouse is my preference but weirdly I'm finding it much less useful these days since I got the habit maximising everything and using apps with tabbing / tiling + keyboard navigation internally. Tried using tabbed window managers a bit and liked it but never found one that quite suited me.
I make very little use of extensions, so I've always assumed that the bloaty behaviour I'd seen from Firefox was largely due to something other than extensions. Mostly, I think the thing which slows my system to a crawl is Flash having a tantrum on a frequent basis, which doesn't change much across browsers. Chrome is good because it makes it easier to kill off Flash. But earlier today the browser to eat my memory sufficiently rapidly that it took about an hour to get access again and kill it properly. I'm blaming Flash for that. *sigh*
They're accurate for me;-) But maybe I'm weird. I don't usually want to see emacs and console and PDF at the same time because if I'm editing the document in emacs the PDF will be out-of-date already. My current solution to this is to have emacs editor plus emacs shell running (I do often do that side-by-side but not always) in maximised emacs, plus a PDF viewer in another maximised window. The sequence is then edit in emacs -> compile in console -> view in PDF viewer. This might vary for others though.
Also, a tabbed workflow is something I'd be more likely to use on my netbook where I'm short on screen space than on my desktop. So it's a bit of a YMMV, I think.
Well, with this tabbing support hopefully that should become really easy for apps that don't implement their own tabbing internally. Especially if they were to add an "Expose all tabbed applications" and "Expose all tabs in this group", that would be very useful. But for things like Firefox and Chrome I guess we'd still want an extension like Windows 7 has that allows tabbed apps to provide previews to show on the taskbar etc.
Emacs 23 can actually embed a PDF viewer, can't it? So perhaps my example wasn't particularly well chosen since one can do it all already:-) FWIW I currently do exactly as you do! On my papers desktop I have a fullscreen emacs and fullscreen Okular and Alt-Tab between them (and sometimes a web browser for looking stuff up).
A more salient example is perhaps how I (used to) use Konqueror as a universal document browser: I'd browse the web in various tabs but also have PDF viewers open in others and file managers in yet others. It had a nice task-grouping property (e.g. windows relating to researching a particular topic, windows relating to generic browsing) that I've lost in recent years since I started playing with Firefox / Chrome. It would be pretty nice to be able to get the grouping ability back without the developers of these applications having to miraculously agree on a style (and embedding API) that fits my workflow. With WM support you could even tab a Windows app running under Wine, or a remote app over X forwarding / NX / xpra, which could be sweet. And when administering virtual machines I'd love to be able to tab together the VNC viewers.
Another thing is that (in *some* ways) it just seems like better software engineering to have tabs implemented outside applications - Windows has an extension to enable tabbed applications to preview their tabs when you mouse-over the taskbar. AFAIK only IE supports that currently. If it's implemented in the window manager then it's trivial to support that for all apps. And the apps themselves can potentially be simpler.
I don't think this would be revolutionary for me (that'll have to wait until there's full tiling support) but I think it'd be somewhat useful - and I can think of various ways it might be extended to make things even better.
KDE's BeOS window manager style allows you to move the tabs around the top of the window with a shift-drag (and/or attempts to avoid tabs getting hidden by moving them itself, not entirely sure how it works), which helps for tabbed-ish operation. Not as nice as a proper tabbing window manager but the BeOS style has been in KDE since forever. Could TWM do this, I wonder? I've seen that themed to look like BeOS although given its level of sophistication as a WM the theme was probably just "made the tabs yellow";-)
I've found the desktop reasonably stable since the KDE 4.1-4.2 era but I have also found that the experience varies a bit between distros. It's worth trying it out under a distro with a known-solid KDE 4 implementation before making a final judgement. (You may have already tried this, just thought it was worth noting).
I personally like the idea of having tabbed windows, however in support of your argument it's worth noting that KDE 4 already supports arbitrary drag-n-drop grouping of apps into named taskbar buttons: http://www.kdedevelopers.org/node/3864
Don't think GNOME has done this, don't know about XFCE. Compiz can do it, plus at least some basic tiling I think. And obviously not on Win or Mac. So KDE it the most "mainstream" desktop to have tabbed windows so far. But it's far from a new invention. There has also been talk of tiling support for KWin, the KDE window manager, which would make it even more useful. Various window managers using tabbing / tiling exist, such as ion, dwm, wmii, Xmonad, etc. They're nice but I missed the integration of having a full DE (though you can get it if you try). Partiwm is a project to create a more DE-friendly tabbing window manager but AFAICS it's gone a bit off track since its creator invented xpra and concentrated on that instead...
Friends of mine have observed that tabbing in the WM makes a lot of sense. Tab together a load of single browser instances and you have a multi-process web browser. OK, so it's not quite Chrome in security features but it's a heck of a lot simpler. Tab a load of terminals together and get a slick multi-terminal app. Tab OpenOffice together with your web browser whilst you're writing a report and researching stuff online. Tab together emacs + console running LaTeX + PDF viewer and get an integrated development environment for scientific papers. Nice.
I'm exaggerating the simplicity slightly but the point is that things are far more flexible if commonly-needed features (how many apps use tabs these days) are provided by the platform where possible.
Military leaders have revealed the existence of a top-secret spy plane. Defense contractors are rumoured to have developed a revolutionary stealth technology based on a combination of smoke and mirrors. The unmanned aircraft is invisible to the naked eye and inaudible in flight. Air force spokesman General McBluster, speaking at a press event, told reporters "This technology will give us the ultimate edge. Some say the military simply pours money into fruitless projects but these are worth every cent at $1B each". The aircraft is fuelled by a new eco-friendly bio-fuel known as Snake Oil, which is manufactured by the plane's designers. Asked about any potential downsides, Gen McBluster commented "Sometimes they are difficult to find. Looking at them, you'd almost think the hangers were completely empty. Believe it or not we've even had engineers accidentally pour fuel on the ground where they thought the aircraft were parked! Actually, it happens rather a lot. So they really are remarkably stealthy. There's currently a glitch with the radios that is preventing communications with the aircraft. But the manufacturers assure us that these are just teething problems."
Slashdot Mirror
I can see his point. But maybe a good mechanic *would* clean his workshop whilst listening to music? Some of what a programmer does is just "housekeeping" and not equivalent to a mechanic at work on a vehicle. I like to switch music on or off depending on whether I need stimulation, relaxation or concentration. For mechanical stuff, or when I'm in "the zone", I think music can help propel me through greater productivity. When I really need to think something through in detail I usually switch it off.
I think the principle of requiring proof is good. However, I suspect a lot of managers would not consider being treated as one of the programmers to be the compliment that the programmers intend it as. If he's managing the programmers and perceives that as "I'm in control of the programmers" rather than "I'm here to help this team get its job done" he might not appreciate being approached to justify a management decision as though he were a peer. Not that your suggestions are a bad idea in all cases, it just requires a bit of care to determine what you can get away with.
I'd think for any given action it would be very helpful if the programmers could act as a group, to minimise the risk that an individual is perceived as "the awkward one", even if he is trying to represent the group's interests.
Well, even they are not concentrating 100% on the thing that keeps them alive - driving the car safely and efficiently. They'll also be managing their sleep cycle, planning fuel stop strategies, reporting back to the pits on the condition of the track and what tyres they need, analysing the car's behaviour to predict potential failures. Also, I would bet real money that many of them thought, at some point or other, "It'll be good when I've won the race" or "It'll be nice to sleep" during the driving.
They do demonstrate that people are capable of extraordinary levels of alertness and concentration for (relatively) short periods of time but even that isn't really 100%, the brain is a very parallel device so it would be surprising if it was only doing one thing!
Whether racing drivers can or cannot concentrate 100% though, I think we'd probably all agree that it's not reasonable to expect workers in a, say, 9am-5:30pm, 5 days a week job to be able to concentrate 100% on something for all their working hours.
There's a dude here: http://lwn.net/Articles/366266/
who suggests that the blob Nouveau currently downloads to the card is reasonably well understood and should get rewritten at some stage.
Another dude: http://lwn.net/Articles/366257/
Suggests that older GeForce cards don't need the blob.
I'm not wanting to contradict you, by the way, just wanted to note what the options are for people wanting to use Nvidia / Nouveau but who are wary of the blobs.
I'd have said the hardest part is reverse engineering how to talk to the hardware. If new hardware comes out they'll have to reverse engineer the new features / changes in that but they shouldn't have to start again from scratch.
This is good stuff, I think. They're not going to shove the whole Nouveau device driver into the kernel, it'll follow the modern X.org / Linux model of having kernel modesetting and a DRM driver in the kernel and a whole load of other stuff living in userspace.
Kernel modesetting (KMS) means that one entity, the kernel, always controls what graphics mode the video card is in. That's useful because pre-KMS, X.org might have changed the mode *without the kernel knowing*. That's one reason Linux can't easily have a Blue Screen Of Death - the kernel doesn't know what it can send to the graphics card to display it. BSOD isn't a feature you want to *see* but if you have a kernel panic, it'd be a lot more useful to actually see it, rather than it being hidden by your frozen X server! I'm not aware of graphical kernel panics currently being supported but at least it could be done in principle now, AIUI. KMS also reduces unnecessary modeswitching "to make sure" that you otherwise get, so switching between console and X should be quicker, as should switching between X sessions (fast user switching). KMS is also what's used by the new bootsplashes, like RedHat's Plymouth (which other distros, e.g. Mandriva) are also moving towards. DRM, in this context, is the Direct Rendering Manager and is how GL apps get direct rendering access to the graphics card, in a controlled way. I don't know so much about that though ;-)
The Nvidia open source driver "nv" doesn't support KMS or any 3D. The Nvidia proprietary driver doesn't support KMS but does support 3D (with good performance). Many distros have tended to use nv by default, some do ship nvidia though. Either way, you don't get the nice boot splash and neater terminal switching that a KMS driver would get you. The Nvidia proprietary driver is good performance-wise but it also tends to lag the open source drivers in terms of features a little; I think Nouveau (at one point? may not still be true) was aiming to support Xrandr features that Nvidia's did not. I've also heard that Nvidia's driver has issues with suspend.
Because of all this, expanding Nouveau support is a good thing. Nouveau are also in the process of reverse engineering for 3D support but they have some way to go. However, I've had the impression that it's getting towards being better than the 2D-only nv driver. So at *least* it will mean that when installing on your system you can expect a decent boot experience and correctly-working basic 2D graphics, with suspend/resume behaving sensible, etc. So it could be making life better for users *soon*. But as 3D support improves, things should get better still.
The Linux kernel devs generally take a stance these days that all kernel code ought to be merged into Linus's tree as soon as possible. It's really rather impressive to see this process working and the kernel devs (mostly) really following through on this.
Another way of putting (some configurations of) DRBD is probably "Networked block device RAID, but more flexible".
You want "make localmodconfig", which I think was also added recently, possibly to 2.6.32 actually. This builds a kernel using a local .config file, except that it only compiles modules that show up in lsmod. So if you boot off your vendor kernel with a squillion modules, let it load the modules you actually *use* then do make localmodconfig, you can make a kernel that only contains those modules. I don't know what it does if module names etc change, maybe you'd need manual fixup then - should still be less work than you currently are doing though.
There's some explanation here, though it might be for an out-of-date version of the patch:
http://linux.derkeiler.com/Mailing-Lists/Kernel/2009-09/msg04230.html
As the other reply said, make oldconfig is also useful to important settings from a previously configured kernel, can save a lot of time.
Doing it in software for purely virtual hardware is useful. I know it's been used to sync disks across the network on Xen hosts, the idea being that if the local and remote copies of the disk are kept in close sync, you can migrate a virtual machine with very low latency. Should be able to do similar tricks with other Linuxy VMMs. Having software available to do this stuff makes it easy to configure this sort of thing quickly, especially if you're budget-constrained, hardware-wise.
I think your comment about asking the engineers to solve the right problem is very insightful.
But I'm curious - did Windows have more fine-grained locking than a single kernel lock at the time Linux introduced SMP support with 2.0? I can imagine Windows may well have been better re locking scalability back then. Both Linux and Windows have been using increasingly fine granularity locking over the years, which is nice. It's somewhat frustrating that the Big Kernel Lock is still hanging around but at least it's not on most / any important critical paths now. And one day hopefully it will go away properly :-)
And I bet the mall owners state "privacy" as the reason and can't see the irony. Classy.
I object to that, sir - in the UK we have the finest stream contrivances money can buy. We've been building "railways" for a number of years and in London they even have horseless carriages! Simply spiffing. I'm typing this on one of our latest teletype machines, which is almost entirely automated. It only requires an office of 50 clerks or so to handle the Slashdot Javascript rendering. Actually, the clerks have something of an attitude time, insisting on collating and compiling the reports I requested on a "Just In Time" basis. I'm off to have a word with them now, toodle pip!
I understand in the UK that there's a specific inspection routine that custom vehicles can go through. This guy made a supermarket ride for children into a road-legal car and had it approved via that process IIRC: http://www.egmcartech.com/2009/05/14/man-builds-worlds-smallest-street-legal-car-gets-70-mpg/
That's the normal way of getting road legality for an unusual vehicle. There might be a less conventional method: I once read that there is a nuclear-powered car somewhere in the UK which, for some bizarre reason, was granted a royal dispensation to allow it on the roads - thereby bypassing the normal regulations. That probably also makes it a bit awkward to revoke; I imagine if it's true we just rely on the person who's got it being a good sport and keeping it in the garage.
" I profoundly believe that. And when you steal from us, you steal from them." That's neat. It's literally true in a sense. But I don't like the implied viewpoint that - because EA pays the artists a cut - the artists and EA's interests are perfectly aligned, cutting EA's profits is bad for artists, EA is a purely good force for the artists. Sounds somewhat similar to the music industry's "But you're stealing from the artists!" line. Some of these people may even believe this stuff but we all know it's a bit of a shaky argument.
I don't usually do this but really the parent has an excellent point. It's bad enough that they have increasingly large amounts of DLC you can't resell or buy on a second hand market. Treating the second hand market as basically piracy is a) bad for consumers and b) stupid. When they find a way to make me pay full price for all games by eliminating second hand sales *I will buy fewer games* and I therefore won't buy the DLC for them. The people who didn't like the games and traded them in so they can afford new games will also be able to afford less. They'll be shooting themselves in the foot but, like the music industry, they'll be doing it in a way which makes themselves *think* they're winning. And if they do that, it'll take ages for sanity to break out, if at all :-(
Maybe I'm being a bit dire but this is the way I imagine stuff going.
Agreed about not using clicking; I'd only really want the tabbing if I could switch using a keyboard shortcut.
It does seem maybe a bit redundant to having both virtual desktops and window tabbing but I think I'd use the extra level of "categorisation" it gives - in some circumstances anyhow.
Focus-follows-mouse is my preference but weirdly I'm finding it much less useful these days since I got the habit maximising everything and using apps with tabbing / tiling + keyboard navigation internally. Tried using tabbed window managers a bit and liked it but never found one that quite suited me.
I make very little use of extensions, so I've always assumed that the bloaty behaviour I'd seen from Firefox was largely due to something other than extensions. Mostly, I think the thing which slows my system to a crawl is Flash having a tantrum on a frequent basis, which doesn't change much across browsers. Chrome is good because it makes it easier to kill off Flash. But earlier today the browser to eat my memory sufficiently rapidly that it took about an hour to get access again and kill it properly. I'm blaming Flash for that. *sigh*
They're accurate for me ;-) But maybe I'm weird. I don't usually want to see emacs and console and PDF at the same time because if I'm editing the document in emacs the PDF will be out-of-date already. My current solution to this is to have emacs editor plus emacs shell running (I do often do that side-by-side but not always) in maximised emacs, plus a PDF viewer in another maximised window. The sequence is then edit in emacs -> compile in console -> view in PDF viewer. This might vary for others though.
Also, a tabbed workflow is something I'd be more likely to use on my netbook where I'm short on screen space than on my desktop. So it's a bit of a YMMV, I think.
Well, with this tabbing support hopefully that should become really easy for apps that don't implement their own tabbing internally. Especially if they were to add an "Expose all tabbed applications" and "Expose all tabs in this group", that would be very useful. But for things like Firefox and Chrome I guess we'd still want an extension like Windows 7 has that allows tabbed apps to provide previews to show on the taskbar etc.
Emacs 23 can actually embed a PDF viewer, can't it? So perhaps my example wasn't particularly well chosen since one can do it all already :-) FWIW I currently do exactly as you do! On my papers desktop I have a fullscreen emacs and fullscreen Okular and Alt-Tab between them (and sometimes a web browser for looking stuff up).
A more salient example is perhaps how I (used to) use Konqueror as a universal document browser: I'd browse the web in various tabs but also have PDF viewers open in others and file managers in yet others. It had a nice task-grouping property (e.g. windows relating to researching a particular topic, windows relating to generic browsing) that I've lost in recent years since I started playing with Firefox / Chrome. It would be pretty nice to be able to get the grouping ability back without the developers of these applications having to miraculously agree on a style (and embedding API) that fits my workflow. With WM support you could even tab a Windows app running under Wine, or a remote app over X forwarding / NX / xpra, which could be sweet. And when administering virtual machines I'd love to be able to tab together the VNC viewers.
Another thing is that (in *some* ways) it just seems like better software engineering to have tabs implemented outside applications - Windows has an extension to enable tabbed applications to preview their tabs when you mouse-over the taskbar. AFAIK only IE supports that currently. If it's implemented in the window manager then it's trivial to support that for all apps. And the apps themselves can potentially be simpler.
I don't think this would be revolutionary for me (that'll have to wait until there's full tiling support) but I think it'd be somewhat useful - and I can think of various ways it might be extended to make things even better.
KDE's BeOS window manager style allows you to move the tabs around the top of the window with a shift-drag (and/or attempts to avoid tabs getting hidden by moving them itself, not entirely sure how it works), which helps for tabbed-ish operation. Not as nice as a proper tabbing window manager but the BeOS style has been in KDE since forever. Could TWM do this, I wonder? I've seen that themed to look like BeOS although given its level of sophistication as a WM the theme was probably just "made the tabs yellow" ;-)
I've found the desktop reasonably stable since the KDE 4.1-4.2 era but I have also found that the experience varies a bit between distros. It's worth trying it out under a distro with a known-solid KDE 4 implementation before making a final judgement. (You may have already tried this, just thought it was worth noting).
I personally like the idea of having tabbed windows, however in support of your argument it's worth noting that KDE 4 already supports arbitrary drag-n-drop grouping of apps into named taskbar buttons: http://www.kdedevelopers.org/node/3864
Don't think GNOME has done this, don't know about XFCE. Compiz can do it, plus at least some basic tiling I think. And obviously not on Win or Mac. So KDE it the most "mainstream" desktop to have tabbed windows so far. But it's far from a new invention. There has also been talk of tiling support for KWin, the KDE window manager, which would make it even more useful. Various window managers using tabbing / tiling exist, such as ion, dwm, wmii, Xmonad, etc. They're nice but I missed the integration of having a full DE (though you can get it if you try). Partiwm is a project to create a more DE-friendly tabbing window manager but AFAICS it's gone a bit off track since its creator invented xpra and concentrated on that instead...
Friends of mine have observed that tabbing in the WM makes a lot of sense. Tab together a load of single browser instances and you have a multi-process web browser. OK, so it's not quite Chrome in security features but it's a heck of a lot simpler. Tab a load of terminals together and get a slick multi-terminal app. Tab OpenOffice together with your web browser whilst you're writing a report and researching stuff online. Tab together emacs + console running LaTeX + PDF viewer and get an integrated development environment for scientific papers. Nice.
I'm exaggerating the simplicity slightly but the point is that things are far more flexible if commonly-needed features (how many apps use tabs these days) are provided by the platform where possible.
Military leaders have revealed the existence of a top-secret spy
plane. Defense contractors are rumoured to have developed a
revolutionary stealth technology based on a combination of smoke and
mirrors. The unmanned aircraft is invisible to the naked eye and
inaudible in flight. Air force spokesman General McBluster, speaking
at a press event, told reporters "This technology will give us the
ultimate edge. Some say the military simply pours money into
fruitless projects but these are worth every cent at $1B each". The
aircraft is fuelled by a new eco-friendly bio-fuel known as Snake Oil,
which is manufactured by the plane's designers. Asked about any
potential downsides, Gen McBluster commented "Sometimes they are
difficult to find. Looking at them, you'd almost think the hangers
were completely empty. Believe it or not we've even had engineers
accidentally pour fuel on the ground where they thought the aircraft
were parked! Actually, it happens rather a lot. So they really are
remarkably stealthy. There's currently a glitch with the radios that
is preventing communications with the aircraft. But the manufacturers
assure us that these are just teething problems."