Debian's package management system is just that, a system, not a single app. You've got a few different apps involved. dpkg handles the package installation/removal, apt-get (or aptitude, or synaptic, or whatever) handles repository stuff like dependency management (dpkg checks them, but can't resolve them), apt-file and other utilities provide other functionality.
1) WINE is not Linux. The only reason you need to use WIN* is because you delegated the choice of OS to some software company that doesn't give a rat's arse about your computer security.
Irrelevant. It's the only way to run some apps I want or need on Linux, and it doesn't run them well. It's still a reason why I'm not using Linux. If WINE was perfect, wouldn't be a reason.
2) There is a GUI for mplayer. Actually, 3 or 4 of them. Don't confuse the screen candy with the functionality.
Care to name any? I named one, smplayer, which was the only one I could find that seemed decent, but it was horrendously buggy at the time. mplayer alone is not a media player, it can't even play some formats without changing commandline parameters between files. DVDs with bare mplayer is a headache, for example. Particularly the lack of menu support, something that MPC (and even VLC) does well.
3) You have desktop environment choice. OK, so if you don't like any of them, that's a valid excuse. But is it because they don't replicate Windows exactly, which is what you want because it is the only one you have experience with? Tell more.
Well, I've got experience with a bunch. I've used GNOME 2 extensively. About a year on my personal laptop, about a year worth of full time employment at a company using it. I used KDE 3.5 for a few months. Xfce for a few months. GNOME 2 was OK, but lacked the polish that modern Windows has. But GNOME 2 doesn't really exist anymore. Mint bringing traditional UI elements to GNOME 3 is hopeful, but I'm still not sure if the polish will be there.
4) Talk to the game developers. That's not a Linux decision.
Yes, it is a Linux decision. The point of this article is "things keeping me from using Linux". The lack of games is one thing. Talking to game developers (who will say "no market share, sorry" doesn't change that. I want to play games on my computer. I can't do that if I run Linux. It's very simple. As I said, WINE isn't really an option. I tried that approach. It took enormous amounts of tweaking just to get games working, and few games ran well enough to be considered "working".
5) Flash is (and always was) broken... period. But if you really want a working Flash, you need to take that up with Adobe. Only they know why their programmers can write portable code.
This repeats my previous point. At this point in time, Flash is needed on a desktop for a proper web experience, and it sucks on Linux (OS X is a bit wonky too, but it got a bit better). Yeah, it's not perfect on Windows, but again, it's something that was a headache on Linux, and it's another factor in why I'm not using Linux as my desktop OS (but am using it as my server OS).
6) Bugs in proprietary drivers are the responsibility of the developers. You did file a formal complaint with them for their incompetence, right?
Yeah, I complained to the vendors. I think I even made a GPL complaint to one of them. They don't listen. But this is again a point of, I don't want to take the effort. Windows 7 works very well, why should I want to go through all sorts of hoops with the hope of someday getting a system that works well when I have a system that works well already?
7) I don't really know what the issue was with that.
Well, I understand it just fine, it wasn't ready. Didn't change the fact that Chrome was my browser of choice, and wasn't available on what was at the time the platform I wanted to use.
8) This is an Apple issue. But it's not unexpected since they do want you to shell out money for their other products. That's how business works. You won't Linux doing that.
Right. It's an Apple issue. And by using Windows, it's none of my concern... It doesn't really matter whose fault any of this stuff is, because
I ran Ubuntu for something like a year. Don't remember the dates, I think it was somewhere around 2008. The issues I had then, that eventually pushed me back to Windows:
1) WINE sucks. I appreciate the effort and all, and they've done great work, but almost nothing runs perfectly, and few apps run well without tweaking. Very few games run well enough that I wouldn't feel like I was getting an inferior experience.
2) There were no good media players. mplayer has no GUI, VLC was bloated, smplayer was promising but extremely buggy... I wanted an equivalent of MPC-HC and there wasn't anything. VLC was the closest, but I'm not a big fan.
3) The UIs these days make me scratch my head. Unity? GNOME 3? KDE 4? I'd have to use something like Xfce of LXDE, and those don't have near the same polish as Win7.
4) No games. See point #1.
5) Flash is (or was, at least) broken on Linux. It was worse when I had to use 32-bit flash in a 64-bit browser using a wrapper that constantly crashed, but even after Flash went 64-bit native, it was just dog slow compared to Windows.
6) Multi-monitor support in Linux is non-existent. The only way to get decent multimon support is by using proprietary software from nVidia or ATI, and nVidia's was buggy and still required X reboots. Why the hell does Linux make me effectively reboot my machine just to show a video on my TV?
7) At the time, there was no Chrome for Linux, although this is resolved now, so consider this one historical.
8) No iTunes means I need to reboot to manage my iPhone. This is easier now that Apple has gone all cloudy (like iOS updates can be done sorta-OTA), but I still need to boot into Windows to do other stuff with it.
9) I always had lots of problem with sleep/hibernate/etc.
10) I've had dedicated or virtual Linux servers for almost a decade (since none of my complaints apply to servers, so I think it's the logical OS to use there), so anything I might need Linux to do can be done on my VPS at Linode sitting on a phat pipe. I've also got my file server at home running Solaris (FreeBSD soon) if I need *nix at home. Or the odd live CD. I do like parted magic.
11) Netflix. Doesn't work on Linux. This may change down the road once they work out adaptive bitrate streaming and DRM with HTML5, allowing Netflix to switch, but it'll probably be a while before this happens.
12) General lack of polish. Windows has a lot of it (not as much as OS X, but still decent), and Windows tends to be more reliable than Linux for desktop use. When I was a desktop (well, laptop) Linux user, stuff seemed to randomly break on its own, often. Windows 7 generally tends to be less troublesome.
Consider that the PowerVR SGX543MP products support up to 16 cores, but nobody has shipped one with more than 2 (Sony's PS Vita will be the first with 4). I believe the Mali-400 in the SGS2 is a 4-core part.
Considering that PS3 -> 2013 (when the ARM GPU is supposed to come out) is seven years, so we should see ~4 doublings, or 16x the performance that we saw in 2006 when the PS3 came out.
If we make the out-of-my-ass assumption that a 4-core mali-400 uses 2W of power at full load, and a 16-core T-658 will use 8W, we get the equivalent of a 128W GPU from 2006.
Yeah, I'm doing a lot of making up numbers, handwaving, and bullshitting, but I'm just trying to illustrate the point that a laptop-class ARM GPU from 2013 will probably be able to match the performance of the 2006 PS3's GPU. Moore's law and all that.
If that behaviour is killing your SSD, then you've bought an amazingly crappy SSD.
My first-gen Intel X25-m can do about 1.6 petabytes of writes over its lifespan. If I assume 10 megabit video, and watched streaming Flash video 24/7 (so that it would be constantly writing/erasing the video on the disk), it would take about 40 years to wear out my SSD. And that's assuming I streamed unique content 24/7...
Except Adobe is clearly moving towards killing off flash. They've announced the death of mobile flash in favour of HTML5, and it seems likely that desktop flash will eventually follow.
After all, it doesn't really matter to Adobe what runtime is used. They don't make money off people downloading the Flash runtime. They do make money off selling the flash designer software, which costs hundreds of dollars, and the streaming server solutions. What difference does it make if Adobe's software is spitting out stuff that runs in the flash runtime or HTML5?
My concern is that HTML5 is clearly not up to the level where it has feature parity (or stability/consistency) with Flash or Silverlight. For example, HTML5 currently has no agreed upon standard for dynamic audio. Sure, it can play a sound or music file, but if you want to actually generate or process audio, that's impossible. Mozilla and WebKit both have their own proprietary competing APIs to do this, but neither is final yet, and certainly not a standard.
Given a few years, HTML5 will probably be able to replace Flash/Silverlight, but clearly not yet!
The development branch of 8-stable, sure. But it's never going to make it into an actual release of 8.x. However, since FreeBSD 9.0 should be out within a few weeks, it's not that big a deal. Just means I have to wait.
And looking into it in more depth, it looks like Delphix and others are actively working on non-Oracle ZFS. They've switched from using version numbers to using named feature flags to differentiate, so that there are no version number collisions, and so that different groups can introduce different features. Of course, there's still the concern about fragmentation, and the loss of compatibility with Oracle isn't going to be fun.
Uh, no, how it compares to available Linux filesystems is not at all relevant to me, because my file server's ZFS v28 filesystem won't run on a version of FreeBSD that only supports v15. I've got to wait until FreeBSD 9.0 to migrate.
ZFS. Seriously, if you haven't used it then you [...]
The problem is that FreeBSD's implementation of ZFS in stable builds is extremely out of date. FreeBSD currently supports ZFS v15 (current closed-source is v33), which means you're missing a lot of the features. No triple-parity RAID, no deduplication, no encryption, no snapshot diffs, etc.
The good news is that FreeBSD 9.0 will bring this up to v28, the version used in the last release of OpenSolaris. My home file server is running OpenSolaris with a ZFS v28 storage pool, and I'm planning on trying to migrate to FreeBSD 9.0 as soon as it's out (RC2 should be out any day now, so close...)
Of course, the downside to all this is that ZFS is now effectively closed-source, and I'm not sure if we'll ever get anything newer than v28, unless it forks...
And that's what I had to do, although I couldn't get the 3TB WD drives I needed before they doubled in cost, I did manage to get 2TB drives after they had only increased 40-50%.
The article doesn't make that clear, although 7% is consistent with the size of the HDD market. But the impact on the market is more severe than that. Anybody using WD as a primary drive brand (for uniformity in enterprise applications) is going to get stuck with massive price increases: The cost of WD drives has more than doubled over the past few weeks, with a 3TB drive going from ~$130 to ~$280. Companies that aren't forced to stick with WD (computer vendors perhaps) will still face the task of validating new drives. And on top of all this, there is the hit in component manufacturing. Nidec makes 75% of all HDD spindle motors globally, and one quarter of their production capacity was in Thailand. That has been heavily disrupted, although they're starting to start those factories back up again. But the hiccup in production is certainly being felt.
Yes, companies can turn up additional production capacity elsewhere, but that takes time, and everybody is feeling the pinch in the meantime, with the inability to get drives.
In terms of a consumer trying to purchase drives, pretty much all stores (in Canada, at least) are limiting consumers to one or two drives per person (putting a real crimp on my plan to add another five disks to my fileserver), and ASUS is reporting that they'll run out of drives by the end of the month, at which point they won't be able to keep up with demand for their computers. So yes, it's pretty dire, although it's not exactly earth-shattering.
There is software for the iPhone that supports Ogg Vorbis. fstream, for example, can stream Vorbis over IceCast just fine (and does not require a jailbreak).
Unfortunately, it seems like playing back individual Vorbis files requires a jailbreak, or that you were lucky to grab VLC for the iPhone when it was available; I've got it, so I can play Vorbis on my iPhone (without jailbreaking), but that does't help others.
Well, Google was selling it to carriers as little as 6 months ago, so users that bought their device from a carrier like Videotron or Mobilicity are finding themselves without updates a mere 6 months later.
The article (and you) are implying that adding power (even internal) defeats much of the purpose, and puts us into BIGDOG or ALPHADOG type territory; would there not still be a large efficiency gain over traditional walking robots, such that an internal power source is much more feasible than it would otherwise have been?
Sort of; the default seems to be to put them beside the address bar, and the option is to move them below; in either configuration, all the empty space above the address bar is wasted and unusable.
Microsoft's value/effort ratio might have been a bit different, though. On a 1280x1024 display, IE9's default tab bar has room for only *two* tabs before they start shrinking. That makes the option a lot more useful than in Chrome, where you can have 5 tabs before they start shrinking, and moving it wouldn't change that much (might add one extra tab)
You'd have to be moving that mouse around pretty darned fast for that to happen, but I'll concede that it'd be something that takes relatively little code to implement. However something like an option to move the tab bar isn't so simple. It would require extra clutter in the options and a ton of code changes to get around the fact that this would break a whole bunch of assumptions (both code and design assumptions), themes, etc. It would be a large amount of work for a limited benefit for a very small number of users.
Chrome has an experimental option to move the tab bar to the side of the browser. They've been working on it for quite some time. Last time I tried it, early on, it was pretty buggy, indicating that this is really not a trivial change.
But if you're dragging a tab and accidentally tear it off, you can simply move the mouse back into the tab bar and it reattaches itself. This is no more effort than it would have taken to drag the tab into the correct position without tearing off, since you'd probably need to move the mouse back there anyhow.
Forking over inconsequential nonsense hurts the opensource ecosystem by spreading developers and duplicating effort. Everybody loses as a result. Forking should be reserved for important stuff.
Slashdot Mirror
First:
apt-file list {package_name}
Second:
apt-file search {file_name}
Debian's package management system is just that, a system, not a single app. You've got a few different apps involved. dpkg handles the package installation/removal, apt-get (or aptitude, or synaptic, or whatever) handles repository stuff like dependency management (dpkg checks them, but can't resolve them), apt-file and other utilities provide other functionality.
1) WINE is not Linux. The only reason you need to use WIN* is because you delegated the choice of OS to some software company that doesn't give a rat's arse about your computer security.
Irrelevant. It's the only way to run some apps I want or need on Linux, and it doesn't run them well. It's still a reason why I'm not using Linux. If WINE was perfect, wouldn't be a reason.
2) There is a GUI for mplayer. Actually, 3 or 4 of them. Don't confuse the screen candy with the functionality.
Care to name any? I named one, smplayer, which was the only one I could find that seemed decent, but it was horrendously buggy at the time. mplayer alone is not a media player, it can't even play some formats without changing commandline parameters between files. DVDs with bare mplayer is a headache, for example. Particularly the lack of menu support, something that MPC (and even VLC) does well.
3) You have desktop environment choice. OK, so if you don't like any of them, that's a valid excuse. But is it because they don't replicate Windows exactly, which is what you want because it is the only one you have experience with? Tell more.
Well, I've got experience with a bunch. I've used GNOME 2 extensively. About a year on my personal laptop, about a year worth of full time employment at a company using it. I used KDE 3.5 for a few months. Xfce for a few months. GNOME 2 was OK, but lacked the polish that modern Windows has. But GNOME 2 doesn't really exist anymore. Mint bringing traditional UI elements to GNOME 3 is hopeful, but I'm still not sure if the polish will be there.
4) Talk to the game developers. That's not a Linux decision.
Yes, it is a Linux decision. The point of this article is "things keeping me from using Linux". The lack of games is one thing. Talking to game developers (who will say "no market share, sorry" doesn't change that. I want to play games on my computer. I can't do that if I run Linux. It's very simple. As I said, WINE isn't really an option. I tried that approach. It took enormous amounts of tweaking just to get games working, and few games ran well enough to be considered "working".
5) Flash is (and always was) broken ... period. But if you really want a working Flash, you need to take that up with Adobe. Only they know why their programmers can write portable code.
This repeats my previous point. At this point in time, Flash is needed on a desktop for a proper web experience, and it sucks on Linux (OS X is a bit wonky too, but it got a bit better). Yeah, it's not perfect on Windows, but again, it's something that was a headache on Linux, and it's another factor in why I'm not using Linux as my desktop OS (but am using it as my server OS).
6) Bugs in proprietary drivers are the responsibility of the developers. You did file a formal complaint with them for their incompetence, right?
Yeah, I complained to the vendors. I think I even made a GPL complaint to one of them. They don't listen. But this is again a point of, I don't want to take the effort. Windows 7 works very well, why should I want to go through all sorts of hoops with the hope of someday getting a system that works well when I have a system that works well already?
7) I don't really know what the issue was with that.
Well, I understand it just fine, it wasn't ready. Didn't change the fact that Chrome was my browser of choice, and wasn't available on what was at the time the platform I wanted to use.
8) This is an Apple issue. But it's not unexpected since they do want you to shell out money for their other products. That's how business works. You won't Linux doing that.
Right. It's an Apple issue. And by using Windows, it's none of my concern... It doesn't really matter whose fault any of this stuff is, because
I ran Ubuntu for something like a year. Don't remember the dates, I think it was somewhere around 2008. The issues I had then, that eventually pushed me back to Windows:
1) WINE sucks. I appreciate the effort and all, and they've done great work, but almost nothing runs perfectly, and few apps run well without tweaking. Very few games run well enough that I wouldn't feel like I was getting an inferior experience.
2) There were no good media players. mplayer has no GUI, VLC was bloated, smplayer was promising but extremely buggy... I wanted an equivalent of MPC-HC and there wasn't anything. VLC was the closest, but I'm not a big fan.
3) The UIs these days make me scratch my head. Unity? GNOME 3? KDE 4? I'd have to use something like Xfce of LXDE, and those don't have near the same polish as Win7.
4) No games. See point #1.
5) Flash is (or was, at least) broken on Linux. It was worse when I had to use 32-bit flash in a 64-bit browser using a wrapper that constantly crashed, but even after Flash went 64-bit native, it was just dog slow compared to Windows.
6) Multi-monitor support in Linux is non-existent. The only way to get decent multimon support is by using proprietary software from nVidia or ATI, and nVidia's was buggy and still required X reboots. Why the hell does Linux make me effectively reboot my machine just to show a video on my TV?
7) At the time, there was no Chrome for Linux, although this is resolved now, so consider this one historical.
8) No iTunes means I need to reboot to manage my iPhone. This is easier now that Apple has gone all cloudy (like iOS updates can be done sorta-OTA), but I still need to boot into Windows to do other stuff with it.
9) I always had lots of problem with sleep/hibernate/etc.
10) I've had dedicated or virtual Linux servers for almost a decade (since none of my complaints apply to servers, so I think it's the logical OS to use there), so anything I might need Linux to do can be done on my VPS at Linode sitting on a phat pipe. I've also got my file server at home running Solaris (FreeBSD soon) if I need *nix at home. Or the odd live CD. I do like parted magic.
11) Netflix. Doesn't work on Linux. This may change down the road once they work out adaptive bitrate streaming and DRM with HTML5, allowing Netflix to switch, but it'll probably be a while before this happens.
12) General lack of polish. Windows has a lot of it (not as much as OS X, but still decent), and Windows tends to be more reliable than Linux for desktop use. When I was a desktop (well, laptop) Linux user, stuff seemed to randomly break on its own, often. Windows 7 generally tends to be less troublesome.
Consider that the PowerVR SGX543MP products support up to 16 cores, but nobody has shipped one with more than 2 (Sony's PS Vita will be the first with 4). I believe the Mali-400 in the SGS2 is a 4-core part.
Considering that PS3 -> 2013 (when the ARM GPU is supposed to come out) is seven years, so we should see ~4 doublings, or 16x the performance that we saw in 2006 when the PS3 came out.
If we make the out-of-my-ass assumption that a 4-core mali-400 uses 2W of power at full load, and a 16-core T-658 will use 8W, we get the equivalent of a 128W GPU from 2006.
Yeah, I'm doing a lot of making up numbers, handwaving, and bullshitting, but I'm just trying to illustrate the point that a laptop-class ARM GPU from 2013 will probably be able to match the performance of the 2006 PS3's GPU. Moore's law and all that.
If that behaviour is killing your SSD, then you've bought an amazingly crappy SSD.
My first-gen Intel X25-m can do about 1.6 petabytes of writes over its lifespan. If I assume 10 megabit video, and watched streaming Flash video 24/7 (so that it would be constantly writing/erasing the video on the disk), it would take about 40 years to wear out my SSD. And that's assuming I streamed unique content 24/7...
And silverlight has pretty good bitrate scaling support, I'm not sure if Flash supported that when Netflix was first implementing this stuff.
Except Adobe is clearly moving towards killing off flash. They've announced the death of mobile flash in favour of HTML5, and it seems likely that desktop flash will eventually follow.
After all, it doesn't really matter to Adobe what runtime is used. They don't make money off people downloading the Flash runtime. They do make money off selling the flash designer software, which costs hundreds of dollars, and the streaming server solutions. What difference does it make if Adobe's software is spitting out stuff that runs in the flash runtime or HTML5?
My concern is that HTML5 is clearly not up to the level where it has feature parity (or stability/consistency) with Flash or Silverlight. For example, HTML5 currently has no agreed upon standard for dynamic audio. Sure, it can play a sound or music file, but if you want to actually generate or process audio, that's impossible. Mozilla and WebKit both have their own proprietary competing APIs to do this, but neither is final yet, and certainly not a standard.
Given a few years, HTML5 will probably be able to replace Flash/Silverlight, but clearly not yet!
The development branch of 8-stable, sure. But it's never going to make it into an actual release of 8.x. However, since FreeBSD 9.0 should be out within a few weeks, it's not that big a deal. Just means I have to wait.
And looking into it in more depth, it looks like Delphix and others are actively working on non-Oracle ZFS. They've switched from using version numbers to using named feature flags to differentiate, so that there are no version number collisions, and so that different groups can introduce different features. Of course, there's still the concern about fragmentation, and the loss of compatibility with Oracle isn't going to be fun.
Uh, no, how it compares to available Linux filesystems is not at all relevant to me, because my file server's ZFS v28 filesystem won't run on a version of FreeBSD that only supports v15. I've got to wait until FreeBSD 9.0 to migrate.
Yes, there are laws against that. The government has a website with more information about it:
http://fightspam.gc.ca/
And the raw law (it was bill C-28) if you're interested:
http://www.parl.gc.ca/HousePublications/Publication.aspx?DocID=4547728
ZFS. Seriously, if you haven't used it then you [...]
The problem is that FreeBSD's implementation of ZFS in stable builds is extremely out of date. FreeBSD currently supports ZFS v15 (current closed-source is v33), which means you're missing a lot of the features. No triple-parity RAID, no deduplication, no encryption, no snapshot diffs, etc.
The good news is that FreeBSD 9.0 will bring this up to v28, the version used in the last release of OpenSolaris. My home file server is running OpenSolaris with a ZFS v28 storage pool, and I'm planning on trying to migrate to FreeBSD 9.0 as soon as it's out (RC2 should be out any day now, so close...)
Of course, the downside to all this is that ZFS is now effectively closed-source, and I'm not sure if we'll ever get anything newer than v28, unless it forks...
And that's what I had to do, although I couldn't get the 3TB WD drives I needed before they doubled in cost, I did manage to get 2TB drives after they had only increased 40-50%.
My Windows directory on my work machine is 24GB, and my home machine, which has been around a bit longer, is up over 30GB at this point.
The article doesn't make that clear, although 7% is consistent with the size of the HDD market. But the impact on the market is more severe than that. Anybody using WD as a primary drive brand (for uniformity in enterprise applications) is going to get stuck with massive price increases: The cost of WD drives has more than doubled over the past few weeks, with a 3TB drive going from ~$130 to ~$280. Companies that aren't forced to stick with WD (computer vendors perhaps) will still face the task of validating new drives. And on top of all this, there is the hit in component manufacturing. Nidec makes 75% of all HDD spindle motors globally, and one quarter of their production capacity was in Thailand. That has been heavily disrupted, although they're starting to start those factories back up again. But the hiccup in production is certainly being felt.
Yes, companies can turn up additional production capacity elsewhere, but that takes time, and everybody is feeling the pinch in the meantime, with the inability to get drives.
In terms of a consumer trying to purchase drives, pretty much all stores (in Canada, at least) are limiting consumers to one or two drives per person (putting a real crimp on my plan to add another five disks to my fileserver), and ASUS is reporting that they'll run out of drives by the end of the month, at which point they won't be able to keep up with demand for their computers. So yes, it's pretty dire, although it's not exactly earth-shattering.
DealExtreme has knockoffs for as low as a buck sixty, with free worldwide shipping.
This is no longer the case with iOS 5, which can activate devices without a PC.
There is software for the iPhone that supports Ogg Vorbis. fstream, for example, can stream Vorbis over IceCast just fine (and does not require a jailbreak).
Unfortunately, it seems like playing back individual Vorbis files requires a jailbreak, or that you were lucky to grab VLC for the iPhone when it was available; I've got it, so I can play Vorbis on my iPhone (without jailbreaking), but that does't help others.
the N1 is a fairly old device at this point
Well, Google was selling it to carriers as little as 6 months ago, so users that bought their device from a carrier like Videotron or Mobilicity are finding themselves without updates a mere 6 months later.
The article (and you) are implying that adding power (even internal) defeats much of the purpose, and puts us into BIGDOG or ALPHADOG type territory; would there not still be a large efficiency gain over traditional walking robots, such that an internal power source is much more feasible than it would otherwise have been?
Sort of; the default seems to be to put them beside the address bar, and the option is to move them below; in either configuration, all the empty space above the address bar is wasted and unusable.
Microsoft's value/effort ratio might have been a bit different, though. On a 1280x1024 display, IE9's default tab bar has room for only *two* tabs before they start shrinking. That makes the option a lot more useful than in Chrome, where you can have 5 tabs before they start shrinking, and moving it wouldn't change that much (might add one extra tab)
Really? I don't see an option in IE to move the tab bar around, although I didn't particularly look all that hard either.
You'd have to be moving that mouse around pretty darned fast for that to happen, but I'll concede that it'd be something that takes relatively little code to implement. However something like an option to move the tab bar isn't so simple. It would require extra clutter in the options and a ton of code changes to get around the fact that this would break a whole bunch of assumptions (both code and design assumptions), themes, etc. It would be a large amount of work for a limited benefit for a very small number of users.
Chrome has an experimental option to move the tab bar to the side of the browser. They've been working on it for quite some time. Last time I tried it, early on, it was pretty buggy, indicating that this is really not a trivial change.
But if you're dragging a tab and accidentally tear it off, you can simply move the mouse back into the tab bar and it reattaches itself. This is no more effort than it would have taken to drag the tab into the correct position without tearing off, since you'd probably need to move the mouse back there anyhow.
Forking over inconsequential nonsense hurts the opensource ecosystem by spreading developers and duplicating effort. Everybody loses as a result. Forking should be reserved for important stuff.