Fear: When you see B8 00 4C CD 21 and know what it means
If I'm not mistaken, that would be x86 machine code for "MOV AX, 4C00; INT 21;", which, as far as I can recall, is the DOS "system call" for terminating the current program, right? Maybe I'm being clueless, but what does that have with fear to do? It's not exactly a virus or a F00F bug or anything...
Well, actually, NTFS has supported journalling for a long time. What they've added this time is quite much cooler. It's actually the one thing that has me a bit excited about Vista. The transactions of which the article speaks are the ability to group multiple high-level file operations into one transaction. Such as "change this file and delete that lockfile" in one, atomic transaction. If one of the operations fail for one reason or another (insufficient process permissions, network outage, disk I/O error...), the entire transaction will be rolled back. Likewise, it will be rolled back if the system crashes during the commit.
I've been thinking about incorporating this kind of stuff into a filesystem myself for a quite some time now, and then I read about it coming into Vista. I'm very excited to see what kind of API they've put in place for it (though I don't hold too high expectations -- being Microsoft, I wouldn't be surprised if it's too complex and demanding to actually use anyway).
Oh, who'd have guessed? I mean, I would be as happy as anyone to see Microsoft release yet another buggy product rather than something which could actually compete on technical merits, but let's face it: It is a beta version. I'd be more surprised if it really were working well.
It's the same thing with Linux distros, too. I've tried the Fedora Core "test" versions more than a few times, and they are incredibly buggy (it hangs, programs segfault left and right, there are UI elements that just don't do anything, hand-drawn scribbles for certain icons, etc.). There isn't anything bad about it either, because that's the entire purpose of a beta release. Get it out and have people report the bugs.
(By the way, I really don't think that the brand of your computer case would affect the stability of Vista;)
Surely, you cannot expect people to actually know what Web 2.0 means? I mean, that tends to be rather hard about words which have no defined meaning...
Seriously though, there is an enormous amount of people who only use their computers to browse the web, IM, send mail and possibly edit the occasional document. I don't think that it would be unfeasible to implement an operating system and application programs in VHDL and burn it into an ASIC for those people. That way, they would definitely be safe from all kinds of malware, with the slight inconvenience of being unable to install new programs.
Most importantly, however, during normal usage, most idle time to due hard drive latency isn't spent waiting for the hard drive to transfer the data off the platters, but waiting for the head arm to seek to the right cylinder. Seek time ain't going to be reduced just by switching to another bus interface.
Sorry to be replying to my own post, but looking at THG's IOMeter benchmarks, my theory seems to be verifiable.
File server scenario: eSATA performance is above USB2, but at most points only very slightly.
Web server scenario: Performance of eSATA and USB2 is virtually identical.
Database scenario: eSATA is slight better than USB2. Interestingly, though, RAID1 is a lot better than RAID0.
Workstation scenario: Fluctuates greatly. At some points, eSATA even performs worse than USB2.
You're extremely unlikely to get any 3x speed improvement anyway. Sure, S-ATA may be able to sustain a much larger bandwidth than USB2 (although even then, 3 times USB2 seems very much... last I read, USB2 is clocked to 480 Mbps, i.e. 60 MBps), but to begin with, most hard drives can read no more than 60 MBps sustained.
Most importantly, however, during normal usage, most idle time to due hard drive latency isn't spent waiting for the hard drive to transfer the data off the platters, but waiting for the head arm to seek to the right cylinder. Seek time ain't going to be reduced just by switching to another bus interface.
Good evening. The last scene was interesting from the point of view of a professional logician because it contained a number of logical fallacies; that is, invalid propositional constructions and syllogistic forms, of the type so often committed by my wife.
'All wood burns,' states Sir Bedevere. 'Therefore,' he concludes, 'all that burns is wood.' This is, of course, pure bullshit. Universal affirmatives can only be partially converted: all of Alma Cogan is dead, but only some of the class of dead people are Alma Cogan. 'Oh yes,' one would think. However, my wife does not understand this necessary limitation of the conversion of a proposition; consequently, she does not understand me, for how can a woman expect to appreciate a professor of logic, if the simplest cloth-eared syllogism causes her to flounder?
And so on. I think my point is made without going onto the next section, which should, indeed, be avoided in this context for obvious reasons.
No, that is utterly incorrect. Iron-56 is the most stable isotope (of great natural abundance -- Nickel-62 is actually the most stable isotope, but due to circumstances unknown to me it is not produced in stellar fusion), but it is by no means the only stable isotope. In this context, X being "more stable" than Y means that the binding energy per nucleon is greater in X than in Y. Being "stable", however, means that the subject does not undergo spontaneous decay. If you want to prove me wrong, how about presenting me with the half-life of e.g. Hydrogen-1? (And no, proton decay does not count -- primarily because it would affect Iron as well)
There may not be any current solutions, but there are tons of scientific groups looking at methods for burning the waste further, until it becomes stable lead, at a gain. Even though none have succeeded right now, I think there's little doubt that someone will succeed within the next couple of hundred years (a relatively short time), and then all that needs to be done is dig up the waste again and use it to generate more energy. Thus, I don't think nuclear waste is a great problem.
I think there's some kind of misunderstanding. Since they also say that it is a "competition between two viable alternatives", I'm sure they aren't talking about IE vs. FF. Maybe it was Opera or Safari they compared to Firefox? Either way, I think it is rather weird of them to make a study of the browser wars without mentioning IE (unless, maybe, they don't consider IE to be a browser, which I could understand), but who am I to speak? I've never understood marketing studies anyway. Or IE, for that matter.
Open source simply means that anyone may look at the code, not necessarily that you have complete rights to do anything with it.
I have no idea whence you got that idea, but the people who coined the term "Open Source" seem to think otherwise. From The Open Source Definition:
3. Derived Works
The license must allow modifications and derived works, and must allow them to be distributed under the same terms as the license of the original software.
Somehow, I'm just having a very hard time believing this to its full extent. DRM needs a full chain of trust, complete down to the BIOS, to even stand a chance of being effective, and since GStreamer (and GNOME) are seldomly, if ever, run on "trusted" kernels, it just seems completely and utterly pointless to implement DRM in GStreamer.
That being said, I can't deny that the work is being done. If anyone's interested, here's a link to the announcement. The short version seems to be that "we have no control over if people will be using DRM anyway, so we're implementing it in the hopes that it will never be used, just so that GStreamer won't be dumped over lack of DRM support".
You get a icon on your desktop that you can click, then browse software categories. When you find something you like, just click the install button and voila, CNR downloads it, puts an entry in the Start menus and puts an icon on the desktop. No other distro that I know of does this with such ease.
I'm not saying you're not right, but from that description it sounds exactly like both Ubuntu's "light-weight" synaptic (I don't know what they call it, apart from "Add or remove software") and FC's Pirut. Is there any significant difference?
Really? Is it not weird, then, that Sun's octuple-core T1 processor outclassed the competition with at least 2:1, and normally closer to 3:1, in the last SPECweb round?
How does that correspond to the theory that the Sun has lived about half of its projected lifespan? Does the process accelerate, or does the Sun die when it's made up of 50% Helium?
But truth is, I sincerily hope all my software is obsolete by the time I die!
Well, I can understand your reluctance to upgrade to Vista (or any later version of Windows), but you know, there are other operating systems out there.
How about/proc/*/maps? It indicates that only the heap of a program such as calctool is as large as 2,670,592 bytes (652 pages). And mind you, that's just the heap. I haven't counted bss segments, private mappings of SO.data segments or the stack, or any other process-private mappings. How about the fact that it takes 0.68 seconds of CPU time just to start and exit gcalctool? And that's just for tiny programs like gcalctool -- don't get me started on programs like evo (or, especially, Firefox, only that's not GNOME).
How about the fact that when I log out from GNOME, my free memory increases by around 500 MiB? I would think that is about as accurate a measure as you can get. And indeed, I then count the free pages excluding blockdev caches. I don't really mind that it uses much memory, if only it weren't for the fact that I'll have to wait 10 seconds every now and then to get a process paged in from swap.
I mean, don't get me wrong and all -- I use GNOME and I like it, but it is terribly resource-inefficient, and the vast part of all dynamic memory allocated in GNOME programs is most likely not allocated by glib or gtk (since I have written programs myself using Gtk that don't use at all as much memory as even gcalctool). I just can't help thinking that it should be possible to optimize it a lot. (Of course, before you mention it, I have been thinking about taking the task upon myself, but I just have too many other things to do.) I read that one of the large pieces of news about GNOME 2.14 was that it's supposed to be heavily optimized, but I haven't tried it yet (it's not unmasked on portage yet).
Of course, that wasn't what I meant. They probably had a good (or at least more than good enough) architecture from the outset, only it wasn't very well implemented. After that, they've gone around the place several times and made lots of optimizations.
I agree with you on one point: Windows XP is probably faster in many respects than running GNOME on FC5 or Ubuntu 5.10, but I'd have to respectfully disagree with basically everything else you write.
because while hardware advanced, Linux still uses technology from 20 years ago to talk with the graphics card (X11)
The performance problems in no way lie in X11. It's more than well possible to run X11 programs that are exactly as fast as Windows programs. The problems lie in GNOME and KDE using far too much memory (and probably CPU cycles as well). That's GNOME and KDE, though -- not X11. Try any alternative window manager/desktop environment, and you're likely to see a vast speed improvement.
still lacks kernel audio mixing
So does Windows, though. Neither Windows nor Linux uses kernel audio mixing -- they rely on hardware mixing instead. All somewhat modern sound cards have several PCM subchannels that operating systems use in order to play several sounds simultaneously, and, yes, it is perfectly supported by Linux. Last I tried (admittedly, that was some time ago, but I can't remember just how long), using Windows with a single-channel sound card meant that I could only play one sound at a time.
still lacks in PnP department (removing a "mounted" USB flash stick anyone?)
I really don't know what you're getting at here. There are no problems whatsoever with removing a mounted USB flash stick, and there hasn't been since at least FC3.
still has abysmal support for various multimedia devices
So does Windows, though. Basically all those multimedia devices are supported on Windows by third-party drivers as well, only that they are written by the device manufacturers instead of independent reverse-engineering developers. Since that doesn't matter in practice, though, I cede this point to you.
I'm not even going to mention hibernation because I know that doesn't work properly in Linux.
That's weird, because I happen to know that it does work properly in Linux (at least on those computers that have proper ACPI hardware implementations). I both suspend and (periodically) hibernate my Linux laptop (a Dell Latitude CP), and I've never run into a problem with it. And yes, 15 seconds sounds about right.
And when Vista comes out this or next year (who cares when, it will be awesome anyway) - it will be even faster.
Well, that remains to be seen. I definitely wouldn't bet on it.
Re:Apple Provides SOME Legacy Support
on
Why Windows is Slow
·
· Score: 4, Informative
Moreover, when Apple releases an OS upgrade, your old machine gets faster, not slower.
Nothing bad about Apple, but that really isn't because their programmers are so much better than Microsoft programmers. It's just that Mac OS X was designed so horribly at first that there's really no way to go but up in each incremental release. It was only in 10.4 that they implemented somewhat fine-grained kernel locking (10.3 used two kernel locks: one for the networking code, and one for the rest of the kernel, while 10.2 and earlier only had one Big Kernel Lock).
If you ask me, though, that's the right design decision. First you make a system that works according the specifications, and only after that should you worry about optimizations.
If Microsoft would obey the law, they wouldn't have to pay these fines and settlements and whatnot.
I don't know the exact phrasing, but there is a quote which goes along the lines of "When the law is unjust, the just ones are in prison". Somehow, it seems applicable here. I don't want to seem like a Microsoft apologist (I, too, hate Windows and has been using Linux exclusively the last five years), but I really don't like the fact that Microsoft aren't allowed to decide for themselves what their products are supposed to contain. No matter how much I would like to see Microsoft eradicated from the face of the Earth, I still think this is the wrong way to go about it.
A monopoly is a company which is sponsored by the state in such a manner that others are prohibited by law to compete with them. Hence, Microsoft is not a monopoly. There are competitors, and they are alive and well. Even though Microsoft does engage in many activities which I would deem immoral, I don't think mere immorality should be outlawed. There are both Macintoshes and PCs that are sold with Linux pre-installed -- still, people buy computers with Windows pre-installed, because that is what they want, for whatever reasons.
Slashdot Mirror
I've been thinking about incorporating this kind of stuff into a filesystem myself for a quite some time now, and then I read about it coming into Vista. I'm very excited to see what kind of API they've put in place for it (though I don't hold too high expectations -- being Microsoft, I wouldn't be surprised if it's too complex and demanding to actually use anyway).
It's the same thing with Linux distros, too. I've tried the Fedora Core "test" versions more than a few times, and they are incredibly buggy (it hangs, programs segfault left and right, there are UI elements that just don't do anything, hand-drawn scribbles for certain icons, etc.). There isn't anything bad about it either, because that's the entire purpose of a beta release. Get it out and have people report the bugs.
(By the way, I really don't think that the brand of your computer case would affect the stability of Vista ;)
It's also "Hello World" in APL.
Surely, you cannot expect people to actually know what Web 2.0 means? I mean, that tends to be rather hard about words which have no defined meaning...
Seriously though, there is an enormous amount of people who only use their computers to browse the web, IM, send mail and possibly edit the occasional document. I don't think that it would be unfeasible to implement an operating system and application programs in VHDL and burn it into an ASIC for those people. That way, they would definitely be safe from all kinds of malware, with the slight inconvenience of being unable to install new programs.
Oh, you youngsters with your bitmapped CRT displays. I'm sitting here with my Model 37 TTY, typing away at 150 baud. And I'm happy with that!
Most importantly, however, during normal usage, most idle time to due hard drive latency isn't spent waiting for the hard drive to transfer the data off the platters, but waiting for the head arm to seek to the right cylinder. Seek time ain't going to be reduced just by switching to another bus interface.
No, that is utterly incorrect. Iron-56 is the most stable isotope (of great natural abundance -- Nickel-62 is actually the most stable isotope, but due to circumstances unknown to me it is not produced in stellar fusion), but it is by no means the only stable isotope. In this context, X being "more stable" than Y means that the binding energy per nucleon is greater in X than in Y. Being "stable", however, means that the subject does not undergo spontaneous decay. If you want to prove me wrong, how about presenting me with the half-life of e.g. Hydrogen-1? (And no, proton decay does not count -- primarily because it would affect Iron as well)
There may not be any current solutions, but there are tons of scientific groups looking at methods for burning the waste further, until it becomes stable lead, at a gain. Even though none have succeeded right now, I think there's little doubt that someone will succeed within the next couple of hundred years (a relatively short time), and then all that needs to be done is dig up the waste again and use it to generate more energy. Thus, I don't think nuclear waste is a great problem.
I think there's some kind of misunderstanding. Since they also say that it is a "competition between two viable alternatives", I'm sure they aren't talking about IE vs. FF. Maybe it was Opera or Safari they compared to Firefox? Either way, I think it is rather weird of them to make a study of the browser wars without mentioning IE (unless, maybe, they don't consider IE to be a browser, which I could understand), but who am I to speak? I've never understood marketing studies anyway. Or IE, for that matter.
That being said, I can't deny that the work is being done. If anyone's interested, here's a link to the announcement. The short version seems to be that "we have no control over if people will be using DRM anyway, so we're implementing it in the hopes that it will never be used, just so that GStreamer won't be dumped over lack of DRM support".
Really? Is it not weird, then, that Sun's octuple-core T1 processor outclassed the competition with at least 2:1, and normally closer to 3:1, in the last SPECweb round?
How about the fact that when I log out from GNOME, my free memory increases by around 500 MiB? I would think that is about as accurate a measure as you can get. And indeed, I then count the free pages excluding blockdev caches. I don't really mind that it uses much memory, if only it weren't for the fact that I'll have to wait 10 seconds every now and then to get a process paged in from swap.
I mean, don't get me wrong and all -- I use GNOME and I like it, but it is terribly resource-inefficient, and the vast part of all dynamic memory allocated in GNOME programs is most likely not allocated by glib or gtk (since I have written programs myself using Gtk that don't use at all as much memory as even gcalctool). I just can't help thinking that it should be possible to optimize it a lot. (Of course, before you mention it, I have been thinking about taking the task upon myself, but I just have too many other things to do.) I read that one of the large pieces of news about GNOME 2.14 was that it's supposed to be heavily optimized, but I haven't tried it yet (it's not unmasked on portage yet).
Of course, that wasn't what I meant. They probably had a good (or at least more than good enough) architecture from the outset, only it wasn't very well implemented. After that, they've gone around the place several times and made lots of optimizations.
If you ask me, though, that's the right design decision. First you make a system that works according the specifications, and only after that should you worry about optimizations.
A monopoly is a company which is sponsored by the state in such a manner that others are prohibited by law to compete with them. Hence, Microsoft is not a monopoly. There are competitors, and they are alive and well. Even though Microsoft does engage in many activities which I would deem immoral, I don't think mere immorality should be outlawed. There are both Macintoshes and PCs that are sold with Linux pre-installed -- still, people buy computers with Windows pre-installed, because that is what they want, for whatever reasons.
[/rant]