I might agree with your other points, but I cannot say that I consider it being bad netiquette. After all, when using no-listing, everything is still RFC-compliant. Servers are fully allowed to reject mails for whatever reason they choose to, and MX records must be tried in order of priority, so I don't at all see how it would be bad netiquette.
I fully agree with your second point, though, but the same goes for greylisting, which I'm using with quite some success (although many spammers are already learning to bypass it, it is still decreasing my spam rate compared to previously by a lot).
While I agree with your third point as well, the "actual problems" of which you speak are most likely unsolvable, and I think that you, too, realize that. In light of that, what choice does one have except solving the symptoms?
By no means. Just because the sides in a debate aren't neutral, does not mean that reporting on a debate cannot be done from a neutral point of view. If you look at the Wikipedia page on the matter, you'll see that (except the things which remain to be cleaned up) it mostly writes about facts -- such as "OpenDocument is defined by OASIS, a not-for-profit, international consortium that drives the development, convergence, and adoption of e-business standards." -- and for the opinionated points, present them as being someone's opinions -- such as "Alex Hudson, J. David Eisenberg, Bruce D'Arcus and Daniel Carrera of the OpenDocument Fellowship wrote an article published by the online journal GrokLaw that argues OpenDocument has several technical advantages over Office Open XML (Hudson, 2005[1]). The article examined some problems based on the original draft of the Office Open XML standard (which has since been superseded), and claims the following differences:".
I think that you, too, miss the point. Even though corporations may be pseudo-persons with no sense of morality and with the sole purpose of amassing money and power, I think it is important to note that they are, indeed, pseudo-persons. They have not actually any consciousness, will or deciding ability in themselves. Normally, that task is carried out by actual humans.
Somewhere, deep inside the twisted corridors in Redmond, some person must have actually thought of the idea to hire third parties to edit Wikipedia. He must also have presented it to his boss (unless it was some boss who thought of the idea himself), who in turn must have ordered someone to carry out the plan. Shouldn't an obviously unethical plan such as this have been stopped at some point in this chain? Shouldn't that boss figure have some kind of conscience which should have stopped him from doing this? Another problem may be the current inability (real or imagined) of "peons" in a corporation to themselves stop such plans when being ordered to carry them out. Generally, I believe that the lack of personal responsibility for actions being carried out "in the name of a corporation" is the real culprit.
Also, aside from the ethical standpoint, must they not have realized that this would leak out?! I mean, this cannot be considered positive PR, right?
Isn't it possible to use HDMI without HDCP? As far as I thought I knew, it is required for HDMI-compliant devices to implement HDCP, but not to always use it, just that the source may (or may not, depending on the source) choose to reduce quality if HDCP is not used. If this is the case, then isn't it possible to just turn HDCP off, or isn't it the case?
If you change the velocity the bullet exits the muzzle of the cannon (or the railgun or whatever), you are making the bullet go higher/lower at the other side, and then hitting the cannon faster/slower when it returns. That is, unless it reaches escape velocity (it'll never return) or hits the planet.
I haven't done the math on this yet, but it sounds to me as though there is an energy threshold, below which the projectile will return and hit the cannon, and above which it has reach escape velocity and increases altitude forever. Then what would happen if one fires the projectile at (somewhat) exactly that threshold? Wouldn't it go into a non-cannon-intercepting (if maybe somewhat unstable and hard to maintain) orbit?
I would argue that FreeBSD does, in fact, not run on x86 compatible CPUs in general, but rather i386 compatible CPUs. If I'm not vastly misled, x86 means 8086 and forward (16 bit, real mode, no MMU), while i386 basically means IA32.
I've seen many people so far in this article speak for the greatness of the Fahrenheit for daily use, and that's one thing I don't get. The greatness of centigrades is the placement of the zero, which is fantastically practical when speaking about the weather. If it's above zero, there's no ice, frost, snow or hail, and if it's below zero, the converse is true. When it comes to virtually anything else, I'd say they are equal in utility -- it's just a matter of being used to it. Being raised with centigrades, I can tell perfectly fine whether I'm going to feel alright or not if I hear on the forecast that it's going to be -20, -10, 0, 10, 20 or 30 degrees tomorrow. The placement of zero degrees Celsius, however, makes a lot of difference when measuring temperatures in daily life.
Likewise, if I were to measure my room, I would simply walk across it normally, and it would be about as many meters as the number of steps I took. My steps are almost exactly 1 meter long, and it works pretty reliably (and mind you, taking 5 steps is even less work than placing your feet in front of each other 16 times).
However, the greatest issue isn't the base units themselves, but rather conversions between them. As the GP said, moving the decimal comma is a lot easier than multiplying by 12 and then 16, and switching to hexadecimal still wouldn't help, since the factor 3 would still remain in a lot of internal conversions. Not to mention that the same base unit is used for length, area and volume measurements.
I find the greatest problem with the imperial units to rather be the interdependencies between them. With the metric system, those are standardized for all units. Just as you have kilogram and milligram, you have kilometer, millimeter, kilosecond, millisecond, and so on, and they are always apart by even powers of ten. You also have meters, square meters and cubic meters, with the special case of the liter being a cubic decimeter. In the imperial system (I don't even remember it all), you have 1 mile, being 1760 yards, each being 3 feet, each being 12 inch. I always see smaller things being measured in units like 1/3, 1/12 or 1/16 inch. Then there is 231 cubic inches = 1 gallon, each being 4 quarts, each being 2 pints, each being 16 fluid ounce.
Even if all the world standardized their feet on being a U.S. foot, that problem wouldn't disappear.
Just wait until you start receiving MS Word e-mail (now that it will probably be possible). In that case, storage space won't be your problem anymore...
I think there might be some slight difference in that software actually doesn't require any money to develop, while hardware development requires pretty many potentially quite expensive parts (those FPGAs and radio amplifiers and what not cost far too much to play with as a hobby, at least). Add to that the fact that if you were to screw up along the line, the risk isn't just that you'll have to reinstall your operating system, but that you might fry your motherboard.
The FHF might be a possibility when we all have nanoassemblers...
my personal belief is that the future, the nebulous area Stroustroupe outlines as "better concurrency," is really going to be implemented at a platform level as this kind of deeply nested transactional data structuring, where instead of overwriting your object to change its state, you simply append the new state in a new part of memory. thus each object accumulates address space (referentiability) as it changes across time. i'll leave the full details implementation & ramifications of Copy on Update up to the user for now.
Yeah! After all, just modifying a single word and allocating a new structure, changing it, and updating all pointers to it are probably just 2 or 3 orders of magnitude apart in performance. Now that we have these fast CPUs, we have to put all those cycles to use somewhere, don't we?
ZFS gets away with it since it's a filesystem driver, and secondary memory is still many more orders of magnitude apart and has a much lesser frequency of access, but to do it with all data structures in memory seems like taking us back to the time of ~1e6 Hz CPUs. I'd say semaphores will be the way to go until the day this can somehow be implemented in hardware. It does have its place here and there (and is used to some extent in the Linux kernel, under the name of RCU (Read, Copy, Update)), but going as far as using it on all shared structures is another thing entirely. Seems like a disaster for caching as well.
Well, one probably shan't read too much into it, but I did find the phrasing "Novell and Microsoft have commissioned a survey to prove that customers love their interoperability and patent deal" a bit funny. If I were them, I would have phrased it something like "...have commissioned a survey to find out whether customers love...", but maybe that's just me.
Just because the bug is severe doesn't mean it's easy to reproduce.
I would argue against that. If it is a bug that requires the system to be reinstalled whenever it is installed on, say a hard drive with an even number of sectors, then it is a severe bug. If, however, it has to be reinstalled whenever it is installed on, say, a system which combines the rare 82LM3X chip from Foobar electronics, an nVidia GeForce 5900GT with 256 MB of VRAM and it was installed on the leap day under a full moon, then it is indeed not a severe bug, and even less a showstopper, even though the symptoms are exactly the same. Of course the reproducability counts as a prioritizer for a bug.
Correct me if I'm wrong, but isn't this the same guy who first bitched about the UAC always asking for passwords to do anything administrative? (read: mimics SU, only more annoyingly)
I'm not saying that this is your opinion of choice of words, but I've heard it elsewhere as well, and I just wanted to clear it up a bit -- UAC has never been like su/sudo. su never automagically captures your terminal whenever you, as a normal user, attempt to run "rm -rf/" and asks you to type your password. Rather, you'll get a permission error would you ever try, and then you'd need to manually run su/sudo to get a root session.
If you ask me, the way su/sudo works is far better, and I always thought Microsoft should design Vista the same way. For example, a li'l button in the quick launch field that you click, type in your password, and get your privileges escalated. Then you click it again to revoke those privileges. The largest problem I see with UAC is that phishingware can fake the dialog, and make you give it your password. Since the UAC box pops up "on demand", you won't know if it is Windows which pops it up for you or if it is a phishing program that pops up one which looks just like it. Probably 90% of all users (and I wouldn't put many corporate administrators beyond it either) will just type in their password without thinking twice. In particular, think of what would happen if the next Gator included such a thing in their installation program -- people will be used to get a UAC prompt during installation anyway. su, on the other hand, will never ask for your password without you asking it to do so.
Amen to that. I just wanted to add to this a mention of Autopackage, which has done wonders, not only for amazing ease of installation (did you say "a few clicks"? Autopackage is one-click), but also for wide-spread binary compatibility under GNU/Linux. GNU/Linux certainly is capable of binary compatibility, it's just that most distros don't care anyway (why should they when there's APT/Yum/Portage/whathaveyou?). The Autopackage project has developed some rather sophisticated wrapper scripts around the GNU toolchain to build binaries that link to ABIs that are available on as wide a range of distros as possible.
I don't mean to troll, but I'm honestly wondering -- what niche does NetBSD have anymore, really? From all I can see, {Free,Open}BSD have lots more capabilities, and from what I hear, NetBSD doesn't really work on that many archs these days anyway (Linux probably works on more archs, doesn't it?). Is anyone using NetBSD for any good reason and able to tell me why?
I'm no quantum psysicist, but if there were particles (virtual or no) of negative mass, would they not be repelled from a black hole, which has positive mass? Which would in turn mean that only the particles of positive mass could ever enter a black hole.
I agree that there may be a problem with lack of device drivers, although even that has become much better as of late (in fact, whenever I pop the Ubuntu 6.06 Live CD in a laptop, WiFi seems to be working in the (maybe even vast) majority of cases). In the cases that it doesn't work out of the box, I haven't ever found that ndiswrapper doesn't make it.
But regardless of that, I don't see what grudges you hold against "WiFi in general", or WPA. I've played a lot with NetworkManager on Ubuntu, and I've yet to encounter a single problem, basically. WPA and WEP works, with the keys beautifully stored in the gnome-keyring-manager. Scanning works. I can't think of anything that doesn't work. Care to enlighten me?
Not that I don't agree with you that Microsoft products tned to be rushed and not ready by the time they're released, but the arguments you make for it aren't exactly convincing. Please mention a single software product ever that hasn't had later version with bugfixes.
Slashdot Mirror
"People in the 21st century must have really loved these grey characters on blue background" -- A 31st century archeologist.
I fully agree with your second point, though, but the same goes for greylisting, which I'm using with quite some success (although many spammers are already learning to bypass it, it is still decreasing my spam rate compared to previously by a lot).
While I agree with your third point as well, the "actual problems" of which you speak are most likely unsolvable, and I think that you, too, realize that. In light of that, what choice does one have except solving the symptoms?
By no means. Just because the sides in a debate aren't neutral, does not mean that reporting on a debate cannot be done from a neutral point of view. If you look at the Wikipedia page on the matter, you'll see that (except the things which remain to be cleaned up) it mostly writes about facts -- such as "OpenDocument is defined by OASIS, a not-for-profit, international consortium that drives the development, convergence, and adoption of e-business standards." -- and for the opinionated points, present them as being someone's opinions -- such as "Alex Hudson, J. David Eisenberg, Bruce D'Arcus and Daniel Carrera of the OpenDocument Fellowship wrote an article published by the online journal GrokLaw that argues OpenDocument has several technical advantages over Office Open XML (Hudson, 2005[1]). The article examined some problems based on the original draft of the Office Open XML standard (which has since been superseded), and claims the following differences:".
Somewhere, deep inside the twisted corridors in Redmond, some person must have actually thought of the idea to hire third parties to edit Wikipedia. He must also have presented it to his boss (unless it was some boss who thought of the idea himself), who in turn must have ordered someone to carry out the plan. Shouldn't an obviously unethical plan such as this have been stopped at some point in this chain? Shouldn't that boss figure have some kind of conscience which should have stopped him from doing this? Another problem may be the current inability (real or imagined) of "peons" in a corporation to themselves stop such plans when being ordered to carry them out. Generally, I believe that the lack of personal responsibility for actions being carried out "in the name of a corporation" is the real culprit.
Also, aside from the ethical standpoint, must they not have realized that this would leak out?! I mean, this cannot be considered positive PR, right?
Isn't it possible to use HDMI without HDCP? As far as I thought I knew, it is required for HDMI-compliant devices to implement HDCP, but not to always use it, just that the source may (or may not, depending on the source) choose to reduce quality if HDCP is not used. If this is the case, then isn't it possible to just turn HDCP off, or isn't it the case?
I wouldn't be surprised if it's OOo Writer or MS Word. Is there any other tool out there which matches the output quality of LaTeX?
I would argue that FreeBSD does, in fact, not run on x86 compatible CPUs in general, but rather i386 compatible CPUs. If I'm not vastly misled, x86 means 8086 and forward (16 bit, real mode, no MMU), while i386 basically means IA32.
Likewise, if I were to measure my room, I would simply walk across it normally, and it would be about as many meters as the number of steps I took. My steps are almost exactly 1 meter long, and it works pretty reliably (and mind you, taking 5 steps is even less work than placing your feet in front of each other 16 times).
However, the greatest issue isn't the base units themselves, but rather conversions between them. As the GP said, moving the decimal comma is a lot easier than multiplying by 12 and then 16, and switching to hexadecimal still wouldn't help, since the factor 3 would still remain in a lot of internal conversions. Not to mention that the same base unit is used for length, area and volume measurements.
Even if all the world standardized their feet on being a U.S. foot, that problem wouldn't disappear.
Just wait until you start receiving MS Word e-mail (now that it will probably be possible). In that case, storage space won't be your problem anymore...
The FHF might be a possibility when we all have nanoassemblers...
ZFS gets away with it since it's a filesystem driver, and secondary memory is still many more orders of magnitude apart and has a much lesser frequency of access, but to do it with all data structures in memory seems like taking us back to the time of ~1e6 Hz CPUs. I'd say semaphores will be the way to go until the day this can somehow be implemented in hardware. It does have its place here and there (and is used to some extent in the Linux kernel, under the name of RCU (Read, Copy, Update)), but going as far as using it on all shared structures is another thing entirely. Seems like a disaster for caching as well.
Well, one probably shan't read too much into it, but I did find the phrasing "Novell and Microsoft have commissioned a survey to prove that customers love their interoperability and patent deal" a bit funny. If I were them, I would have phrased it something like "...have commissioned a survey to find out whether customers love...", but maybe that's just me.
I hate saying this (especially on Slashdot...), but this is actually an argument in favor of TPM.
If you ask me, the way su/sudo works is far better, and I always thought Microsoft should design Vista the same way. For example, a li'l button in the quick launch field that you click, type in your password, and get your privileges escalated. Then you click it again to revoke those privileges. The largest problem I see with UAC is that phishingware can fake the dialog, and make you give it your password. Since the UAC box pops up "on demand", you won't know if it is Windows which pops it up for you or if it is a phishing program that pops up one which looks just like it. Probably 90% of all users (and I wouldn't put many corporate administrators beyond it either) will just type in their password without thinking twice. In particular, think of what would happen if the next Gator included such a thing in their installation program -- people will be used to get a UAC prompt during installation anyway. su, on the other hand, will never ask for your password without you asking it to do so.
Amen to that. I just wanted to add to this a mention of Autopackage, which has done wonders, not only for amazing ease of installation (did you say "a few clicks"? Autopackage is one-click), but also for wide-spread binary compatibility under GNU/Linux. GNU/Linux certainly is capable of binary compatibility, it's just that most distros don't care anyway (why should they when there's APT/Yum/Portage/whathaveyou?). The Autopackage project has developed some rather sophisticated wrapper scripts around the GNU toolchain to build binaries that link to ABIs that are available on as wide a range of distros as possible.
I don't mean to troll, but I'm honestly wondering -- what niche does NetBSD have anymore, really? From all I can see, {Free,Open}BSD have lots more capabilities, and from what I hear, NetBSD doesn't really work on that many archs these days anyway (Linux probably works on more archs, doesn't it?). Is anyone using NetBSD for any good reason and able to tell me why?
I'm no quantum psysicist, but if there were particles (virtual or no) of negative mass, would they not be repelled from a black hole, which has positive mass? Which would in turn mean that only the particles of positive mass could ever enter a black hole.
But regardless of that, I don't see what grudges you hold against "WiFi in general", or WPA. I've played a lot with NetworkManager on Ubuntu, and I've yet to encounter a single problem, basically. WPA and WEP works, with the keys beautifully stored in the gnome-keyring-manager. Scanning works. I can't think of anything that doesn't work. Care to enlighten me?
Not that I don't agree with you that Microsoft products tned to be rushed and not ready by the time they're released, but the arguments you make for it aren't exactly convincing. Please mention a single software product ever that hasn't had later version with bugfixes.
Whoever modded this informative rather than funny really, really needs to revalidate his geek license. Soon.