Just on one point: I think you are seriously misrepresenting Singer's position.
His argument is not at all based on considerations of directing evolution or eugenics, despite some strange arguments to the contrary [1]. On the contrary, it follows logically from two propositions: firstly that we should minimize suffering, all other things being equal; and secondly, that a foetus is not a person in the sense of a it being a rational self-conscious being, and so questions of its abortion do not infringe upon other ethical considerations that are restricted to that class of beings.
That abortion should be performed in order to shape the genetic profile of a population simply doesn't come in to it.
[1] One argument for example is that eugenicists predicated their arguments on a reconsideration of the sanctity of human life based on scientific and utilitarian ideas, and that Singer's stance on abortion is also predicated on such a questioning. To say that this means the two stances are the same is quite clearly fallacious.
It really is the least flawed of any MMORPG I've come across, and indeed has these great features. Sidekicking has got to be the best thing I've seen so far; a good balance between being upped enough to be useful in your friends' group, and being sufficiently weaker to encourage you to acquire new powers and catch up.
The hover power at level 6 is very handy, and one can technically fly with it, but I'd definitely call it 'technically' given the speed.
The missions though are not world-changing, and often you'll get the same mission over and over. There is still a lot of genericism in the game, and even the culmination of a long story arc doesn't have any effect on the rest of the world. People will join a team to complete a mission because the act itself is fun, not because their contribution is going to affect their own character's storyline in the future, or affect the world at large.
All that said, CoH is (for me) much more fun than any other MMORPG in the traditional style.
(Initial caveat: posting while drunk, apologies in advance.)
Grinding is dull and pointless. It's a competition between players: who has the most patience? who has the least life outside a computer game? It was dull in LPMUDs and it's dull in the modern generation of online RPGs. There is more than enough tedium in the real world. Why on Earth do you want to do it in a game that is supposed to be fun?
In every online game I've played or seen so far, one starts off being totally incompetent. The 'mangy rat' is a challenge. Who wants to play someone who has trouble dealing with mangy rats? Those who persist and reach the end-game are orders of magnitude more powerful. Their in-game skills are incomparably better than the starting character. This is their reward for sheer bloody-mindedness.
Characters should develop and change over time, but they should start off being able to affect the world. To matter. To be someone. Otherwise it's either dull or a huge stretch of the imagination.
Case in point: take City of Heroes. You play a superhero. Yet when you start you have 3 powers, which are probably two attacks and a defence. You then run fleeing from all but the most innocuous of petty street thugs. By the end of the game, one is fighting off alien invasions and world-destroying foes, but at the start you are decidedly un-heroic. No one can even fly until 14th level. And City of Heroes is one of the least bad offenders. What's with this?
The big problem is that designing and implementing a world where people's online alter-egos can actually matter, is really hard. AI for the NPCs is not up to the task of creating a community in which the PC can shine, and it's unreasonable to expect a starting player to immediately make a splash in the community of other PCs. That coupled with possibly hundreds of thousands of players makes it especially difficult.
Nonetheless, until people can feel like they matter, the MMORPG is going to have limited appeal. The policies against auto-leveling and other forms of programmatical advancement simply exclude another class of players who would rather a machine take care of all the tedious aspects so that they instead can concentrate on the bits worth playing. Why is it so bad that someone skip 17 hours of mindless clicking? Where is the appeal?
Until AI tech and automated story-telling is vastly improved, MMORPGs seem stuck with this terrible grinding aspect. I'll play them as long as the other rewards make up for this huge deficiency, but the technology just isn't there yet to appeal to players who don't have a certain masochistic streak.
One huge exception comes to mind: Yohoho! Puzzle Pirates. There your character doesn't improve, you improve. You the player get better at doing the puzzles, and your ability in the game improves conmesurably. This is extremely cool, and is one of the reasons why Y!PP is such a damn fine game.
I suspect this is not so serious and more like a troll. But nonetheless.
Before any technical arguments, realise that a key difference between gcc and vc.net is that the former is copyleft and the latter is not. gcc runs on a large number of operating systems, and compiles a number of different languages (eg ada, fortran, c, objective c, c++).
As regards standards compliance, vc.net is not a c99 compiler; using c99 constructs which are not c++-ish will just fail under vc.net. vc.net is not an ada, fortran or objective-c compiler either.
Regarding news, the long sought-after fortran 95 compiler replacing the fortran 77 compiler is certainly newsworthy. In addition, a major architectural change -- the integreation of the tree-ssa branch -- heralds significant future optimization possibilities and is the culmination of two years (or more?) work. It's hard to think what would be more relevant to report.
Why would people care about gcc reports? Given that it is the standard compiler on a number of very popular operating systems, and that the quality of gcc is fundamental to the quality of compiled software on these platforms, and that similarly limitations of gcc in turn limit software on these platforms, it is obvious that gcc is a very important project. People couldn't turn to vc.net (or icc, or...) for the basis of linux kernel and application development even if they wanted to.
Summary: gcc news is relevant to almost all free operating system developers and (indirectly) users, and this news is definitely gcc news.
They tout it for scientific research applications... but still no ECC support!
It's hard to imagine any sort of serious scientific computing that is making use of the advertised features - such as 64-bit optimised libraries, 8GB of memory, etc - which wouldn't also feel the lack of any real confidence that the results aren't contaminated by bit errors.
The newest Xserve supports ECC. Whyever do they not support it on their workstations?! It boggles the mind.
Sony, much like every other company, want to maximise their profits. Sony, unlike most companies, is large enough to set standards and influence government policy on matters such as copyright so that they can engage in discriminatory pricing (discriminatory in the sense that are able to discriminate between markets and charge in each market according to demand and local pricings.)
Proprietay hardware and software formats. Pointless (from the user's point of view) restrictions on using their hardware. Lobbying for legislation to declare re-importation of CDs as a copyright violation. Region ecoding on DVDs. These are all part of the same policy.
The fact that they make sexy looking hardware is no reason to buy into this scheme unnecessarily.
Given the asinine restrictions and unnecessary hoops I would have to jump through to use a Net-MD player or its ilk, there is no way I'd ever buy an MD player with the intent of connecting it to a computer. I'm sorry to hear that yet another aspect of their profit-maximisation policy has claimed a victim.
PS: Sony warranty support is also one of the worst I've ever had to deal with in a professional capacity. Again, they're big enough that they don't need to care.
PPS: It's a real shame that the fruits of Fujitsu's very cool MO technology look like they will be primarily found only in Sony's product. The 2.3GB MO 3.5" format was amazing, but just never caught on. There was a collaboration with Sony and this looks like the result.
If the G5s had support for ECC memory, then we could use them for modelling without taking a gamble that a flipped bit screws the results.
When days of computation go into making a calculation, the last thing you want to do is to run it again because there's a non-negligable chance that there was an uncaught bit error.
Luckily Apple have since seen the light and the new Xserve G5 at least supports ECC RAM. Before that, for affordable scientific computing, dual Opteron machines had no real competition.
Digital released the DEC Mulita in 1995. Definitely 64-bit. Ran Windows NT. Was targetted for the PC niche (not home niche though) - it was designed to be small and cheap, even using a 2.5" disk drive. It was a PC.
Also regarding Apple's claim, the Opteron had been out for a while, and it's hard to think of any good metrics for distinguishing between workstations and PCs that would exclude Opteron-based machines and include the G5.
Lastly, the G5 may be a 64-bit processor, but one can't call OS X a 64-bit operating system, at least not yet.
Just a heads-up on an error in the summary: the Wired articles states than (all) the rats which received the combination treatment regained 70% of their walking function, not that 70% of the injured rats became able to walk.
Given that the improvement was over a period of just eight weeks, this is possibly even more promising than the mangled statistic in the summary.
what's wrong with making firearm ownership a right?
Because it seems to have very negative consequences.
The US seems quite similar to many Western countries in culture: similar degrees of exposure to violence, even (with a bit of a stretch) similar levels of religious fervour. But the US has a terrifyingly high level of gun-related deaths.
The big difference is the degree of gun ownership. There is a correlation. Is it meaningful? Many will argue it is, and personally speaking, these arguments are convincing.
If you have an idealised model, but which in practice isn't working out, then it needs to be tweaked. A right to gun ownership, acting as a check on government abuse, seems to be a case like this. It doesn't seem to prevent or stop the abuse, and the resulting widespread availability of guns is killing and maiming thousands upon thousands of people every year (I believe on the order of 15000 non-suicide deaths, and over 50000 injuries).
If a theory isn't working out in practice, then it's time for a new theory.
I'm wondering how it is faster to pull a cookie from the browser, compute its checksum, compair, if they match, decrypt, then decode. Surely that can't be faster than a properly cached local database query.
Given that the limiting resource is server resources as opposed to customer waiting time or network bandwidth, and given how much seriously faster CPU is over disk access, it looks like a win to me.
Once your data gets larger than 8k or so, you begin to seriously annoy people on modem connections, so I'm assuming the cookie is smaller than this. Checksumming and decrypting 8kbytes of data on a modern machine really ought to be very quick indeed. For order of magnitude estimates, I'd guess the process takes about 15 clock cycles per byte of cookie as an upper bound, coming to significantly less than a milisecond on a modern CPU. This is much less than the cost of a disk access.
That's in direct contradiction to documented evidence. Do note that some wildly inflated figures have been circulating, for which there is absolutely no support.
Note that it is very much in the interest of Ukraine to promote very high numbers of casualties in order to obtain increased compensation from Russia.
But the consequences of nuclear power station failure are more severe than any other category of civilian accident.
No, that's simply not true.
Union Carbide in Bhopal: 3000 to 8000 dead; over 100000 injured.
Chisso Corporation at Minimata: mercury poisoning kills hundreds, with at least 3000 people afflicted.
The Grandcamp in Texas: Fertilizer explosion kills nearly 600; over 3500 injured.
Chernobyl: fewer than 100 deaths to date; fewer than 1500 known attributable radiation-related illnesses. Potential premature deaths due to excess radiation exposure estimated to be 3000, but we'll have to wait and see.
Nuclear power is dangerous, but there's a lot worse out there. Look up deaths attributable to coal-fired power plants sometime.
It's more properly a ground for saying that we can choose different versions of set theory which are extensions of ZF or ZFC and further, have something to say about the Contiuum Hypothesis.
To say 'for ever undecidable' is strange. What is the 'for ever' for? It makes it sound like the problem has an answer, but one which will forever be outside our capability to calculate. On the contrary, Goedel and Cohen have resolved the issue. So DFW's phrase is -- at least a little -- misleading.
The Japanese version of Windows XP is designed for people who want their OS to be in Japanese. If you want it in English you should have bought the English version.
Try getting the English version in Japan. Not easy. On top of that, when you buy your laptop, you pay for Windows in whatever localization is installed. So we should be paying twice just for a different message catalog? Great!
Or you could change your system font settings so they stopped having uglified small fonts, but maybe that would be too much effort for you.
Strangely enough, changing those settings do not seem to affect these applications. How about that? Weird, eh.
You do it in exactly the same way you enter full-width katakana in any application - switch to Japanese input and, like, type it?
In your Japanese localized version of Windows XP, start a command window. Watch your IME selector bar transform to 'dumb' mode.
Since have learnt that pressing alt together with zenkaku switches into IME mode in the command prompt. Yay! Why do you need to press the alt key? You don't elsewhere. Where is this documented? What can't I use the IME selector to do this?
Use the keyboard shortcuts, then.
Sure. You can't deny that this is merely an alternative to dealing with something that is broken.
So ah, tell me. How do I convince my Japanese version of Windows XP Home to display OS messages in English?
The last time I tried to search for an answer, the only one I found was: buy an English version, back-up everything, install English version, restore from backup. This is a far cry from setting your LANG envvar.
While I'm at it, I can complain that under this version of the OS, a whole bunch of English-language software seems to have uglified small fonts, sometimes to the point of illegibility.
Oh, and how do you enter full-width katakana from the command prompt? It seems impossible. Which makes entering the localized name of the Local Area Connection tricky in netsh. (As far as I can tell, you have to cut and paste the name.)
Not to mention the way that the IME taskbar widget regularly gets obscured by other taskbar widgets making changing input settings a pain in the neck. Microsoft's own deskswitcher application is a particular offender in this regard.
Sure it has localization, but it doesn't appear to be especially well thought out (infamously bad translations aside), and certainly its internationalization quality leaves something to be desired.
Sometimes they'd get 'stuck' if they were left on for a long time (like a year or two), then turned off. At this point they wouldn't spin up, or make a half-hearted attempt to.
If they couldn't be coaxed into moving, taking it out the enclosure and letting it drop four centimetres or so flat onto a wooden table often got them unstuck enough to grab the data and back it up.
That said, have had some success with the same trick with newer drives with different modes of failure. Of course try the least invasive approaches first and work up, but if the drive is otherwise dead, then there's little left to lose. Unless you want to spend big dollars with a professional data recovery mob.
The thing that gets me, is the tying of products to regions.
On one hand, we have all the big media companies pushing for uniform (and to them, highly favourable) IP laws around the world. On the other, they engage in what is essentially price fixing by charging differing amounts in different markets, and then seeking technical and legal means to prevent the free trade of their own products.
This current story would be a storm in a tea cup if there were no issues in importing games from other regions.
Highlighting this sort of hypocrisy is the recent move in Japan of the music publishing industry to restrict through changes in copyright law the importation of CDs of Japanese artists' music from overseas. These CDs, containing pretty much the same music, sell for a third to a half the cost in South Korea and Taiwan, and after importation, can be about 1000 yen (circa US$10) in Japan. How did these CDs get to be printed legally? Because these very same companies sell the rights to do so to the foreign publishers in the first place.
It must be great to have enough money that you can buy laws that grant you even more.
Slashdot Mirror
Just on one point: I think you are seriously misrepresenting Singer's position.
His argument is not at all based on considerations of directing evolution or eugenics, despite some strange arguments to the contrary [1]. On the contrary, it follows logically from two propositions: firstly that we should minimize suffering, all other things being equal; and secondly, that a foetus is not a person in the sense of a it being a rational self-conscious being, and so questions of its abortion do not infringe upon other ethical considerations that are restricted to that class of beings.
That abortion should be performed in order to shape the genetic profile of a population simply doesn't come in to it.
[1] One argument for example is that eugenicists predicated their arguments on a reconsideration of the sanctity of human life based on scientific and utilitarian ideas, and that Singer's stance on abortion is also predicated on such a questioning. To say that this means the two stances are the same is quite clearly fallacious.
Do you think it would work in the US?
I play CoH too much actually :)
It really is the least flawed of any MMORPG I've come across, and indeed has these great features. Sidekicking has got to be the best thing I've seen so far; a good balance between being upped enough to be useful in your friends' group, and being sufficiently weaker to encourage you to acquire new powers and catch up.
The hover power at level 6 is very handy, and one can technically fly with it, but I'd definitely call it 'technically' given the speed.
The missions though are not world-changing, and often you'll get the same mission over and over. There is still a lot of genericism in the game, and even the culmination of a long story arc doesn't have any effect on the rest of the world. People will join a team to complete a mission because the act itself is fun, not because their contribution is going to affect their own character's storyline in the future, or affect the world at large.
All that said, CoH is (for me) much more fun than any other MMORPG in the traditional style.
... the alternatives don't yet seem viable.
(Initial caveat: posting while drunk, apologies in advance.)
Grinding is dull and pointless. It's a competition between players: who has the most patience? who has the least life outside a computer game? It was dull in LPMUDs and it's dull in the modern generation of online RPGs. There is more than enough tedium in the real world. Why on Earth do you want to do it in a game that is supposed to be fun?
In every online game I've played or seen so far, one starts off being totally incompetent. The 'mangy rat' is a challenge. Who wants to play someone who has trouble dealing with mangy rats? Those who persist and reach the end-game are orders of magnitude more powerful. Their in-game skills are incomparably better than the starting character. This is their reward for sheer bloody-mindedness.
Characters should develop and change over time, but they should start off being able to affect the world. To matter. To be someone. Otherwise it's either dull or a huge stretch of the imagination.
Case in point: take City of Heroes. You play a superhero. Yet when you start you have 3 powers, which are probably two attacks and a defence. You then run fleeing from all but the most innocuous of petty street thugs. By the end of the game, one is fighting off alien invasions and world-destroying foes, but at the start you are decidedly un-heroic. No one can even fly until 14th level. And City of Heroes is one of the least bad offenders. What's with this?
The big problem is that designing and implementing a world where people's online alter-egos can actually matter, is really hard. AI for the NPCs is not up to the task of creating a community in which the PC can shine, and it's unreasonable to expect a starting player to immediately make a splash in the community of other PCs. That coupled with possibly hundreds of thousands of players makes it especially difficult.
Nonetheless, until people can feel like they matter, the MMORPG is going to have limited appeal. The policies against auto-leveling and other forms of programmatical advancement simply exclude another class of players who would rather a machine take care of all the tedious aspects so that they instead can concentrate on the bits worth playing. Why is it so bad that someone skip 17 hours of mindless clicking? Where is the appeal?
Until AI tech and automated story-telling is vastly improved, MMORPGs seem stuck with this terrible grinding aspect. I'll play them as long as the other rewards make up for this huge deficiency, but the technology just isn't there yet to appeal to players who don't have a certain masochistic streak.
One huge exception comes to mind: Yohoho! Puzzle Pirates. There your character doesn't improve, you improve. You the player get better at doing the puzzles, and your ability in the game improves conmesurably. This is extremely cool, and is one of the reasons why Y!PP is such a damn fine game.
There are worse things than violent FPSs.
Much, much worse.
I suspect this is not so serious and more like a troll. But nonetheless.
...) for the basis of linux kernel and application development even if they wanted to.
Before any technical arguments, realise that a key difference between gcc and vc.net is that the former is copyleft and the latter is not. gcc runs on a large number of operating systems, and compiles a number of different languages (eg ada, fortran, c, objective c, c++).
As regards standards compliance, vc.net is not a c99 compiler; using c99 constructs which are not c++-ish will just fail under vc.net. vc.net is not an ada, fortran or objective-c compiler either.
Regarding news, the long sought-after fortran 95 compiler replacing the fortran 77 compiler is certainly newsworthy. In addition, a major architectural change -- the integreation of the tree-ssa branch -- heralds significant future optimization possibilities and is the culmination of two years (or more?) work. It's hard to think what would be more relevant to report.
Why would people care about gcc reports? Given that it is the standard compiler on a number of very popular operating systems, and that the quality of gcc is fundamental to the quality of compiled software on these platforms, and that similarly limitations of gcc in turn limit software on these platforms, it is obvious that gcc is a very important project. People couldn't turn to vc.net (or icc, or
Summary: gcc news is relevant to almost all free operating system developers and (indirectly) users, and this news is definitely gcc news.
They tout it for scientific research applications ... but still no ECC support!
It's hard to imagine any sort of serious scientific computing that is making use of the advertised features - such as 64-bit optimised libraries, 8GB of memory, etc - which wouldn't also feel the lack of any real confidence that the results aren't contaminated by bit errors.
The newest Xserve supports ECC. Whyever do they not support it on their workstations?! It boggles the mind.
This is standard operating procedure for Sony.
Sony, much like every other company, want to maximise their profits. Sony, unlike most companies, is large enough to set standards and influence government policy on matters such as copyright so that they can engage in discriminatory pricing (discriminatory in the sense that are able to discriminate between markets and charge in each market according to demand and local pricings.)
Proprietay hardware and software formats. Pointless (from the user's point of view) restrictions on using their hardware. Lobbying for legislation to declare re-importation of CDs as a copyright violation. Region ecoding on DVDs. These are all part of the same policy.
The fact that they make sexy looking hardware is no reason to buy into this scheme unnecessarily.
Given the asinine restrictions and unnecessary hoops I would have to jump through to use a Net-MD player or its ilk, there is no way I'd ever buy an MD player with the intent of connecting it to a computer. I'm sorry to hear that yet another aspect of their profit-maximisation policy has claimed a victim.
PS: Sony warranty support is also one of the worst I've ever had to deal with in a professional capacity. Again, they're big enough that they don't need to care.
PPS: It's a real shame that the fruits of Fujitsu's very cool MO technology look like they will be primarily found only in Sony's product. The 2.3GB MO 3.5" format was amazing, but just never caught on. There was a collaboration with Sony and this looks like the result.
This is true.
On the other hand, Linux, which also ran (and of course still does run) on the Multia was 64-bit.
If the G5s had support for ECC memory, then we could use them for modelling without taking a gamble that a flipped bit screws the results.
When days of computation go into making a calculation, the last thing you want to do is to run it again because there's a non-negligable chance that there was an uncaught bit error.
Luckily Apple have since seen the light and the new Xserve G5 at least supports ECC RAM. Before that, for affordable scientific computing, dual Opteron machines had no real competition.
Also regarding Apple's claim, the Opteron had been out for a while, and it's hard to think of any good metrics for distinguishing between workstations and PCs that would exclude Opteron-based machines and include the G5.
Lastly, the G5 may be a 64-bit processor, but one can't call OS X a 64-bit operating system, at least not yet.
Just a heads-up on an error in the summary: the Wired articles states than (all) the rats which received the combination treatment regained 70% of their walking function, not that 70% of the injured rats became able to walk.
Given that the improvement was over a period of just eight weeks, this is possibly even more promising than the mangled statistic in the summary.
Even Microsoft don't use SourceSafe.
It's that bad.
While the end of the trailer used United Future Organizaton's "The Planet Plan", from their truly great album "3rd Perspective".
UFO do good stuff. Highly recommend the album.
The US seems quite similar to many Western countries in culture: similar degrees of exposure to violence, even (with a bit of a stretch) similar levels of religious fervour. But the US has a terrifyingly high level of gun-related deaths.
The big difference is the degree of gun ownership. There is a correlation. Is it meaningful? Many will argue it is, and personally speaking, these arguments are convincing.
If you have an idealised model, but which in practice isn't working out, then it needs to be tweaked. A right to gun ownership, acting as a check on government abuse, seems to be a case like this. It doesn't seem to prevent or stop the abuse, and the resulting widespread availability of guns is killing and maiming thousands upon thousands of people every year (I believe on the order of 15000 non-suicide deaths, and over 50000 injuries).
If a theory isn't working out in practice, then it's time for a new theory.
Given that the limiting resource is server resources as opposed to customer waiting time or network bandwidth, and given how much seriously faster CPU is over disk access, it looks like a win to me.
Once your data gets larger than 8k or so, you begin to seriously annoy people on modem connections, so I'm assuming the cookie is smaller than this. Checksumming and decrypting 8kbytes of data on a modern machine really ought to be very quick indeed. For order of magnitude estimates, I'd guess the process takes about 15 clock cycles per byte of cookie as an upper bound, coming to significantly less than a milisecond on a modern CPU. This is much less than the cost of a disk access.
That's in direct contradiction to documented evidence. Do note that some wildly inflated figures have been circulating, for which there is absolutely no support. Note that it is very much in the interest of Ukraine to promote very high numbers of casualties in order to obtain increased compensation from Russia.
Some links include a summary of other published work, the report from the UN IAEA, and a summary from a nuclear power specialist.
Regarding this specific point: from the 'executive summary' report on a recent conference organised by the IAEA on Chernobyl,
No, that's simply not true.
Union Carbide in Bhopal: 3000 to 8000 dead; over 100000 injured.
Chisso Corporation at Minimata: mercury poisoning kills hundreds, with at least 3000 people afflicted.
The Grandcamp in Texas: Fertilizer explosion kills nearly 600; over 3500 injured.
Chernobyl: fewer than 100 deaths to date; fewer than 1500 known attributable radiation-related illnesses. Potential premature deaths due to excess radiation exposure estimated to be 3000, but we'll have to wait and see.
Nuclear power is dangerous, but there's a lot worse out there. Look up deaths attributable to coal-fired power plants sometime.
Sorry -- my mistake. The reviewer was of course referring to the text of different authors (Robert Kaplan and Ellen Kaplan), not DFW.
It's more properly a ground for saying that we can choose different versions of set theory which are extensions of ZF or ZFC and further, have something to say about the Contiuum Hypothesis.
To say 'for ever undecidable' is strange. What is the 'for ever' for? It makes it sound like the problem has an answer, but one which will forever be outside our capability to calculate. On the contrary, Goedel and Cohen have resolved the issue. So DFW's phrase is -- at least a little -- misleading.
The Japanese version of Windows XP is designed for people who want their OS to be in Japanese. If you want it in English you should have bought the English version.
Try getting the English version in Japan. Not easy. On top of that, when you buy your laptop, you pay for Windows in whatever localization is installed.
So we should be paying twice just for a different message catalog? Great!
Or you could change your system font settings so they stopped having uglified small fonts, but maybe that would be too much effort for you.
Strangely enough, changing those settings do not seem to affect these applications. How about that? Weird, eh.
You do it in exactly the same way you enter full-width katakana in any application - switch to Japanese input and, like, type it?
In your Japanese localized version of Windows XP, start a command window. Watch your IME selector bar transform to 'dumb' mode.
Since have learnt that pressing alt together with zenkaku switches into IME mode in the command prompt. Yay! Why do you need to press the alt key? You don't elsewhere. Where is this documented? What can't I use the IME selector to do this?
Use the keyboard shortcuts, then.
Sure. You can't deny that this is merely an alternative to dealing with something that is broken.
So ah, tell me. How do I convince my Japanese version of Windows XP Home to display OS messages in English?
The last time I tried to search for an answer, the only one I found was: buy an English version, back-up everything, install English version, restore from backup. This is a far cry from setting your LANG envvar.
While I'm at it, I can complain that under this version of the OS, a whole bunch of English-language software seems to have uglified small fonts, sometimes to the point of illegibility.
Oh, and how do you enter full-width katakana from the command prompt? It seems impossible. Which makes entering the localized name of the Local Area Connection tricky in netsh. (As far as I can tell, you have to cut and paste the name.)
Not to mention the way that the IME taskbar widget regularly gets obscured by other taskbar widgets making changing input settings a pain in the neck. Microsoft's own deskswitcher application is a particular offender in this regard.
Sure it has localization, but it doesn't appear to be especially well thought out (infamously bad translations aside), and certainly its internationalization quality leaves something to be desired.
Back in the days of 1GB SCSI drives ...
Sometimes they'd get 'stuck' if they were left on for a long time (like a year or two), then turned off. At this point they wouldn't spin up, or make a half-hearted attempt to.
If they couldn't be coaxed into moving, taking it out the enclosure and letting it drop four centimetres or so flat onto a wooden table often got them unstuck enough to grab the data and back it up.
That said, have had some success with the same trick with newer drives with different modes of failure. Of course try the least invasive approaches first and work up, but if the drive is otherwise dead, then there's little left to lose. Unless you want to spend big dollars with a professional data recovery mob.
The thing that gets me, is the tying of products to regions.
On one hand, we have all the big media companies pushing for uniform (and to them, highly favourable) IP laws around the world. On the other, they engage in what is essentially price fixing by charging differing amounts in different markets, and then seeking technical and legal means to prevent the free trade of their own products.
This current story would be a storm in a tea cup if there were no issues in importing games from other regions.
Highlighting this sort of hypocrisy is the recent move in Japan of the music publishing industry to restrict through changes in copyright law the importation of CDs of Japanese artists' music from overseas. These CDs, containing pretty much the same music, sell for a third to a half the cost in South Korea and Taiwan, and after importation, can be about 1000 yen (circa US$10) in Japan. How did these CDs get to be printed legally? Because these very same companies sell the rights to do so to the foreign publishers in the first place.
It must be great to have enough money that you can buy laws that grant you even more.