You're right, of course, but it still seems incredibly stupid that we're using a DOS filesystem because nobody had the presence of mind to distribute something useful with the first USB mass storage drivers.
Yup. Goes to show how lack of standardization and the NIH syndrome can hold back things.
The mainboard manufacturers have figured out how to do this safely. So can the GPU manufacturers. Fr example, two flash chips and an emergency ROM would make for a very smooth upgrade:
1. The card flashes the upgrade onto the flash chip it's currently not using.
2. The card tells the OS it's going to switch to the new driver.
3. The card switches over to the new driver, reinitializes and tells the OS it's ready.
4. The OS displays a dialog box like the one Windows shows when switching resolutions: "The new driver has been loaded. Does the graphics card work correctly? [Yes] [No] This dialog will automatically select 'No' in 30 seconds."
5. If the OS signifies user satisfaction the new driver is used, otherwise the card loads the old one and reinitializes.
Should the card be unable to work with either driver it falls back to the backup ROM where a known-good (probably reduced-functionality) driver resides. Combined with a smart protocol that allows the card and the OS to tell each other about/change their state and a graphics infrastructure that allows dynamic reloading of the driver this could mean that you can update the GPU driver and switch to the new one without rebooting or even restarting X11.
It might even possible (although only useful for power users) to install two alternative drivers on the two chips and switch between them on the fly - for example one driver which minimizes power consumption by powering down parts of the card and one which gives good gaming performance. Of course, both of these would have to come from the manufacturer.
And suppose I have a portable hard drive, which I want to make sure is readable everywhere -- all I have to do is make a separate FAT32 partition with all the relevant software on it for Windows and OS X to read the other partition.
Tried it, didn't work. In order to do anything meaningful with a non-FAT/NTFS drive under Windows you need to install an IFS. Few people are going to let you install some random driver on their notebook just to access your USB drive. That's also the reason why MacDrive and ext2-IFS are no solutions: While they give you portability under laboratory conditions (between all of your computers), they are useless for exchanging data with others. A FAT killer needs to run out of the box everywhere, period.
That's aso why I don't believe that FUSE will solve the problem. It's in Ubuntu but what about Debian or CentOS or Sabayon (this being three distros I encountered lately on random computers)? And, of course, under OS X it's always a separate install (two even, IIRC), putting it solidly in "why should I install some random driver" territory.
The reason we put up with FAT32 is not high compatibility, it's absolute compatibility. You can stick a FAT32-formatted USB drive into any random comouter you encounter and it will definitely mount with no kind of setup neccessary. Any filesystem that does not meet those criteria is simply no competitor.
Actually, if you want a filesystem usable by everyone it will definitely have to come from Microsoft. MS is not going to adopt someone else's FS unless they really have to, so the only serious contenders are FAT32, NTFS and exFAT (things like UDF are highly unlikely as Microsoft probably won't support it for anything but DVD images).
FAT32 is the current portability king. However, it's also old and extremely limited and people stopped liking it half a decade ago.
NTFS works... kinda. For example, you can mount a Vista NTFS disk unter XP but you're going to lose some metadata. XP NTFS is supported on every computer equipped with FUSE and ntfs-3g and on every computer running a new-ish Windows. Everywhere else it's useless and ntfs-3g is hardly a default package in most distros so no dice.
exFAT is essentially FAT64. It supports large files and has some improvements but in general it's very much FAT32's successor, being lightweight and relatively uncomplicated. While this sounds ideal for a widely supported filesystem there are no free exFAT implementations yet and people aren't sure whether Microsoft is going to allow the existence of one. So this one's not a winner, either.
Everything else will remain *nix-only so, yes, FAT32 is still the only option if you want complete portability (UDF if you can live with jumping through hoops). And it's going to stay that way for the forseeable future.
In fact, this is why it's in the interest of the manufacturers to put part of the drivers into the device itself: It becomes easier to reach new parts of the market - in fact if all problematic parts of the driver are hardware-side one needs to just provide some "harmless" specs to the OSS community and let them create their own driver at zero cost. Also, if those driver parts are sufficiently OS-agnostic one can easily support new versions of Windows... which should make this interesting for at least Nvidia.
No, video game trailers are different from ads. Ads inform you about a product that has or shortly will hit the market - in any case it's market-ready. Trailers inform you of what the company is working on and are often released rather early during production. One important difference is that the company pays mass media to show their ads (as they already know they're going to make money with the product shortly) while trailers are often just released to certain sites and then mirrored throughout the internet.
Content brings revenue, not slimy deals with crappy spammers.
How? Think of the sites you commonly use - with how many of those sites do you pay for a subscription? Those are the sites where you generate revenue. On all other sites you only generate cost and they have to get their money from elsewhere. There are few reliable ways of making a website work financially - essentially you either require your users to pay for subscriptions, you show ads or you lose money.
You could of course demand that every web site is operated at a loss because you neither want to pay for everything nor want to see ads - but that would mean that 95% of all bigger websites would have to shut down or go subscribers-only, including Slashdot.
Unless you can find some way to reduce traffic and hosting costs to zero there needs to be some reliable way for websites to generate revenue.
You appear to have different scent ads than Germany does. In Germany, scent ads are usually exercises in dadaism, completely devoid of any discernible content. The palette goes from women randomly gyrating with trippy special effects overlaid to a man rambling about adventure before driving off on a moped to weird disjointed nonsense that ends with a man walking away from a cyan-tinted Union Jack. In fact, most of them are simply disjointed nonsense.
The easiest way to spot a scent ad is to look if it appears to be completely nonsensical. The less sense it makes, the more likely it's a scent ad.
Makes me wonder who they're targeting with those ads. People who buy stuff simply because some yellow-hued women stumble around on TV while "Heart of Glass" is playing? How the hell does that work?
I once saw a documentation about how they make the photos you see on convenience food packaging. The tomato soup with a cream swirl was actually 100% toxic-if-ingested wall paint. Other dishes were either made by cooks (of course using completely different recipes) or weren't food at all. Don't think only models get airbrushed; food does, as well. With clear varnish, during the shoot.
I mean, some car manufacturer recently ran a TV ad in Germany where they deconstructed the usual car ad by gradually switching off the humans (all professional models), the beautiful scenery (completely computer-generated), the brilliant highlights on the car (ditto), the majestic music and finally the street. I don't remember which car it was but the ad strikes me as insightful - it shows just how much of the ads you see has to do with the actual car (not much at all, not even the car's appearance is realistic).
Ads lie. Ads lie all the time. Do not expect anything you hear in an ad to be remotely true, apart fom "product XYZ exists".
Well, "/m - open-mid front unrounded vowel - m - alveolar approximant - near-close near-front unreounded vowel - st - mid-central vowel/" is certainly much more complicated to read and write than the IPA string it's trying to represent. By the way, I wrote "memristor" there. I know, one could use SAMPA... but isn't SAMPA essentially ASCII art that tries to emulate IPA on legacy systems?
Also, if every decision is made towards making life difficult for obnoxious users - why do we still allow anonymous posting? There's more to message board quality than troll management and sometimes the net gain from implementing a possibly-abused feature is positive.
Yes it is... However one should assume that in 2008 a tech website would be able to support the technology X-SAMPA crudely reimplements. Also, SAMPA doesn't have nearly the penetration of IPA - while one can expect virtually anyone who has learned a foreign language to have at least encountered IPA, SAMPA is useful only to those who do phonetic discussions in a long-outdated encoding.
Yes. Remember the IDN homoglyph attack?
Hm, that's funny. I always assumed they had some kind of format IDNs are supposed to be transmitted as. You know, some kind of puny code that can be algorithmically derived... Oh, wait! They have such a format! And I'd be willing to bet that it's entirely possible to automatically mangle URLs entered to the appropriate Punycode representation (not to mention that all major browser already do that in the status bar).
€ produces €.
But the appropriate Unicode glyph doesn't. And that's another thing wrong with Slashdot's weird filter: Some glyphs work when entered directly, some only work when entered as HTTP entities and yet others don't work either way. IPA glyphs get converted to HTML entities but those entities are then filtered. It's impossible to determine beforehand whether Slash accepts a glyph and in which form it does.
Correct. It's easier to whitelist Latin-1 than to comb the entire Basic Multilingual Plane looking for anything that's not part of a bidirectional or complex script. If something won't increase SourgeForge, Inc.'s ad revenue, it's not worth spending time on.
And that's probably why Digg is the superior platform, community and content aside. "We won't do anything more than the absolute minimum required to keep the site running unless it significantly increases our bottom line" is how corporations show that they run the platform not for the community but for the ad partners. Of course the weird filtering is not the only problem - the CSS has been broken for ages (especially in Idle, even though that doesn't hurt much) but apparently there's no business case for getting that right, either.
I'm waiting for the day when Slashdot changes its "we won't delete posts unless legally forced to" policy to "we won't delete posts unless it's in SourceForge, Inc.'s financial interest".
In that case the question is: "Why does a forum for geeks not allow anything outside what's covered by ISO Latin-1?"
Is IPA somehow dangerous? Cyrillic? Mathematical symbols? Or the Euro sign? This is a forum for technical people who, due to their geek nature, prefer to use the appropriate notation to communicate things not easily communicated in ASCII. For example, it's extremely difficult to accurately and concisely communicate the pronounciation of something without using IPA. That's what IPA was designed for. That's why it's in Unicode.
I get the feeling that the whitelist was put in place by someone who doesn't really know Unicode (and/or didn't want to spend time with it) and thus opted to just keep Latin-1 since that's what he knows. There are a lot of characters in Unicode that could safely be whitelisted; depending on how well-written the filter is this should be a matter of one afternoon with a Unicode character list. I mean, we don't demand Linear-B; the safe parts of the BMP (all non-combining non-control code points except those between D800 and F8FF (FFFF if you don't want to include Han)) would be everything 99.999% of all posters need.
I won't link to a screenshot because I don't want to melt someone's server but a Google image search for slashdot omg ponies will give you a lot of examples.
Perhaps not mumbling would make you sound clearer. I seriously don't see how "Kibibyte" sounds like mumbling unless, well, you mumble it. Likewise I don't see how "memristor" should slow one down. [MISSING1*] doesn't seem too bad - [MISSING2*] certainly is possible in the English language even if it doesn't (commonly?) appear in a single word; I can say things like "damn ring" without pausing.
* I couldn't properly use the following strings since IPA is filtered out by Slashdot even in HTML entity notation.
MISSING1:/ˈmɛmɹɪstə/
MISSING2:/mɹ/
Actually, the scroll ball is a pain to clean but that's mostly an artifact of the Mighty's "no maintenance possible" design.
I think Apple should try to use inductive fields to replace the ball. Sense where the finger is moving and use that to scroll. They probably have already tried it and it doesn't work but it would be very slick and completely dirt-proof.
That's not because she was femals, that's because she was Deanna. Come on, this is the woman that used telepathic abilities to "detect hostility" in people who were currently busy getting a torpedo lock on the Enterprise...
I thought that was what American culture is all about. CYA is the highest maxim.
An example: If in the USA someone fell and hurt themselves and I was nearby I wouldn't help them (past calling an ambulance) without second thought if there was a chance of getting away unnoticed unless there are plenty reliable-looking witnesses nearby which would testify in a court of law that I tried to help that person with only the purest of intentions. Why? Because if I help and I make a mistake or the person misconstrues my intentions, I'm looking at a potential lawsuit. Helping someone in need without a solid legal defense at the ready is just too dangerous. If the person in need hapens to be female and underage I could end up on trial for attempted child molestation and I'm certainly not going to risk that without a dozen witnesses on my side.
Actually, perhaps I'd skip the ambulance, as well - if I call an ambulance but don't help the person myself I might get sued for not helping, whereas I might also get sued if I help; it appears to be a "damned if you do, damned if you don't" kind of situation. But I'd probably take my chances and call that ambulance.
In Germany I'd call an ambulance and then walk over and try to help that person to the best of my abilities. Helpless persons are relatively safe over here.
Seeing how many entry-level positions in local job boards read "XYZ Consultant" I'd think that you can trust a consultant about as much as any other junior-level code monkey.
Unfortunately I am immune to all forms of connery so I will not tell you as you didn't ask for people who can not, in fact, be conned. Sorry. You do seem very trustworthy.
Slashdot Mirror
Yup. Goes to show how lack of standardization and the NIH syndrome can hold back things.
The mainboard manufacturers have figured out how to do this safely. So can the GPU manufacturers. Fr example, two flash chips and an emergency ROM would make for a very smooth upgrade:
1. The card flashes the upgrade onto the flash chip it's currently not using.
2. The card tells the OS it's going to switch to the new driver.
3. The card switches over to the new driver, reinitializes and tells the OS it's ready.
4. The OS displays a dialog box like the one Windows shows when switching resolutions: "The new driver has been loaded. Does the graphics card work correctly? [Yes] [No] This dialog will automatically select 'No' in 30 seconds."
5. If the OS signifies user satisfaction the new driver is used, otherwise the card loads the old one and reinitializes.
Should the card be unable to work with either driver it falls back to the backup ROM where a known-good (probably reduced-functionality) driver resides. Combined with a smart protocol that allows the card and the OS to tell each other about/change their state and a graphics infrastructure that allows dynamic reloading of the driver this could mean that you can update the GPU driver and switch to the new one without rebooting or even restarting X11.
It might even possible (although only useful for power users) to install two alternative drivers on the two chips and switch between them on the fly - for example one driver which minimizes power consumption by powering down parts of the card and one which gives good gaming performance. Of course, both of these would have to come from the manufacturer.
Tried it, didn't work. In order to do anything meaningful with a non-FAT/NTFS drive under Windows you need to install an IFS. Few people are going to let you install some random driver on their notebook just to access your USB drive. That's also the reason why MacDrive and ext2-IFS are no solutions: While they give you portability under laboratory conditions (between all of your computers), they are useless for exchanging data with others. A FAT killer needs to run out of the box everywhere, period.
That's aso why I don't believe that FUSE will solve the problem. It's in Ubuntu but what about Debian or CentOS or Sabayon (this being three distros I encountered lately on random computers)? And, of course, under OS X it's always a separate install (two even, IIRC), putting it solidly in "why should I install some random driver" territory.
The reason we put up with FAT32 is not high compatibility, it's absolute compatibility. You can stick a FAT32-formatted USB drive into any random comouter you encounter and it will definitely mount with no kind of setup neccessary. Any filesystem that does not meet those criteria is simply no competitor.
Actually, if you want a filesystem usable by everyone it will definitely have to come from Microsoft. MS is not going to adopt someone else's FS unless they really have to, so the only serious contenders are FAT32, NTFS and exFAT (things like UDF are highly unlikely as Microsoft probably won't support it for anything but DVD images).
FAT32 is the current portability king. However, it's also old and extremely limited and people stopped liking it half a decade ago.
NTFS works... kinda. For example, you can mount a Vista NTFS disk unter XP but you're going to lose some metadata. XP NTFS is supported on every computer equipped with FUSE and ntfs-3g and on every computer running a new-ish Windows. Everywhere else it's useless and ntfs-3g is hardly a default package in most distros so no dice.
exFAT is essentially FAT64. It supports large files and has some improvements but in general it's very much FAT32's successor, being lightweight and relatively uncomplicated. While this sounds ideal for a widely supported filesystem there are no free exFAT implementations yet and people aren't sure whether Microsoft is going to allow the existence of one. So this one's not a winner, either.
Everything else will remain *nix-only so, yes, FAT32 is still the only option if you want complete portability (UDF if you can live with jumping through hoops). And it's going to stay that way for the forseeable future.
In that cast someone should start some kind of organization that fights pollution. Greenpeace certainly doesn't appear to.
In fact, this is why it's in the interest of the manufacturers to put part of the drivers into the device itself: It becomes easier to reach new parts of the market - in fact if all problematic parts of the driver are hardware-side one needs to just provide some "harmless" specs to the OSS community and let them create their own driver at zero cost. Also, if those driver parts are sufficiently OS-agnostic one can easily support new versions of Windows... which should make this interesting for at least Nvidia.
No, video game trailers are different from ads. Ads inform you about a product that has or shortly will hit the market - in any case it's market-ready. Trailers inform you of what the company is working on and are often released rather early during production. One important difference is that the company pays mass media to show their ads (as they already know they're going to make money with the product shortly) while trailers are often just released to certain sites and then mirrored throughout the internet.
Great, now we have Pastafarianism: Terror From The Deep, the same as regular Pastafarianism but underwater and with a lot of bugs.
How? Think of the sites you commonly use - with how many of those sites do you pay for a subscription? Those are the sites where you generate revenue. On all other sites you only generate cost and they have to get their money from elsewhere. There are few reliable ways of making a website work financially - essentially you either require your users to pay for subscriptions, you show ads or you lose money.
You could of course demand that every web site is operated at a loss because you neither want to pay for everything nor want to see ads - but that would mean that 95% of all bigger websites would have to shut down or go subscribers-only, including Slashdot.
Unless you can find some way to reduce traffic and hosting costs to zero there needs to be some reliable way for websites to generate revenue.
You appear to have different scent ads than Germany does. In Germany, scent ads are usually exercises in dadaism, completely devoid of any discernible content. The palette goes from women randomly gyrating with trippy special effects overlaid to a man rambling about adventure before driving off on a moped to weird disjointed nonsense that ends with a man walking away from a cyan-tinted Union Jack. In fact, most of them are simply disjointed nonsense.
The easiest way to spot a scent ad is to look if it appears to be completely nonsensical. The less sense it makes, the more likely it's a scent ad.
Makes me wonder who they're targeting with those ads. People who buy stuff simply because some yellow-hued women stumble around on TV while "Heart of Glass" is playing? How the hell does that work?
I once saw a documentation about how they make the photos you see on convenience food packaging. The tomato soup with a cream swirl was actually 100% toxic-if-ingested wall paint. Other dishes were either made by cooks (of course using completely different recipes) or weren't food at all. Don't think only models get airbrushed; food does, as well. With clear varnish, during the shoot.
I mean, some car manufacturer recently ran a TV ad in Germany where they deconstructed the usual car ad by gradually switching off the humans (all professional models), the beautiful scenery (completely computer-generated), the brilliant highlights on the car (ditto), the majestic music and finally the street. I don't remember which car it was but the ad strikes me as insightful - it shows just how much of the ads you see has to do with the actual car (not much at all, not even the car's appearance is realistic).
Ads lie. Ads lie all the time. Do not expect anything you hear in an ad to be remotely true, apart fom "product XYZ exists".
Well, "/m - open-mid front unrounded vowel - m - alveolar approximant - near-close near-front unreounded vowel - st - mid-central vowel/" is certainly much more complicated to read and write than the IPA string it's trying to represent. By the way, I wrote "memristor" there. I know, one could use SAMPA... but isn't SAMPA essentially ASCII art that tries to emulate IPA on legacy systems?
Also, if every decision is made towards making life difficult for obnoxious users - why do we still allow anonymous posting? There's more to message board quality than troll management and sometimes the net gain from implementing a possibly-abused feature is positive.
Apart from me not speaking Perl, I'm not sure whether that patch would have a chance of actually being used.
Yes it is... However one should assume that in 2008 a tech website would be able to support the technology X-SAMPA crudely reimplements. Also, SAMPA doesn't have nearly the penetration of IPA - while one can expect virtually anyone who has learned a foreign language to have at least encountered IPA, SAMPA is useful only to those who do phonetic discussions in a long-outdated encoding.
Hm, that's funny. I always assumed they had some kind of format IDNs are supposed to be transmitted as. You know, some kind of puny code that can be algorithmically derived... Oh, wait! They have such a format! And I'd be willing to bet that it's entirely possible to automatically mangle URLs entered to the appropriate Punycode representation (not to mention that all major browser already do that in the status bar).
But the appropriate Unicode glyph doesn't. And that's another thing wrong with Slashdot's weird filter: Some glyphs work when entered directly, some only work when entered as HTTP entities and yet others don't work either way. IPA glyphs get converted to HTML entities but those entities are then filtered. It's impossible to determine beforehand whether Slash accepts a glyph and in which form it does.
And that's probably why Digg is the superior platform, community and content aside. "We won't do anything more than the absolute minimum required to keep the site running unless it significantly increases our bottom line" is how corporations show that they run the platform not for the community but for the ad partners. Of course the weird filtering is not the only problem - the CSS has been broken for ages (especially in Idle, even though that doesn't hurt much) but apparently there's no business case for getting that right, either.
I'm waiting for the day when Slashdot changes its "we won't delete posts unless legally forced to" policy to "we won't delete posts unless it's in SourceForge, Inc.'s financial interest".
In that case the question is: "Why does a forum for geeks not allow anything outside what's covered by ISO Latin-1?"
Is IPA somehow dangerous? Cyrillic? Mathematical symbols? Or the Euro sign? This is a forum for technical people who, due to their geek nature, prefer to use the appropriate notation to communicate things not easily communicated in ASCII. For example, it's extremely difficult to accurately and concisely communicate the pronounciation of something without using IPA. That's what IPA was designed for. That's why it's in Unicode.
I get the feeling that the whitelist was put in place by someone who doesn't really know Unicode (and/or didn't want to spend time with it) and thus opted to just keep Latin-1 since that's what he knows. There are a lot of characters in Unicode that could safely be whitelisted; depending on how well-written the filter is this should be a matter of one afternoon with a Unicode character list. I mean, we don't demand Linear-B; the safe parts of the BMP (all non-combining non-control code points except those between D800 and F8FF (FFFF if you don't want to include Han)) would be everything 99.999% of all posters need.
I won't link to a screenshot because I don't want to melt someone's server but a Google image search for slashdot omg ponies will give you a lot of examples.
Perhaps not mumbling would make you sound clearer. I seriously don't see how "Kibibyte" sounds like mumbling unless, well, you mumble it. Likewise I don't see how "memristor" should slow one down. [MISSING1*] doesn't seem too bad - [MISSING2*] certainly is possible in the English language even if it doesn't (commonly?) appear in a single word; I can say things like "damn ring" without pausing.
/ˈmɛmɹɪstə/ /mɹ/
* I couldn't properly use the following strings since IPA is filtered out by Slashdot even in HTML entity notation.
MISSING1:
MISSING2:
It's also the year of desktop Linux.
Actually, the scroll ball is a pain to clean but that's mostly an artifact of the Mighty's "no maintenance possible" design.
I think Apple should try to use inductive fields to replace the ball. Sense where the finger is moving and use that to scroll. They probably have already tried it and it doesn't work but it would be very slick and completely dirt-proof.
Except for the Habu, which is built by Razer.
That's not because she was femals, that's because she was Deanna. Come on, this is the woman that used telepathic abilities to "detect hostility" in people who were currently busy getting a torpedo lock on the Enterprise...
I thought that was what American culture is all about. CYA is the highest maxim.
An example: If in the USA someone fell and hurt themselves and I was nearby I wouldn't help them (past calling an ambulance) without second thought if there was a chance of getting away unnoticed unless there are plenty reliable-looking witnesses nearby which would testify in a court of law that I tried to help that person with only the purest of intentions. Why? Because if I help and I make a mistake or the person misconstrues my intentions, I'm looking at a potential lawsuit. Helping someone in need without a solid legal defense at the ready is just too dangerous. If the person in need hapens to be female and underage I could end up on trial for attempted child molestation and I'm certainly not going to risk that without a dozen witnesses on my side.
Actually, perhaps I'd skip the ambulance, as well - if I call an ambulance but don't help the person myself I might get sued for not helping, whereas I might also get sued if I help; it appears to be a "damned if you do, damned if you don't" kind of situation. But I'd probably take my chances and call that ambulance.
In Germany I'd call an ambulance and then walk over and try to help that person to the best of my abilities. Helpless persons are relatively safe over here.
None. There are no mice with my left big toe.
Seeing how many entry-level positions in local job boards read "XYZ Consultant" I'd think that you can trust a consultant about as much as any other junior-level code monkey.
This seems like a perfectly harmless request.
Unfortunately I am immune to all forms of connery so I will not tell you as you didn't ask for people who can not, in fact, be conned. Sorry. You do seem very trustworthy.