What happens is you can post a Tweet with astral-plane glyphs and it all appears to work fine, but mysteriously --- a week or so later --- the astral-plane glyphs just vanish. (I don't know if this happens to basic-plane glyphs; I haven't tested it.) I suspect what's happening is that they have different short-term and long-term storage systems, and the long-term systems don't handle Unicode properly.
For example, see this message. That one lasted for about two weeks before the last word vanished. I should probably go hunting for a bug report form...
Couldn't that $1.4 billion have been better spent buying Valium for the rampant xenophobes in Congress? Just trank 'em all out and stop them from worrying about a non-problem.
One of L. Sprague de Camp's fantasy novels features a tribe of barbarians who discuss all political issues twice: once sober, once drunk.
I think this is an excellent idea. Who's going to lobby Congress?
I would think the ability to regenerate body parts on demand would be an evolutionary advantage, wouldn't it?
Not necessarily. A lot of small animals are pretty much disposable: they're sufficiently fragile that there's only a very narrow boundary between a trivial injury and a fatal one. (And anyone who's kept small birds and animals will know that if they're hurt beyond a certain point they'll simply go into shock and die.)
So it's entirely plausible that the gene might have been caused by a spot mutation very early on while all mammals were basically mice, and it then had a sufficiently small effect on actual survivability that the trait didn't get bred out. Later, once the small, disposable animals turned into large, expensive ones, it was too late.
It is interesting that both birds and animals appear to lack this trait, though. We both descend from much the same sort of lizards but in different directions. Finding out exactly where this gene sequence appeared might be productive.
(Of course, I want to know when we'll be able to get gene therapy to suppress the gene. Assuming it works in humans, and that the gene doesn't do anything else critical, it might even be fairly straightforward! But probably won't happen soon and I'm certainly not volunteering to be the guinea pig...)
Nah, I'm more professional than that --- I discovered that it worked, went and asked permission, and then used it to download all the development tools that had been preloaded onto the laptop but had now been erased, while we were waiting for their beaurocracy to enable my account.
All in all it took about a week and a half of my very expensive time (mostly hotel costs) just to get into a position where I could start to get my job done. Hurrah for security.
I've been on the other end of this: I got sent out to work with a Korean customer on-site. They wouldn't allow our machine to be connected to their LAN without installing some security software (name and shame: Waterwall).
The installation goes halfway through and fails obscurely. Three hours of debugging later they finally realise that I have IE8, but Waterwall only works on IE6 or IE7.
So I try to install IE7. I can't, because IE8 is installed. I try to uninstall IE8. I can't, because our sysadmin is, like, competent, and had set up the laptop with a DVD image with IE8 slipstreamed into it.
I eventually had to borrow an XP disk from the customer and reinstall Windows. Then I installed Microsoft Security Essentials and removed the virus that was on their XP disk.
The real joke? Waterwall blocks web access, enforces encryption on USB keys and recordable media, etc. (It's intended to stop 'information leakage'.) The internet? Wide open. ssh worked fine...
I stand corrected. Thanks for the link --- now I have something to point people at (because they are very cool).
Interestingly they're considerably cheaper here in the UK. I know the expensive part is the tritium --- they come in two brightnesses which differ only in terms of how much gas they put in. I wonder which of us is being ripped off...
We also get them in pink, blue and orange as well as green. Green's brightest.
...because tritium's really expensive to make and they're wasting it.
A few years back I bought a bunch of glow-in-the-dark keyrings as stocking fillers for my family. These are little tubes containing tritium. The tritium produces very low energy beta particles, which excite phosphor on the inside of the tube, which cause them to glow. They have a half-life of 12 years, which in effect means that they glow usefully for about five or six years before they need replacing. (I should probably get them new ones.)
Let me repeat that: it's a little glowing thing that will glow for six years, continuously. They don't need recharging, they don't need their batteries changed, they don't need exposure to sunlight. They're fantastic for safety-critical things like exit signs. My father sails, and he has his tied to the end of the emergency torch on his boat --- it means that if he needs it in a hurry in the dark, he can find it. I know a nurse who uses them to find things in bags of equipment. They're really handy.
Naturally, they're banned in the US, because they're atomic.
(Tritium, being hydrogen and really hard to contain, will slowly diffuse out through the walls of the glass tube and into the environment. However there's a tiny, tiny amount of the stuff, and the radioactivity they emit is so weak it won't penetrate six millimetres of air, let alone anything solid. I suppose it is possible to absorb the stuff into the body --- we are largely made of hydrogen, after all --- but the low energies, short half-life and tiny quantities means that you're probably more likely to get radiation damage from Bikini Atoll than your tritium keyring.)
Incidentally, did you know that after the Chalk River reactor in Canada was shut down in 2009 due to overreaction, there is now a worldwide shortage of medical isotopes? There are only five reactors worldwide, sorry, four now, that produce the stuff. I wonder how many people that shutdown has killed?
So, is there any actual evidence backing all this up, or is it just more anti-Chinese vilification?
(Remember, we have always been at war with Eastasia.)
Re:But will it get you high when you snort it?
on
Spray-On Liquid Glass
·
· Score: 1
Unless you think this process includes popping the coated object in a forge, again I say: Apples -> Oranges.
And again I say, evidence please. You are making a statement based on no data other than your intuition. Yes, this coating is different from a big slab of glass. Does this then imply that all this bad stuff you're talking about will happen? No.
The burden of proof here is on you. You must justify your statements, or else they will be rightly ignored.
Re:But will it get you high when you snort it?
on
Spray-On Liquid Glass
·
· Score: 2, Insightful
But I do toss stuff onto my counters all the time. Abrasive stuff like my keys, heavy stuff like cast iron pots, and even just sharp and pointy stuff like a pile of silverware.
You mean similar to the way people slam cast iron pots and piles of silverware onto glass cooker tops, cutting boards, etc?
We use glass a hell of a lot, and silicosis is not a problem. You're going to need evidence to back up your statement.
Re:But will it get you high when you snort it?
on
Spray-On Liquid Glass
·
· Score: 1
A glass shell over your counter-top is going to be silicon dust in the air in a few months of use, if it lasts that long.
My house is full of windows. It's not full of silicon dust.
I agreed right up to the point you suggested the Moon as a good training ground. The Moon is far more harsh than Mars.
The Moon has the huge advantage in that if everything goes pear-shaped you can do a crash return and get home. It's not hard to build an escape vehicle that can get from the lunar surface to Earth in a few days.
On Mars, though, if things go wrong, you die.
This makes the Moon an ideal place to get started on the hideously difficult job of setting up a permanent off-world base. Not only can we get people home if things go wrong, but we can also resupply on a short-term basis as needed --- and it will be needed, because as a first-attempt engineering project, things will always go wrong!
Do you remember Skylab, the very first space station ever? The launch went badly wrong, and it sustained major damage, including the loss of the solar heat shield. The first crew had to be launched in a hurry to do repairs or the station would have overheated and released poisonous gases inside, rendering it uninhabitable! Had the station not been close enough to Earth that it was possible to reconfigure the manned mission to include the appropriate repair equipment, the station would have been a write-off.
This link provides a one page summary of each attempted Debian port...
Don't forget the unofficial Debian ports; one really interesting one is Debian Interix, which is a Debian userland on top of Microsoft's Unix layer for Windows. It hasn't had a lot of activity recently because it's mainly run by one guy, and Interix doesn't get a lot of bugfix love from Microsoft (surprise surprise), but compared to the abomination that is Cygwin it's already light-years ahead. All it needs is a decent community...
Their FreeBSD filesystem-heavy benchmarks look extremely dubious --- UFS and ext3 aren't that different. I find myself wondering whether they actually remembered to turn SoftUpdates on.
You are not ready for the US American cell phone companies. Packets are transported by the souls of the damned who sign service contracts. Your firstborn is not payment enough; you must provide the souls of all your friends and family.
You'd be much better off in the EU. Here, all calls are accompanied by the heavenly sound of singing angels, and whenever you call customer service an avatar of your preferred sex materialises in front of you to give you sexual satisfaction while you're on hold.
The one true programming font. No other font better manages the compromise between legibility and compactness,
and being a well-crafted bitmap font, it is crisper and clearer than ever on modern LCD screens.
X11 got it right 25 odd years ago, and now with near-full Unicode support, it's only gotten better.
Alas, Gnome and KDE both try really hard to hide the mere fact of bitmap fonts' existence from you. It appears to be impossible to tell apps like Eclipse or gnome-terminal to use fixed.
I've always been rather partial to Unifont. It's a bit chunky compared to fixed, by it supports all characters in Unicode BMP (programming: it's not just in ASCII any more), and there's a TTF version. Of course, the TTF file renders each character as a sequence of little squares, which means that unless you like 12pt you'll get a migraine, but at least it works...
As the author, you are entitled to relicense the software as you see fit. But if you have ever received a patch from anyone, you are no longer the sole author. This means that the only parts you can relicense are the bits you wrote. You can't relicense any of the bits other people wrote without their approval.
Having multiple authors in a projects makes your life significantly more complicated, legally speaking. It's vitally important to keep a record of any contributions you've received to a project. The FSF actually require all contributors to their official projects to sign over copyright to them so that they legally own the contributions, simply to avoid the problem of having multiple authors.
This has been a problem with some semi-open-source projects. I remember that the sparse compiler frontend, which had a rather strange and legally dubious license, is now unable to relicense the source to something more sensible because some of the authors no longer exist, which means their approval cannot be obtained.
So you need to be very careful here. Producing a commercial license simply may not be an option for you.
$40 million dollars is a lot of money, and will do a great deal of good in Haiti. It's great that people are willing to donate to help people --- goodness knows they need it, and we can all spare it.
But to put matters into perspective, the war in Iraq and Afghanistan is spending that amount every three and a half hours (based on the Congressional Research Service figure of $2 billion a week, which comes out to about $12 million an hour).
<pedant> 92 employees isn't such a big number. And who's the 0.3521 of an employee? Did someone fail to get out of the way fast enough when closing the tape vault? </pedant>
It's not thought to be "space junk" any more: it was thought it might be an old booster segment but apparently based on its path there's no rocket launch that it could've come from.
Incoming alien spaceship, perhaps? 130000km is a good distance to approach to; it's safely clear of our geostationary satellite belt, but still close enough for a good look. Plus, our puny earth technology can't actually get that far without about six months' lead time. 10m isn't big by our standards but any spacefaring civilisation is unlikely to be crewing their ships with canned primates anyway so there's no real grounds for comparison.
Alas, there's no way of estimating what sort of drive it's got, and therefore when it would start needing to brake; but we don't even know whether it's going into orbit or not. I mean, we can probably assume that these guys are intelligent, so why would they want to come here? I don't want to be here, and I evolved here...
Other gotchas - not really...if you want to control access, the recipient has to have a Live ID. If you want to make it public, they don't. No trial, no cost, just free with advertisements.
To me the big gotcha is that there's no access API. I'd love to use SkyDrive as an online backup service --- with encrypted files, natch --- but there's just no way of getting stuff there other than via the web interface.
What I'd really like is one of these services to adopt rsync, but it ain't going to happen...
Except that IE8 is perfectly capable of emulating both IE6's and IE7's standards-noncompliance modes, in addition to rendering in a proper (albeit lacking some newer features) standards-compliant mode.
I recently spent some time in Korea, working on-site for a customer who you will have heard of.
We had to set up our machine with their ghastly intranet security software. After realising that their intranet portal only works with IE due to stupid stuff like missing Javascript onClick handlers, we started the installation procedure and the requisite four reboots. It failed weirdly after reboot #3.
After some time trying to make it work, we discovered that their security software is not compatible with IE8.
Unfortunately our sysadmin is quite efficient, which means that the Windows installation had IE8 slipstreamed into it. This meant it couldn't be removed. And you can't install IE7 on a machine with IE8 on it. Which meant that the only way to progress was to reinstall Windows, from scratch, using an XP CD that the customer lent us.
Why the hell can't cell phones be this way, instead of the current quagmire where they're hopelessly entangled with what the carrier wants? I want a cellular carrier that charges a fair price for service (per byte and per minute, or whatever), and then lets me use whatever device I want to use that service. If I can stick a radio into a TI-89 and make it speak CDMA, let me make phone calls with it.
Because you're in America, the land of the fee.
More seriously, CDMA is a large part of the problem. Most CDMA phones aren't designed to work with multiple carriers. The phone ID is hard-coded at build time and tied to a particular carrier. This means that it's really hard to change them to another carrier.
GSM phones work differently. The network ID, the bit that is tied to a particular carrier, is actually housed on a smartcard that plugs into the phone. You can remove the smartcard and insert it into another phone, and presto, that phone adopts the smartcard's ID and logs on to the appropriate carrier.
While you still get subsidised phones with GSM that are locked to one particular carrier, and will refuse to work with a different SIM, the fact that this is possible and easy has encouraged a whole industry of unlocked phones and SIMs. You can go into any supermarket and buy a SIM in a box (that one is $7 and contains $15 worth of credit). If you need a phone you can either buy a cheap SIM-less phone (that one costs $10!), but they'll work in any unlocked GSM phone. The end result is that I, living in the UK, can spend about $30 a year on mobile phone service. That includes data.
(If you hunt around you can actually find SIM-only options for GSM phones in America, but of course this requires you to live in a GSM area; plus, the terms are usually terrible with unpleasant features like evaporating credit if you don't use it.)
There is apparently a standard for a similar CDMA smartcard system, but it's now too late and nobody cares.
Slashdot Mirror
Actually, Twitter is not Unicode-safe.
What happens is you can post a Tweet with astral-plane glyphs and it all appears to work fine, but mysteriously --- a week or so later --- the astral-plane glyphs just vanish. (I don't know if this happens to basic-plane glyphs; I haven't tested it.) I suspect what's happening is that they have different short-term and long-term storage systems, and the long-term systems don't handle Unicode properly.
For example, see this message. That one lasted for about two weeks before the last word vanished. I should probably go hunting for a bug report form...
Couldn't that $1.4 billion have been better spent buying Valium for the rampant xenophobes in Congress? Just trank 'em all out and stop them from worrying about a non-problem.
One of L. Sprague de Camp's fantasy novels features a tribe of barbarians who discuss all political issues twice: once sober, once drunk.
I think this is an excellent idea. Who's going to lobby Congress?
Not necessarily. A lot of small animals are pretty much disposable: they're sufficiently fragile that there's only a very narrow boundary between a trivial injury and a fatal one. (And anyone who's kept small birds and animals will know that if they're hurt beyond a certain point they'll simply go into shock and die.)
So it's entirely plausible that the gene might have been caused by a spot mutation very early on while all mammals were basically mice, and it then had a sufficiently small effect on actual survivability that the trait didn't get bred out. Later, once the small, disposable animals turned into large, expensive ones, it was too late.
It is interesting that both birds and animals appear to lack this trait, though. We both descend from much the same sort of lizards but in different directions. Finding out exactly where this gene sequence appeared might be productive.
(Of course, I want to know when we'll be able to get gene therapy to suppress the gene. Assuming it works in humans, and that the gene doesn't do anything else critical, it might even be fairly straightforward! But probably won't happen soon and I'm certainly not volunteering to be the guinea pig...)
Nah, I'm more professional than that --- I discovered that it worked, went and asked permission, and then used it to download all the development tools that had been preloaded onto the laptop but had now been erased, while we were waiting for their beaurocracy to enable my account.
All in all it took about a week and a half of my very expensive time (mostly hotel costs) just to get into a position where I could start to get my job done. Hurrah for security.
I've been on the other end of this: I got sent out to work with a Korean customer on-site. They wouldn't allow our machine to be connected to their LAN without installing some security software (name and shame: Waterwall).
The installation goes halfway through and fails obscurely. Three hours of debugging later they finally realise that I have IE8, but Waterwall only works on IE6 or IE7.
So I try to install IE7. I can't, because IE8 is installed. I try to uninstall IE8. I can't, because our sysadmin is, like, competent, and had set up the laptop with a DVD image with IE8 slipstreamed into it.
I eventually had to borrow an XP disk from the customer and reinstall Windows. Then I installed Microsoft Security Essentials and removed the virus that was on their XP disk.
The real joke? Waterwall blocks web access, enforces encryption on USB keys and recordable media, etc. (It's intended to stop 'information leakage'.) The internet? Wide open. ssh worked fine...
Mention my Slashdot UID while ordering, and receive a FREE stay in the Federal prison of your choice!
I stand corrected. Thanks for the link --- now I have something to point people at (because they are very cool).
Interestingly they're considerably cheaper here in the UK. I know the expensive part is the tritium --- they come in two brightnesses which differ only in terms of how much gas they put in. I wonder which of us is being ripped off...
We also get them in pink, blue and orange as well as green. Green's brightest.
...because tritium's really expensive to make and they're wasting it.
A few years back I bought a bunch of glow-in-the-dark keyrings as stocking fillers for my family. These are little tubes containing tritium. The tritium produces very low energy beta particles, which excite phosphor on the inside of the tube, which cause them to glow. They have a half-life of 12 years, which in effect means that they glow usefully for about five or six years before they need replacing. (I should probably get them new ones.)
Let me repeat that: it's a little glowing thing that will glow for six years, continuously. They don't need recharging, they don't need their batteries changed, they don't need exposure to sunlight. They're fantastic for safety-critical things like exit signs. My father sails, and he has his tied to the end of the emergency torch on his boat --- it means that if he needs it in a hurry in the dark, he can find it. I know a nurse who uses them to find things in bags of equipment. They're really handy.
Naturally, they're banned in the US, because they're atomic.
(Tritium, being hydrogen and really hard to contain, will slowly diffuse out through the walls of the glass tube and into the environment. However there's a tiny, tiny amount of the stuff, and the radioactivity they emit is so weak it won't penetrate six millimetres of air, let alone anything solid. I suppose it is possible to absorb the stuff into the body --- we are largely made of hydrogen, after all --- but the low energies, short half-life and tiny quantities means that you're probably more likely to get radiation damage from Bikini Atoll than your tritium keyring.)
Incidentally, did you know that after the Chalk River reactor in Canada was shut down in 2009 due to overreaction, there is now a worldwide shortage of medical isotopes? There are only five reactors worldwide, sorry, four now, that produce the stuff. I wonder how many people that shutdown has killed?
So, is there any actual evidence backing all this up, or is it just more anti-Chinese vilification?
(Remember, we have always been at war with Eastasia.)
Unless you think this process includes popping the coated object in a forge, again I say: Apples -> Oranges.
And again I say, evidence please. You are making a statement based on no data other than your intuition. Yes, this coating is different from a big slab of glass. Does this then imply that all this bad stuff you're talking about will happen? No.
The burden of proof here is on you. You must justify your statements, or else they will be rightly ignored.
But I do toss stuff onto my counters all the time. Abrasive stuff like my keys, heavy stuff like cast iron pots, and even just sharp and pointy stuff like a pile of silverware.
You mean similar to the way people slam cast iron pots and piles of silverware onto glass cooker tops, cutting boards, etc?
We use glass a hell of a lot, and silicosis is not a problem. You're going to need evidence to back up your statement.
A glass shell over your counter-top is going to be silicon dust in the air in a few months of use, if it lasts that long.
My house is full of windows. It's not full of silicon dust.
I agreed right up to the point you suggested the Moon as a good training ground. The Moon is far more harsh than Mars.
The Moon has the huge advantage in that if everything goes pear-shaped you can do a crash return and get home. It's not hard to build an escape vehicle that can get from the lunar surface to Earth in a few days.
On Mars, though, if things go wrong, you die.
This makes the Moon an ideal place to get started on the hideously difficult job of setting up a permanent off-world base. Not only can we get people home if things go wrong, but we can also resupply on a short-term basis as needed --- and it will be needed, because as a first-attempt engineering project, things will always go wrong!
Do you remember Skylab, the very first space station ever? The launch went badly wrong, and it sustained major damage, including the loss of the solar heat shield. The first crew had to be launched in a hurry to do repairs or the station would have overheated and released poisonous gases inside, rendering it uninhabitable! Had the station not been close enough to Earth that it was possible to reconfigure the manned mission to include the appropriate repair equipment, the station would have been a write-off.
This link provides a one page summary of each attempted Debian port...
Don't forget the unofficial Debian ports; one really interesting one is Debian Interix, which is a Debian userland on top of Microsoft's Unix layer for Windows. It hasn't had a lot of activity recently because it's mainly run by one guy, and Interix doesn't get a lot of bugfix love from Microsoft (surprise surprise), but compared to the abomination that is Cygwin it's already light-years ahead. All it needs is a decent community...
Their FreeBSD filesystem-heavy benchmarks look extremely dubious --- UFS and ext3 aren't that different. I find myself wondering whether they actually remembered to turn SoftUpdates on.
You are not ready for the US American cell phone companies. Packets are transported by the souls of the damned who sign service contracts. Your firstborn is not payment enough; you must provide the souls of all your friends and family.
You'd be much better off in the EU. Here, all calls are accompanied by the heavenly sound of singing angels, and whenever you call customer service an avatar of your preferred sex materialises in front of you to give you sexual satisfaction while you're on hold.
The one true programming font. No other font better manages the compromise between legibility and compactness, and being a well-crafted bitmap font, it is crisper and clearer than ever on modern LCD screens. X11 got it right 25 odd years ago, and now with near-full Unicode support, it's only gotten better.
Alas, Gnome and KDE both try really hard to hide the mere fact of bitmap fonts' existence from you. It appears to be impossible to tell apps like Eclipse or gnome-terminal to use fixed.
I've always been rather partial to Unifont. It's a bit chunky compared to fixed, by it supports all characters in Unicode BMP (programming: it's not just in ASCII any more), and there's a TTF version. Of course, the TTF file renders each character as a sequence of little squares, which means that unless you like 12pt you'll get a migraine, but at least it works...
Have you ever received any contributions?
As the author, you are entitled to relicense the software as you see fit. But if you have ever received a patch from anyone, you are no longer the sole author. This means that the only parts you can relicense are the bits you wrote. You can't relicense any of the bits other people wrote without their approval.
Having multiple authors in a projects makes your life significantly more complicated, legally speaking. It's vitally important to keep a record of any contributions you've received to a project. The FSF actually require all contributors to their official projects to sign over copyright to them so that they legally own the contributions, simply to avoid the problem of having multiple authors.
This has been a problem with some semi-open-source projects. I remember that the sparse compiler frontend, which had a rather strange and legally dubious license, is now unable to relicense the source to something more sensible because some of the authors no longer exist, which means their approval cannot be obtained.
So you need to be very careful here. Producing a commercial license simply may not be an option for you.
$40 million dollars is a lot of money, and will do a great deal of good in Haiti. It's great that people are willing to donate to help people --- goodness knows they need it, and we can all spare it.
But to put matters into perspective, the war in Iraq and Afghanistan is spending that amount every three and a half hours (based on the Congressional Research Service figure of $2 billion a week, which comes out to about $12 million an hour).
Have you ever _used_ Vista?
Has anyone?
What else should I run my Windows ME emulator on?
...about 3.10^4 employees worldwide...
<pedant> 92 employees isn't such a big number. And who's the 0.3521 of an employee? Did someone fail to get out of the way fast enough when closing the tape vault? </pedant>
It's not thought to be "space junk" any more: it was thought it might be an old booster segment but apparently based on its path there's no rocket launch that it could've come from.
Incoming alien spaceship, perhaps? 130000km is a good distance to approach to; it's safely clear of our geostationary satellite belt, but still close enough for a good look. Plus, our puny earth technology can't actually get that far without about six months' lead time. 10m isn't big by our standards but any spacefaring civilisation is unlikely to be crewing their ships with canned primates anyway so there's no real grounds for comparison.
Alas, there's no way of estimating what sort of drive it's got, and therefore when it would start needing to brake; but we don't even know whether it's going into orbit or not. I mean, we can probably assume that these guys are intelligent, so why would they want to come here? I don't want to be here, and I evolved here...
Other gotchas - not really...if you want to control access, the recipient has to have a Live ID. If you want to make it public, they don't. No trial, no cost, just free with advertisements.
To me the big gotcha is that there's no access API. I'd love to use SkyDrive as an online backup service --- with encrypted files, natch --- but there's just no way of getting stuff there other than via the web interface.
What I'd really like is one of these services to adopt rsync, but it ain't going to happen...
Except that IE8 is perfectly capable of emulating both IE6's and IE7's standards-noncompliance modes, in addition to rendering in a proper (albeit lacking some newer features) standards-compliant mode.
I recently spent some time in Korea, working on-site for a customer who you will have heard of.
We had to set up our machine with their ghastly intranet security software. After realising that their intranet portal only works with IE due to stupid stuff like missing Javascript onClick handlers, we started the installation procedure and the requisite four reboots. It failed weirdly after reboot #3.
After some time trying to make it work, we discovered that their security software is not compatible with IE8.
Unfortunately our sysadmin is quite efficient, which means that the Windows installation had IE8 slipstreamed into it. This meant it couldn't be removed. And you can't install IE7 on a machine with IE8 on it. Which meant that the only way to progress was to reinstall Windows, from scratch, using an XP CD that the customer lent us.
...which turned out to have a virus on it.
</pissed off>
Why the hell can't cell phones be this way, instead of the current quagmire where they're hopelessly entangled with what the carrier wants? I want a cellular carrier that charges a fair price for service (per byte and per minute, or whatever), and then lets me use whatever device I want to use that service. If I can stick a radio into a TI-89 and make it speak CDMA, let me make phone calls with it.
Because you're in America, the land of the fee.
More seriously, CDMA is a large part of the problem. Most CDMA phones aren't designed to work with multiple carriers. The phone ID is hard-coded at build time and tied to a particular carrier. This means that it's really hard to change them to another carrier.
GSM phones work differently. The network ID, the bit that is tied to a particular carrier, is actually housed on a smartcard that plugs into the phone. You can remove the smartcard and insert it into another phone, and presto, that phone adopts the smartcard's ID and logs on to the appropriate carrier.
While you still get subsidised phones with GSM that are locked to one particular carrier, and will refuse to work with a different SIM, the fact that this is possible and easy has encouraged a whole industry of unlocked phones and SIMs. You can go into any supermarket and buy a SIM in a box (that one is $7 and contains $15 worth of credit). If you need a phone you can either buy a cheap SIM-less phone (that one costs $10!), but they'll work in any unlocked GSM phone. The end result is that I, living in the UK, can spend about $30 a year on mobile phone service. That includes data.
(If you hunt around you can actually find SIM-only options for GSM phones in America, but of course this requires you to live in a GSM area; plus, the terms are usually terrible with unpleasant features like evaporating credit if you don't use it.)
There is apparently a standard for a similar CDMA smartcard system, but it's now too late and nobody cares.