Hell, yeah. Whenever I see the alternative I realise just how good the BBC is. Sky is dire by comparison, and US TV is just unwatchable.
One thing people tend not to realise is that because the BBC has its own guaranteed source of income, they can't be put under pressure by their sponsors. There's an old story of a car company whose latest product had just been slammed by Top Gear, the BBC's excellent motoring programme (although less excellent than it was. Sigh). And when Top Gear doesn't like something, they're not subtle about it... the story goes is that the CEO watched the review, said, "I'll teach them to talk like that about us. Pull all our advertising from that channel. Now." And his secretary said, "Um..."
You would not believe what a difference this makes. During the last Gulf War I watched some CNN and MSNBC. It was embarrassing. A lot of it was cultural differences, but the blatant jingoism and emotionalism made, to me, a complete mockery of the whole concept of independent journalism ---I found it hard to accept what I was watching as being anything but outright propaganda.
(Incidentally, the BBC is not government funded, and their charter clearly makes them independent from government interference. If the government tries to pressure them, most BBC journalists shout 'Hurrah!' and it tends to make the news.)
I don't like the heavy-handed way the license fee is collected --- they use scare tactics a lot. "This man didn't pay his license fee. Now he's bankrupt, his wife has left him, his kids are drug addicted hookers, and we shot his dog. Don't let this happen to you." They also have a lot of trouble believing that some people don't have TVs. If they'd be nicer about it, I'd be much happier paying for it.
Actually, Mathematicians don't say that. Mathematicians say that a closed curve is homeomorphic to S^1, and a line to R^1, ie, there exists a bijective, bicontinuous mapping between the sets.
A topologist is someone who can't work out whether to dip his doughnut into his coffee mug, or vice versa...
Warm reboots don't erase memory. Cold reboots usually don't erase memory, either. (There are still fragments of what was left before after doing a cold boot.)
Standard DRAM will maintain its state --- mostly --- for a remarkably long time without refreshing. Unfortunately, it doesn't do so in a useful state.
I once was working on an embedded device that had VGA out. The development cycle was power on, boot from TFTP, run system, wait until it crashed, power off, repeat. When the system switched on, one of the first things the boot loader did was to initialise the video chipset, but without clearing the video memory.
If the board had been off for less than about five minutes, you could still see the last display that had been there when the board crashed.
Without refreshes, the data would gradually fade; the image was always corrupted with snow. The longer you left it switched off for, the worse the snow got. Different RAM chips lasted different lengths of time --- there was one band across the middle that would become completely unintelligable in about 30s, while another one could hold an image for about two minutes.
I suppose you could use this to store data for short periods during a power down, but you'd have to use so much redundancy to ensure that the data would survive the inevitable corruption that it probably wouldn't be worth it, but I'm sure someone, somewhere, could come up with a Nifty Trick(TM)... You couldn't do it at all on PCs, of course --- on boot, they wipe all their RAM, video or otherwise.
Why don't you just pick up the phone and ask wtf they are and what they want. Answering won't hurt you any way so whats the deal?
The caller's probably from the US, where they have this bizarre system were the recipient of a mobile phone call pays.
+ in a phone number usually indicates an international number (you're supposed to dial your international call prefix plus the number, you see) but there are no country codes beginning with 0. Assuming that, again, he's in the US and the + is in fact erroneous, 001819 would dial a number in Sherbooke, Quebec, Canada.
And to make things even better, the energy the process requires comes from natural gas produced in the later stages of the thermal depolymerization process. The only energy a TDP plant needs is an initial shot of natural gas to get things going, and an electical supply for such things as controlling valves and running sensors.
And that's really impressive. A decade from now we're probably going to see miniaturised versions of this being used as standard sewage treatment plants.
(Which will really piss off the village in Scotland where my parents live; they've just spent vast amounts of money building a sewage treatment plant only to discover that due to incompetence it's been located slightly below the high tide mark. Which means that every time there's a spring tide, unpleasant stuff happens.)
I wonder if it would be economically viable to funnel all the sewage produced by some random big city into one of these things, producing biodiesel and electricity that are fed back into the city?
It will be a welcome change from the jingoism and neo-conservative hate-mongering that is currently blanketing the US.
Maybe you could use it to transmit data? If you hooked up, say, your average Bush voter to a blood-pressure machine and then had someone several miles away talk about how bad the war in Iraq was, I'm sure you'd see a spike on the graph. By using carefully timed conversations and statistical analysis, you should be able to get at least some bandwidth.
The problem would be isolating the signal from the noise. A passing planeload of French tourists would produce so much interference that you'd probably have to give up and resend. And if a Michael Moore film was shown in the same state, you'd probably burn out your surge protectors...
In my personal experience, most usenet programs (especially OE) were fine for getting me registered in groups and posting, but I'll be damned if I'm going to retread everything to find one little post I made that I want to read the replies to.
Any newsreader worth its salt should be able to find your own articles at the push of a button. This is something I like doing, too.
Secondly, Slashdot is already experimenting with an NNTP based feed...
Really? Where? The FAQ entry still says its not going to happen, partly for advertising reasons...
The lecturer was, apparently, talking about the problems in writing mission-critical embedded devices, and at one point he asks his audience: "You all write embedded systems software. Tell me honestly; if your company wrote the software for a 747, how many of you would actually feel comfortable on board one?"
One hand goes up.
"You, sir! You're so confident in your software you'd trust your life to it?"
"Hell, no," comes the reply. "But any plane running my team's software would never crash, because it'd never get off the ground..."
I am confident in the level of safety given by running Windows on a mission-critical device.
would increasing the size of the actual chip help any?
It would --- but there would be other problems.
The first one is the most simple: silicon's expensive. Really expensive. The more units you can slice off that wafer the cheaper the units are. Making the die bigger simply for thermal reasons isn't going to wash with the chip manufacturers. They already glue the die to a metal backing plate, which gives you much the same effect anyway.
The second one, however, is the most crucial one. Electricity is slow. Electrical impulses travel at about 2/3 c through copper and a touch less through silicon (IIRC, I can't find the figures to check). This means that the bigger your die is, the longer it takes the impulses to travel from one side of it to the other.
A 1GHz clock fires every 10^-9 seconds; since the speed of light is 3x10^8 m/s, this means that the impulses are going to travel about twenty centimetres between clock pulses. For a 4GHz clock, it'll be about 5cm. There's a lot more wiring than that folded up inside the die; and it gets worse --- particular things happen at particular times throughout the clock cycle, and where you are in the clock cycle now depends on how long the wire is that connects you to the clock. Making sure everything happens in sync is a nightmare.
There are solutions to all of this; asynchronous designs which don't use clocks, offloading functionality to special-purpose processors like GPUs so you don't need as fast a main processor, radically different approaches like Cell, optical transports so you can route signals through each other, etc, but basically there are loads of good reasons why you need the die to be as small as humanly possible.
Like I said in a previous comment, I'm working on Camaelon 2, a pixmap theme engine that lets you have pretty things.. it should be officially released before the end of this month (it already works, I just want to clean up things).
Cool; will it become the default?
(Incidentally, would it be particularly hard to implement all the low-level GNUstep widgets by peering them off GTK or Qt widgets instead? Seeing as they've already done the work, it would make sense to take the shortcut...)
Well, I didn't show that part, but that works exactly in Gorm like on InterfaceBuilder on OSX
Well, I haven't used InterfaceBuilder, so I'll take your word for it.
The demo's very impressive; I particularly like the connections feature for setting up relationships between different objects. I wish Glade had that.
I do, however, have two minor criticisms.
Firstly, please, please update the look-and-feel. If you want to be taken seriously, don't look like a reject from the 80s. Given GNUsteps modularity, this should be easy enough to do. So, do it. (Tip: application icons should always have labels, because since they're supposed to be unique you can pretty much guarantee they're going to be unfamiliar to someone.)
Secondly, I didn't see any support for layout management in Gorm --- that application was constructed by just placing absolute-sized objects at absolute positions in a window. Please tell me this isn't how you design all applications... because that way leads to inflexible, unscalable, uncustomisable applications, and there's no excuse for that any more. Fixed layouts mean you can't let the user change fonts, because different fonts are different shapes (you can't just scale linearly). Fixed layouts mean wasted screen estate (remember the old Mac file browser dialogues that would float a tiny, eight-line scrollable list in the middle of a 21" monitor?). Fixed layouts are just wrong.
Whether or not the guy had a stake in the towing company is of little consequence. The point being the government is now becoming directly in volved as a competitor to these towing companies at the tax payers expense. Also, could anyone use these yellow trucks services or did they check id to ensure you paid for the service?
Then again, maybe not... if it was free, you could argue that it's a public service to get stuck cars moving again as quickly as possible during rush hour: the amount of money that everyone else on the road is saving is far greater than you might get if you charged for the service; and that, IMO, is the mark of whether something should be a public service or not.
Because the other person in the car is exposed to the same environment you are, and you are aware of each other's body language. It's a small matter, but a crucial one: it requires far less attention to communicate with someone who is physically present than with someone who's a disconnected voice on the other end of a telephone line.
For example, if a truck suddenly pulls out in front of you, you will suddenly focus on it; your passenger will tend to notice this and stop talking. Someone on the other end of a phone won't.
Sodding Salon's sodding registration thing doesn't work --- the 'free pass' button isn't clickable. On my browser, at least. Someone hurry up and post the text.
(And people can't understand why these jump-through-a-hoop-to-read-the-article systems annoy people so much... half the time they don't work! I've simply given up trying to register the NYT, for example.)
For what the article claims is a cultural phenomenon, I've never heard of it. Is this just a US thing?
FWIW, it sounds awfully similar to Frozen Bubble --- was FB based on Snood?
Re:Maybe you should read a book / the spec
on
The CSS Anthology
·
· Score: 1
Your logo is "cowlark.com" in text. Thats 11 elements(characters). You can just set the size as 11em, and the browser figured out how much 1em is based on the size of the font.
Actually, an em is the height of the font, from the lowest descender to the highest ascender. (It used to be the width of the letter 'M' in that font, hence the name --- you'll never guess what an en is.)
This means that 11em is going to be roughly the width of:
MMMMMMMMMMM
Which is much longer than my logo, in a proportional font:
cowlark.com
So, it still doesn't help, I'm afraid.
Re:Maybe you should read a book / the spec
on
The CSS Anthology
·
· Score: 2, Insightful
<div class="foo bar">
I didn't know about that; useful, although not quite what I'm looking for. Ta.
(The thing I object to here is that it requires the HTML content to have detailed knowledge of the mechanism of the markup, which is evil --- there's no need for the content to have to state that that particular div is a foo and a bar. It should just be able to state that it's a bar. The relationship to foo should be in the style sheet.)
Big enough to fit what string? A container can fit any string with a proper overflow property.
Big enough to fit a string without overflowing, of course.
Take that hoary old example, the columned layout. On my website I have a columned layout made up of tables. I want the left-hand column to be as small as possible without wrapping the 'cowlark.com' logo in the top-left corner, and the right-hand column to occupy everything else.
I cannot find a way of doing this with CVS. The problem is working out how wide to make the left-hand column. With tables, the logo does not contain any spaces, and so the table layout algorithm will try quite hard not to wrap it; try resizing the font and you see it all still works (pretty much). With CSS, I have to hard-code the logo's width, and I can't tell what that is, because it depends on the fonts being used.
I have only recently discovered display:table-row and friends, and I could probably use those instead, but it's still basically tables. There should be a better way.
CSS has lots of limitations (some of which will be addressed in V3, some of which aren't)
<plaintively> Can we have decent support for CSS2, please? Leaving aside the aberration that is IE's attempt at CSS implementation, Gecko doesn't even do counters yet...
Basically, I find that the biggest problem with CSS is that it's like most of the other of the W3Cs standards: a really good idea implemented in a really half-assed way. (XML --- plus: namespaces, ASCII, structured data, extensibility. Minus: repetitive, bloaty, unseekable, attributes vs real nodes. XSLT --- plus: general purpose XML transformation language. Minus: it's written in XML; it's not a real programming language; it's stupidly limited in a lot of ways (variables that won't vary?). XPATH --- don't get me started. etc.)
I'll leave aside whether it's a good idea to have a dynamic web page made up of three *different* scripting systems in one document, and just go on to mention two of the things that bug me most:
Firstly, it does not seem to be possible (unless I just haven't found it yet, please feel free to correct me) to say that I want style FOO to be the same as style BAR except with these changes. i.e., true hierarchical styles. Any word processor worthy of the name supports this. It allows you to make sure that each important style property is defined in exactly one, central place.
Secondly, CSS' styling system is very, very limited. How do you say, I want this container to be big enough to fit this string into? You can't. This means that any kind of layout where you have sized objects with text in them --- such as columns, or a header --- has to be specified in fixed values. You can't say, I want a graphic followed by a line of text followed by something that fills all the rest of the space. (You've got ems and ens and points but frankly, they're not useful.)
I do think that CSS is currently the best thing around for formatting HTML; I've also used it with moderate success for formatting raw XML. It just makes me cringe when I think how good it could be, and how lousy it actually is... I keep having this urge to write my own formatting engine in Javascript, and that's never good.
The IDE controller isn't the master, the words master and slave are used to distinguish from up to two IDE devices
Sorry, I didn't mean that (it's too late, I've just played through the Darwinia demo [which rules] and I need caffeine). I mean that the interface has a master end (the computer) and a slave end (the drive). USB works the same way. Firewire and SCSI, however, are peer-to-peer in that there's the same interface on all devices, regardless of what they do.
(You can plug multiple computers in on the same SCSI bus; there are very-high-speed Ethernet implementations that do that. In fact, that may be a viable way of solving the original poster's problem.)
IDE master/slave drive select is a ghastly hack, as well as being confusing terminology.
Sounds like you need an IDE adaptor designed to work as a slave, not as a master: i.e., the same interface that IDE hard disks have. Unfortunately I have no idea if such a thing exists. Probably not.
However, you could do this with Firewire, because there isn't a master/slave distinction with Firewire. You could buy a cheap-and-nasty computer, put all your memory into it, hook it up to your server and use it as a (very big) solid-state hard disk. Firewire's about the same speed as IDE.
Unfortunately, your slave computer would need software to make it act like a hard disk, and while I'm sure such a thing exists, I don't know where...
Slashdot Mirror
Hell, yeah. Whenever I see the alternative I realise just how good the BBC is. Sky is dire by comparison, and US TV is just unwatchable.
One thing people tend not to realise is that because the BBC has its own guaranteed source of income, they can't be put under pressure by their sponsors. There's an old story of a car company whose latest product had just been slammed by Top Gear, the BBC's excellent motoring programme (although less excellent than it was. Sigh). And when Top Gear doesn't like something, they're not subtle about it... the story goes is that the CEO watched the review, said, "I'll teach them to talk like that about us. Pull all our advertising from that channel. Now." And his secretary said, "Um..."
You would not believe what a difference this makes. During the last Gulf War I watched some CNN and MSNBC. It was embarrassing. A lot of it was cultural differences, but the blatant jingoism and emotionalism made, to me, a complete mockery of the whole concept of independent journalism ---I found it hard to accept what I was watching as being anything but outright propaganda.
(Incidentally, the BBC is not government funded, and their charter clearly makes them independent from government interference. If the government tries to pressure them, most BBC journalists shout 'Hurrah!' and it tends to make the news.)
I don't like the heavy-handed way the license fee is collected --- they use scare tactics a lot. "This man didn't pay his license fee. Now he's bankrupt, his wife has left him, his kids are drug addicted hookers, and we shot his dog. Don't let this happen to you." They also have a lot of trouble believing that some people don't have TVs. If they'd be nicer about it, I'd be much happier paying for it.
There's quite a good writeup on the BBC's journalism here.
A topologist is someone who can't work out whether to dip his doughnut into his coffee mug, or vice versa...
Yeah. Just take a look at this...
Alas, it doesn't seem to be, uh, compatible with my hardware, but I can't help wondering --- does it work?
Standard DRAM will maintain its state --- mostly --- for a remarkably long time without refreshing. Unfortunately, it doesn't do so in a useful state.
I once was working on an embedded device that had VGA out. The development cycle was power on, boot from TFTP, run system, wait until it crashed, power off, repeat. When the system switched on, one of the first things the boot loader did was to initialise the video chipset, but without clearing the video memory.
If the board had been off for less than about five minutes, you could still see the last display that had been there when the board crashed.
Without refreshes, the data would gradually fade; the image was always corrupted with snow. The longer you left it switched off for, the worse the snow got. Different RAM chips lasted different lengths of time --- there was one band across the middle that would become completely unintelligable in about 30s, while another one could hold an image for about two minutes.
I suppose you could use this to store data for short periods during a power down, but you'd have to use so much redundancy to ensure that the data would survive the inevitable corruption that it probably wouldn't be worth it, but I'm sure someone, somewhere, could come up with a Nifty Trick(TM)... You couldn't do it at all on PCs, of course --- on boot, they wipe all their RAM, video or otherwise.
The caller's probably from the US, where they have this bizarre system were the recipient of a mobile phone call pays.
+ in a phone number usually indicates an international number (you're supposed to dial your international call prefix plus the number, you see) but there are no country codes beginning with 0. Assuming that, again, he's in the US and the + is in fact erroneous, 001819 would dial a number in Sherbooke, Quebec, Canada.
And that's really impressive. A decade from now we're probably going to see miniaturised versions of this being used as standard sewage treatment plants.
(Which will really piss off the village in Scotland where my parents live; they've just spent vast amounts of money building a sewage treatment plant only to discover that due to incompetence it's been located slightly below the high tide mark. Which means that every time there's a spring tide, unpleasant stuff happens.)
I wonder if it would be economically viable to funnel all the sewage produced by some random big city into one of these things, producing biodiesel and electricity that are fed back into the city?
Maybe you could use it to transmit data? If you hooked up, say, your average Bush voter to a blood-pressure machine and then had someone several miles away talk about how bad the war in Iraq was, I'm sure you'd see a spike on the graph. By using carefully timed conversations and statistical analysis, you should be able to get at least some bandwidth.
The problem would be isolating the signal from the noise. A passing planeload of French tourists would produce so much interference that you'd probably have to give up and resend. And if a Michael Moore film was shown in the same state, you'd probably burn out your surge protectors...
Any newsreader worth its salt should be able to find your own articles at the push of a button. This is something I like doing, too.
Secondly, Slashdot is already experimenting with an NNTP based feed ...
Really? Where? The FAQ entry still says its not going to happen, partly for advertising reasons...
The lecturer was, apparently, talking about the problems in writing mission-critical embedded devices, and at one point he asks his audience: "You all write embedded systems software. Tell me honestly; if your company wrote the software for a 747, how many of you would actually feel comfortable on board one?"
One hand goes up.
"You, sir! You're so confident in your software you'd trust your life to it?"
"Hell, no," comes the reply. "But any plane running my team's software would never crash, because it'd never get off the ground..."
I am confident in the level of safety given by running Windows on a mission-critical device.
It would --- but there would be other problems.
The first one is the most simple: silicon's expensive. Really expensive. The more units you can slice off that wafer the cheaper the units are. Making the die bigger simply for thermal reasons isn't going to wash with the chip manufacturers. They already glue the die to a metal backing plate, which gives you much the same effect anyway.
The second one, however, is the most crucial one. Electricity is slow. Electrical impulses travel at about 2/3 c through copper and a touch less through silicon (IIRC, I can't find the figures to check). This means that the bigger your die is, the longer it takes the impulses to travel from one side of it to the other.
A 1GHz clock fires every 10^-9 seconds; since the speed of light is 3x10^8 m/s, this means that the impulses are going to travel about twenty centimetres between clock pulses. For a 4GHz clock, it'll be about 5cm. There's a lot more wiring than that folded up inside the die; and it gets worse --- particular things happen at particular times throughout the clock cycle, and where you are in the clock cycle now depends on how long the wire is that connects you to the clock. Making sure everything happens in sync is a nightmare.
There are solutions to all of this; asynchronous designs which don't use clocks, offloading functionality to special-purpose processors like GPUs so you don't need as fast a main processor, radically different approaches like Cell, optical transports so you can route signals through each other, etc, but basically there are loads of good reasons why you need the die to be as small as humanly possible.
Cool; will it become the default?
(Incidentally, would it be particularly hard to implement all the low-level GNUstep widgets by peering them off GTK or Qt widgets instead? Seeing as they've already done the work, it would make sense to take the shortcut...)
Well, I didn't show that part, but that works exactly in Gorm like on InterfaceBuilder on OSX
Well, I haven't used InterfaceBuilder, so I'll take your word for it.
I'll download it all and have a play...
I do, however, have two minor criticisms.
Firstly, please, please update the look-and-feel. If you want to be taken seriously, don't look like a reject from the 80s. Given GNUsteps modularity, this should be easy enough to do. So, do it. (Tip: application icons should always have labels, because since they're supposed to be unique you can pretty much guarantee they're going to be unfamiliar to someone.)
Secondly, I didn't see any support for layout management in Gorm --- that application was constructed by just placing absolute-sized objects at absolute positions in a window. Please tell me this isn't how you design all applications... because that way leads to inflexible, unscalable, uncustomisable applications, and there's no excuse for that any more. Fixed layouts mean you can't let the user change fonts, because different fonts are different shapes (you can't just scale linearly). Fixed layouts mean wasted screen estate (remember the old Mac file browser dialogues that would float a tiny, eight-line scrollable list in the middle of a 21" monitor?). Fixed layouts are just wrong.
I'll remind you of this old story; which if you come to think of it, is quite an advertisement for Novell products...
Then again, maybe not... if it was free, you could argue that it's a public service to get stuck cars moving again as quickly as possible during rush hour: the amount of money that everyone else on the road is saving is far greater than you might get if you charged for the service; and that, IMO, is the mark of whether something should be a public service or not.
For example, if a truck suddenly pulls out in front of you, you will suddenly focus on it; your passenger will tend to notice this and stop talking. Someone on the other end of a phone won't.
Yay! Thanks.
(And people can't understand why these jump-through-a-hoop-to-read-the-article systems annoy people so much... half the time they don't work! I've simply given up trying to register the NYT, for example.)
FWIW, it sounds awfully similar to Frozen Bubble --- was FB based on Snood?
Actually, an em is the height of the font, from the lowest descender to the highest ascender. (It used to be the width of the letter 'M' in that font, hence the name --- you'll never guess what an en is.)
This means that 11em is going to be roughly the width of:
MMMMMMMMMMM
Which is much longer than my logo, in a proportional font:
cowlark.com
So, it still doesn't help, I'm afraid.
I didn't know about that; useful, although not quite what I'm looking for. Ta.
(The thing I object to here is that it requires the HTML content to have detailed knowledge of the mechanism of the markup, which is evil --- there's no need for the content to have to state that that particular div is a foo and a bar. It should just be able to state that it's a bar. The relationship to foo should be in the style sheet.)
Big enough to fit what string? A container can fit any string with a proper overflow property.
Big enough to fit a string without overflowing, of course.
Take that hoary old example, the columned layout. On my website I have a columned layout made up of tables. I want the left-hand column to be as small as possible without wrapping the 'cowlark.com' logo in the top-left corner, and the right-hand column to occupy everything else.
I cannot find a way of doing this with CVS. The problem is working out how wide to make the left-hand column. With tables, the logo does not contain any spaces, and so the table layout algorithm will try quite hard not to wrap it; try resizing the font and you see it all still works (pretty much). With CSS, I have to hard-code the logo's width, and I can't tell what that is, because it depends on the fonts being used.
I have only recently discovered display:table-row and friends, and I could probably use those instead, but it's still basically tables. There should be a better way.
CSS has lots of limitations (some of which will be addressed in V3, some of which aren't)
<plaintively> Can we have decent support for CSS2, please? Leaving aside the aberration that is IE's attempt at CSS implementation, Gecko doesn't even do counters yet...
I'll leave aside whether it's a good idea to have a dynamic web page made up of three *different* scripting systems in one document, and just go on to mention two of the things that bug me most:
Firstly, it does not seem to be possible (unless I just haven't found it yet, please feel free to correct me) to say that I want style FOO to be the same as style BAR except with these changes. i.e., true hierarchical styles. Any word processor worthy of the name supports this. It allows you to make sure that each important style property is defined in exactly one, central place.
Secondly, CSS' styling system is very, very limited. How do you say, I want this container to be big enough to fit this string into? You can't. This means that any kind of layout where you have sized objects with text in them --- such as columns, or a header --- has to be specified in fixed values. You can't say, I want a graphic followed by a line of text followed by something that fills all the rest of the space. (You've got ems and ens and points but frankly, they're not useful.)
I do think that CSS is currently the best thing around for formatting HTML; I've also used it with moderate success for formatting raw XML. It just makes me cringe when I think how good it could be, and how lousy it actually is... I keep having this urge to write my own formatting engine in Javascript, and that's never good.
Unfortunately, given the current state of our Art, there's no way we could keep the cup up.
I mean, come on. We have reality TV. Forget the shoe-shop event horizon, we're really screwed...
Now, now. The full saying to the shmup motto is, IIRC:
"If it moves, kill it. If it doesn't move, shoot it until it does. Then kill it."
Unfortunately I can't find a source --- anyone?
Sorry, I didn't mean that (it's too late, I've just played through the Darwinia demo [which rules] and I need caffeine). I mean that the interface has a master end (the computer) and a slave end (the drive). USB works the same way. Firewire and SCSI, however, are peer-to-peer in that there's the same interface on all devices, regardless of what they do.
(You can plug multiple computers in on the same SCSI bus; there are very-high-speed Ethernet implementations that do that. In fact, that may be a viable way of solving the original poster's problem.)
IDE master/slave drive select is a ghastly hack, as well as being confusing terminology.
However, you could do this with Firewire, because there isn't a master/slave distinction with Firewire. You could buy a cheap-and-nasty computer, put all your memory into it, hook it up to your server and use it as a (very big) solid-state hard disk. Firewire's about the same speed as IDE.
Unfortunately, your slave computer would need software to make it act like a hard disk, and while I'm sure such a thing exists, I don't know where...