Actually, it's not laborious. What he's doing, though he doesn't explain it, is building a style sheet. (Or perhaps "document template" as I think Word calls it now.) Once you've done that, you just tag a paragraph with the appropriate style (one click) and you're done.
I wish.
I'm afraid that while this does work for very simplistic documents, as soon as you start going anywhere near structured text it all breaks down. The problem is that there's no containment model on OpenOffice; you represent a section containing subsections as: section header, body text, subsection header, body text, subsection header, body text... In order to do anything useful with it programmatically, you need: section { header, body text, subsection { header, body text}, subsection { header, body text },...}
This means that as soon as you start getting non-trivial documents in OpenOffice, you lose the ability to usefully transform or manage the document via an external tool, because you don't have the structure information you need to do this. At a simpler level, it also means that any block of text can have at most one paragraph style and one character style --- you can't have a chunk of 'code' inside a bigger block of 'URL', for example.
I've been trying for years to find a WYSIWYG structured text editor to let me write, among other things, Docbook. Periodically I go back to OpenOffice and Abiword and KWord in the hope that they've got better, struggle for a bit, and then give up --- they're just inadequate for such things. Apps like Lyx and Texmacs are better, but they're really weird and have poor Docbook support. Currently what I'm using is the Vex plugin for Eclipse, which while being clunky and difficult to use, does at least provide a decent WYSIWYG view of my document. (No, I don't want to use a text editor.)
There's a big hole here; there seem to be very few applications that provide this kind of thing, I suspect because it's Really Hard. It doesn't mean I don't still want one, though... any suggestions?
Nope. He got up, and with in a week was running 10 miles or more a day. He lost essentially no bone density. Freaking fantastic shape he's in. All the astronaut core is like that. It's all about constant exercise and having impact exercise.
I remember there being some talk, not that long ago, that it would be better if astronauts weren't be so fit. The idea was that because you inevitably get out of shape in space --- partly due to the free fall, partly due to being in a very small metal box with little opportunity to exercise --- it's very bad for the astronauts morale. The idea is that if you send up people who are in decent shape but not ultra-fit and obsessed about staying in shape, you'll end up with a much happier crew without there being much overall change in performance. Also, the kind of people who are willing to exercise for hours every day tend to be strong-willed extroverts who probably aren't the best choice for being cooped up for months in said small metal box.
I don't know if anything came of it; it was a while ago, and I don't have any references...
...while UTF-8 is great even though it doesn't degrade gracefully for 8-bit characters? That's absurd. If all you care about is the 7-bit characters that UTF-8 supports, why not just use ASCII?
Because when you're parsing text, you're usually only interested in a few special characters --- control codes, spaces, etc. These are all in the ASCII range. These means that all the UTF-8 extended characters will just pass straight through, unchanged, correctly. You don't need to worry about them. Because UTF-8 consists of ASCII (with the top bit of each byte clear), and extended characters (with the top bit of each byte set), there's no chance of corrupting an extended character or misinterpreting part of an extended character as an ASCII character. It's astonishingly convenient to write code for.
The reader cares about the extended characters. You, the one writing the parsing code, usually don't.
But it has degraded gracefully --- you can read it, can't you? If they'd sent it in, say, UTF-16 instead it certainly wouldn't be readable; you'd have an incomprehensible block of Base64 encoded binary data.
Unfortunately, a lot of early Java class libraries did not implement it right.
In fact, Java --- and Windows --- got it so catastrophically wrong (using 16-bit values for characters, instead of 32-bit value) that it was found easier to change the Unicode specification to prohibit most characters that wouldn't fit in a 16-bit value!
There is a standard in place for encoding such things using 16-bit values; it's UTF-16, and given that it's a variable-length-character encoding like UTF-8, it rather defeats the whole purpose of using 16-bit characters in the first place. Most apps you'll meet just ignore it, and will break horribly when they come across a UTF-16 extended character.
Your best bet is simply to do everything in UTF-8. It degrades nicely into ASCII, which means that all your old friends like strcpy() and strtok() will Just Work in the vast majority of cases that you'd be interested in. Admittedly, to find the nth character in a string you've got to start at the beginning and read (and discard) n characters, but you'll be surprised how infrequently you need to do this.
...the idea was to build a big pipe that ran up and over the land, but was completely sealed. So it would basically act like a siphon...
I don't know much about the geography at Nicaragua, but syphons can only lift water ten metres --- is it flat enough to make that work? I have heard of plans to build turbines on the floor of, for example, the Straits of Gibraltar.
One intriguing power generation idea is to build big vertical pipes out at sea, to connect two differing levels of ocean. Slight differences in temperature and salinity cause a current flow through the pipe, which you can use to drive a turbine. You are, in effect, using the surface of the sea as a big solar collector. The environmental impact is minimal, they scale well, and if you're cunning you can actually use the warm/saline water to do something useful afterwards, such as running fish farms or other ocean farming...
Gas turbines are the quickest to bring online, taking only minutes to spool up, and are often used for peak load times (i.e. the afternoons of hot sunny days).
It's possible to bring up a gas turbine in seconds if you're prepared for it; you leave the turbine spinning but with no actual load.
There's also a special type of hydroelectric plant called a pumped storage power station. What you do is to connect two lakes at different levels via a set of turbines. When you have excess power on the grid, you pump water uphill; when you need power, you let it run downhill. They don't have a great deal of capacity, but you can bring them online from cold in only a slightly longer time than a hot gas turbine. The one I've visited, the Ben Cruachan power station, can generate 440MW for 22 hours and can come online in two minutes.
A while back I remembered seeing proposals for storing excess electricity during off-peak hours in huge supercooled superconducting storage rings, but I haven't heard any more about it in years, and don't even know how such a scheme would work.
The problem with superconducting storage rings is that if anything goes wrong all the energy gets liberated as heat... very, very suddenly. If you had a storage ring the size of the pumped storage station described above, you'd end up dissapating 6x10^11 joules of energy... about the equivalent of 150 kilotonnes. Yum!
2025 Neurological research finally leads to an understanding of all the senses, and direct inputs become possible, bypassing eyes, ears, skin, etc. The inevitable result is the metal "Braincap" of which the 20th century's Walkman was a primitive precursor. Anyone wearing this helmet, fitting tightly over the skull, can enter a whole universe of experience real or imaginary - and even merge in real-time with other minds. Apart from its use for entertainment and vicarious adventure, the Braincap is a boon to doctors, who can now experience their patients symptoms (suitably attenuated). It also revolutionizes the legal profession; deliberate lying is impossible. As the Braincap can only function properly on a completely bald head, wig-making becomes a major industry.
Two of them have composite inputs, one only has cable RF.
S-Video to composite connectors cost about 10p each in bulk --- they consist of two or three resistors wired together. I'll possibly agree with you that the RF-only TV, but if you really needed to get it working, video modulators are only a little more expensive --- remember all those 8-bit micros that had them?
Recently -- like within the last six months or so, I've noticed an alarming number of domains that aren't receiving my emails. And no, I haven't been blackholed or otherwise put on anyone's shit list, nor am I running an open relay. The mailserver is perfectly well-behaved, standards compliant, and only relays from within my home LAN.
It's also, I'm afraid, going to be automatically blackholed --- as you're finding out --- because you're inside a slum IP block. Nobody trusts mail sent from a residential IP range any more. The solution's easy, and is to use your ISP's own mail server as a smarthost; you can still receive mail --- assuming your ISP doesn't block incoming port 25 connections --- but if send directly, nobody will listen.
If you talk about 'free will' you are, in essence, talking about souls --- because there's an implicit assumption when talking about free will that something is making the choice to do a particular action other than the actual hardware. You're right, it's not a term that's particularly useful outside the context of discussions of animism vs. materialism.
You need to look more closely at how rates are ratios of even unlike units.
Rates are not ratios. They're rates, a fundamentally different thing. Rates have a specific unit; metres per second, joules per kelvin, etc. Ratios don't; they're unitless, like logarithms.
But you made the claim that "when you give a ratio as, say, 10:1, this means ten parts out of eleven: you add both sides to give the total number of parts". That's what I'm responding to.
Yes, you do.
You're right, it's not a useful concept when dealing with things such as mass ratios, which is why mass ratios are usually expressed as a fraction and not using ratio notation. But that doesn't escape the fact that ratios must deal with two values of the same type.
How do you explain a P/E ratio then? The sum "price + earnings" is meaningless. Same for the mass ratio of a rocket: the sum "dry mass + wet mass" is equally meaningless.
price + earnings: both values are monetary quantities.
When people use the "/" symbol, it's hard to distinguish from its "divided by" meaning. Especially when the ratio is being used in arithmetic, as it was in the conversion I used, the ":" symbol is more clear, less ambiguous.
When people describe a unit as, say, m/s (metres per second), that does mean division. One metre, divided by one second. That's the definition.
And I'm afraid that using : is simply incorrect, because units such as m/s are not ratios. Ratios only make sense when the numbers on both sides can be added. You can't add metres and seconds --- they're not the same type and the result makes no sense. The reason for this is that when you give a ratio as, say, 10:1, this means ten parts out of eleven: you add both sides to give the total number of parts. (One reason why ratio notation is generally not used in science. They're confusing. Fractions or decimal notation are usually preferred.)
for lightness and performance. At least as far as scripting languages go.
*nods*
I never use anything else any more --- it's small (compiling into an 200kB binary with no dependencies on my platform), it's fast (faster than Python!), it's simple (you can easily understand the entire language), it's elegant (closures, coroutines, a superb callout interface...), and it's flexible (there's enough functionality under the surface that you can, e.g., rewrite the OO system to better suit your needs). It's also BSD licensed, which means that there are no legal hurdles to using it in your project; if you play games, you've probably already used Lua without realising it.
I will admit to not being overly enamoured with its syntax --- it uses Pascalish if...then...end style rather than C's if () {} style --- but I can easily live with that.
Testimonial: I wrote a gaim plugin not long ago for the Citadel BBS. It was easier to bolt the Lua engine onto gaim and write the logic in gaim rather than try and figure out how to do it in gaim. Lua's coroutines support allowed me to turn gaim's callback-based API into a callout-based structure, which in turn allowed me to invert all my nasty complex state machines, which made the whole thing an order of magnitude less complex. Good stuff.
Someone has probably already said this but you can flash the ROM on your DVD player and skip those inane advertisements. It'll also unlock the region encoding and you can play pirated movies from Bangkok or some crap like that, but I've never been interested in that.
My cheapo no-name far eastern piece of imported junk does all this out of the box --- for most disks, loading the disk and pressing STOP STOP PLAY will cause it to immediately start playing from the beginning of track 1, bypassing all the unskippable crap, menus, etc. It's amazingly convenient.
It's also trivial to change or disable the region encoding, and it'll play a wide variety of different disk types, including my own badly-encoded out-of-spec SVCDs. I haven't yet found a way of disabling Macrovision, but I haven't really needed to yet.
Remember, the Asian consumer electronics manufacturers know where the money is.
Indisputably if you ignore the Motorola 6800 and the MOS 6502 and the Z80. Even if you did ignore them, you still wouldn't end up with 100% microprocessor market share for the 8080.
Dude, he wasn't talking about the 8080. He was talking about the 4004. And trust me, that processor really did have 100% market share.
What's the bet a few light globe manufacturers will get together, buy the rights and then put it away in the archives?
Because if they could do this, they'd have already done it for fluorescent tubes, which can be up to about 60% efficient (compared to 10% for incandescent bulbs)?
Slashdot Mirror
I wish.
I'm afraid that while this does work for very simplistic documents, as soon as you start going anywhere near structured text it all breaks down. The problem is that there's no containment model on OpenOffice; you represent a section containing subsections as: section header, body text, subsection header, body text, subsection header, body text... In order to do anything useful with it programmatically, you need: section { header, body text, subsection { header, body text}, subsection { header, body text }, ...}
This means that as soon as you start getting non-trivial documents in OpenOffice, you lose the ability to usefully transform or manage the document via an external tool, because you don't have the structure information you need to do this. At a simpler level, it also means that any block of text can have at most one paragraph style and one character style --- you can't have a chunk of 'code' inside a bigger block of 'URL', for example.
I've been trying for years to find a WYSIWYG structured text editor to let me write, among other things, Docbook. Periodically I go back to OpenOffice and Abiword and KWord in the hope that they've got better, struggle for a bit, and then give up --- they're just inadequate for such things. Apps like Lyx and Texmacs are better, but they're really weird and have poor Docbook support. Currently what I'm using is the Vex plugin for Eclipse, which while being clunky and difficult to use, does at least provide a decent WYSIWYG view of my document. (No, I don't want to use a text editor.)
There's a big hole here; there seem to be very few applications that provide this kind of thing, I suspect because it's Really Hard. It doesn't mean I don't still want one, though... any suggestions?
I remember there being some talk, not that long ago, that it would be better if astronauts weren't be so fit. The idea was that because you inevitably get out of shape in space --- partly due to the free fall, partly due to being in a very small metal box with little opportunity to exercise --- it's very bad for the astronauts morale. The idea is that if you send up people who are in decent shape but not ultra-fit and obsessed about staying in shape, you'll end up with a much happier crew without there being much overall change in performance. Also, the kind of people who are willing to exercise for hours every day tend to be strong-willed extroverts who probably aren't the best choice for being cooped up for months in said small metal box.
I don't know if anything came of it; it was a while ago, and I don't have any references...
Because when you're parsing text, you're usually only interested in a few special characters --- control codes, spaces, etc. These are all in the ASCII range. These means that all the UTF-8 extended characters will just pass straight through, unchanged, correctly. You don't need to worry about them. Because UTF-8 consists of ASCII (with the top bit of each byte clear), and extended characters (with the top bit of each byte set), there's no chance of corrupting an extended character or misinterpreting part of an extended character as an ASCII character. It's astonishingly convenient to write code for.
The reader cares about the extended characters. You, the one writing the parsing code, usually don't.
But it has degraded gracefully --- you can read it, can't you? If they'd sent it in, say, UTF-16 instead it certainly wouldn't be readable; you'd have an incomprehensible block of Base64 encoded binary data.
In fact, Java --- and Windows --- got it so catastrophically wrong (using 16-bit values for characters, instead of 32-bit value) that it was found easier to change the Unicode specification to prohibit most characters that wouldn't fit in a 16-bit value!
There is a standard in place for encoding such things using 16-bit values; it's UTF-16, and given that it's a variable-length-character encoding like UTF-8, it rather defeats the whole purpose of using 16-bit characters in the first place. Most apps you'll meet just ignore it, and will break horribly when they come across a UTF-16 extended character.
Your best bet is simply to do everything in UTF-8. It degrades nicely into ASCII, which means that all your old friends like strcpy() and strtok() will Just Work in the vast majority of cases that you'd be interested in. Admittedly, to find the nth character in a string you've got to start at the beginning and read (and discard) n characters, but you'll be surprised how infrequently you need to do this.
I don't know much about the geography at Nicaragua, but syphons can only lift water ten metres --- is it flat enough to make that work? I have heard of plans to build turbines on the floor of, for example, the Straits of Gibraltar.
One intriguing power generation idea is to build big vertical pipes out at sea, to connect two differing levels of ocean. Slight differences in temperature and salinity cause a current flow through the pipe, which you can use to drive a turbine. You are, in effect, using the surface of the sea as a big solar collector. The environmental impact is minimal, they scale well, and if you're cunning you can actually use the warm/saline water to do something useful afterwards, such as running fish farms or other ocean farming...
It's possible to bring up a gas turbine in seconds if you're prepared for it; you leave the turbine spinning but with no actual load.
There's also a special type of hydroelectric plant called a pumped storage power station. What you do is to connect two lakes at different levels via a set of turbines. When you have excess power on the grid, you pump water uphill; when you need power, you let it run downhill. They don't have a great deal of capacity, but you can bring them online from cold in only a slightly longer time than a hot gas turbine. The one I've visited, the Ben Cruachan power station, can generate 440MW for 22 hours and can come online in two minutes.
A while back I remembered seeing proposals for storing excess electricity during off-peak hours in huge supercooled superconducting storage rings, but I haven't heard any more about it in years, and don't even know how such a scheme would work.
The problem with superconducting storage rings is that if anything goes wrong all the energy gets liberated as heat... very, very suddenly. If you had a storage ring the size of the pumped storage station described above, you'd end up dissapating 6x10^11 joules of energy... about the equivalent of 150 kilotonnes. Yum!
From Arthur C Clarke's future history:
I can visualise the Penny Arcade strip already...
http://www.penny-arcade.com/comic/2005/05/27
S-Video to composite connectors cost about 10p each in bulk --- they consist of two or three resistors wired together. I'll possibly agree with you that the RF-only TV, but if you really needed to get it working, video modulators are only a little more expensive --- remember all those 8-bit micros that had them?
People who already have a TV?
Are all Linux distros based on the GNU userspace? Are there any based on, for example, the BSD userspace instead?
It's also, I'm afraid, going to be automatically blackholed --- as you're finding out --- because you're inside a slum IP block. Nobody trusts mail sent from a residential IP range any more. The solution's easy, and is to use your ISP's own mail server as a smarthost; you can still receive mail --- assuming your ISP doesn't block incoming port 25 connections --- but if send directly, nobody will listen.
If you talk about 'free will' you are, in essence, talking about souls --- because there's an implicit assumption when talking about free will that something is making the choice to do a particular action other than the actual hardware. You're right, it's not a term that's particularly useful outside the context of discussions of animism vs. materialism.
NEVER!
Uninformed and Inaccurate Alarmism, by Michael Crichton?
Rates are not ratios. They're rates, a fundamentally different thing. Rates have a specific unit; metres per second, joules per kelvin, etc. Ratios don't; they're unitless, like logarithms.
Yes, you do.
You're right, it's not a useful concept when dealing with things such as mass ratios, which is why mass ratios are usually expressed as a fraction and not using ratio notation. But that doesn't escape the fact that ratios must deal with two values of the same type.
price + earnings: both values are monetary quantities.
dry mass + wet mass: both values are masses.
When people describe a unit as, say, m/s (metres per second), that does mean division. One metre, divided by one second. That's the definition.
And I'm afraid that using : is simply incorrect, because units such as m/s are not ratios. Ratios only make sense when the numbers on both sides can be added. You can't add metres and seconds --- they're not the same type and the result makes no sense. The reason for this is that when you give a ratio as, say, 10:1, this means ten parts out of eleven: you add both sides to give the total number of parts. (One reason why ratio notation is generally not used in science. They're confusing. Fractions or decimal notation are usually preferred.)
1:10 == 1/11 == 0.090909...
1:9 == 1/10 == 0.1 (exactly)
*nods*
I never use anything else any more --- it's small (compiling into an 200kB binary with no dependencies on my platform), it's fast (faster than Python!), it's simple (you can easily understand the entire language), it's elegant (closures, coroutines, a superb callout interface...), and it's flexible (there's enough functionality under the surface that you can, e.g., rewrite the OO system to better suit your needs). It's also BSD licensed, which means that there are no legal hurdles to using it in your project; if you play games, you've probably already used Lua without realising it.
I will admit to not being overly enamoured with its syntax --- it uses Pascalish if...then...end style rather than C's if () {} style --- but I can easily live with that.
Testimonial: I wrote a gaim plugin not long ago for the Citadel BBS. It was easier to bolt the Lua engine onto gaim and write the logic in gaim rather than try and figure out how to do it in gaim. Lua's coroutines support allowed me to turn gaim's callback-based API into a callout-based structure, which in turn allowed me to invert all my nasty complex state machines, which made the whole thing an order of magnitude less complex. Good stuff.
My cheapo no-name far eastern piece of imported junk does all this out of the box --- for most disks, loading the disk and pressing STOP STOP PLAY will cause it to immediately start playing from the beginning of track 1, bypassing all the unskippable crap, menus, etc. It's amazingly convenient.
It's also trivial to change or disable the region encoding, and it'll play a wide variety of different disk types, including my own badly-encoded out-of-spec SVCDs. I haven't yet found a way of disabling Macrovision, but I haven't really needed to yet.
Remember, the Asian consumer electronics manufacturers know where the money is.
Dude, he wasn't talking about the 8080. He was talking about the 4004. And trust me, that processor really did have 100% market share.
Because if they could do this, they'd have already done it for fluorescent tubes, which can be up to about 60% efficient (compared to 10% for incandescent bulbs)?
$936 < $800?
I really, really hope that was a typo.
Unless you happen to be female...