And if I scuff my feet while walking across the room, I can generate TWENTY! THOUSAND! VOLTS! OF! ENERGY! Someone hook me up to the power grid!
Slight correction: the Culture isn't "far future"
on
Matter
·
· Score: 2, Informative
The description of the Culture as "far future" isn't correct - it's intended to be roughly contemporary with the present time, as evidenced by "State of the Art" and some of the timelines given in Consider Phlebas. "Technologically advanced" is a more accurate description.
This incident of nitpickery has been brought to you by the letters "E" and "Schwa" and the number needle-nardle-noo.
Actually, it's not hard to improve on Huffman coding. Arithmetic coding has been around for a long time, and can do much better on most kinds of data. In fact, I'd suspect that this is how they're doing this trick: undo the Huffman coding and redo it with arithmetic coding. In fact, it's possible (though unlikely) that all they're doing is recoding a Huffman-coded JPEG file to an arithmetic-coded JPEG file - that's right, the original JPEG standard has variants that use arithmetic coding. You'll never see these in the wild because they're not part of "baseline JPEG" which is what everyone uses. Arithmetic coding is part of the standard, but nobody uses it. Arithmetic coding takes more CPU than Huffman coding, and at the time JPEG came out even Huffman-coded JPEG files took a long time to encode/decode. Also, the patent situation for arithmetic coding is a lot muddier than for Huffman coding.
This page has a lot of information on arithmetic coding. Very briefly, it's a way of using fractional bits to encode symbols - Huffman coding encodes each symbol to some integral number of bits (more bits for infrequent symbols, fewer bits for common symbols); arithmetic coding does the same, but each symbol can map to a non-integral number of bits; you save a fraction of a bit per symbol, which can really add up. It's not easy at first to see how this works, but the math works out.
Arithmetic coding has another advantage over Huffman coding: with Huffman coding, you first collect symbol frequency information, then you build your coding table based on that frequency information. You then have to somehow encode the coding table in your bitstream (so the decoder knows how to decode the symbols). The coding table is based on an average over (often) the whole file, so it can't adapt to changes in the symbol frequencies: if one part of the file contains just numbers, and the other part contains just letters, their statistics get mixed together so you end up with a Huffman coding table that's not optimal for either part. Adaptive Huffman coding (changing the codewords as you encode the file, based on changing statistics) is possible but painful. On the other hand, adaptive arithmetic coding is very very easy.
Well, the power that's being extracted by fission of uranium is power that was put into the uranium atom when it was formed by fusion, so in that sense it's fusion-powered (powered by energy stored, rather than released, in a past fusion). True, that energy itself came from gravitational collapse, so I guess we're back to the Big Bang again.
And radioactive decay (say, of uranium) is really fusion-powered. The uranium (in fact, all the elements past iron) was formed by fusion in some long-ago supernova. The heat of formation... I guess that's Big Bang-powered.
That reminds me - it's about time to replace all my bookshelves. All the ones I've got now have just about run out of energy, and you just can't get good bookshelf rechargers these days.
Hint: energy = force * distance. Holding something stationary in a force field (like a gravity or magnetic field) takes zero energy. "Providing sufficient energy to counteract the gravitational force" is nonsense.
I built a small Mini-ITX server that should draw no more than about 30W. It's not the fastest machine on the block (600MHz Via CPU; hell, it's not even the fastest machine on the desk) but it's very quiet and has more than enough CPU power for what it has to do. I wrote about it a while ago; see http://slashdot.org/comments.pl?sid=83510&cid=7305 504
for useful info like the parts list.
"The" is an article, not a preposition. (Those would be words like "on", "to", "with", etc.)
D'oh! Blame the new baby and consequent lack of sleep.
Sorting "a" with diaresis (umlaut if you prefer) along with "a" is not correct for Swedish, but the rules differ for each particular language. German is not the same as Swedish, for example; as for Finnish, I don't know. But for English, sorting it as an "a" is correct, since accents are, to monolingual anglophones, essentially invisible, or just decoration at best.
Right - it would have to be locale-specific, as would the article-stripping. It's all just Perl, so anything is possible. Still, given the current message on slimdevices.com: "2003-12-03 Squeezebox is currently shipping in volume. However, due to extraordinary demand, orders placed today will ship in ten days, in the order that they were received.", it sounds like they have their hands full with just the US market right now.
I got my Squeezebox last night and played with it for a while. Setting up the server software (on a Fedora Core 1 machine) was as simple as installing an RPM. All I had to do then was to plug in the Squeezebox, tell it my ESSID and WEP key, and tell the server where to find the music. The server then spent several minutes scanning the music directory (80GB, over 10K tracks, and yes, they're all ripped from CDs I own). Once that was done, the Squeezebox found the server and was ready to go.
I had been worried that it would be too painful to find music from as large a collection as mine using only the remote control's numeric keypad, and that I'd end up having to use the Web interface to control playlists, which is less convenient. However, the browsing and searching functionality built into the Squeezebox worked much better than I'd expected. Browsing by artist is quick - you zoom down to the right section of the alphabet with a few keypresses on the remote's numeric pad (e.g., press "7" once for P, twice for Q, and so on), then use the up/down buttons to scroll to the right artist. You can then browse the list of albums or tracks. You can also do a search for keywords in the title.
One feature that surprised me, but that I quite like having seen it, is that the indexing software ignores prepositions in artist names. So the list of artists starting with "S" went something like "Sarah McLachlan", "The Seekers", "Severe Tire Damage",..., "Sting", "The Strawbs", "Sunday's Well". It also handled accented characters without a glitch: "äaut" was treated the same as "a" in terms of sorting and searching. Neither of these features is really appropriate outside the English-speaking world (in Swedish, "äaut" shouldn't sort with "a", but at the end of the alphabet), but they work great for me. (Yes, those should be real a-with-umlaut characters but Slashcode seems to strip them out if I enter them properly. Sigh.)
The display is bright and easy to read, and if you're too far away, one button press on the remote switches it to double-size characters, which can be read from across the room.
$300 is a little expensive for a toy like this, but it's going to make a huge difference to the way that we listen to music at home. We'd already got a dedicated 24/7 home file server holding the music collection, and the Squeezebox is the perfect complement to that. We're already discussing whether to get a second one for another room.
Just this morning I put in an order for the parts required to make a new server for our home network. The principal requirement was that it be low-wattage: living in California (home of the gouging power companies), I didn't want to leave a 100+ watt machine turned on all the time.
After reading a lot of info about the various mini-ITX boards, cases, and so on, I settled on this configuration:
VIA EPIA ME6000 fanless mini-ITX motherboard (has audio, 10/100 ethernet, USB 2.0, 1394)
Morex 2699 mini-ITX case
512M PC2100 DDR memory
120GB disk
Slimline CD-ROM
The total was less than $500, and I could have reduced it some more if I'd been willing to place orders with 3 suppliers, rather than getting everything from one place (logicsupply.com).
While this machine is underpowered for a lot of computing tasks, and is a joke for playing games on, it should do just fabulously as a SMB/NFS file server, web server for pictures of the new baby, and so on. I'm downloading the Fedora beta (Severn) as we speak.
The total power draw for this machine ought to be about 30W. Even at inflated California prices, that's less than $5/month to run. Plus, since the motherboard and case are both fanless, it should run very very quietly, and should be small enough to just tuck away on a shelf somewhere.
Now I get to wait anxiously and see if my expectations match reality.
They used to use a 2.1.24 (plus mods) kernel, so they should fall outside SCO's demands. I don't know if the latest TiVos are using 2.4-based kernels, but I'd be surprised if they are.
I'd like to see a diff between the Final Decree and the DOJ's proposed settlement - it looks like the decree doesn't go much beyond what the settlement said, but that's based on a very quick read of the decree and a vague memory of the settlement. Hope dies hard, but it dies nonetheless.
Many replies are along the lines of "Even if you don't agree to the EULA, you're still bound by it if you use the software" or "It's illegal to use the software if you haven't agreed to the EULA". This is flat-out wrong.
If I buy a book, then I have bought a copy of some information, embedded in a physical medium.
If I buy a CD, then I have bought a copy of some information, embedded in a physical medium.
If I buy a piece of software then I have bought a copy of some information, embedded in a physical medium. In the first two cases, I was clearly an owner of a copy of the information; it's pretty hard to argue that in this case, I'm not an owner of a copy of the information.
As the owner of a copy of some information, I have certain rights granted by law that are explicitly NOT infringing of the copyright of the one that created that information. For software, these rights EXPLICITLY include the right to copy it to a hard drive, to copy it into memory for the purposes of running the software, and to make backups.
So: after I walk out of the store carrying my box, I am at that moment entitled to copy the contents of that CD to a hard drive, and to execute the software. If I get home and do so, using my own home-grown installer that copies the bits off the CD, then I have a copy of the software, installed, which I have the right to run. At that point, I am fully entitled to run the software, and I have not agreed to any EULA, nor have I violated any law - I have used only the rights explicitly granted to me by the copyright law. It would be absure to think that I am now bound by the EULA.
This article describes a method for installing software that's slightly more convenient than manually unpacking it, creating registry keys, and so on, but has the same effect: the software is installed on your hard drive and you have not agreed to the EULA.
This of course only applies to cases where you bought a copy of the software (as in, you exchanged money for a physical copy of the software, with no other terms imposed). Something like Oracle isn't sold like this: you get a copy of Oracle after signing a contract with Oracle, and that contract includes many of the normal EULA terms. That's a completely different situation: you didn't get the copy of the software until AFTER you agreed to some limitations on your use of that software; you are bound by those limitations. Software downloads are similar: you're often required to agree to the EULA before the download begins. It's the case where you got your copy of the software BEFORE agreeing to the limitations that's the more interesting (and much more common) one - and in that one, you're entitled to install and use the software without agreeing to the EULA.
It boils down to this: The law that the software developers are attempting to use to make EULAs binding is copyright law: the unspoken claim is that it is illegal to make a copy of the software without agreeing to the EULA. This is just not true.
I don't think there'a ny need for AT&T to buy out and close down PARC - Xerox seems to be doing a good enough job of that. They've been trying to sell it to venture capitalists for a while, with a notable lack of success. I don't think that PARC will last another 12 months, which is very sad.
Xerox also has (had?) a research lab in Cambridge, colloquially known as EuroPARC. I visited there a few times and saw some quite neat stuff.
Bob Shaw. Light of Other Days. It's most commonly seen as a short story, but also appeared as the centrepiece in a novel called (big surprise) "Slow Glass".
Our cats like to filet their minions (i.e., us) as well. Those back claws were designed to eviscerate, after all.
My favourite NetHack message
on
Nethack 3.4.0
·
· Score: 1
"The bite-covered troll rises from the dead."
Trolls regenerate, even after you kill them. You can kill a troll, turn away, and it will come back to life ans start beating on you. You have to destroy the body completely. There are a number of ways to do this (of course) but the simplest is to eat it. Eating it takes time, and sometimes the troll regenerates first...
1)Have you ever seen another computer company do stuff like that? I have not.
Many years ago, I bought one of the first Amigas available. It came with AmigaDOS 0.9. This was, let's just say, rough... the hardware was clearly ready before the software was. The first Amigas were supposed to have the OS in ROM, but it was so unready that they added 256K of write-protectable RAM to the machine. This was a LOT at the time (the system RAM was also 256K), and it must have been quite expensive for them to do.
A couple of months after buying the machine, I received a free copy of AmigaDOS 1.0 - as I recall it was mailed directly to me. So this kind of behaviour isn't unprecedented.
While the first lines appear to be genuine Nostradamus, not something put
together today (remember the "village idiot will lead" fake Nostradamus
from the last US election?), the last line ("The third big war") was almost certainly written
today. Nostradamus' verses were quatrains, and that line makes
five; furthermore no poet or
translator would repeat "big" like that.
See here for commentary on
Nostradamus that happens to use this exact quatrain as an example of a
prophecy that's vague enough that a huge number of possible events could
seem to fit it. Of the four specifics ("City of God", "two
brothers", "fortress endures", "great leader will succumb") only two are
actually even close ("City of God" for New York, or maybe Washington or
maybe both - seems like a very poor fit; no leaders have succumbed).
This prophecy would apply just as well (in fact, better) to something like
political/ecclesiastical family infighting in Renaissance Rome: City of God
(seat of the Church), two brothers (family conflict, and remember that the
Popehood was often controlled by the powerful political families), fortress
endures (easy enough to come up with interpretations like "the glory of the
Church endures"), great leader will succumb (the struggle causes the fall
of one of the rulers).
If I remember correctly, the letter that the MPAA sends to the ISP is required by the DMCA to contain language like "Under penalty of perjury, we declare that these facts are true". This goes along with the assertions in the letter that IP A.B.C.D uploaded a file and that file contained material copyrighted by an MPAA member, so the ISP should suspend A.B.C.D.
If they're abusing the power granted to them by the MPAA by sending out such letters without performing competent investigations, then whoever signed that letter should by hauled into court and charged with perjury. The problem is that in cases like this, it's the word of insignificant nobodies against the word of a powerful entity - no prosecutor will ever bother to take such a case. If, on the other hand, anyone ever send such a letter alleging a DMCA violation on the part of the MPAA, you can bet they'd be in court so fast it would make their head spin.
The minimal safeguards that the DMCA puts on the power to accuse and convict merely by sending a letter (by requiring it to be sent "under penalty of perjury") are in practice worthless.
A few years ago I was attending a JPEG meeting where the Emmy that JPEG had just won was being proudly shown around. It's a recognition of contributions to the hard technical work of TV production.
Slashdot Mirror
And if I scuff my feet while walking across the room, I can generate TWENTY! THOUSAND! VOLTS! OF! ENERGY! Someone hook me up to the power grid!
The description of the Culture as "far future" isn't correct - it's intended to be roughly contemporary with the present time, as evidenced by "State of the Art" and some of the timelines given in Consider Phlebas. "Technologically advanced" is a more accurate description.
This incident of nitpickery has been brought to you by the letters "E" and "Schwa" and the number needle-nardle-noo.
This page has a lot of information on arithmetic coding. Very briefly, it's a way of using fractional bits to encode symbols - Huffman coding encodes each symbol to some integral number of bits (more bits for infrequent symbols, fewer bits for common symbols); arithmetic coding does the same, but each symbol can map to a non-integral number of bits; you save a fraction of a bit per symbol, which can really add up. It's not easy at first to see how this works, but the math works out.
Arithmetic coding has another advantage over Huffman coding: with Huffman coding, you first collect symbol frequency information, then you build your coding table based on that frequency information. You then have to somehow encode the coding table in your bitstream (so the decoder knows how to decode the symbols). The coding table is based on an average over (often) the whole file, so it can't adapt to changes in the symbol frequencies: if one part of the file contains just numbers, and the other part contains just letters, their statistics get mixed together so you end up with a Huffman coding table that's not optimal for either part. Adaptive Huffman coding (changing the codewords as you encode the file, based on changing statistics) is possible but painful. On the other hand, adaptive arithmetic coding is very very easy.
Well, the power that's being extracted by fission of uranium is power that was put into the uranium atom when it was formed by fusion, so in that sense it's fusion-powered (powered by energy stored, rather than released, in a past fusion). True, that energy itself came from gravitational collapse, so I guess we're back to the Big Bang again.
And radioactive decay (say, of uranium) is really fusion-powered. The uranium (in fact, all the elements past iron) was formed by fusion in some long-ago supernova. The heat of formation... I guess that's Big Bang-powered.
That reminds me - it's about time to replace all my bookshelves. All the ones I've got now have just about run out of energy, and you just can't get good bookshelf rechargers these days.
Hint: energy = force * distance. Holding something stationary in a force field (like a gravity or magnetic field) takes zero energy. "Providing sufficient energy to counteract the gravitational force" is nonsense.
I built a small Mini-ITX server that should draw no more than about 30W. It's not the fastest machine on the block (600MHz Via CPU; hell, it's not even the fastest machine on the desk) but it's very quiet and has more than enough CPU power for what it has to do. I wrote about it a while ago; see http://slashdot.org/comments.pl?sid=83510&cid=7305 504
for useful info like the parts list.
I had been worried that it would be too painful to find music from as large a collection as mine using only the remote control's numeric keypad, and that I'd end up having to use the Web interface to control playlists, which is less convenient. However, the browsing and searching functionality built into the Squeezebox worked much better than I'd expected. Browsing by artist is quick - you zoom down to the right section of the alphabet with a few keypresses on the remote's numeric pad (e.g., press "7" once for P, twice for Q, and so on), then use the up/down buttons to scroll to the right artist. You can then browse the list of albums or tracks. You can also do a search for keywords in the title.
One feature that surprised me, but that I quite like having seen it, is that the indexing software ignores prepositions in artist names. So the list of artists starting with "S" went something like "Sarah McLachlan", "The Seekers", "Severe Tire Damage", ..., "Sting", "The Strawbs", "Sunday's Well". It also handled accented characters without a glitch: "äaut" was treated the same as "a" in terms of sorting and searching. Neither of these features is really appropriate outside the English-speaking world (in Swedish, "äaut" shouldn't sort with "a", but at the end of the alphabet), but they work great for me. (Yes, those should be real a-with-umlaut characters but Slashcode seems to strip them out if I enter them properly. Sigh.)
The display is bright and easy to read, and if you're too far away, one button press on the remote switches it to double-size characters, which can be read from across the room.
$300 is a little expensive for a toy like this, but it's going to make a huge difference to the way that we listen to music at home. We'd already got a dedicated 24/7 home file server holding the music collection, and the Squeezebox is the perfect complement to that. We're already discussing whether to get a second one for another room.
After reading a lot of info about the various mini-ITX boards, cases, and so on, I settled on this configuration:
The total was less than $500, and I could have reduced it some more if I'd been willing to place orders with 3 suppliers, rather than getting everything from one place (logicsupply.com).
While this machine is underpowered for a lot of computing tasks, and is a joke for playing games on, it should do just fabulously as a SMB/NFS file server, web server for pictures of the new baby, and so on. I'm downloading the Fedora beta (Severn) as we speak.
The total power draw for this machine ought to be about 30W. Even at inflated California prices, that's less than $5/month to run. Plus, since the motherboard and case are both fanless, it should run very very quietly, and should be small enough to just tuck away on a shelf somewhere.
Now I get to wait anxiously and see if my expectations match reality.
They used to use a 2.1.24 (plus mods) kernel, so they should fall outside SCO's demands. I don't know if the latest TiVos are using 2.4-based kernels, but I'd be surprised if they are.
I'd like to see a diff between the Final Decree and the DOJ's proposed settlement - it looks like the decree doesn't go much beyond what the settlement said, but that's based on a very quick read of the decree and a vague memory of the settlement. Hope dies hard, but it dies nonetheless.
The IBM 1602 was known as the "CADET". This stood for "Can't Add, Doesn't Even Try". No, really.
Many replies are along the lines of "Even if you don't agree to the EULA, you're still bound by it if you use the software" or "It's illegal to use the software if you haven't agreed to the EULA". This is flat-out wrong.
If I buy a book, then I have bought a copy of some information, embedded in a physical medium.
If I buy a CD, then I have bought a copy of some information, embedded in a physical medium.
If I buy a piece of software then I have bought a copy of some information, embedded in a physical medium. In the first two cases, I was clearly an owner of a copy of the information; it's pretty hard to argue that in this case, I'm not an owner of a copy of the information.
As the owner of a copy of some information, I have certain rights granted by law that are explicitly NOT infringing of the copyright of the one that created that information. For software, these rights EXPLICITLY include the right to copy it to a hard drive, to copy it into memory for the purposes of running the software, and to make backups.
So: after I walk out of the store carrying my box, I am at that moment entitled to copy the contents of that CD to a hard drive, and to execute the software. If I get home and do so, using my own home-grown installer that copies the bits off the CD, then I have a copy of the software, installed, which I have the right to run. At that point, I am fully entitled to run the software, and I have not agreed to any EULA, nor have I violated any law - I have used only the rights explicitly granted to me by the copyright law. It would be absure to think that I am now bound by the EULA.
This article describes a method for installing software that's slightly more convenient than manually unpacking it, creating registry keys, and so on, but has the same effect: the software is installed on your hard drive and you have not agreed to the EULA.
This of course only applies to cases where you bought a copy of the software (as in, you exchanged money for a physical copy of the software, with no other terms imposed). Something like Oracle isn't sold like this: you get a copy of Oracle after signing a contract with Oracle, and that contract includes many of the normal EULA terms. That's a completely different situation: you didn't get the copy of the software until AFTER you agreed to some limitations on your use of that software; you are bound by those limitations. Software downloads are similar: you're often required to agree to the EULA before the download begins. It's the case where you got your copy of the software BEFORE agreeing to the limitations that's the more interesting (and much more common) one - and in that one, you're entitled to install and use the software without agreeing to the EULA.
It boils down to this: The law that the software developers are attempting to use to make EULAs binding is copyright law: the unspoken claim is that it is illegal to make a copy of the software without agreeing to the EULA. This is just not true.
I don't think there'a ny need for AT&T to buy out and close down PARC - Xerox seems to be doing a good enough job of that. They've been trying to sell it to venture capitalists for a while, with a notable lack of success. I don't think that PARC will last another 12 months, which is very sad.
Xerox also has (had?) a research lab in Cambridge, colloquially known as EuroPARC. I visited there a few times and saw some quite neat stuff.
Bob Shaw. Light of Other Days. It's most commonly seen as a short story, but also appeared as the centrepiece in a novel called (big surprise) "Slow Glass".
Our cats like to filet their minions (i.e., us) as well. Those back claws were designed to eviscerate, after all.
"The bite-covered troll rises from the dead."
Trolls regenerate, even after you kill them. You can kill a troll, turn away, and it will come back to life ans start beating on you. You have to destroy the body completely. There are a number of ways to do this (of course) but the simplest is to eat it. Eating it takes time, and sometimes the troll regenerates first...
Ah, finally an explanation for "The Emperor's New Mind"!
Look at Mojonation for an implementation of exactly what you just described.
10.1 was a lot like that: "oh, CD burning? sorry... didn't get done in time".
Many years ago, I bought one of the first Amigas available. It came with AmigaDOS 0.9. This was, let's just say, rough... the hardware was clearly ready before the software was. The first Amigas were supposed to have the OS in ROM, but it was so unready that they added 256K of write-protectable RAM to the machine. This was a LOT at the time (the system RAM was also 256K), and it must have been quite expensive for them to do.
A couple of months after buying the machine, I received a free copy of AmigaDOS 1.0 - as I recall it was mailed directly to me. So this kind of behaviour isn't unprecedented.
See here for commentary on Nostradamus that happens to use this exact quatrain as an example of a prophecy that's vague enough that a huge number of possible events could seem to fit it. Of the four specifics ("City of God", "two brothers", "fortress endures", "great leader will succumb") only two are actually even close ("City of God" for New York, or maybe Washington or maybe both - seems like a very poor fit; no leaders have succumbed).
This prophecy would apply just as well (in fact, better) to something like political/ecclesiastical family infighting in Renaissance Rome: City of God (seat of the Church), two brothers (family conflict, and remember that the Popehood was often controlled by the powerful political families), fortress endures (easy enough to come up with interpretations like "the glory of the Church endures"), great leader will succumb (the struggle causes the fall of one of the rulers).
If I remember correctly, the letter that the MPAA sends to the ISP is required by the DMCA to contain language like "Under penalty of perjury, we declare that these facts are true". This goes along with the assertions in the letter that IP A.B.C.D uploaded a file and that file contained material copyrighted by an MPAA member, so the ISP should suspend A.B.C.D.
If they're abusing the power granted to them by the MPAA by sending out such letters without performing competent investigations, then whoever signed that letter should by hauled into court and charged with perjury. The problem is that in cases like this, it's the word of insignificant nobodies against the word of a powerful entity - no prosecutor will ever bother to take such a case. If, on the other hand, anyone ever send such a letter alleging a DMCA violation on the part of the MPAA, you can bet they'd be in court so fast it would make their head spin.
The minimal safeguards that the DMCA puts on the power to accuse and convict merely by sending a letter (by requiring it to be sent "under penalty of perjury") are in practice worthless.
A few years ago I was attending a JPEG meeting where the Emmy that JPEG had just won was being proudly shown around. It's a recognition of contributions to the hard technical work of TV production.