I agree with you that in 99% of cases the e-mail warning will have no effect. Nonetheless it should still be sent, because:
a) 1% of users will act on it and thus avoid an inconvenient disconnection. This may not make a big difference to the outside world but it will make a difference to them.
b) You then have an excuse to use with the remaining irate 99%. "We tried e-mailing you but got no response. We were left with no option but to disconnect you."
It's a question of being seen to do the right thing.
Why any kind of drive? These boards include a NetBoot ROM with the in-built NIC so they work very nicely without a local hard disc drive. For a cluster you're surely better off with one (or two if you need redundancy) boot location rather than having to update one boot device per node.
> WWW - Got it's start as DARPAnet -US Dept of > Defense or did you mean surfing with a browser > (Mosaic, University of Illinois)
You're confusing the web and the Internet - they're not the same thing. The Internet did indeed originate as DARPAnet/ARPAnet but the WWW originated at Cern in Switzerland.
"South Canadian power outage"!?! Funny, from the reports we got here it seemed to have affected small parts of the USA as well. Or do you mean that most Spammers are in "South Canada" so that's the only bit that was relevant?
If true, this is very unfortunate. The last thing the Open Source community needs in its fight against SCO (and indeed, in general) is to be associated with virus writers.
> The people who spell it "an historic" aren't > pronouncing the "h".
But they are, and do! That's what's so silly about it. It seems to be a fashion amongst meeja types in particular to say, "an" before any word beginning with h, regardless of how it's pronounced.
For a word like "honour" (or "honor" for left-ponders), practically no-one pronounces the h so "an honour" is perfectly sensible.
For the word "hotel", there is a school of thought which pronounces it the French way, without the h and so for them, "an hotel" is perfectly sensible.
If you happen to come from the north of England and call a four legged creature like an outsize pony an "'orse" then saying "an 'orse" is perfectly sensible.
What's just plain dumb (and, if you accept any rules at all in language, just plain wrong) is twisting your tongue to use the indefinite article "an" in front of a word where you also pronounce the leading h - I've heard "an historic", "an horse", "an house" and lots of others, all with the h clearly pronounced.
> I say it and spell it the way > you do, but AFAIK they're both valid > pronounciations.
Indeed, this isn't an argument about pronounciation. If you don't pronounce the h then "an" is the sensible one to use; if you do pronounce the h then "an" is just silly.
> Unless Lycoris is referring to the GPL when they > are talking about the "prior agreement", it is > impossible for them to have another agreement with > SCO: the GPL simply does not permit redistribution > of code under side-agreements.
The way I read the notice on the Lycoris web-site, it's precisely the GPL they're referring to. They've been to SCO's ftp site, got a fresh copy of the kernel from there, thus making sure that all components in which SCO holds copyright are explicitly licensed to them under the GPL and are now re-distributing those components under the terms of the GPL.
As the author of the above quote I feel I should clarify. I wasn't attempting to make an issue out of it, just to answer the question, "Why haven't you got cheap CDs of Red Hat Linux 8.0?"
Quite right, there is no reason why Red Hat have to make their material available under the GPL. OTOH, there seems little point in fragmenting the market by having a lot of near-clones of Red Hat, each of which differs only in that it has a different logo.
If I were Red Hat I'd do it differently. Perhaps have a different name and image for their download edition, and (like Debian do) require it to be distributed entirely un-amended if that name ("Basic Red Hat Linux", "Free Red Hat Linux", "Download Red Hat Linux") is to be retained. That deals with the problem of someone potentially tarnishing Red Hat's reputation by distributing something different under the same name.
I tend to see it from the other direction. Red Hat put a lot of effort into producing a quality Linux distribution. It's silly that they don't get the credit for it.
> Its arguably the tiniest PC around...for the smallest form factor. I have a PC on my desk which is 15cm x 15cm x 4.5cm. (About the footprint of a CD case, but taller.) They're readily available under a host of different names. See www.paysan.co.uk for one supplier.
> I bought a 100GB drive last spring and it's not > even half full yet!
I bought a 120GB drive a month ago and it's 90% full.
There again, if I will go mirroring Debian...
Re:You think YOU had a rounding bug????
on
Pet Bugs?
·
· Score: 2, Interesting
> OK- top that.
Some years ago I was working on Prestel software running on the GEC 4000 series mini-computers. One particular problem affected the system at startup (when it was extremely busy for about 30 seconds) and took an awful lot of tracking down.
The GEC 4000 series was (is?) a real-time system with inter-process message passing built in to the CPU. You load values into registers (one of which will cause a chunk of memory to be passed to the other process) then execute a SEND instruction and away it goes. You can do a GOFREE to accept incoming messages from any source, a WAIT to accept them from one nominated source or a SENDWAIT to send a message and wait for the response. When a message arrives the register values are automatically loaded into the registers for the receiving process and then the process continues.
At startup the Prestel system had about 200 processes all frantically sending messages to each other. On odd occasions one process would crash, having apparently received a garbage response from a system process. Lots of heavy debugging, (including stopping the whole system and printing out large chunks of memory on the console (TI Silent 700 thermal paper - remember those) seemed to confirm that the system process was sending back garbage in response to a request.
I reported it to the OS guys who took a lot of convincing. After a lot of pressure they agreed to investigate and reluctantly agreed that it wasn't an application fault. In the end it turned out that it wasn't an OS fault either - it was a bug in the CPU. Under heavy load when executing a SENDWAIT the GEC 4160 would very occasionally do neither the SEND nor the WAIT, but allow the process to continue with whatever values were in the registers before.
Explained like that it sounds simple. Working from the sharp end in the field it was anything but.
> I wonder how they can pretend that a digital copy > can have a better sound than another digital copy, > if both are identical bit-per-bit
If you have audio files stored on a CD-ROM then a copied CD-R of it will sound just the same. Copying audio CDs is more complex and the software used *can* make a difference to the sound. It isn't just a question of a bit-for-bit copy. You can't, for instance, do a verify of a copied audio CD in the way you can verify a copied CD-ROM.
Not just simplistic but utterly useless. If you make a rule specifying the quantity of comments then what you'll get is the required quantity of utterly useless verbiage. Like those Windows programs where someone has clearly been told that they must write a help page for every input field. You get to a field labelled "Throoble quotient", hit F1 and up pops a help thingy saying, "In this field you should enter the throoble quotient".
There seems to be a growing misconception that, because a business model has worked in the past it therefore must be entitled to legal protection in order to ensure it continues to work in the future.
Consider the position of being, say, a musician a few hundred years ago. You could make a living (probably not a very good one) by composing and playing music for other people but, much like a plumber today you couldn't apply any multipliers to that. You play music for one evening - you get paid (or fed or something) a corresponding amount. If you want to be paid again tomorrow, make sure you have another gig lined up. The only way of avoiding that would be to find a rich sponsor.
Along came printing - suddenly there was a way for musicians (and others) to get the multiplication factor in. Write a piece of music and then *sell* it. You only have to write it once but you can sell it lots of times.
Along came audio recording - an even bigger multiplier. Now you don't even have to play it for each listener. Play it once (all right - I know - several times), record it, then sell it lots of times. You're not guaranteed to make lots of money that way but the potential is there and it's a perfectly reasonable thing to do (and it's perfectly reasonable to insist that others comply by the restrictions you choose to put on your material when you sell it - copyright).
What is *not* reasonable is then to expect legislation simply to preserve your business model from other perfectly legitimate business models. If you're producing and selling recorded music you have absolutely no right to insist that others can't distribute *their* music in a different way, even if it blows your business model right out of the water.
Similarly with the question of commercial TV channels. 100 years ago there were no commercial TV channels (bliss!). A particular combination of available technologies made them feasible (TVs available at prices consumers can afford; cameras and broadcasting kit available at prices consumers definitely can't afford; limited broadcast bandwidth available etc.) Now the technology position is moving on. Lots of new equipment is available and people may not be willing to make the same trade-off as before ("I'll watch your irritating adverts because I want to watch the program in the gaps"), particularly as the quality of both programs and adverts goes through the floor. Perhaps an entirely new business model will have to arise but there is absolutely no possible justification for legislation to protect an existing business model just because its window of opportunity is closing.
> How many people here USE Linux for more than
> a firewall or the kewlness factor?
I use Linux for everything and run my business on it. Web server, firewall, in-house server, desktops. It does everything we need and does it with far less hassle than the obvious alternative.
> I used to, but Windows is easier
Funny - that's precisely the opposite of the reason I use Linux. Windows is just *so* much harder to get things done on. You spend all your time fighting the system where with Linux you tell it what you want done and it does it.
Of course, it helps if you can speak the language.
Slashdot Mirror
I agree with you that in 99% of cases the e-mail warning will have no effect. Nonetheless it should still be sent, because:
a) 1% of users will act on it and thus avoid an inconvenient disconnection. This may not make a big difference to the outside world but it will make a difference to them.
b) You then have an excuse to use with the remaining irate 99%. "We tried e-mailing you but got no response. We were left with no option but to disconnect you."
It's a question of being seen to do the right thing.
John
Why any kind of drive? These boards include a NetBoot ROM with the in-built NIC so they work very nicely without a local hard disc drive. For a cluster you're surely better off with one (or two if you need redundancy) boot location rather than having to update one boot device per node.
> I thought the QL was 16-bit, not 32-bit...
It was (is?) based on the 68008 which is a 32 bit processor with an 8-bit external data bus.
John
> Don't forget that the facsimile was an
> Italian invention
Scottish I think you'll find.
> installed first in France
correct.
John
> WWW - Got it's start as DARPAnet -US Dept of
> Defense or did you mean surfing with a browser
> (Mosaic, University of Illinois)
You're confusing the web and the Internet - they're not the same thing. The Internet did indeed originate as DARPAnet/ARPAnet but the WWW originated at Cern in Switzerland.
HTH
John
> they say you could take it on daytrips to france
> as it has a range of 50 miles (the english
> channel is 22miles long)
The English channel is 22 miles *wide* (at its narrowest point). It's quite a bit longer.
> so you can drive there, drive 6 miles, and
> drive back,
Amazingly, there are reports that even as 3rd world a country as France now has a few petrol stations.
John
> Fax Machine
Which amazingly was invented before the telephone!
> Lightbulbs
> World Wide Web
I thought that was more a product of Swiss industry (albeit by an Englishman).
John
"South Canadian power outage"!?! Funny, from the reports we got here it seemed to have affected small parts of the USA as well. Or do you mean that most Spammers are in "South Canada" so that's the only bit that was relevant?
If true, this is very unfortunate. The last thing the Open Source community needs in its fight against SCO (and indeed, in general) is to be associated with virus writers.
John
> Unlike the Titanic, however, SCO actually steered
> itself into the ice berg.
You mean in the Titanic's case, the iceberg leaped out in front of it?
John
> "That's quoit un 'istorik evant!"
Dick van Dyke is an exceptional case and not subject to the usual rules.
John
> The people who spell it "an historic" aren't
> pronouncing the "h".
But they are, and do! That's what's so silly about it. It seems to be a fashion amongst meeja types in particular to say, "an" before any word beginning with h, regardless of how it's pronounced.
For a word like "honour" (or "honor" for left-ponders), practically no-one pronounces the h so "an honour" is perfectly sensible.
For the word "hotel", there is a school of thought which pronounces it the French way, without the h and so for them, "an hotel" is perfectly sensible.
If you happen to come from the north of England and call a four legged creature like an outsize pony an "'orse" then saying "an 'orse" is perfectly sensible.
What's just plain dumb (and, if you accept any rules at all in language, just plain wrong) is twisting your tongue to use the indefinite article "an" in front of a word where you also pronounce the leading h - I've heard "an historic", "an horse", "an house" and lots of others, all with the h clearly pronounced.
> I say it and spell it the way
> you do, but AFAIK they're both valid
> pronounciations.
Indeed, this isn't an argument about pronounciation. If you don't pronounce the h then "an" is the sensible one to use; if you do pronounce the h then "an" is just silly.
John
> Unless Lycoris is referring to the GPL when they
> are talking about the "prior agreement", it is
> impossible for them to have another agreement with
> SCO: the GPL simply does not permit redistribution
> of code under side-agreements.
The way I read the notice on the Lycoris web-site, it's precisely the GPL they're referring to. They've been to SCO's ftp site, got a fresh copy of the kernel from there, thus making sure that all components in which SCO holds copyright are explicitly licensed to them under the GPL and are now re-distributing those components under the terms of the GPL.
It's easy - anyone can do it.
John
> Anyone notice that SCO's stock slipped another 11% today? heh.
Can anyone recommend a good place to track SCO's price? It could be interesting to watch today.
With everyone busily making sure they're covered with a nice new copy of the GPL from SCO it's going to amount to a DDoS attack on their ftp server.
As the author of the above quote I feel I should clarify. I wasn't attempting to make an issue out of it, just to answer the question, "Why haven't you got cheap CDs of Red Hat Linux 8.0?"
Quite right, there is no reason why Red Hat have to make their material available under the GPL. OTOH, there seems little point in fragmenting the market by having a lot of near-clones of Red Hat, each of which differs only in that it has a different logo.
If I were Red Hat I'd do it differently. Perhaps have a different name and image for their download edition, and (like Debian do) require it to be distributed entirely un-amended if that name ("Basic Red Hat Linux", "Free Red Hat Linux", "Download Red Hat Linux") is to be retained. That deals with the problem of someone potentially tarnishing Red Hat's reputation by distributing something different under the same name.
I tend to see it from the other direction. Red Hat put a lot of effort into producing a quality Linux distribution. It's silly that they don't get the credit for it.
> Its arguably the tiniest PC around ...for the smallest form factor. I have a PC on my desk which is 15cm x 15cm x 4.5cm. (About the footprint of a CD case, but taller.) They're readily available under a host of different names. See www.paysan.co.uk for one supplier.
> I bought a 100GB drive last spring and it's not
> even half full yet!
I bought a 120GB drive a month ago and it's 90% full.
There again, if I will go mirroring Debian...
> OK- top that.
Some years ago I was working on Prestel software running on the GEC 4000 series mini-computers. One particular problem affected the system at startup (when it was extremely busy for about 30 seconds) and took an awful lot of tracking down.
The GEC 4000 series was (is?) a real-time system with inter-process message passing built in to the CPU. You load values into registers (one of which will cause a chunk of memory to be passed to the other process) then execute a SEND instruction and away it goes. You can do a GOFREE to accept incoming messages from any source, a WAIT to accept them from one nominated source or a SENDWAIT to send a message and wait for the response. When a message arrives the register values are automatically loaded into the registers for the receiving process and then the process continues.
At startup the Prestel system had about 200 processes all frantically sending messages to each other. On odd occasions one process would crash, having apparently received a garbage response from a system process. Lots of heavy debugging, (including stopping the whole system and printing out large chunks of memory on the console (TI Silent 700 thermal paper - remember those) seemed to confirm that the system process was sending back garbage in response to a request.
I reported it to the OS guys who took a lot of convincing. After a lot of pressure they agreed to investigate and reluctantly agreed that it wasn't an application fault. In the end it turned out that it wasn't an OS fault either - it was a bug in the CPU. Under heavy load when executing a SENDWAIT the GEC 4160 would very occasionally do neither the SEND nor the WAIT, but allow the process to continue with whatever values were in the registers before.
Explained like that it sounds simple. Working from the sharp end in the field it was anything but.
> I wonder how they can pretend that a digital copy
> can have a better sound than another digital copy,
> if both are identical bit-per-bit
If you have audio files stored on a CD-ROM then a copied CD-R of it will sound just the same. Copying audio CDs is more complex and the software used *can* make a difference to the sound. It isn't just a question of a bit-for-bit copy. You can't, for instance, do a verify of a copied audio CD in the way you can verify a copied CD-ROM.
I must apologise - I attached that comment to the wrong parent. It was meant to be a comment on the idea of judging comments by their volume.
Not just simplistic but utterly useless. If you make a rule specifying the quantity of comments then what you'll get is the required quantity of utterly useless verbiage. Like those Windows programs where someone has clearly been told that they must write a help page for every input field. You get to a field labelled "Throoble quotient", hit F1 and up pops a help thingy saying, "In this field you should enter the throoble quotient".
There seems to be a growing misconception that, because a business model has worked in the past it therefore must be entitled to legal protection in order to ensure it continues to work in the future.
Consider the position of being, say, a musician a few hundred years ago. You could make a living (probably not a very good one) by composing and playing music for other people but, much like a plumber today you couldn't apply any multipliers to that. You play music for one evening - you get paid (or fed or something) a corresponding amount. If you want to be paid again tomorrow, make sure you have another gig lined up. The only way of avoiding that would be to find a rich sponsor.
Along came printing - suddenly there was a way for musicians (and others) to get the multiplication factor in. Write a piece of music and then *sell* it. You only have to write it once but you can sell it lots of times.
Along came audio recording - an even bigger multiplier. Now you don't even have to play it for each listener. Play it once (all right - I know - several times), record it, then sell it lots of times. You're not guaranteed to make lots of money that way but the potential is there and it's a perfectly reasonable thing to do (and it's perfectly reasonable to insist that others comply by the restrictions you choose to put on your material when you sell it - copyright).
What is *not* reasonable is then to expect legislation simply to preserve your business model from other perfectly legitimate business models. If you're producing and selling recorded music you have absolutely no right to insist that others can't distribute *their* music in a different way, even if it blows your business model right out of the water.
Similarly with the question of commercial TV channels. 100 years ago there were no commercial TV channels (bliss!). A particular combination of available technologies made them feasible (TVs available at prices consumers can afford; cameras and broadcasting kit available at prices consumers definitely can't afford; limited broadcast bandwidth available etc.) Now the technology position is moving on. Lots of new equipment is available and people may not be willing to make the same trade-off as before ("I'll watch your irritating adverts because I want to watch the program in the gaps"), particularly as the quality of both programs and adverts goes through the floor. Perhaps an entirely new business model will have to arise but there is absolutely no possible justification for legislation to protect an existing business model just because its window of opportunity is closing.
It is a right foot - you're just looking at it from below.
> How many people here USE Linux for more than
> a firewall or the kewlness factor?
I use Linux for everything and run my business on it. Web server, firewall, in-house server, desktops. It does everything we need and does it with far less hassle than the obvious alternative.
> I used to, but Windows is easier
Funny - that's precisely the opposite of the reason I use Linux. Windows is just *so* much harder to get things done on. You spend all your time fighting the system where with Linux you tell it what you want done and it does it.
Of course, it helps if you can speak the language.
John