All kernel modules included in Red Hat Linux 8.0 are compiled with GCC 3.2; however, when using third-party modules it is important to make sure that every module and its dependent objects, in their entirety, is compiled with GCC 3.2. The modutils programs insmod and modprobe have rudimentary checks for this and will prevent loading of modules in the case of compiler version mismatches; these modules can be forced to load via the -f parameter.
I don't know which compiler NVidia used for their module. But from what I understand from the excerpt above, it will probably not work. Somebody tried it already? You can probably bet that an updated NVidia module will be out shortly with support for RH 8.0.
The GPL, thank god, means that Red Hat DOES have to give away their SRPMS, at least to any code in their OS that is GPL'd. Their installer doesn't have to be GPL'd. Their makefiles and build scripts don't have to be GPL'd. They could legally give away nothing but the actual source code they used to build the finished product. That satisifes the GPL, both in letter and spirit.
The GPL says that if you distribute a compiled version of a GPL'd software, you have to offer the source at the same people. So if RH asked $1000000 per boxed set, and included the source code in it, they wouldn't have to offer their SRPMS for anybody to download. Another thing the GPL says is that if you received GPL'd software, you have the right to redistribute it under the GPL. So if I bought one of those $1000000 boxed set, I could then redistribute the GPL'd content on my FTP server.
Red Hat offers their binary CDs as free downloads, so they have to offer their source code as well (could be only the SRPMS rather than an iso full of SRPMS). But if they didn't distributed freely downloadable isos, they wouldn't have to provide SRPMS to everybody, only those to whom they distributed the binaries.
This means more games, more office software, more of the general desktop software that many people say is missing from Linux.
Have you checked the specs of the box? Not much current games will run smoothly on that, Windows or Linux. It's a box designed to do basic things: power on, get on AOL, power down.
Re:an interesting bit of trivia
on
RC5-64 Success
·
· Score: 2
The short answer is that to be able to track all blocks, the keyserver concentrates on a small portion of the keyspace at a time. So if 90% of that small part is returned, the balance will be resent, in an attempt of getting the results faster and be able to switch to another part of the keyspace.
Search on the distributed.net website for more details, I recall a graph saying that 95%+ blocks were returned after 4 days, and after that the return rate was nearing zero, so it makes sense to resend those after a few days.
There's also the possibility that it was in a random generated block: when your client cannot reach the keyserver, it processes a block at random (actually, I think all the random blocks are close to each other). That would explain the 3 reports, at almost 1 year interval each.
Re:Portion of Internet's data
on
RC5-64 Success
·
· Score: 2
Since the keys are sequential, you just need to give a 64 bits value to clients, plus the length that the client should check. Same thing on the return trip: start, length, result (yes/no), who (for stats). Of course it was all encrypted, so it was a bit more than those values per packet, but nowhere 105GB/s or some other insane numbers.
All in all, it's a quite small portion of the total Internat traffic. I recall an article on Slashdot about a guy in Atlanta who secretly installed the client on state's computer, and was fined for that. IIRC, it amounted to something around 100K$ per MB, or something like that (of course the fine wasn't only for the bandwidth used).
Re:Let's check the math...
on
RC5-64 Success
·
· Score: 2, Interesting
So, for one computer to crack RC5-64 in a matter of weeks (less than five) would require a computer about 460 times faster than what we have now; assuming moore's law keeps going, we'll get those in about 13 years (2015).
You forget THE major point of Distributed.net: distributed computing. If you put 2 computers to the task, you already cut by half the time needed. Have more money? Put 3000 CPUs (go read the nVidia and ATI tour at Anandtech to see if somebody can afford those now) through it, and the time will shrink by the same amount.
And regarding the time needed to crack it, I get a couple orders of magnitude greater than 384000 computer*hours. More akin to (quoting the PR) 46000*790*24=872 million computer*hour (using an Athlon XP 2GHz). A single CPU computer wouldn't be able to do it on a human scale time (would be about 100000 years), you absolutely need more than one computer to live to see the result.
For a realisitic cracking scenerio, let's say our cracker has ten computers and wants to crack the code in a week... he'd still have to wait 8 years to be able to do it, and who'd want to bother with 13 year old data for cracking, anyway?
I probably miss something about why the 8 years becomes 13, but there are some things that don't change in time, and could be used by somebody even in a few years. My credit card number hasn't changed since I first got it, same thing for my bank account. The goal is not for it to be secure only now, but also in the future. You may think about other examples involving national security if you prefer.
Not really. If you consider that over 5 years, the average keyrate is 105.5 GKeys/sec, and the latest day averages were somewhere around 180 GKeys/sec, it means the same thing could have been finished in almost half the time, if it was started now with today's computers. Moore's law being what it is, if it really was started again now, it would take around half that time again, because more powerful CPUs are to be unveiled in that timeframe.
By their own estimates, it would take ~46000 Athlon XP 2GHz (now, where are you to find those right now?) to have 270 GKeys/sec (their peak rate in 5 years), which gives completing the keyspace in 790 days. Who would buy that much CPUs? Good question. With 2 dual MP motherboards in 1U (too lazy to find a link, I know somebody offers something like that), it would only take about 300 40U racks. Would you bet future national security on it? I don't think I would (and I'm not even american).
What it really shows is that brute-force can succeed, given enough time. But of course the more effective way to attack an encrytion algorithm is on the algorithmic side, because it helps you to find not only one cleartext, but all cleartexts encrypted with that algorithm.
Here (Canada, so obviously different laws apply), my Uni does collect that info, if only to send us the necessary things to attend classes (registration, etc.). I don't recall anything about selling those informations to other entities though.
The students of my faculty do have a student directory. The old way of getting the info was by passing sheets in each class, where people could modify their current information at the beginning of the semester. So if you didn't want to be listed, you just didn't filled it (opt-in). Now, that has changed a bit: the info for the directory is channeled from the Uni's info to the paper directory. But again, sheets are passed at the beginning of each semester to get the approbation of students to be listed in the directory.
To get back closer to the subject, I recall that some insurance company was sponsoring an event for coop students, so they had to send a letter to each of them. The way it was handled was that the insurance company gave us what they wanted us (the student's association) to send, and we handled the "put it in an envelope, put the address sticker and repeat" part. They never had access to the list of students, or to their info.
Re:Replacing the wallwart by a battery should be e
on
Portable Hubs?
·
· Score: 2
That's right off the scale. Transformers don't work on DC.
My knowledge about transformers is, I confess, rather limited. That's why I put the 7808 before in my list of possible solutions.
Of course, thinking more about it, since it's DC there's not change in the voltage in the input coil, so no current is inducted in the core of the transformer. No core current, no voltage at the output coil. So yes, the transformer was a very bad idea, unless you want to drain the battery rather quickly:)
Thank you for pointing out the obvious.
Another point against 5 1.5V AA batteries is that the energy they have is somewhat limited. So not only their output voltage will drop, but even if they keep their output voltage somehow they won't be able to power the hub (or anything else) for as long as your gel cell battery.
Bonus question: why do they need 7.5V? I've handled a couple hubs/switches (5 to 8 ports, 10 or 10/100Mbps), and they all need 7.5V. Normal chips need 5V or lower. Is it to drive the Ethernet signals?
Replacing the wallwart by a battery should be easy
on
Portable Hubs?
·
· Score: 2
The wall wart of my 5 ports Linksys 10/100 switch gives 7.5VDC at a maximum of 700mA, which is about 5W of power. I'd have to check with a multimeter or use one of the DC power supplies at school, but under normal usage I'm sure it's way lower than that maximum. Even with a 12Ah 12V battery, plus a voltage regulator, you'd still have enough juice to outrun the laptops.
That's something just thought about: your laptops will drain faster than the hub. So even a smaller battery than the 12Ah I mentionned would do the trick. Pick a rechargeable battery if you don't wan't to throw away too much metal in the trash.
The only thing you'll need to do is a small package to transform the 12V to 7.5 (7808 or even a small transformer), and wire it to the right DC plug for the hub, and there you go!
There's more than the ASUS A7M266D (I have one too) with the 760MPX chipset. Tyan's Tiger MPX, MSI' K7D Master (L), Epox's M762A and Gigabyte's GA-7DPXDW are the other 760MPX boards I know of. If you want to go back to the 760MP, Tyan's Thunder MP and Tiger MP are the only boards sporting that chipset.
The various 760MPX boards vary by what's onboard. Some have almost nothing, some others have sound only, and some have Ethernet (with an Intel chip for the MSI). Depending on your cooling solution, some have the 4 holes around the socket A, some don't. But if you chose the ASUS, I guess you don't use them (they can be drilled though).
I don't recall a metallic smell the first times I powered it. It did got hot though:)
Then, you'll need some type of converter (hence increased losses) at the production sites. Remember, most of electricity produced in the US is from a tubine coupled with an alternator. It's very easy to have the axle of a turbo-alternator to go faster than the resulting electrical phase, but much difficultier to do the opposite.
In a common turbo-alternator, IIRC, n=60*f/p, where n is the rotational speed (RPM), and p is the number of pair of poles on the rotor, and 60 is to have the speed in RPM rather than RPS (or Hz). So if you want some insane number for f (like 20kHz), your standard run-of-the-mill turbine (2 pair of poles, which normally runs at 1800RPM) will need to resist to 600000RPM. Of course, if you want it to actually hold together, you'll need to make it run slower (I remind you that a Dremel is ~30000RPM max), so let's say you choose 10000RPM (which is still way higher than current equipments). This means you'll need 120 pairs of poles on the rotor, which is a lot. If you need to make it spin at 10000RPM, each one of your poles will need to be really small (or you have a very small radius for your alternator, which mean the same thing).
Just for comparison, the biggest turbo-alternator I know of (doesn't mean there aren't some bigger ones) are for the Three-Gorges dam in China. Something like 20m radius (60ft). Of course it's not supposed to turn at even 100RPM, much probably less.
So having your turbos making 20kHz AC is not very possible. The other solution would be to have some kind of converter between the production and the use. Problem is, each converter (frequency) or transformer (voltage) will have some losses, because of the current running through it, and because of other reasons as well (inductance, magnetic losses, etc.). One of the things electricity producers and distributers hate the most are losses: they don't get money for that electricity, which must be produced anyway. And usually, the losses are estimated to a tenth or a hundredth of a percent before the construction of a power plant. So I don't see the change to 20kHz as really feasible (not to mention the huge installed base and the potential skin effect problems as others have already pointed out).
Plus, there's talk that the Inspiron 8200's might be upgradable, video-wise:)
So, you own one, and you're still not sure if it's upgradeable or not? Or are you unsure about which videocard you can actually put in it?
Personnally, that's one of the reason I'm not likely to buy a laptop: you're not told up front what you can and cannot change once you bought it. In a normal self-made tower, you know you can easily change or add any part (except what you chose to be integrated on the MB). And I'm still not sold to an LCD, even if some people like their expericence with them.
(Last thing: thirty degrees of freedom in either direction? What is it supposed to mean in the context of a laptop?)
I agree. It was the feature I liked most in WP, and that I missed most in Word when my father installed Word from Office 4.2 (Windows 3.1) rather than WordPerfect for Windows, which was quite buggy for the first few versions.
It really let you specify where each formatting would begin or end, and could be hidden if you didn't like it. And Word still doesn't have something comparable (Word is too WYSIWYG to include it now, it would mean they should've included it earlier).
I remember that symbol (the tachometer) from a couple years back (1998). It was (and still is) the logo of the RPM format. RH used it for the cover of their book "RPM Unleashed" or "Max RPM" (I don't recall the exact title, but it was along those lines). I didn't knew about the recursive acronym, though.
Could you be more specific about what's bad in *.rpm? File format (cpio archive)? Dependancy tracking? Name (RedHat Package Manager)?
Personnally (but then I only used RH, since 4.2, circa 1998), I like that format. It's easy to download and install bugfixes or security updates from RH. It's easy to bundle something myself, if I need to. It's easy to upgrade the whole distribution (even though I usually don't do that the "recommended" way).
A little more precision would be helpful for me, to understand what you dislike in it, and maybe have a different look on the packaging system.
Maybe I'm not in the right neighborhood or I don't know the right people, but none of my friends nor myself own a PVR, so I don't see the automatic skipping of ads as a big threat (for now) to TV stations. Usually, when I watch TV, it's to put my brain in a "don't think too much" mode (except when I watch the news). Ads are just another part of the TV programmation, although one I don't mind missing by going to the toilet or getting a drink.
I second this. The default syntax highlighting modes include C, C++, bash, java, html, Makefile, perl, latex, and that's just what I use on a regular basis. The only thing I'd like to have is some kind of auto-complete, and that's because I'm starting in Java so I'm not totally sure of some method names in the API.
And for the file manager... I've been a Norton Commander (and clones) convert since my dad introduced me to it, circa 1988.
Back on subject... You _can_ save a your work from Word as a textfile, but nobody I know does this, because it's not the right tool for the job. Notepad (when you've got nothing else as a real text editor) is better in this respect (if only it could be made to understand Unix EOL...).
No, the water is not hot enough, even with a dual Athlon XP1800+. I know, that's the setup I have, although it's from another kit than the one in this article.
Now on to the explanation: heat goes from a colder place to a hotter place. If you want to make coffee from the water, it needs to be at least 80-90C. Now, that's the max (read: if you reach that, pull the plug!!) temperature supported by the XPs, and I think Intel's are a bit lower (75C). If your cooling fluid is hotter than your CPU, it's not cooling anymore, so you have a big problem (actually, the temperature of the CPU will just raise until either it melts or it's hotter enough than your cooling fluid to create a new thermal equilibrium). So even if you wanted to have a coffee tap, all you would end up with is a fried chip.
Thanks for the info! My involvement with FORTRAN is too young (only a few years ago, while maintaining some code on a Windows platform) to know all those details.
May I ask you how FORTRAN is a) integrated with it's OS (whatever that really means in the case of FORTRAN), or b) integrating the development environment with the software under development?
Actually, the P6 has already been released. That was the Pentium Pro. So either Intel will change naming convention before that (not very likely), or they will have 2 different products with the same name (I'm sure it wouldn't be the first time).
Doesn't it feel weird that a 8 years old chip has a name "higher" than one not yet released?
OTOH, I don't recall for sure if the original Pentium (or derivatives, like P54C) were known as "P5". If they had, then it's already time for a name change for the successor of the P4.
Oh, and about those commercials.... in the one where the blue aliens modify some digital pictures (remove nose, etc.), doesn't it seem bizarre that when they add color, it's done in a slow, progressive pass? I would have thought that Intel's P4 was faster than that and could do so in the blink of an eye. Thoughts? (I know it's probably some marketing thingy, it's just odd to me)
Slashdot Mirror
In the release notes of RH 8.0:
I don't know which compiler NVidia used for their module. But from what I understand from the excerpt above, it will probably not work. Somebody tried it already? You can probably bet that an updated NVidia module will be out shortly with support for RH 8.0.
The GPL, thank god, means that Red Hat DOES have to give away their SRPMS, at least to any code in their OS that is GPL'd. Their installer doesn't have to be GPL'd. Their makefiles and build scripts don't have to be GPL'd. They could legally give away nothing but the actual source code they used to build the finished product. That satisifes the GPL, both in letter and spirit.
The GPL says that if you distribute a compiled version of a GPL'd software, you have to offer the source at the same people. So if RH asked $1000000 per boxed set, and included the source code in it, they wouldn't have to offer their SRPMS for anybody to download. Another thing the GPL says is that if you received GPL'd software, you have the right to redistribute it under the GPL. So if I bought one of those $1000000 boxed set, I could then redistribute the GPL'd content on my FTP server.
Red Hat offers their binary CDs as free downloads, so they have to offer their source code as well (could be only the SRPMS rather than an iso full of SRPMS). But if they didn't distributed freely downloadable isos, they wouldn't have to provide SRPMS to everybody, only those to whom they distributed the binaries.
This means more games, more office software, more of the general desktop software that many people say is missing from Linux.
Have you checked the specs of the box? Not much current games will run smoothly on that, Windows or Linux. It's a box designed to do basic things: power on, get on AOL, power down.
The short answer is that to be able to track all blocks, the keyserver concentrates on a small portion of the keyspace at a time. So if 90% of that small part is returned, the balance will be resent, in an attempt of getting the results faster and be able to switch to another part of the keyspace.
Search on the distributed.net website for more details, I recall a graph saying that 95%+ blocks were returned after 4 days, and after that the return rate was nearing zero, so it makes sense to resend those after a few days.
There's also the possibility that it was in a random generated block: when your client cannot reach the keyserver, it processes a block at random (actually, I think all the random blocks are close to each other). That would explain the 3 reports, at almost 1 year interval each.
Since the keys are sequential, you just need to give a 64 bits value to clients, plus the length that the client should check. Same thing on the return trip: start, length, result (yes/no), who (for stats). Of course it was all encrypted, so it was a bit more than those values per packet, but nowhere 105GB/s or some other insane numbers.
All in all, it's a quite small portion of the total Internat traffic. I recall an article on Slashdot about a guy in Atlanta who secretly installed the client on state's computer, and was fined for that. IIRC, it amounted to something around 100K$ per MB, or something like that (of course the fine wasn't only for the bandwidth used).
So, for one computer to crack RC5-64 in a matter of weeks (less than five) would require a computer about 460 times faster than what we have now; assuming moore's law keeps going, we'll get those in about 13 years (2015).
You forget THE major point of Distributed.net: distributed computing. If you put 2 computers to the task, you already cut by half the time needed. Have more money? Put 3000 CPUs (go read the nVidia and ATI tour at Anandtech to see if somebody can afford those now) through it, and the time will shrink by the same amount.
And regarding the time needed to crack it, I get a couple orders of magnitude greater than 384000 computer*hours. More akin to (quoting the PR) 46000*790*24=872 million computer*hour (using an Athlon XP 2GHz). A single CPU computer wouldn't be able to do it on a human scale time (would be about 100000 years), you absolutely need more than one computer to live to see the result.
For a realisitic cracking scenerio, let's say our cracker has ten computers and wants to crack the code in a week... he'd still have to wait 8 years to be able to do it, and who'd want to bother with 13 year old data for cracking, anyway?
I probably miss something about why the 8 years becomes 13, but there are some things that don't change in time, and could be used by somebody even in a few years. My credit card number hasn't changed since I first got it, same thing for my bank account. The goal is not for it to be secure only now, but also in the future. You may think about other examples involving national security if you prefer.
Not really. If you consider that over 5 years, the average keyrate is 105.5 GKeys/sec, and the latest day averages were somewhere around 180 GKeys/sec, it means the same thing could have been finished in almost half the time, if it was started now with today's computers. Moore's law being what it is, if it really was started again now, it would take around half that time again, because more powerful CPUs are to be unveiled in that timeframe.
By their own estimates, it would take ~46000 Athlon XP 2GHz (now, where are you to find those right now?) to have 270 GKeys/sec (their peak rate in 5 years), which gives completing the keyspace in 790 days. Who would buy that much CPUs? Good question. With 2 dual MP motherboards in 1U (too lazy to find a link, I know somebody offers something like that), it would only take about 300 40U racks. Would you bet future national security on it? I don't think I would (and I'm not even american).
What it really shows is that brute-force can succeed, given enough time. But of course the more effective way to attack an encrytion algorithm is on the algorithmic side, because it helps you to find not only one cleartext, but all cleartexts encrypted with that algorithm.
Here (Canada, so obviously different laws apply), my Uni does collect that info, if only to send us the necessary things to attend classes (registration, etc.). I don't recall anything about selling those informations to other entities though.
The students of my faculty do have a student directory. The old way of getting the info was by passing sheets in each class, where people could modify their current information at the beginning of the semester. So if you didn't want to be listed, you just didn't filled it (opt-in). Now, that has changed a bit: the info for the directory is channeled from the Uni's info to the paper directory. But again, sheets are passed at the beginning of each semester to get the approbation of students to be listed in the directory.
To get back closer to the subject, I recall that some insurance company was sponsoring an event for coop students, so they had to send a letter to each of them. The way it was handled was that the insurance company gave us what they wanted us (the student's association) to send, and we handled the "put it in an envelope, put the address sticker and repeat" part. They never had access to the list of students, or to their info.
That's right off the scale. Transformers don't work on DC.
My knowledge about transformers is, I confess, rather limited. That's why I put the 7808 before in my list of possible solutions.
Of course, thinking more about it, since it's DC there's not change in the voltage in the input coil, so no current is inducted in the core of the transformer. No core current, no voltage at the output coil. So yes, the transformer was a very bad idea, unless you want to drain the battery rather quickly :)
Thank you for pointing out the obvious.
Another point against 5 1.5V AA batteries is that the energy they have is somewhat limited. So not only their output voltage will drop, but even if they keep their output voltage somehow they won't be able to power the hub (or anything else) for as long as your gel cell battery.
Bonus question: why do they need 7.5V? I've handled a couple hubs/switches (5 to 8 ports, 10 or 10/100Mbps), and they all need 7.5V. Normal chips need 5V or lower. Is it to drive the Ethernet signals?
The wall wart of my 5 ports Linksys 10/100 switch gives 7.5VDC at a maximum of 700mA, which is about 5W of power. I'd have to check with a multimeter or use one of the DC power supplies at school, but under normal usage I'm sure it's way lower than that maximum. Even with a 12Ah 12V battery, plus a voltage regulator, you'd still have enough juice to outrun the laptops.
That's something just thought about: your laptops will drain faster than the hub. So even a smaller battery than the 12Ah I mentionned would do the trick. Pick a rechargeable battery if you don't wan't to throw away too much metal in the trash.
The only thing you'll need to do is a small package to transform the 12V to 7.5 (7808 or even a small transformer), and wire it to the right DC plug for the hub, and there you go!
There's more than the ASUS A7M266D (I have one too) with the 760MPX chipset. Tyan's Tiger MPX, MSI' K7D Master (L), Epox's M762A and Gigabyte's GA-7DPXDW are the other 760MPX boards I know of. If you want to go back to the 760MP, Tyan's Thunder MP and Tiger MP are the only boards sporting that chipset.
The various 760MPX boards vary by what's onboard. Some have almost nothing, some others have sound only, and some have Ethernet (with an Intel chip for the MSI). Depending on your cooling solution, some have the 4 holes around the socket A, some don't. But if you chose the ASUS, I guess you don't use them (they can be drilled though).
I don't recall a metallic smell the first times I powered it. It did got hot though :)
Then, you'll need some type of converter (hence increased losses) at the production sites. Remember, most of electricity produced in the US is from a tubine coupled with an alternator. It's very easy to have the axle of a turbo-alternator to go faster than the resulting electrical phase, but much difficultier to do the opposite.
In a common turbo-alternator, IIRC, n=60*f/p, where n is the rotational speed (RPM), and p is the number of pair of poles on the rotor, and 60 is to have the speed in RPM rather than RPS (or Hz). So if you want some insane number for f (like 20kHz), your standard run-of-the-mill turbine (2 pair of poles, which normally runs at 1800RPM) will need to resist to 600000RPM. Of course, if you want it to actually hold together, you'll need to make it run slower (I remind you that a Dremel is ~30000RPM max), so let's say you choose 10000RPM (which is still way higher than current equipments). This means you'll need 120 pairs of poles on the rotor, which is a lot. If you need to make it spin at 10000RPM, each one of your poles will need to be really small (or you have a very small radius for your alternator, which mean the same thing).
Just for comparison, the biggest turbo-alternator I know of (doesn't mean there aren't some bigger ones) are for the Three-Gorges dam in China. Something like 20m radius (60ft). Of course it's not supposed to turn at even 100RPM, much probably less.
So having your turbos making 20kHz AC is not very possible. The other solution would be to have some kind of converter between the production and the use. Problem is, each converter (frequency) or transformer (voltage) will have some losses, because of the current running through it, and because of other reasons as well (inductance, magnetic losses, etc.). One of the things electricity producers and distributers hate the most are losses: they don't get money for that electricity, which must be produced anyway. And usually, the losses are estimated to a tenth or a hundredth of a percent before the construction of a power plant. So I don't see the change to 20kHz as really feasible (not to mention the huge installed base and the potential skin effect problems as others have already pointed out).
Plus, there's talk that the Inspiron 8200's might be upgradable, video-wise :)
So, you own one, and you're still not sure if it's upgradeable or not? Or are you unsure about which videocard you can actually put in it?
Personnally, that's one of the reason I'm not likely to buy a laptop: you're not told up front what you can and cannot change once you bought it. In a normal self-made tower, you know you can easily change or add any part (except what you chose to be integrated on the MB). And I'm still not sold to an LCD, even if some people like their expericence with them.
(Last thing: thirty degrees of freedom in either direction? What is it supposed to mean in the context of a laptop?)
I agree. It was the feature I liked most in WP, and that I missed most in Word when my father installed Word from Office 4.2 (Windows 3.1) rather than WordPerfect for Windows, which was quite buggy for the first few versions.
It really let you specify where each formatting would begin or end, and could be hidden if you didn't like it. And Word still doesn't have something comparable (Word is too WYSIWYG to include it now, it would mean they should've included it earlier).
I remember that symbol (the tachometer) from a couple years back (1998). It was (and still is) the logo of the RPM format. RH used it for the cover of their book "RPM Unleashed" or "Max RPM" (I don't recall the exact title, but it was along those lines). I didn't knew about the recursive acronym, though.
Could you be more specific about what's bad in *.rpm? File format (cpio archive)? Dependancy tracking? Name (RedHat Package Manager)?
Personnally (but then I only used RH, since 4.2, circa 1998), I like that format. It's easy to download and install bugfixes or security updates from RH. It's easy to bundle something myself, if I need to. It's easy to upgrade the whole distribution (even though I usually don't do that the "recommended" way).
A little more precision would be helpful for me, to understand what you dislike in it, and maybe have a different look on the packaging system.
Maybe I'm not in the right neighborhood or I don't know the right people, but none of my friends nor myself own a PVR, so I don't see the automatic skipping of ads as a big threat (for now) to TV stations. Usually, when I watch TV, it's to put my brain in a "don't think too much" mode (except when I watch the news). Ads are just another part of the TV programmation, although one I don't mind missing by going to the toilet or getting a drink.
62.3% of Linux users have a (warezed) Windows 2000 partition so that they can play Warcraft III
No need for that. Wine (and Winex) will play Warcraft III.
And according to Blizzard and their crushing of bnetd, it's rather the opposite (62.3% of Linux users have a (warezed) version of Warcraft III)
I second this. The default syntax highlighting modes include C, C++, bash, java, html, Makefile, perl, latex, and that's just what I use on a regular basis. The only thing I'd like to have is some kind of auto-complete, and that's because I'm starting in Java so I'm not totally sure of some method names in the API.
And for the file manager... I've been a Norton Commander (and clones) convert since my dad introduced me to it, circa 1988.
Back on subject... You _can_ save a your work from Word as a textfile, but nobody I know does this, because it's not the right tool for the job. Notepad (when you've got nothing else as a real text editor) is better in this respect (if only it could be made to understand Unix EOL...).
No, the water is not hot enough, even with a dual Athlon XP1800+. I know, that's the setup I have, although it's from another kit than the one in this article.
Now on to the explanation: heat goes from a colder place to a hotter place. If you want to make coffee from the water, it needs to be at least 80-90C. Now, that's the max (read: if you reach that, pull the plug!!) temperature supported by the XPs, and I think Intel's are a bit lower (75C). If your cooling fluid is hotter than your CPU, it's not cooling anymore, so you have a big problem (actually, the temperature of the CPU will just raise until either it melts or it's hotter enough than your cooling fluid to create a new thermal equilibrium). So even if you wanted to have a coffee tap, all you would end up with is a fried chip.
Thanks for the info! My involvement with FORTRAN is too young (only a few years ago, while maintaining some code on a Windows platform) to know all those details.
May I ask you how FORTRAN is a) integrated with it's OS (whatever that really means in the case of FORTRAN), or b) integrating the development environment with the software under development?
Not just the test. The digits reversing part is also based on base-10. Could be interesting to mix the bases for the test and the reversing.
:P
As for the use of base-10 rather than another base... probably because we have 10 fingers?
Actually, the P6 has already been released. That was the Pentium Pro. So either Intel will change naming convention before that (not very likely), or they will have 2 different products with the same name (I'm sure it wouldn't be the first time).
Doesn't it feel weird that a 8 years old chip has a name "higher" than one not yet released?
OTOH, I don't recall for sure if the original Pentium (or derivatives, like P54C) were known as "P5". If they had, then it's already time for a name change for the successor of the P4.
Oh, and about those commercials.... in the one where the blue aliens modify some digital pictures (remove nose, etc.), doesn't it seem bizarre that when they add color, it's done in a slow, progressive pass? I would have thought that Intel's P4 was faster than that and could do so in the blink of an eye. Thoughts? (I know it's probably some marketing thingy, it's just odd to me)
Yes, of course. Thank you.
Didn't checked back the article a second time, and now it's back to byte me.