You can resell the GPLed software that RH puts out, but you are not permitted to use their trademark, logo and other copyrighted items. Those are not GPLed.
Hence if you take RH software, you can put it on a CD and sell it as the GPL says you can, but you cannot use the RH trademark, and therefore you cannot call it Red Hat Linux.
Think of the Coke recipe that Coca-cola published some time ago. You can take the recipe and put the result in bottle and selll it but you can't call it Coke.
> Europeans see putting a check US power as paramount;
Agreed, and this may be short-sighted
> while Americans see ending the tyrany of brutal dicatator as most important.
No, America actually helped and continues to help local tyrants whenever it suits them (From Chile to Afghanistan right now). Has America ever installed a genuine democracy anywhere? Don't say Japan, because Japan's democracy is not working very well. The same party has been in power since WWII. Whenever democracy has flourished in a country it was in spite of the US and not thanks to them, such as in Europe and South Korea.
> Most Americans are dismayed why we deserve such mistrust.
Looking around on the web some more, Mt Pinatubo released very little C02, what it did was release a lot of ash which diffused the solar radiation and made it possible for plants to absorb atmospheric C02 for some reason.
So it had the inverse effect to what you are claiming. Well done.
I'm tired of this argument. Yes, people HAVE to use Microsoft products even if they don't want to, generally because of lack of choice and deliberate undermining of the competition by anti-competitive means.
You don't have to believe me, just read the court transcripts and findings, it's written there black on white. It was upheld on appeal so this is not even a discussion point anymore.
You can be as successful and agressive as you want until you become a monopoly, then the rules change.
> reduce the problems caused by the poor > design of C itself.
C was designed to be a simple, powerful, low-level language, as easy to use and with as compact a compiler as Pascal, but with improved speed and flexibility.
With the territory came the ability to write more subtle and deadly bug, but that was understood.
So what language would you recommend to rewrite the Linux kernel in, say ?
> I'm enjoying the idea that this interchange hasn't > digressed into a ridiculous flame war. It's nice > to have a reason to reply.
Yeah, same here. If you hadn't posted as an AC you would be on my friends list by now:-)
I agree with you on the stability of the 2.4 series, it is not so bad now since about 2.4.18, and the RH kernel which I use were never affected because of Alan Cox' sterling work.
The BSD vs. SYSV stuff, appart from the classic vs. newish tree layout and the various different way to get things done as far as sysadmin is concerned, what is important is the API. I suspect (but I'm not sure) that the BSDs all support most of the sysv system calls such as shared memory, semaphores, etc, and probably the sysv signals too. They all are fine features, so again all of that is a matter of shades of grey.
Normally Linux `aims' for POSIX compliance, which I think is fine. In my apps I always try to code to ISO C/C++ and then Posix, and all the rest goes into the "annoying non-portable stuff", which includes the sysv-specific calls, win32 calls, and BSD specific when they exist.
LSB is not a pipe dream, it's there and all major distributions are compliant. The problem is that it is not enough, it formalizes only a tiny portion of a Linux system and there is still a lot to do, and making so many vendor agree is a major pain.
I forgot to agree with you earlier with the fact that a Solaris box makes a splendid NFS server and excellent NFS clients too (caching NFS is great).
For the companion CD, being out of date for bash or ls is of no consequence to me either, but being up to date with gcc and gdb is very very important to me as I code in C++, and gcc has been making progress at a huge rate with that language in the last couple of years.
Also lately I've really appreciated being able to keep up with the KDE developments. The difference between KDE 2 and 3 is between unusable ugliness and a usable fine desktop, and with KDE3, one really really needs those updates as soon as they come, and recompiling KDE is an impossible task for me, not worth the effort.
About the `linux for the masses' with support for crappy hardware and the fact that Linux is not innovative, this is true to a degree, but there are Linux true innovation. My favourite example are the Linux virtual consoles. The BSDs have them too now but I believe they originated with Linux and the commercial Unices never had them AFAIK. They are incredibly useful (multiple X sessions are also a godsend). The whole development process is also an innovation, with an open team and a very visible process. Historically both GNU and BSDs had more closed teams (not necessarily a disadvantage).
Myself I don't really care if Microsoft continue to dominate 95% of the desktop as long as the remaining 5% are viable, i.e. as long as I can install some kind of Unix on the DELL my workplace puts on my desk. I've always been lucky to be able to have a Unix desktop machine for the last 15 years or so (NeXT, Sun, DEC, and Linux), and really I don't want to go back. If being viable means that we have to compete on usability and we have to put up with some degree of zealotry and marketspeak, then so be it because I don't think Microsoft will be playing nice and allow any kind of competition to survive if they have their way.
The question I'm asking myself is would have the BSDs been in the position they are in now, including Darwin, without Linux? Would FreeBSD or some other fork have taken Linux's position had Linus not posted his kernel on a mailing list? or would have this whole movement not gelled at all and we would all be running Windows now?
As far as Apple is concerned, I was lucky enough to have a NeXT workstation in the early 90s and really this was at the time the Unix with the prettiest face and most consistent and innovative design. Apple simply inherited that wholesale and added very little. They had a head start. Also Apple does
Fedora is not just a change of name, it comes with no company-backed guarantees whatsoever (just community support), it's just a showcase and a beta distribution to get the enterprise packages tested in the community. Before the free RH had a small, but sufficient guarantee of support. Fedora doesn't even have the name.
For tech-savvy individuals it does not matter too much although they might fear the constant upgrade treadmill and the potential unstability, but these guys are a minority. For the real tech-savvy individual there is no shortage of choice and Fedora is just one of them anyway.
For not-super-rich corporations and institutions such as colleges, it is a disaster. They cannot afford the unfriendly per-seat licensing scheme of the RH enterprise products (even the cheaper ones), they loathe the EULA (it makes them auditable), and they've just lost the PHB-friendly support from RH.
Note this: it does not matter that Fedora provides updates of the highest quality. The PHBs will see this as an amateurish effort at best, easily hijacked at worst and will simply forbid this to run in their enterprise. Note that you cannot buy a small number of RHEL licenses and install it everywhere, the licensing agreement forbids it.
In other words this is the end of RedHat everywhere. People will be better off running stable Debian or *BSD because they have a track record of reliability whereas Fedora has nothing.
Soon the Enterprise solutions will follow them in the dump because no one will bother learning RH anymore. Current RHCEs are pissed off and will be angry at RH for devaluating their effort.
There is a high degree of probability that RH is throwing the baby with the bathwater and will be finding itself in the same league as the proprietary Unix vendors such as BSDI and SCO.
Myself I plan to evaluate Fedora when it come out, at home, but I won't touch my work RH9 installations until shortly before EOL. Then I'll probably move to something else, SUSE being a strong candidate, unless I am proved wrong with Fedora.
Fedora has a *very* short time to prove itself worthy.
You comment highlights the difference between the BSD and the Linux philosophies, even though you are using Microsoft as an example.
> Not to pat Microsoft on the back, but I've > never heard of them releasing a product > where they need to experimentally change the > kernel and have an end user "try it out."
I'm sure they do that in RC mode, just like Linux. If you've ever had an incident report with Microsoft, once out of trivial ideas they get you to try experimental patches too. With Microsoft everybody is a beta tester, except you get to pay for the product too.
In general this is a different development model, with Linux this sort of thing is condoned,certainly in development kernels: "release early, release often", etc. I grant you that the BSD philosophy is different, there is a core group that try to get things polished before they are released. Conversely it's reputedly hard to get even small changes accepted in the BSD kernels. My experience with sending patches to the Linux kernel has been very positive, although I haven't done that recently.
Really however this is just a matter of degrees. FreeBSD has had its share of bugs and sub-standard behaviours, and Linux stable vs. unstable kernel is the same thing. Normally you would only see bug fixes in stable kernels.
Also let's not confuse Linux the kernel and the Linux distributions. Some are more BSD-ish and some are more SysV-ish and a lot are in-between. Yes it can be a mess but frankly (speaking about any particular distribution) you just have to know where things are and things are slowly getting better with the LSB. The BSDs of today are very different to the BSDs of the 90's, so it's not as if there is One True Way of doing things.
Anyway, there is a lot in your post that is true. I don't believe that Linux is superior to BSD or vice-versa, I think both lines have benefited from one-another. Certainly Linux has integrated a lot of BSD code (possibly re-written later) and certainly the BSD core group is technically very competent, but so is the Linux core group. Linux stole the limelight and the mindshare because of its more democratic philosophy and just slightly better timing, but now if I had to run an absolutely secure server I would run OpenBSD and if I were running an ISP I would certainly choose FreeBSD, and if I needed my old Mac to run some kind of Unix I would use OpenBSD, but my main machine is a DELL desktop, for better of for worse there is better desktop support with Linux, currrently.
Now all my sparc machines run Solaris and there is no way I would consider running anything else on them, but frankly I don't find the Solaris system exciting. Yes I know of the companion CDs, my experience with that sort of things is that if it comes on a CD in a commercial package, it's too old.
All Solaris provides is the bare system, not even with a C compiler these days. I know there is a repository of free solaris pre-compiled package, but from Australia it is unusable (downloads at 0.5kb, not kidding). I don't know of any working mirror. It gets quickly terribly old to compile everyting.
In spite of the quality of the O/S, Sun is not doing terribly well these days, so there must be something wrong with their model too, don't you think?
Having worked with all sorts of Unices including the BSDs and the commercials, I find Linux the nicest because it doesn't have any silly limits. Tried to use your `free' x86 Solaris on an SMP machine? How is SMP on FreeBSD? What about IA-64 or x86-64 support, any of that?
While *BSD is Free, this is not the case for Solaris. End of story for many people.
The whole argument bandied around here is that sysadmin have been installing RHE because it's PHB approved and pretty much the same as the RHL 7.x they have at home.
Now the `at home' bit is going. We'll see in a few months time if that was a smart decision.
> If we ever get one of those babies, the world > changes completely overnight.
Probably not, because indications are that quantum computers will be very hard to program, and we already have pretty good methods for solving many NP problems approximately. They will probably be expensive too, at least in the beginning.
And who would pay for the thousands of servers, the huge quantity of RAM, the electricity bill, the rent, etc? Google is a lot more than just clever algorithm, it's also a very fast caching system of pretty much the whole internet, and the two go together.
I don't think a distributed effort like SETI@Home would cut it there.
And that is somehow anti-semitism on the part of the French, is it?
Mahatir has made these comments many times. If you listen to them instead of jumping on the outrage bandwagon, you'll hear that he urges the Muslim world to do the SAME thing as Israel: develop technology, push science and the arts, become educated, shake off the traditional fatalism.
He is saying that the Jews have an amazingingly effective lobby group (which is obvious in the US and elsewhere). This has allowed them to survive for centuries despite low numbers and general hostility.
Because he framed that as a struggle between the Muslim and the Jewish world (it's not as if there isn't a conflict at the moment), he should be condemned?
In the middle ages the Muslim world was a beacon of peace and culture, there is no reason why it couldn't be one now.
Mahatir doesn't like Israel, that much is certainly true. Note that when Israel uses its military power to kill a few Hamas leaders and a lot of innocents bystanders, the US never condemns them either.
> The trick is telling if a C is slightly off when > heard in isolation to other notes/instruments.
That's what perfect tone is. Normally trained musicians learn to do dictation which is basically the ability to write down what you ear, and one can definitely train for that. I've done it for years. In advances courses you get to write down multiple voices, chords, etc.
In some courses pupils are trained to hear comma differences (that's a 9th of a full note, or the difference between a C# and a Dflat). Violonist do it all the time precisely because a C# and a Dflat sound different and in an orchestra they must play the right one or it sounds bad.
Since you are a pipe organist then you must know aobut the different ways to tune an organ BTW. Organs often aren't tempered like pianos.
However telling a note in isolation just like you said is not considered something one can train for (although recent studies dispute that). One can either do it or not. This is what happened that day and it has been the first and only time when I've seen someone do it.
Tuning one's guitar is easy once the top E is correct.
Thanks for taking the time to respond, first of all.
By brilliant I mostly meant that the *result* of the code, the final product, shows something rarely seen in run-of-the-mill projects. Perhaps one very good idea in GUI design or an enlightening feature.
Many people disparage open-source as being a kind of a copy exercise (say KDE copying the Windows look, etc). In fact the very best open-source projects (say Perl or Gcc) have very unique qualities.
To some degree you can find those qualities in lesser projects. For example the guy who designed FLTK, a very simple GUI toolkit (www.fltk.org) did this on company time (Digital Domain, they did the rendering of Titanic). He had a very clear idea of what a GUI toolkit should be like and how it should be implemented thanks to his extensive experience with other toolkits in previous employments. The result is a simple, compact and yes, brilliant product. The style of the code itself is irrelevant. He designed this system because at the time (before GTK and QT became the things they are now) DD was rolling out *hundreds* of linux workstations to do 3D work and they needed a toolkit that would work the same on Irix and Linux.
To me it shows this guy is super-competent. If you saw his CV, there might just be the line `designed GUI system for DD' and you would maybe start thinking: why didn't this guy use MFC or something?
However I take your points, they are perfectly sensible. Thanks for reading so far.
I'm thinking something along the line of the commonest PC you can buy with DELL, say. That Amstrad was nothing special, litterally the cheapest thing I could buy at the time. I also could have bought an industrial-strength computer at the time, but that's not what this was.
What about the monitor? I've had my share of monitor failures and even one caught fire. This PC still has the original monitor and still works well (very very greasy ON button though).
To date the outlook viruses seem to be a greater and more annoying threat than bugs in recent versions of Bind.
Personnally I've recently been extremely annoyed with the latest outlook virus, even though I run 100% Linux and I have nothing to to with Outlook or Exchange.
The problem is that if someone's mail box (on an insecure win9x) machine is compromised then I'm starting to receive huge amounts of unwanted email if I'm on that guy's address list. That's something I can do exactly nothing about.
I'm not so sure that Bind is a problem today. It has a poor history but Bind 9 was a complete rewrite. Of course whoever wrote that alternative version of dns had only bad things to say about it but after all it's not as if he had no vested interest. To date his software is not running on top-level servers, is it?
Certainly sendmail is a bit similar in the sense that it has a poor history of hackability, but again it's not as if it has not seen a lot of usage, something which is very very hard to buy.
Slashdot Mirror
You can resell the GPLed software that RH puts out, but you are not permitted to use their trademark, logo and other copyrighted items. Those are not GPLed.
Hence if you take RH software, you can put it on a CD and sell it as the GPL says you can, but you cannot use the RH trademark, and therefore you cannot call it Red Hat Linux.
Think of the Coke recipe that Coca-cola published some time ago. You can take the recipe and put the result in bottle and selll it but you can't call it Coke.
Hi,
> Europeans see putting a check US power as paramount;
Agreed, and this may be short-sighted
> while Americans see ending the tyrany of brutal dicatator as most important.
No, America actually helped and continues to help local tyrants whenever it suits them (From Chile to Afghanistan right now). Has America ever installed a genuine democracy anywhere? Don't say Japan, because Japan's democracy is not working very well. The same party has been in power since WWII. Whenever democracy has flourished in a country it was in spite of the US and not thanks to them, such as in Europe and South Korea.
> Most Americans are dismayed why we deserve such mistrust.
See above.
Looking around on the web some more, Mt Pinatubo released very little C02, what it did was release a lot of ash which diffused the solar radiation and made it possible for plants to absorb atmospheric C02 for some reason.
So it had the inverse effect to what you are claiming. Well done.
Care to cite your source?
According to this paper, Mt Pinatubo's eruption had zero net effect on CO2 atmosphere content.
If it had the effect that you describe, you can bet your bottom dollar that everyone would know about it.
History teaches us that the greatest thieves and criminals run the world. Do you need an example?
I'm tired of this argument. Yes, people HAVE to use Microsoft products even if they don't want to, generally because of lack of choice and deliberate undermining of the competition by anti-competitive means.
You don't have to believe me, just read the court transcripts and findings, it's written there black on white. It was upheld on appeal so this is not even a discussion point anymore.
You can be as successful and agressive as you want until you become a monopoly, then the rules change.
> reduce the problems caused by the poor
> design of C itself.
C was designed to be a simple, powerful, low-level language, as easy to use and with as compact a compiler as Pascal, but with improved speed and flexibility.
With the territory came the ability to write more subtle and deadly bug, but that was understood.
So what language would you recommend to rewrite the Linux kernel in, say ?
Because the use of C was mandated by his employer.
Hi,
:-)
> I'm enjoying the idea that this interchange hasn't
> digressed into a ridiculous flame war. It's nice
> to have a reason to reply.
Yeah, same here. If you hadn't posted as an AC you would be on my friends list by now
I agree with you on the stability of the 2.4 series, it is not so bad now since about 2.4.18, and the RH kernel which I use were never affected because of Alan Cox' sterling work.
The BSD vs. SYSV stuff, appart from the classic vs. newish tree layout and the various different way to get things done as far as sysadmin is concerned, what is important is the API. I suspect (but I'm not sure) that the BSDs all support most of the sysv system calls such as shared memory, semaphores, etc, and probably the sysv signals too. They all are fine features, so again all of that is a matter of shades of grey.
Normally Linux `aims' for POSIX compliance, which I think is fine. In my apps I always try to code to ISO C/C++ and then Posix, and all the rest goes into the "annoying non-portable stuff", which includes the sysv-specific calls, win32 calls, and BSD specific when they exist.
LSB is not a pipe dream, it's there and all major distributions are compliant. The problem is that it is not enough, it formalizes only a tiny portion of a Linux system and there is still a lot to do, and making so many vendor agree is a major pain.
I forgot to agree with you earlier with the fact that a Solaris box makes a splendid NFS server and excellent NFS clients too (caching NFS is great).
For the companion CD, being out of date for bash or ls is of no consequence to me either, but being up to date with gcc and gdb is very very important to me as I code in C++, and gcc has been making progress at a huge rate with that language in the last couple of years.
Also lately I've really appreciated being able to keep up with the KDE developments. The difference between KDE 2 and 3 is between unusable ugliness and a usable fine desktop, and with KDE3, one really really needs those updates as soon as they come, and recompiling KDE is an impossible task for me, not worth the effort.
About the `linux for the masses' with support for crappy hardware and the fact that Linux is not innovative, this is true to a degree, but there are Linux true innovation. My favourite example are the Linux virtual consoles. The BSDs have them too now but I believe they originated with Linux and the commercial Unices never had them AFAIK. They are incredibly useful (multiple X sessions are also a godsend). The whole development process is also an innovation, with an open team and a very visible process. Historically both GNU and BSDs had more closed teams (not necessarily a disadvantage).
Myself I don't really care if Microsoft continue to dominate 95% of the desktop as long as the remaining 5% are viable, i.e. as long as I can install some kind of Unix on the DELL my workplace puts on my desk. I've always been lucky to be able to have a Unix desktop machine for the last 15 years or so (NeXT, Sun, DEC, and Linux), and really I don't want to go back. If being viable means that we have to compete on usability and we have to put up with some degree of zealotry and marketspeak, then so be it because I don't think Microsoft will be playing nice and allow any kind of competition to survive if they have their way.
The question I'm asking myself is would have the BSDs been in the position they are in now, including Darwin, without Linux? Would FreeBSD or some other fork have taken Linux's position had Linus not posted his kernel on a mailing list? or would have this whole movement not gelled at all and we would all be running Windows now?
As far as Apple is concerned, I was lucky enough to have a NeXT workstation in the early 90s and really this was at the time the Unix with the prettiest face and most consistent and innovative design. Apple simply inherited that wholesale and added very little. They had a head start. Also Apple does
Yes RedHat has really dropped Free Linux.
Fedora is not just a change of name, it comes with no company-backed guarantees whatsoever (just community support), it's just a showcase and a beta distribution to get the enterprise packages tested in the community. Before the free RH had a small, but sufficient guarantee of support. Fedora doesn't even have the name.
For tech-savvy individuals it does not matter too much although they might fear the constant upgrade treadmill and the potential unstability, but these guys are a minority. For the real tech-savvy individual there is no shortage of choice and Fedora is just one of them anyway.
For not-super-rich corporations and institutions such as colleges, it is a disaster. They cannot afford the unfriendly per-seat licensing scheme of the RH enterprise products (even the cheaper ones), they loathe the EULA (it makes them auditable), and they've just lost the PHB-friendly support from RH.
Note this: it does not matter that Fedora provides updates of the highest quality. The PHBs will see this as an amateurish effort at best, easily hijacked at worst and will simply forbid this to run in their enterprise. Note that you cannot buy a small number of RHEL licenses and install it everywhere, the licensing agreement forbids it.
In other words this is the end of RedHat everywhere. People will be better off running stable Debian or *BSD because they have a track record of reliability whereas Fedora has nothing.
Soon the Enterprise solutions will follow them in the dump because no one will bother learning RH anymore. Current RHCEs are pissed off and will be angry at RH for devaluating their effort.
There is a high degree of probability that RH is throwing the baby with the bathwater and will be finding itself in the same league as the proprietary Unix vendors such as BSDI and SCO.
Myself I plan to evaluate Fedora when it come out, at home, but I won't touch my work RH9 installations until shortly before EOL. Then I'll probably move to something else, SUSE being a strong candidate, unless I am proved wrong with Fedora.
Fedora has a *very* short time to prove itself worthy.
Roman philosophers:
Juvenal, Cicero, Epictetus, Marcus Aurelius (also emperor), St Augustine, Hypatia of Alexandria (a woman).
A mathematican:
Proclus, wrote a commentary on Euclid which is the major source of Euclid's writing in modern days.
I'm sure more can be found on Google
You comment highlights the difference between the BSD and the Linux philosophies, even though you are using Microsoft as an example.
> Not to pat Microsoft on the back, but I've
> never heard of them releasing a product
> where they need to experimentally change the
> kernel and have an end user "try it out."
I'm sure they do that in RC mode, just like Linux.
If you've ever had an incident report with Microsoft, once out of trivial ideas they get you to try experimental patches too. With Microsoft everybody is a beta tester, except you get to pay for the product too.
In general this is a different development model, with Linux this sort of thing is condoned,certainly in development kernels: "release early, release often", etc. I grant you that the BSD philosophy is different, there is a core group that try to get things polished before they are released. Conversely it's reputedly hard to get even small changes accepted in the BSD kernels. My experience with sending patches to the Linux kernel has been very positive, although I haven't done that recently.
Really however this is just a matter of degrees. FreeBSD has had its share of bugs and sub-standard behaviours, and Linux stable vs. unstable kernel is the same thing. Normally you would only see bug fixes in stable kernels.
Also let's not confuse Linux the kernel and the Linux distributions. Some are more BSD-ish and some are more SysV-ish and a lot are in-between. Yes it can be a mess but frankly (speaking about any particular distribution) you just have to know where things are and things are slowly getting better with the LSB. The BSDs of today are very different to the BSDs of the 90's, so it's not as if there is One True Way of doing things.
Anyway, there is a lot in your post that is true. I don't believe that Linux is superior to BSD or vice-versa, I think both lines have benefited from one-another. Certainly Linux has integrated a lot of BSD code (possibly re-written later) and certainly the BSD core group is technically very competent, but so is the Linux core group. Linux stole the limelight and the mindshare because of its more democratic philosophy and just slightly better timing, but now if I had to run an absolutely secure server I would run OpenBSD and if I were running an ISP I would certainly choose FreeBSD, and if I needed my old Mac to run some kind of Unix I would use OpenBSD, but my main machine is a DELL desktop, for better of for worse there is better desktop support with Linux, currrently.
Now all my sparc machines run Solaris and there is no way I would consider running anything else on them, but frankly I don't find the Solaris system exciting. Yes I know of the companion CDs, my experience with that sort of things is that if it comes on a CD in a commercial package, it's too old.
All Solaris provides is the bare system, not even with a C compiler these days. I know there is a repository of free solaris pre-compiled package, but from Australia it is unusable (downloads at 0.5kb, not kidding). I don't know of any working mirror. It gets quickly terribly old to compile everyting.
In spite of the quality of the O/S, Sun is not doing terribly well these days, so there must be something wrong with their model too, don't you think?
Having worked with all sorts of Unices including the BSDs and the commercials, I find Linux the nicest because it doesn't have any silly limits. Tried to use your `free' x86 Solaris on an SMP machine? How is SMP on FreeBSD? What about IA-64 or x86-64 support, any of that?
While *BSD is Free, this is not the case for Solaris. End of story for many people.
The whole argument bandied around here is that sysadmin have been installing RHE because it's PHB approved and pretty much the same as the RHL 7.x they have at home.
Now the `at home' bit is going. We'll see in a few months time if that was a smart decision.
Hi
> If we ever get one of those babies, the world
> changes completely overnight.
Probably not, because indications are that quantum computers will be very hard to program, and we already have pretty good methods for solving many NP problems approximately. They will probably be expensive too, at least in the beginning.
You rich person, I raise you a `bc'
7 166243825799642490473 837803842334832839\7 558340890106714439262 837987573438185793\7 654370999834036159013 438371831442807001\6 723346843445866174968 079087058037040712\8 029006686938976881787 785946905630190260\9 644305902501597239986 771421554169383555\0 878118726394964751001 890413490084170616\9 550769983616369411933 015213796825837188\2 555022599830041234478 486259567449219461\3 800876086221028342701 976982023131690176\0 285375124784014907159 135459982790513399\7 288427979155484978295 432353451706522326\6 877828789484406160074 129456749198230505\2 916136926708342856440 730447899971901781\9 599609079946920177462 481771844986745565\8 075682218465717463732 968849128195203174\8 411929804522981857338 977648103126085903\1 113160292078173803343 609024380470834040\
% bc -l
bc 1.06
Copyright 1991-1994, 1997, 1998, 2000 Free Software Foundation, Inc.
This is free software with ABSOLUTELY NO WARRANTY.
For details type `warranty'.
2^4096
1044388881413152506691752710
5390797155745684882681193499
6072632360878513652779459569
1855946226376318839397712745
8404874011860911446797778359
9405995794534328234693030266
9885291486318237914434496734
7509366833385055103297208826
0918336567512213184928463681
7023806505913245610825731835
7800667519548507992163641937
6115517942711068311340905842
9061394905987693002122963395
7164237715481632138063104590
4657634732238502672530598997
9250178329070473119433165550
5700244092661691087414838507
0013024134671897266732164915
3154190336
Note that the result is exactly the same.
And who would pay for the thousands of servers, the huge quantity of RAM, the electricity bill, the rent, etc? Google is a lot more than just clever algorithm, it's also a very fast caching system of pretty much the whole internet, and the two go together.
I don't think a distributed effort like SETI@Home would cut it there.
Cheers.
It takes the better part of 5 minutes to get going with CVS. I don't believe this is an issue.
CVS might be cumbersome for things like branching, but for the casual contributor it is a simple thing
% cvs co
[edit like mad]
% cvs commit
[to make your changes available]
% cvs update
[to get the other contributors changes]
The End.
I don't notice fluo flicker when the tube is new, I do when it gets old, I suspect it has to do with something else than just main grid frequency.
I do notice 60Hz monitor refresh rates and that's unbearable. 75Hz is Okay and 85Hz is perfect (for me). I think the cap at 60Hz might be a bit low.
How long before people work out how to uncap the refresh rate?
And that is somehow anti-semitism on the part of the French, is it?
Mahatir has made these comments many times. If you listen to them instead of jumping on the outrage bandwagon, you'll hear that he urges the Muslim world to do the SAME thing as Israel: develop technology, push science and the arts, become educated, shake off the traditional fatalism.
He is saying that the Jews have an amazingingly effective lobby group (which is obvious in the US and elsewhere). This has allowed them to survive for centuries despite low numbers and general hostility.
Because he framed that as a struggle between the Muslim and the Jewish world (it's not as if there isn't a conflict at the moment), he should be condemned?
In the middle ages the Muslim world was a beacon of peace and culture, there is no reason why it couldn't be one now.
Mahatir doesn't like Israel, that much is certainly true. Note that when Israel uses its military power to kill a few Hamas leaders and a lot of innocents bystanders, the US never condemns them either.
> The trick is telling if a C is slightly off when
> heard in isolation to other notes/instruments.
That's what perfect tone is. Normally trained musicians learn to do dictation which is basically the ability to write down what you ear, and one can definitely train for that. I've done it for years. In advances courses you get to write down multiple voices, chords, etc.
In some courses pupils are trained to hear comma differences (that's a 9th of a full note, or the difference between a C# and a Dflat). Violonist do it all the time precisely because a C# and a Dflat sound different and in an orchestra they must play the right one or it sounds bad.
Since you are a pipe organist then you must know aobut the different ways to tune an organ BTW. Organs often aren't tempered like pianos.
However telling a note in isolation just like you said is not considered something one can train for (although recent studies dispute that). One can either do it or not. This is what happened that day and it has been the first and only time when I've seen someone do it.
Tuning one's guitar is easy once the top E is correct.
Thanks for taking the time to respond, first of all.
By brilliant I mostly meant that the *result* of the code, the final product, shows something rarely seen in run-of-the-mill projects. Perhaps one very good idea in GUI design or an enlightening feature.
Many people disparage open-source as being a kind of a copy exercise (say KDE copying the Windows look, etc). In fact the very best open-source projects (say Perl or Gcc) have very unique qualities.
To some degree you can find those qualities in lesser projects. For example the guy who designed FLTK, a very simple GUI toolkit (www.fltk.org) did this on company time (Digital Domain, they did the rendering of Titanic). He had a very clear idea of what a GUI toolkit should be like and how it should be implemented thanks to his extensive experience with other toolkits in previous employments. The result is a simple, compact and yes, brilliant product. The style of the code itself is irrelevant. He designed this system because at the time (before GTK and QT became the things they are now) DD was rolling out *hundreds* of linux workstations to do 3D work and they needed a toolkit that would work the same on Irix and Linux.
To me it shows this guy is super-competent. If you saw his CV, there might just be the line `designed GUI system for DD' and you would maybe start thinking: why didn't this guy use MFC or something?
However I take your points, they are perfectly sensible. Thanks for reading so far.
> My experience though is that VC++ rejects a lot
> more non-standards compliant code than g++,
In that sense then yes I agree. I always think of a compiler not accepting standard constructs as being more of a problem.
Did DDJ run g++ with `-ansi -pedantic' ? usually that helps a bit for the `type 2' errors that you are mentioning.
I'm thinking something along the line of the commonest PC you can buy with DELL, say. That Amstrad was nothing special, litterally the cheapest thing I could buy at the time. I also could have bought an industrial-strength computer at the time, but that's not what this was.
What about the monitor? I've had my share of monitor failures and even one caught fire. This PC still has the original monitor and still works well (very very greasy ON button though).
To date the outlook viruses seem to be a greater and more annoying threat than bugs in recent versions of Bind.
Personnally I've recently been extremely annoyed with the latest outlook virus, even though I run 100% Linux and I have nothing to to with Outlook or Exchange.
The problem is that if someone's mail box (on an insecure win9x) machine is compromised then I'm starting to receive huge amounts of unwanted email if I'm on that guy's address list. That's something I can do exactly nothing about.
I'm not so sure that Bind is a problem today. It has a poor history but Bind 9 was a complete rewrite. Of course whoever wrote that alternative version of dns had only bad things to say about it but after all it's not as if he had no vested interest. To date his software is not running on top-level servers, is it?
Certainly sendmail is a bit similar in the sense that it has a poor history of hackability, but again it's not as if it has not seen a lot of usage, something which is very very hard to buy.