If you really want "cheap", grab a Coppermine barebones complete with memory, floppy and a CD-ROM for about $200- (see cpusolutions.com, for example). Then migrate any usable spare parts (eg network cards, modems, soundcards etc) to the chassis. Add another $100- for a decent hard drive, and you've got a $300- computer. For $200 extra, you could use an Athlon barebones as a base, and you'd have a $500- computer. (no "after-rebate" smoke and mirrors here either)
Why is it that you believe so strongly in freedom, then proceed to try to restrict the freedoms of others; including that freedom to make business or personal decisions based on whatever criteria one wishes?
Neither of the major political parties care very much about "freedom" unless one of their pet special interest groups cares. There is bipartisan support for eroding the civil liberties of those who fall outside the safety net of a pet special interest.As for the hate crimes, why should any segment of the population, either homosexuals, blacks, hispanics, jews, christians or penguins deserve any special protections under the law. All crimes are hate crimes.
At least try to express your argument in well formed sentences. The "all-crimes-are-hate-crimes" line is false and demonstrates an ignorance of
the issue at hand. And it has nothing to do with
"special protections" for people who "belong to a certain group" (hint: if it did, it would be unconstitutional)
and it has been shown any number of times that the pitiful efforts to roll back greenhouse gas emissions will do nothing to stop the purported global warming.
Utter nonsense. In global warming forecasts presented to senate committees by leading scientists, the primary variable is global policy
on emissions.
This guy gets to be attorney general because the representatives of the majority of Americans decided he was suitable for the job.
ROFL. Not only do you not understand the issues
you beligerantly rant about, you don't understand neither
your own electoral system either.
You're completely missing my point. I'm not bashing KDE. What I'm doing is pointing out that it runs just fine on old hardware if you have enough memory (ie you don't need a 1.4G athlon to run it). It uses a fair amount of resources, but
that's the price you pay for functionality.
Personally, i find all these "desktop environments" are too heavy for older computers..
I've used KDE on pentium I class machines bfore, and it runs quite nicely. However, it's a bit of a memory pig, especially if you want to run (for example) Mozilla, Star Office, and other apps on top of it. To summarize, the desktop environments are memory pigs, but given enough memory, they work nicely.
I've been pricing these machines, having recently purchased a Tiger MP machine. The marginal cost of the second CPU was less than $400-, making it a clear winner. If you're using the machine for anything moderately parallelizable, you will get more bang for your buck with the Athlon MP machines.
There are only two choices of motherboard at present with the Palomino chips. There's the Tyan Thunder with everything on the board (eg dual scsi, dual network, requires a 460 Watt power supply), and a cheaper Tyan Tiger motherboard. IMO the Tyan Thunder is a big waste of money unless you absolutely require two onboard SCSI controllers and/or the onboard network.
... for mainstream acceptance. C++ compilers all have their own name-mangling schemes which makes different compilers fundamentally incompatible. So the compiler may be able to find a niche market among developers of proprietary applications, but it's very difficult to persuade people to switch from g++, since that's what all the preloaded C++ libraries that ship with Linux use.
Not everyone surfs to look for porn- there's a hell of a lot of information out there, and being able to do research in "real time," actually during a discussion or lecture can bring a lot of value to it.
Your point may have some merit for some classes, but it's certainly not true for all classes all the time. This is not an all-or-nothing proposition, the plan is to block for *some* classes.
Not really an option, most introductory classes are required attendance, even if it's not worth your time and you don't pay attention.
Then the right thing to complain about is required attendance. However, it's just plain dumb to expect the school to provide resources such as computers for your entertainment. And it's good policy on part of the schools to avoid placing diversions under students noses.
Do you think students get annoyed by professors who just read out of the book, don't make the lectures even worthwhile to hear?
Sure, but most professors don't do this. If a professor is awful, and forces you to attend, then that in itself is the problem. Complaints about the lack of entertainment material are again irrelevant and pointless, because the university is not there to entertain you. The real problem in this instance is with the quality of teaching.
For freshman just out of college that may be true, but a lot of people in college now have something called a life
In the end, it's better to actually try to improve classes and the college, so that the tendancy to do these "distracting" things would be reduced or even eliminated.
You're playing a false dichotomy card here. Why not do both -- improve quality of teaching, and eliminate distractions from classrooms ? Both are worthwhile goals. I disagree with your view that an interesting class will prevent students goofing off. Waving an easy diversion under someones nose is a great way to distract them, even if they're interested and motivated.
MhZ is not analogous to resistance. You want to sub 1/2.2 for resistance. You end up getting 1/4.4 for your answer which isn't surprising -- the "current" in each branch is the same if the flow rate is equal.
Asking for an ISO C++ compiler is asking for trouble, because such a thing does not exist. For example, how many compilers support export ? However, it's important to draw the line somewhere on features. Where you draw the line depends on how important it is to support broken compilers.
There is a secondary issue, regarding work-arounds for small annoyances. Autoconf does a good job of taking care of this. For example, an autoconf macro in configure.in like
this can be used to tell me whether I need to install an sstream header for old gcc versions, or if one is already installed.
Sorry, had to snip it, because of the lame "lameness" filter.
One can check other conditions and define macros,
or automatically edit Makefiles based on outcomes of this and similar tests. GNOME and KDE packages
ship with a good collection of autoconf macros.
I have found these very useful.
I test regularly on g++ versions 2.95 and up. Less frequently, I try to build against g++ 2.91xx. The compiler is portable, and using the same compiler makes life simpler. Even within this narrow framework, there are portability issues. For example, earlier gcc versions do not ship with the sstream header. The streams library is broken in early versions. For examploe,
int is used instead of appropriate types. The best way to manage these subtle annoyances is to use autoconf. Libtool is also essential, for a different reason: the commands used to link vary wildly from platform to platform.
I think your problems partly answer the question. You need to make sure that your software runs on your users platforms. If you can't find shell accounts on a given OS, it's probably because your users aren't using it. If your users are using a given OS, but can't/won't give you a shell account, let them test/port to that OS.
I'd think that if you've got big endian/little endian platforms, and SysV/BSD, that's a pretty good start.
Did I mention, use autoconf, and it will make porting to anything else much easier...
My problem is that I *never* have whatever library the program I'm trying to install requires, and all too often, I have no idea where I might *find* the library.
www.rpmfind.net. You'll also find that a lot of the problems are caused by binary incompatibility. Binary compatibility problems don't have any easy solution.
Why don't developers offer a version of their packages *with* every libary you may need?
Because such a package would be enormous.
A better approach would be to just link to the required libraries.
Of course, then you have to worry about how well that script works, and if it'll overwrite my libraries with older / non-working libraries
Again, binary compatibility. If this is a big problem, just compile from the source. Or get versions of the library that were compiled against your distribution.
I guess I'm just confused...what have you wanted to do with, say, MS Dev Studio or KDevelop, that these tools wouldn't let you do?
Create and use dl-modules. Almost impossible to get this right with automake, let-alone from KDevelop.
Or maybe I'm missing the point. I like having an IDE manage the makefile.
autoconf/automake takes care of a lot of this. Automation doesn't require a GUI.
And really, when it comes to designing a window or a dialog, having a graphical tool to do that layout is really nice.
Depends on what you're doing. The trivial stuff (ie anything where you know the layout at compile time) is
very easy to do with a GUI builder. But it's also very easy to code by hand. I've found most of my time in GUI programming is spent coding dynamically generated widgets.
or coding, you need a more modular language, something less complex. The semi-linguistic grunts and signs of a Neanderthal, or Koko the signing ape,may be more useful.
I believe it's already been done. See the posix utilities. A pretty good vocab: (try pronouncing these as spelled) awk, grep, ls, su, tar, stty, zcat, rm, mknod, cp, rmdir, chmod
Then there's the functions: fchmod, msgsnd, msgop, msgget, mmap,... etc.
CORBA is NOT heavywieght. Or at least, it doens't have to be. It's about the same overhead of a virtual function call in C++, if done right.
KDE were using CORBA, but found that it was too big and slow. But that was because they were using a big and slow ORB!
The size of the ORB was not the problem, I believe the size of the resulting executables was. Mico is STL based, and I think the problem may have something to do with template bloat (the STL classes that ship with gcc compile to 50-100k *per instance*)
But MonsterHut argued that they should be allowed to continue unless 2% of the 69 million emails sent came back as complaints!
This argument is obviously bogus, and I wish the judge would see through it. The problem is that most people obviously won't complain. This is actually a back-door opt-out scheme -- unless users explicitly complain, it's assumed that they've opted in.
It would be interesting to take a random sample of say 1000 people on the mailing lists, ask whether they solicited mail from monster-hut, and see if more than 98% of the replies assert that they did !
Since I'm writing code, the Linux kernel compilation benchmark is a pretty good measure of the performance I can expect. And last I checked, AMD was beating the pants off intel here. I suppose an informed consumer needs to look at what they're actually going to do with the chip, and which one will produce the best results for them (of course, if they're playing games, the card is as important as the CPU)
What could be more wasteful than letting that connection sit all day doing nothing?
That's not wasteful at all. How can not using something be wasteful ?
No, there is nothing the cable company can do if you are using NAT or masq.
I don't think they'd care, as long as they don't see a lot of bandwidth absorbed by your connection. If your machine is sucking up all the bandwidth, they might notice something.
The better analogy would be you are walking down the street, you see a door is unlocked, walk in and take a wad of cash. _Then_ you report the door not being locked to the owners.
THere's a big difference. Files posted on a website are supposed to be downloaded. If they didn't want the perl script downloaded, they should not have posted it on a public website.
I've got a problem with that. I am not too sure your average vendor is capable of doing it right. They will insist on RedHat because it's easier to install, but when you say you want Debian, they look at you cross-eyed.
The vendors I've talked to are pretty flexible about choice of distribution. And most Linux vendors are not "your average vendor". "Your average vendor" is a windows-only shop.
IMO, vendors that are capable of configuring rackmount RAID systems with custom kernels can be trusted to set up a pee-cee.
Slashdot Mirror
If you really want "cheap", grab a Coppermine barebones complete with memory, floppy and a CD-ROM for about $200- (see cpusolutions.com, for example). Then migrate any usable spare parts (eg network cards, modems, soundcards etc) to the chassis. Add another $100- for a decent hard drive, and you've got a $300- computer. For $200 extra, you could use an Athlon barebones as a base, and you'd have a $500- computer. (no "after-rebate" smoke and mirrors here either)
Neither of the major political parties care very much about "freedom" unless one of their pet special interest groups cares. There is bipartisan support for eroding the civil liberties of those who fall outside the safety net of a pet special interest.As for the hate crimes, why should any segment of the population, either homosexuals, blacks, hispanics, jews, christians or penguins deserve any special protections under the law. All crimes are hate crimes.
At least try to express your argument in well formed sentences. The "all-crimes-are-hate-crimes" line is false and demonstrates an ignorance of
the issue at hand. And it has nothing to do with
"special protections" for people who "belong to a certain group" (hint: if it did, it would be unconstitutional)
and it has been shown any number of times that the pitiful efforts to roll back greenhouse gas emissions will do nothing to stop the purported global warming.
Utter nonsense. In global warming forecasts presented to senate committees by leading scientists, the primary variable is global policy
on emissions.
This guy gets to be attorney general because the representatives of the majority of Americans decided he was suitable for the job.
ROFL. Not only do you not understand the issues
you beligerantly rant about, you don't understand neither
your own electoral system either.
Loading apps has nothing to do with the desktop environment. BTW, how long does it take to "switch virtual desktops" on W2k ?
You're completely missing my point. I'm not bashing KDE. What I'm doing is pointing out that it runs just fine on old hardware if you have enough memory (ie you don't need a 1.4G athlon to run it). It uses a fair amount of resources, but
that's the price you pay for functionality.
I've used KDE on pentium I class machines bfore, and it runs quite nicely. However, it's a bit of a memory pig, especially if you want to run (for example) Mozilla, Star Office, and other apps on top of it. To summarize, the desktop environments are memory pigs, but given enough memory, they work nicely.
There are only two choices of motherboard at present with the Palomino chips. There's the Tyan Thunder with everything on the board (eg dual scsi, dual network, requires a 460 Watt power supply), and a cheaper Tyan Tiger motherboard. IMO the Tyan Thunder is a big waste of money unless you absolutely require two onboard SCSI controllers and/or the onboard network.
... for mainstream acceptance. C++ compilers all have their own name-mangling schemes which makes different compilers fundamentally incompatible. So the compiler may be able to find a niche market among developers of proprietary applications, but it's very difficult to persuade people to switch from g++, since that's what all the preloaded C++ libraries that ship with Linux use.
Cheers,
Your point may have some merit for some classes, but it's certainly not true for all classes all the time. This is not an all-or-nothing proposition, the plan is to block for *some* classes.
Sure it's your choice. But it's also the schools choice whether to provide students with entertaining diversions or not.
Then the right thing to complain about is required attendance. However, it's just plain dumb to expect the school to provide resources such as computers for your entertainment. And it's good policy on part of the schools to avoid placing diversions under students noses.
Do you think students get annoyed by professors who just read out of the book, don't make the lectures even worthwhile to hear?
Sure, but most professors don't do this. If a professor is awful, and forces you to attend, then that in itself is the problem. Complaints about the lack of entertainment material are again irrelevant and pointless, because the university is not there to entertain you. The real problem in this instance is with the quality of teaching.
For freshman just out of college that may be true, but a lot of people in college now have something called a life
In the end, it's better to actually try to improve classes and the college, so that the tendancy to do these "distracting" things would be reduced or even eliminated.
You're playing a false dichotomy card here. Why not do both -- improve quality of teaching, and eliminate distractions from classrooms ? Both are worthwhile goals. I disagree with your view that an interesting class will prevent students goofing off. Waving an easy diversion under someones nose is a great way to distract them, even if they're interested and motivated.
MhZ is not analogous to resistance. You want to sub 1/2.2 for resistance. You end up getting 1/4.4 for your answer which isn't surprising -- the "current" in each branch is the same if the flow rate is equal.
There is a secondary issue, regarding work-arounds for small annoyances. Autoconf does a good job of taking care of this. For example, an autoconf macro in configure.in like this can be used to tell me whether I need to install an sstream header for old gcc versions, or if one is already installed. Sorry, had to snip it, because of the lame "lameness" filter.
One can check other conditions and define macros, or automatically edit Makefiles based on outcomes of this and similar tests. GNOME and KDE packages ship with a good collection of autoconf macros. I have found these very useful.
I test regularly on g++ versions 2.95 and up. Less frequently, I try to build against g++ 2.91xx. The compiler is portable, and using the same compiler makes life simpler. Even within this narrow framework, there are portability issues. For example, earlier gcc versions do not ship with the sstream header. The streams library is broken in early versions. For examploe, int is used instead of appropriate types. The best way to manage these subtle annoyances is to use autoconf. Libtool is also essential, for a different reason: the commands used to link vary wildly from platform to platform.
I'd think that if you've got big endian/little endian platforms, and SysV/BSD, that's a pretty good start.
Did I mention, use autoconf, and it will make porting to anything else much easier
www.rpmfind.net. You'll also find that a lot of the problems are caused by binary incompatibility. Binary compatibility problems don't have any easy solution.
Why don't developers offer a version of their packages *with* every libary you may need?
Because such a package would be enormous.
A better approach would be to just link to the required libraries.
Of course, then you have to worry about how well that script works, and if it'll overwrite my libraries with older / non-working libraries
Again, binary compatibility. If this is a big problem, just compile from the source. Or get versions of the library that were compiled against your distribution.
Create and use dl-modules. Almost impossible to get this right with automake, let-alone from KDevelop.
Or maybe I'm missing the point. I like having an IDE manage the makefile.
autoconf/automake takes care of a lot of this. Automation doesn't require a GUI.
And really, when it comes to designing a window or a dialog, having a graphical tool to do that layout is really nice.
Depends on what you're doing. The trivial stuff (ie anything where you know the layout at compile time) is
very easy to do with a GUI builder. But it's also very easy to code by hand. I've found most of my time in GUI programming is spent coding dynamically generated widgets.
I believe it's already been done. See the posix utilities. A pretty good vocab: (try pronouncing these as spelled) awk, grep, ls, su, tar, stty, zcat, rm, mknod, cp, rmdir, chmod
Then there's the functions: fchmod, msgsnd, msgop, msgget, mmap,
KDE were using CORBA, but found that it was too big and slow. But that was because they were using a big and slow ORB!
The size of the ORB was not the problem, I believe the size of the resulting executables was. Mico is STL based, and I think the problem may have something to do with template bloat (the STL classes that ship with gcc compile to 50-100k *per instance*)
This argument is obviously bogus, and I wish the judge would see through it. The problem is that most people obviously won't complain. This is actually a back-door opt-out scheme -- unless users explicitly complain, it's assumed that they've opted in.
It would be interesting to take a random sample of say 1000 people on the mailing lists, ask whether they solicited mail from monster-hut, and see if more than 98% of the replies assert that they did !
It's a good start, and the goatse.cx link is a winner. However, I was disappointed by the absence of references to "Natalie Portman" and "hot grits".
That's not wasteful at all. How can not using something be wasteful ?
No, there is nothing the cable company can do if you are using NAT or masq.
I don't think they'd care, as long as they don't see a lot of bandwidth absorbed by your connection. If your machine is sucking up all the bandwidth, they might notice something.
THere's a big difference. Files posted on a website are supposed to be downloaded. If they didn't want the perl script downloaded, they should not have posted it on a public website.
TeX is free, and it's certainly very original
The vendors I've talked to are pretty flexible about choice of distribution. And most Linux vendors are not "your average vendor". "Your average vendor" is a windows-only shop.
IMO, vendors that are capable of configuring rackmount RAID systems with custom kernels can be trusted to set up a pee-cee.
My solution is to forget it all
See: "False dichotomy".