Sip has NO documentation, but it has the advantage that it's used for the Python Qt bindings. It's reasonably easy to use, though it really could do with some documentation.
But why use STL if you're already using Qt ? Qt already includes several container classes, and their cost per instance is considerably smaller than that of STL
If the university owns the copyright -- which, under their statutes, they do -- then I can't release the software under any license.
If the software is copyrighted under a free license, then there's no such problem. The real question is whether or not the university is willing to allow you to do this.
1. How often does this theft really occur? I mean, does it really happen all of the time?
Not very often, I don't think. Most of the faculty are decent people.
That is, will people (mentors, Ph.D.'s, etc.) be more reluctant to take on students knowing that students might actually turn the tables and claim rights to their work?
There are already more PhD students than we need, to the point where some schools "import" students because they can't get enough US nationals to apply. This isn't going to destroy PhD programs.
3. Would there have been a better non-legal solution?
A university policy that is consistent with the law would certainly help.
Which means that I immediately have to stop work on any Free Software projects, because by licensing my work Freely I would be violating the university statutes.
I'm not clear that this follows. Why couldn't you release the software under your favourite license, and assign the copyright to the university, for example ? I believe there are others at Oxford who have released software, btw. (I develop fMRI software, and IIRC there's a group at Oxford that do the same) You may want to look into this, and contact them. Ah, here it is:http://www.fmrib.ox.ac.uk/~mark/
Adding increased levels of bureaucracy to this process just guarantees students will be passed over, and more experienced researchers used.
But the graduate schools put our too many PhDs now anyway, because everyone wants to have a graduate program. Some of the schools even recruit primarily foreign students, because there aren't enough US nationals to fill all the spots! I doubt this case will kill the graduate programs.
Shouldn't an artist be able to contract the pressing of their own CD's with out any of these money-hungry acronyms pumping them for money from the sale and distrobution?
They can, but it's a fair bit of work, and most of them delegate to a record company. Still, perhaps they could have a manager as an extra band member to take care of business affairs.
so why not change to the system to undercut all these middle men and make more MONEY.? Answer: Contracts.
Well, they could simply choose not sign! If your proposed business model is so great, more people would try it. It's plausible that more people will try a more direct distribution method. It does seem to have its advantages, and it has worked for some companies (Dell is a good example) I don't think ther's a sinister conspiracy at work, more like a slow embrace of technology.
Oh wait, I can't be taken into custody on something as flimsy as suspicion just because I was pulled over.
Don't be so sure.
Are you familiar with the recent supreme court case, where a woman was arrested on a traffic offense ?
Re:KDE is the environment of choice
on
KDE Wins 3 awards
·
· Score: 1
I don't mean the "skin" so much as I mean the choice of UI gadgets and the way they are arranged.
I haven't seen anyone succesfully demonstrate that kwin is any more like Windows than it is like (for example) CDE. Desktop icons and bars are not unique to Windows, and weren't invented by Microsoft.
Re:KDE is the environment of choice
on
KDE Wins 3 awards
·
· Score: 1
But it *does* look like Windows! Just looks at the screenshots of KDE 2.x at kde.org!
It's configurable. That means it can be made to look like windows. That's very different from saying it "does look like windows".
Re:KDE is the environment of choice
on
KDE Wins 3 awards
·
· Score: 2
Its ironic isn't it that people slag off Windows left right and center on here but bring out a
WM on Linux that looks exactly like it
KDE is not a window manager. It is (a) an application development framework, and (b) a suite of applications that uses that framework.
Repeating these lies that KWM (not KDE, which, AGAIN, is not a Window Manager) "looks like Windows" won't make them come true. kwm is configurable and can be made to look like a whole bunch of different things. One of those things resembles Windows (but them, Windows doesn't have virtual desktops)
Fine, but I don't want to use something that looks like Windows out of the box
Actually, it runs a "setup wizard" out of the box, and how it looks after that depends on your choices. My choices resulted in something that looked a lot like CDE.
and have to hack around for 30 mins trying to
make it look like afterstep when I can just use
afterstep itself!
Well of course. If you want to use afterstep, use afterstep. But don't pretend that KDE is like Windows.
There's no comparison between KDE and fvwm. KDE is an application development framework, and a suite of applications, one of which is a window manager, built on that framework. fvwm is a window manager.
KDE is a much bigger, and much more significant project, which is why it gets more press.
On a number of occasions I've attempted to post code to
do something useful, that was related to the story or a parent comment, and it got rejected because
it compressed too well.
In fact the "lameness" filter does a fantastic job at filtering out code.
I've had the same problem, and I'd rather live with the ASCII art. People who care can always browse at +1...
It's especially frustrating because I really, REALLY try not to spout off unless I've got something
to add to the conversation. So much for choosing quality over quantity.:-(
I haven't noticed an overwhelming correlation between posts of mine that are insightful, and posts that are marked as such. One thing I have observed is a snowball effect -- posts that are marked up to 3 are likely to go up to 5.
The Linux solution beats out Sun solutions in price hands down. You are talking $30,000 per box for the minimal Sun allowed hardware requirement for the Sun Cluster software with the Oracle Parallel DB runtime licenses (this has changed with v3 and so have the hw requirements).
But who says you need a Sun cluster anyway ? Couldn't one just get a single Sun box ? The Sun model seems to be that you get a big machine, and scale it up by pouring CPU juice into it.
Of course, the big-iron machines from Sun are fairly pricy though -- I agree with your main point, that Sun is expensive
I like the way you need to use stupid hacks to get around the "disappearing console" problem. I espeically like the way Borland advocate that CS 101 students hack around the problem using non-standard code.
As for the mandatory MS bashing, it's interesting to note that prior to 3.0, g++ has a hideously broken standard library (namespace std doesn't work correctly, the streams only superficially resemble the standard, it ships without sstream. g++ 3 is a big improvement, I just hope they can iron the bugs out so that the distributions start shipping it as the default.
The answer is obvious -- because it correlates with performance, in fact within the one make of CPU, it's a pretty reliable gauge of CPU performance.
Using clock speeds to compare different types of CPUs however is just plain dumb.
MHz is MHz, regardless of whether or not it's a true measure of
performance.
It can be verified quantitatively with a counter.
First, depending on how the PR is defined, it's possible that this can also be measured (one would hope so)
And if MHZ not a "true measure of performance", is it not misleading to use it to describe the performance of a chip ?
And, as a customer, I can see that they're
not technically lying, but they're trying to pull the wool over my eyes with the fine print.
Encouraging customers to use MHZ instead of independent performance ratings to measure performance of different types of CPUs is pulling the wool over the customers eyes.
Using a performance rating, and moving away from MHZ is not (provided the PR is not loaded or misleading, of course).
Yes but at least Intel is telling the truth about their processors.
So are AMD. The truth is that the AMD XP 1900 performs at least as well as the intel P4 1900
As for the issue about explaining AMDs PR, well when it comes to comparing the XP to the Pentium,
the PR is no more or less arbitrary than the CPU clock speed. Either way, you'd still need to explain which intel chip is comparable to the "XP 1533" . Sounds to me like you consider your friends too stupid to make intelligent buying decisions, and that they should become intel shills instead.
I realize I'm only one, but I'm accusing them of dishonesty--they're taking advantage of what they must hope is ignorance on the part of a large portion of their customer base that will never see their FAQ [amd.com] in which they admit that the "model numbers" aren't really clock speeds.
I don't think they're taking advantage of the ignorance of the market at all. The market expects an "AMD 1900" CPU to perform like an "Intel 1900" CPU. And it does. The "ignorant customers" in question use the number as a measure of performance. If anything, they're doing the "ignorant customers" a favor, by giving them a number that's a reasonably accurate measure of performance-- which is, after all, what the customers are looking for when they look at clock speeds.
But did they really lose credibility ? I don't think anyone's accuing them of dishonesty, or of overstating the numbers. If anything, they've been taken to task for understating them (since the 1800+ beats the P42G on most benchmarks)
Re:Messing things up or using Perl for what it fit
on
Perl6 for Mortals
·
· Score: 1
You can program OO using C, plenty of systems
have used it in the past.
Yes, you can write OO code in C, which doesn't in anyway contradict my point.
Re:Just what Perl needs - more syntax
on
Perl6 for Mortals
·
· Score: 1
WHy does that make Perl a good language?
It makes perl a useful language.
The library could equally rwrite the code in C/C++,
I don't see your point. Libraries do not generally "write code".
I agree with most of your response, and am happy that someone was so level-headed about it.
However, I write cron jobs all the time that use LWP, which would be a major pain if it were not
OO.
You make a good point here -- for small tasks, you can benefit from Perl's OO, just by using OO packages. I suppose the important point is that using perl objects is actually very easy, it's the implementation code that's a bit messy (with all the explicit $self-references)
Re:Messing things up or using Perl for what it fit
on
Perl6 for Mortals
·
· Score: 2
Who says I referred to OO support as an "esoteric feature"?
Using OO is not "programming in a C style". I extrapolated, perhaps incorrectly.
This is why I consider Perl
the wrong tool for large projects involving many programmers.
But not all projects are "large" and involve many programmers. I agree that Perl wouldn't be the best choice for something like KDE. But for me, Perl works nicely for small programs (note emphasis: the use of the word "program" implies the task at hand can't easily be solved by shellscript or sed !) It's also handy for writing a throwaway prorotype prior to coding something in a "ral" programming language.
Slashdot Mirror
Sip has NO documentation, but it has the advantage that it's used for the Python Qt bindings. It's reasonably easy to use, though it really could do with some documentation.
But why use STL if you're already using Qt ? Qt already includes several container classes, and their cost per instance is considerably smaller than that of STL
If the software is copyrighted under a free license, then there's no such problem. The real question is whether or not the university is willing to allow you to do this.
Not very often, I don't think. Most of the faculty are decent people.
That is, will people (mentors, Ph.D.'s, etc.) be more reluctant to take on students knowing that students might actually turn the tables and claim rights to their work?
There are already more PhD students than we need, to the point where some schools "import" students because they can't get enough US nationals to apply. This isn't going to destroy PhD programs.
3. Would there have been a better non-legal solution?
A university policy that is consistent with the law would certainly help.
I'm not clear that this follows. Why couldn't you release the software under your favourite license, and assign the copyright to the university, for example ? I believe there are others at Oxford who have released software, btw. (I develop fMRI software, and IIRC there's a group at Oxford that do the same) You may want to look into this, and contact them. Ah, here it is:http://www.fmrib.ox.ac.uk/~mark/
Cheers,
But the graduate schools put our too many PhDs now anyway, because everyone wants to have a graduate program. Some of the schools even recruit primarily foreign students, because there aren't enough US nationals to fill all the spots! I doubt this case will kill the graduate programs.
They can, but it's a fair bit of work, and most of them delegate to a record company. Still, perhaps they could have a manager as an extra band member to take care of business affairs.
so why not change to the system to undercut all these middle men and make more MONEY.? Answer: Contracts.
Well, they could simply choose not sign! If your proposed business model is so great, more people would try it. It's plausible that more people will try a more direct distribution method. It does seem to have its advantages, and it has worked for some companies (Dell is a good example) I don't think ther's a sinister conspiracy at work, more like a slow embrace of technology.
Don't be so sure. Are you familiar with the recent supreme court case, where a woman was arrested on a traffic offense ?
I haven't seen anyone succesfully demonstrate that kwin is any more like Windows than it is like (for example) CDE. Desktop icons and bars are not unique to Windows, and weren't invented by Microsoft.
It's configurable. That means it can be made to look like windows. That's very different from saying it "does look like windows".
Actually, it runs a "setup wizard" out of the box, and how it looks after that depends on your choices. My choices resulted in something that looked a lot like CDE.
and have to hack around for 30 mins trying to make it look like afterstep when I can just use afterstep itself!
Well of course. If you want to use afterstep, use afterstep. But don't pretend that KDE is like Windows.
do something useful, that was related to the story or a parent comment, and it got rejected because
it compressed too well.
In fact the "lameness" filter does a fantastic job at filtering out code.
I've had the same problem, and I'd rather live with the ASCII art. People who care can always browse at +1
I haven't noticed an overwhelming correlation between posts of mine that are insightful, and posts that are marked as such. One thing I have observed is a snowball effect -- posts that are marked up to 3 are likely to go up to 5.
But who says you need a Sun cluster anyway ? Couldn't one just get a single Sun box ? The Sun model seems to be that you get a big machine, and scale it up by pouring CPU juice into it.
Of course, the big-iron machines from Sun are fairly pricy though -- I agree with your main point, that Sun is expensive
As for the mandatory MS bashing, it's interesting to note that prior to 3.0, g++ has a hideously broken standard library (namespace std doesn't work correctly, the streams only superficially resemble the standard, it ships without sstream. g++ 3 is a big improvement, I just hope they can iron the bugs out so that the distributions start shipping it as the default.
The answer is obvious -- because it correlates with performance, in fact within the one make of CPU, it's a pretty reliable gauge of CPU performance.
Using clock speeds to compare different types of CPUs however is just plain dumb.
MHz is MHz, regardless of whether or not it's a true measure of performance. It can be verified quantitatively with a counter.
First, depending on how the PR is defined, it's possible that this can also be measured (one would hope so) And if MHZ not a "true measure of performance", is it not misleading to use it to describe the performance of a chip ?
And, as a customer, I can see that they're not technically lying, but they're trying to pull the wool over my eyes with the fine print.
Encouraging customers to use MHZ instead of independent performance ratings to measure performance of different types of CPUs is pulling the wool over the customers eyes. Using a performance rating, and moving away from MHZ is not (provided the PR is not loaded or misleading, of course).
So are AMD. The truth is that the AMD XP 1900 performs at least as well as the intel P4 1900
As for the issue about explaining AMDs PR, well when it comes to comparing the XP to the Pentium,
the PR is no more or less arbitrary than the CPU clock speed. Either way, you'd still need to explain which intel chip is comparable to the "XP 1533" . Sounds to me like you consider your friends too stupid to make intelligent buying decisions, and that they should become intel shills instead.
I don't think they're taking advantage of the ignorance of the market at all. The market expects an "AMD 1900" CPU to perform like an "Intel 1900" CPU. And it does. The "ignorant customers" in question use the number as a measure of performance. If anything, they're doing the "ignorant customers" a favor, by giving them a number that's a reasonably accurate measure of performance-- which is, after all, what the customers are looking for when they look at clock speeds.
But did they really lose credibility ? I don't think anyone's accuing them of dishonesty, or of overstating the numbers. If anything, they've been taken to task for understating them (since the 1800+ beats the P42G on most benchmarks)
Yes, you can write OO code in C, which doesn't in anyway contradict my point.
It makes perl a useful language.
The library could equally rwrite the code in C/C++,
I don't see your point. Libraries do not generally "write code".
I agree with most of your response, and am happy that someone was so level-headed about it.
However, I write cron jobs all the time that use LWP, which would be a major pain if it were not
OO.
You make a good point here -- for small tasks, you can benefit from Perl's OO, just by using OO packages. I suppose the important point is that using perl objects is actually very easy, it's the implementation code that's a bit messy (with all the explicit $self-references)
Using OO is not "programming in a C style". I extrapolated, perhaps incorrectly.
This is why I consider Perl the wrong tool for large projects involving many programmers.
But not all projects are "large" and involve many programmers. I agree that Perl wouldn't be the best choice for something like KDE. But for me, Perl works nicely for small programs (note emphasis: the use of the word "program" implies the task at hand can't easily be solved by shellscript or sed !) It's also handy for writing a throwaway prorotype prior to coding something in a "ral" programming language.