Suppose processor A costs $100 and gives 100 units of performance, but processor B costs $200 and gives 150 units of performance. Then the cost per performance is 1.0 unit/$ for processor A and 0.75 unit/$ for processor B. Right? WRONG!
Both of these processors need $500 of ancillary equipment in order to function. Therefore, a system with processor A gives 100 units for $600, or 0.167 units/$, whereas processor B gives 150 units for $650, or 0.231 units/$. This analysis shows that processor B is better value when speccing out a new system
But what about the case where you're just upgrading your cpu? Well, in that case it's moot to compare the AMD with the Intel processors, as you would need a new motherboard too. But simply dividing the performance by the cost of the cpu is meaningless here, too, because staying with your existing processor ($0) would give you a performance/price ratio of infinity.
Conclusion: you have to calculate your total outlay in order to figure out which cpu is the better value.
A system where apps have to be signed in order for the app to be deployable on an OS would make that OS a totally closed and a private shop. No developer would be allowed to develop apps for the OS unless some central body deems it in their interest.
I doesn't have to be a central authority - I'd be happy to get a signature from the Mozilla foundation themselves.
"Ah", you say, "you're back where you started: how do you know that the server where the signature resides hasn't been compromised?"
Well, digital signature files can be validated by another authority. I'm sure Red Hat, Novell, or IBM would be happy to countersign the Mozilla signature file. All they are doing is confirming that they have checked that this is indeed a signature that belongs to Mozilla - they don't have to sign every package.
In any case, a signature is a small file. The Mozilla signature could be published on Red Hat's, IBM's, Novell's websites, and I can check that they all agree. The more websites I check, the more confident I can be that this is an authentic signature file. I can afford to spend a little effort doing this, because I only need to accept the signature once for each application (or each application developer). This signature will not only be valid for the initial installation, but also for each upgrade or patch. Very useful if, say, you're compiling KDE snapshots from dozens of tarballs.
As for any software project who is too small to be on anyone else's radar: contact the developer themself to get the signature, and think hard before installing it system-wide. You can always compile and run it in an unprivileged user account dedicated to this software alone.
Even if the app was signed, it may be that the files were infected as they were being built
I sincerely hope that no developer uses their own machine as an FTP server. A project as important as Mozilla should only permit development on machines that are protected from the internet. If the story is true, it's much more likely that the FTP server was the weak link - it inevitably provides a much larger target to prospective crackers.
You shouldn't be trusting Slashdot articles either, given that this whole story looks like a work of fiction
Whether or not this particular case really happened, the fact remains that there is no safeguard aginst a web/FTP server being hacked. It certainly happened to Debian.
My way of avoiding this sort of problem is just to allow other users to play with new releases of software for a few weeks before installing it myself.
However, this does nothing to protect you if the original is only replaced by the trojaned version after a couple of weeks.
New versions of Firefox, for instance, never contain anything too exciting
Yet another example of the lamentable state of modern computer security. This wouldn't be a problem if operating systems required a trusted signature for software to be installed.
I use a lot of OS software (e.g. Firefox, NeoOffice/J, LyX, R), but the standard installation process on my platform (OS X) does not allow checking for an authentic signature. Why is this not built in? It doesn't have to be this way: for instance, Red Hat signs its own RPMs (though Debian's APT didn't support this last time I looked).
We already have to trust the developers. We shouldn't have to trust every FTP server too.
Indeed, the A320 is not so sophisticated that - when a Russian pilot let his 13 year old son 'take the wheel' - it could prevent him flying it into the ground from 25,000 feet.
Actually, the crash you are refering to was an A310, which is not fly-by-wire; crashes of the A320 series can be found here.
And when classical compilation CDs produced by small publishers (usually recordings of performances by east euro orchestras) many of these dying purists attacked them - again - for "diluting the value of these works."
Many in the IT industry complain about outsourcing to cheaper countries - are they also `dying purists'?
hence my wanting to contrast that trend with the one in the UK
Well, there's a distinction between the patterns in the current sex ratio, and the way that this sex ratio is changing over time. As I understood it, you were talking about the former, and we all agree that biology attracts a higher proportion of girls than other sciences. The instantaneous sex ratio doesn't allow you to distinguish between a sex-linked genetic disposition for the subject, and one which is shaped by society.
However, changes in the sex ratio do tell you something about society, because the biological factors presumably don't change over time. A reasonable null model is that the sex ratio in the sciences will become less male-biased over time, due to society's efforts to encourage women. I would suspect that this is the case in UK maths and physics lectures - i.e., the same trend as seen elsewhere. If you know any stats on this, I'd be interested to see them.
If an already male-biased sex ratio becomes even more so with time, in spite of efforts to remove anti-female sexism, then this is an extremely interesting result - and one which isn't explained by the sexes' natural affinity for the subject alone.
Girls are steered or scared away from Physics and Chemistry because of the reliance of Maths (another reason) and general Science, Technology and Engineering fields
The Story suggests that women are turning away from computer science in particular, while this trend is not reproduced in maths, physics, and chemistry. Other posters in this discussion say that there are women majoring in math but not in compsci.
Sure, the sex ratio is more female-biased in biology than in other sciences, but the issue here is that computer science is becoming still more male-biased with time, whereas other sciences are becoming less so.
Incidentally, the female biology students I know are much more interested in nature and conservation than in `nurturing' and medicine.
There have been a number of posts saying that this is because of girls' biological disposition against science and engineering - boys like lego(s) and girls like dolls, etc. However, the article (also quoted in the story) says that the underrepresentation in computer science is not repeated in other scientific disciplines. So, what is specific to computer science that is unattractive to women, compared to chemistry, maths, etc?
The article contains a quote that
girls--unlike boys--want jobs they believe can make a difference in society. But they don't view high tech as a key to that idealistic path.
but surely biotechnology is also `high tech', and I see no suggestion that women's representation is decreasing in that area.
So, what is it? At the risk of being modded flaimbait, is it perhaps that Physics, Chemistry, and Biology are somehow seen as more noble pursuits, that Computers are intrinsically a means to an end rather than an end in themselves? That CS majors are seen as an inferior type of geek relative to their cancer-curing, drug-designing, atom-smashing counterparts? Yet other branches of engineering (bridge building, rocket science) are also fundamentally concerned with solving practical problems, but somehow they don't carry the same stigma.
Is there, after all, something intrinsically semi-autistic, and therefore testosterone-linked, in fiddling with computers?
I suspect your other doom-laden predictions are equally overstated. Even your own
link to Forbes shows that Apple shares have increased more than $10 over the past 8 months (look at the share prices in the graph compared with the text of the article).
I wanted to see if there were a way to have OS X installed and be able to switch between the normal mac gui (only... aka no XFree86 or X11) to running XFree86/X11 along with some window manager
On Panther, you can run Apple's XFree86 in full screen mode (X11->Preferences->Output->Full Screen Mode). The next time you start X, you will enter full screen mode directly. When you exit X, you will return to Aqua. You can switch between the two without exiting your X session with Command-Option-A
You can run any window manager you like - I use Windowmaker, which is available via Fink - just make sure your.xinitrc file ends with the line `exec $WMANAGER'. You aren't exiting Quartz, as Apple's X11 runs on top of it, but I guess the above does what you want.
Like most public transportation projects I've been on, this one is pretty useless.
I guess you've never been to Helsinki. There's a website that tells me the next connections between any two addresses I like, including a map of the route and any additional walking (and the buses keep reasonablty to time). My season ticket costs me about 2 euros a day, and includes some trains and boats as well as buses and trams. There's no problem finding buses to the airport for early flights or back after late ones. There are night buses to take you home after clubbing. The service is good whether you're living in the centre or in the outskirts. There are plenty of better examples in other countries.
There's no reason in principle why a public transportation system can't provide a convenient and complete service, provided it's assembled by people with a desire for it to work rather than motivated by personal political or economic ambitions.
well the 1% of you can, frankly, fuck off because making things easier for the other 99% is what i care about -- supply and demand, buddy
You illustrate exactly why WMP should indeed be unbundled from Windows. Microsoft use their operating system monopoly to ensure WMP is on almost all computers; providers like you then cater to that overwhelming majority because that's economic good sense; this means that anyone who wants to use the web needs to have Windows (*) because otherwise they can't enjoy all that streaming content. This ensures that there will never be a viable competitor to Windows.
(*) Yes, I know WMP is available for Mac, but how long before Microsoft stops developing it, just as they did with IE?
What Cray's rightfully pointing out is that for most business applications, however, distributed computing is not a viable option. When processing on a transaction basis, the transactions often need to posted in the exact order they were recieved, which means they must be taken serially.
There are indeed many problems which are not parallelizable, but these are typically not the ones that require a supercomputer. The really big problems are often suitable for parallelization, e.g. spatially-extended PDEs (fluid dynamics, weather prediction), image rendering, or stochastic systems where many independent realizations are needed (protein folding, epidemic modelling).
There are high-stakes real-time environments where getting results a few seconds faster makes a difference (stock market predictions, defense applications), and Cray has a niche there. However, the majority of
`supercomputer' use is scientific applications that run for days, weeks or months. It's difficult to think of a problem that needs that much number crunching but isn't suitable for distributed computing.
I don't think that this will fly either unless they either: a) find a company to make a powerpoint alternative which saves to html files b) make the aforementioned software themselves
Openoffice saves to HTML.
Microsoft Office also claims to save to HTML, but you will be warned This presentation contains content that your browser may not be able to show properly. This presentation was optimized for more recent versions of Microsoft Internet Explorer. Sure enough, the presentation doesn't work in Opera, Konqueror, or Mozilla.
Nevermind that we need that energy to go about our daily business whatever the cost so demand isn't reduced anyway
Not true - there is no fixed amount of energy that people `need' to do their business. In countries where energy is more expensive, people have smaller cars and better insulation.
nevermind that those same progressive governments put exactly zilch of that tax revenue back into alternative energy research
Have you really never seen a windfarm?
and nevermind that it doesn't make any difference anyway because the rest of the world is still polluting at least as much as they ever did
The Kyoto Treaty was an attempt to address this, and quite a few countries were happy to sign up until the Americans pulled out. Besides, even if you can't change the worldwide problem, you can at least make the air in your own town more breathable.
I've installed BeOS on every laptop I've run across
Only 2 out of 3 in my case. It would only run in `safe mode' on a laptop with an S3 ViRGE MX+ graphics chip. Support has been added for other ViRGE chips, but not this particular one, which was quite popular for a while.
If you have root access on *any* Unix, you can do whatever you like to any file, period.
Not true. On the BSD's you can set a `secure level' which allows files or filesystems to be unalterable even by root - the only way to change them is to reboot with a lower secure level. Consequently, on a properly configured box you would need physical access too. There's a Linux patch that achieves the same thing.
On Windows, I can copy/paste pretty much anything from any program to any other reasonable program
On the contrary, people migrating from the Microsoft world are already used to inconsistent and unpredictable clipboard behaviour. As I was using Office 2000 yesterday, I couldn't persuade it to paste correctly from one word document to another - text was OK but equations weren't, and formatting, styles, were assigned somewhat arbitrarily.
Slashdot Mirror
Both of these processors need $500 of ancillary equipment in order to function. Therefore, a system with processor A gives 100 units for $600, or 0.167 units/$, whereas processor B gives 150 units for $650, or 0.231 units/$. This analysis shows that processor B is better value when speccing out a new system
But what about the case where you're just upgrading your cpu? Well, in that case it's moot to compare the AMD with the Intel processors, as you would need a new motherboard too. But simply dividing the performance by the cost of the cpu is meaningless here, too, because staying with your existing processor ($0) would give you a performance/price ratio of infinity.
Conclusion: you have to calculate your total outlay in order to figure out which cpu is the better value.
"Ah", you say, "you're back where you started: how do you know that the server where the signature resides hasn't been compromised?"
Well, digital signature files can be validated by another authority. I'm sure Red Hat, Novell, or IBM would be happy to countersign the Mozilla signature file. All they are doing is confirming that they have checked that this is indeed a signature that belongs to Mozilla - they don't have to sign every package.
In any case, a signature is a small file. The Mozilla signature could be published on Red Hat's, IBM's, Novell's websites, and I can check that they all agree. The more websites I check, the more confident I can be that this is an authentic signature file. I can afford to spend a little effort doing this, because I only need to accept the signature once for each application (or each application developer). This signature will not only be valid for the initial installation, but also for each upgrade or patch. Very useful if, say, you're compiling KDE snapshots from dozens of tarballs.
As for any software project who is too small to be on anyone else's radar: contact the developer themself to get the signature, and think hard before installing it system-wide. You can always compile and run it in an unprivileged user account dedicated to this software alone.
I sincerely hope that no developer uses their own machine as an FTP server. A project as important as Mozilla should only permit development on machines that are protected from the internet. If the story is true, it's much more likely that the FTP server was the weak link - it inevitably provides a much larger target to prospective crackers.I use a lot of OS software (e.g. Firefox, NeoOffice/J, LyX, R), but the standard installation process on my platform (OS X) does not allow checking for an authentic signature. Why is this not built in? It doesn't have to be this way: for instance, Red Hat signs its own RPMs (though Debian's APT didn't support this last time I looked).
We already have to trust the developers. We shouldn't have to trust every FTP server too.
Indeed - Linus has already explained the reasoning behind this name.
You're right - the `Die-By-Wire' sytem is probably behind the Airbus 320 family's terrible accident record relative to other aircraft. Oh, wait...
However, changes in the sex ratio do tell you something about society, because the biological factors presumably don't change over time. A reasonable null model is that the sex ratio in the sciences will become less male-biased over time, due to society's efforts to encourage women. I would suspect that this is the case in UK maths and physics lectures - i.e., the same trend as seen elsewhere. If you know any stats on this, I'd be interested to see them.
If an already male-biased sex ratio becomes even more so with time, in spite of efforts to remove anti-female sexism, then this is an extremely interesting result - and one which isn't explained by the sexes' natural affinity for the subject alone.
Sure, the sex ratio is more female-biased in biology than in other sciences, but the issue here is that computer science is becoming still more male-biased with time, whereas other sciences are becoming less so.
Incidentally, the female biology students I know are much more interested in nature and conservation than in `nurturing' and medicine.
The article contains a quote that
but surely biotechnology is also `high tech', and I see no suggestion that women's representation is decreasing in that area.So, what is it? At the risk of being modded flaimbait, is it perhaps that Physics, Chemistry, and Biology are somehow seen as more noble pursuits, that Computers are intrinsically a means to an end rather than an end in themselves? That CS majors are seen as an inferior type of geek relative to their cancer-curing, drug-designing, atom-smashing counterparts? Yet other branches of engineering (bridge building, rocket science) are also fundamentally concerned with solving practical problems, but somehow they don't carry the same stigma.
Is there, after all, something intrinsically semi-autistic, and therefore testosterone-linked, in fiddling with computers?
I suspect your other doom-laden predictions are equally overstated. Even your own link to Forbes shows that Apple shares have increased more than $10 over the past 8 months (look at the share prices in the graph compared with the text of the article).
Conclusion: parent is a troll.
You can run any window manager you like - I use Windowmaker, which is available via Fink - just make sure your .xinitrc file ends with the line `exec $WMANAGER'. You aren't exiting Quartz, as Apple's X11 runs on top of it, but I guess the above does what you want.
There's no reason in principle why a public transportation system can't provide a convenient and complete service, provided it's assembled by people with a desire for it to work rather than motivated by personal political or economic ambitions.
(*) Yes, I know WMP is available for Mac, but how long before Microsoft stops developing it, just as they did with IE?
There are high-stakes real-time environments where getting results a few seconds faster makes a difference (stock market predictions, defense applications), and Cray has a niche there. However, the majority of `supercomputer' use is scientific applications that run for days, weeks or months. It's difficult to think of a problem that needs that much number crunching but isn't suitable for distributed computing.
The PDF file claims to have been made by dvips, so it was written in Latex. It was then converted to PDF using Distiller.