if they just implemented the iPod as a USB mass storage device
You can't update your library, but an iPod is a mass storage device. Microsoft rewrote the MSC driver for Vista and now the power-management doesn't work if you safely remove MSC devices. That's also a problem for other companies. Vista users should "eject" the drive first to get the old "Windows suspends your USB device" behavior.
I think Apple waited for Microsoft to document (or fix) this "feature".
Let's assume that I'm not to lazy to write a helper function to get rid of the casts in the JavaBeans version. Why would I want to use the Buoy event classes? Should I mix Buoy and EventHandler code when I want to use the more powerful EventHandler.create() variants?
Keep in mind that nobody wants to learn new shortcuts for every website. Some users might not be able to navigate a page if their navigation keys are reprogrammed. If you want to make your pages more accessible then make sure that your pages use "strict" xhtml and css. Then visitors who need additional accessibility don't run into problems with their browser-modifications.
Obscure keyboard shortcuts are convenient for blog/cms-administrators. And administrators can tweak the shortcuts if they don't like them. Regular users don't have this option, so they might be annoyed.
The first one hits memory sequentially, which is good for caching as each cache line stores a large chunk of sequential memory.
A sequential access has the disadvantage that modern CPUs can't fill the L2 cache efficiently.
An AthlonXP will cache a 64 byte block of data when you read the first bytes. If you access your data in sequence your code will have to wait for every memory access. Your second loop can be faster on such systems, because more data will be waiting in the cache before you access it.
When you're optimizing never assume that something will be faster. Allways examine a simple solution on your target platform(s), it might already be quite efficient.
Re:The real reason the Euro is BAD NEWS
on
The Euro
·
· Score: 1
The control of this currency rests with the German Government.
I hope that's not a common misbelief. Please note that the German Government didn't even control the Deutschmark. At first it was controlled inderectly through the US. (There was a fixed exchange-rate simmilar to Argentinia.) But after the US used their printing machines excessively in order to finance the vietnam war control was transfered to the "Bundesbank" which is owned but not controlled by the government.
It's not comparable to the ICANN which currently is controlled by the US-government.
There is no such thing as a bundesbank.
Actually the Deutsche Bundesbank still exists and I think there are no plans to close it.
Re:Appreciate, inflate, depreciate...
on
The Euro
·
· Score: 1
> But the bigger threat is inflation.
Actually the Euro already significantly decreased inflation. But now this might even be a problem because people want to get raises "as usual" but the industry can't pay 7% more when the prices only climbed 2%.
And please not that the 2.x% inflation we have in germany are mostly caused by increased fuel taxes.
I heard that Portugal still has some problems with inflation, but this will mostly affect consumers and not the industry: the commercial trades with foreign countries can be calculated in advance and invesestors can rely on a stable currency.
As Germany (the driver of EU commerce), has just
officially gone into recession
IIRC the USA "officially" went into recession. But Germany will probably follow in a few years.
(Then Asia will get stronger and even later Japan will get out of recession, if the analysts are right.)
Britain, by not taking part of the Euro, is best placed to benefit in the short/medium term.
I never heard this opinion before, but then again I'm no financial expert. I thought britain
is afraid to become the looser of a "united europe"
and therefore tries to keep better relations to the
US than to its european neighbours. If not switching to the Euro would be a disadvantage for the other contries their advisors wouldn't have agreed to it.
We're a research-department an we're not supposed to create real products. We're creating several source-analysis and documentation-management tools, which might be "sold" to other departments.
Commonly used OS-projects are e.g. xerces, boost, python and wxpython, where we usually only contribute bugreports or small patches.
Some tools use proprietary libraries that we must not redistribute as sourcecode. (e.g. a C++ backend from EDG) So if we want to call GPL programs (e.g. cvs or some source-documentation systems) we must not make proprietary extensions to it. It doesn't matter if we put the modified program under the GPL, it would still violate the license, because such an extension would be only usable with other non-GPL programs.
The sad thing is, that some people prefer to pay for inferior commercial sourcecode instead of contacting the original authors in order to come to an agreement.
Instead of documenting the subtle relations between dozens of different licenses (as someone suggested) we should try to encourage a personal contact to the developers. If the author allows you to use his code under certain conditions you don't even have to understand the original license.
IMHO the problem is that PHBs are afraid of such agreements, and as long as there's a different solution they don't want to try something new.
OSI has a nice list of OpenSource licenses:
www.opensource.org/licenses/index.html But I suppose it won't be a help for PHBs if the comparison gets more complicated.
Our department quite often uses open-source projects as a basic part for in-house tools.
Co-workers sometimes consider to call an external program but say "It's GPL, we would have to add a commandline-option, we don't want to make things complicated if we want to sell it."
I wish they would at least try to contact the original author and hear his opinion. I suppose most programmers would be willing to cooperate and support (for example) a generic interface to custom extensions.
Your hardware probably uses 32bpp during calculations because it's faster that way.
But if your graphic-card isn't very exotic it only uses 24bpp to actually display a colour and eight bits are ignored.
IIRC Murdock worked for an university and is now the CEO of Progeny.
www.progeny.com
(Progeny Debian 1.0 looks promising, btw)
I suppose Bruce Perens is the connection between Debian and Pixar.
He is a former debian project leader and he worked twelve years at pixar.
www.perens.com/Articles/Bio.html
.. one of his biggest disappointments was the failure to inspire a large community of developers to join up, and that having so many Netscape full-timers on the project may have violated the dynamics of a culture that thrives on volunteer programming..
In my opinion full-timers are a great thing. I'd rather say that unusual methods of development make it hard to participate.
Please have a look at the Portability Guidelines on mozilla.org. Instead of porting our tools (egcs) to uncommon architectures developers are expected to spend additional time on writing the program. That's not globally efficient from my point of view.
It's nice that mozilla.org wants do have a port on every system, and I'm sure they're on the right way, but the typical hobby-developer works for his OS, and not for "planetwide" portability. Blaming the full-time-hackers doesn't help us.
I installed a Debian/Slink beta this week and GPG seems to be well integrated into the mail-user-agent "exmh".
Packages are signed somehow, but I think the sig is only checked when the package-maintainers upload the software. (And I don't know if everyone uses GPG already.)
Please note that GPG can't be an integral part of any free GNU/Linux-System, since US-Gov. doesn't permit reexporting.
What other kind of "integration" might be needed in the near future?
Slashdot Mirror
if they just implemented the iPod as a USB mass storage device
You can't update your library, but an iPod is a mass storage device. Microsoft rewrote the MSC driver for Vista and now the power-management doesn't work if you safely remove MSC devices. That's also a problem for other companies. Vista users should "eject" the drive first to get the old "Windows suspends your USB device" behavior. I think Apple waited for Microsoft to document (or fix) this "feature".
button.addEventLink(CommandEvent.class, this, "buttonPressed");
and every Java programmer should understand JavaBeans code likebutton.addActionListener(((ActionListener.class) EventHandler.create(ActionListener.class, this, "buttonPressed"));
Let's assume that I'm not to lazy to write a helper function to get rid of the casts in the JavaBeans version. Why would I want to use the Buoy event classes? Should I mix Buoy and EventHandler code when I want to use the more powerful EventHandler.create() variants?Keep in mind that nobody wants to learn new shortcuts for every website. Some users might not be able to navigate a page if their navigation keys are reprogrammed. If you want to make your pages more accessible then make sure that your pages use "strict" xhtml and css. Then visitors who need additional accessibility don't run into problems with their browser-modifications.
Obscure keyboard shortcuts are convenient for blog/cms-administrators. And administrators can tweak the shortcuts if they don't like them. Regular users don't have this option, so they might be annoyed.
XML file may be manipulated by applications that understand XML
Hmmm. All we need is a generic tool that manipulates XML files but doesn't understand XML.
The first one hits memory sequentially, which is good for caching as each cache line stores a large chunk of sequential memory.
A sequential access has the disadvantage that modern CPUs can't fill the L2 cache efficiently.
An AthlonXP will cache a 64 byte block of data when you read the first bytes. If you access your data in sequence your code will have to wait for every memory access. Your second loop can be faster on such systems, because more data will be waiting in the cache before you access it.
When you're optimizing never assume that something will be faster. Allways examine a simple solution on your target platform(s), it might already be quite efficient.
I hope that's not a common misbelief. Please note that the German Government didn't even control the Deutschmark. At first it was controlled inderectly through the US. (There was a fixed exchange-rate simmilar to Argentinia.) But after the US used their printing machines excessively in order to finance the vietnam war control was transfered to the "Bundesbank" which is owned but not controlled by the government.
It's not comparable to the ICANN which currently is controlled by the US-government.
There is no such thing as a bundesbank.
Actually the Deutsche Bundesbank still exists and I think there are no plans to close it.
Actually the Euro already significantly decreased inflation. But now this might even be a problem because people want to get raises "as usual" but the industry can't pay 7% more when the prices only climbed 2%.
And please not that the 2.x% inflation we have in germany are mostly caused by increased fuel taxes.
I heard that Portugal still has some problems with inflation, but this will mostly affect consumers and not the industry: the commercial trades with foreign countries can be calculated in advance and invesestors can rely on a stable currency.
As Germany (the driver of EU commerce), has just officially gone into recessionIIRC the USA "officially" went into recession. But Germany will probably follow in a few years.
(Then Asia will get stronger and even later Japan will get out of recession, if the analysts are right.)
Britain, by not taking part of the Euro, is best placed to benefit in the short/medium term.I never heard this opinion before, but then again I'm no financial expert. I thought britain is afraid to become the looser of a "united europe" and therefore tries to keep better relations to the US than to its european neighbours. If not switching to the Euro would be a disadvantage for the other contries their advisors wouldn't have agreed to it.
Currently squid is slower on ReiserFS compared to ext2 because squid is optimized for ext2.
And if you put the squid files on a seperate partition you don't need journaling anyway if you automatically reformat the partition after a crash.
We're a research-department an we're not supposed to create real products. We're creating several source-analysis and documentation-management tools, which might be "sold" to other departments.
Commonly used OS-projects are e.g. xerces, boost, python and wxpython, where we usually only contribute bugreports or small patches.
Some tools use proprietary libraries that we must not redistribute as sourcecode. (e.g. a C++ backend from EDG) So if we want to call GPL programs (e.g. cvs or some source-documentation systems) we must not make proprietary extensions to it. It doesn't matter if we put the modified program under the GPL, it would still violate the license, because such an extension would be only usable with other non-GPL programs.
The sad thing is, that some people prefer to pay for inferior commercial sourcecode instead of contacting the original authors in order to come to an agreement.
Instead of documenting the subtle relations between dozens of different licenses (as someone suggested) we should try to encourage a personal contact to the developers. If the author allows you to use his code under certain conditions you don't even have to understand the original license.
IMHO the problem is that PHBs are afraid of such agreements, and as long as there's a different solution they don't want to try something new.
OSI has a nice list of OpenSource licenses: www.opensource.org/licenses/index.html
But I suppose it won't be a help for PHBs if the comparison gets more complicated.
Our department quite often uses open-source projects as a basic part for in-house tools. Co-workers sometimes consider to call an external program but say "It's GPL, we would have to add a commandline-option, we don't want to make things complicated if we want to sell it."
I wish they would at least try to contact the original author and hear his opinion. I suppose most programmers would be willing to cooperate and support (for example) a generic interface to custom extensions.
Your hardware probably uses 32bpp during calculations because it's faster that way.
But if your graphic-card isn't very exotic it only uses 24bpp to actually display a colour and eight bits are ignored.
IIRC Murdock worked for an university and is now the CEO of Progeny.
www.progeny.com (Progeny Debian 1.0 looks promising, btw)
I suppose Bruce Perens is the connection between Debian and Pixar.
He is a former debian project leader and he worked twelve years at pixar.
www.perens.com/Articles/Bio.html
I'd guess DVD support in Linux sits on one of these zones, in an area of high geological activity between CorpNet, TechNet & CultureNet :-).
Actually it belongs to X-Net ... "forbidden pleasures". ;-)
It doesn't compile on my system (gcc version 2.95.2, glibc2) but you might want to check Illiads entry for this contest.
In my opinion full-timers are a great thing. I'd rather say that unusual methods of development make it hard to participate.
Please have a look at the Portability Guidelines on mozilla.org. Instead of porting our tools (egcs) to uncommon architectures developers are expected to spend additional time on writing the program. That's not globally efficient from my point of view.
It's nice that mozilla.org wants do have a port on every system, and I'm sure they're on the right way, but the typical hobby-developer works for his OS, and not for "planetwide" portability. Blaming the full-time-hackers doesn't help us.
I installed a Debian/Slink beta this week and GPG seems to be well integrated into the mail-user-agent "exmh".
Packages are signed somehow, but I think the sig is only checked when the package-maintainers upload the software. (And I don't know if everyone uses GPG already.)
Please note that GPG can't be an integral part of any free GNU/Linux-System, since US-Gov. doesn't permit reexporting.
What other kind of "integration" might be needed in the near future?