And that brings us to my point: making software compatible with older hardware shouldn't be a goal in and of itself. Why? One need only to venture over to Pricewatch to see that an AMD 1800+ mobo/CPU combo sells for under $300. Systems faster than what anyone could ever need are commodities now.
While this is true in most western countries, it is absolutely wrong in many regions of the world.
The CS school I've been in has many students organizations. One of them, EFREI Aide Humanitaire (french link, use the fish) helps schools in Africa to get CS related stuff, by collecting "old" mobos, towers, harddisks, and so on.
What's old for us is quite new for them. Hey, the whole "desktop computer" thingy is quite new (it's only 25 years old) ! One may argue that old hardware may run old software well, so why don't they use linux 1.x ? I would answer I don't see the reason why they should use buggy software when bugs have been corrected for years. Linux is a modern OS these people can use.
I agree that making backward compatible software should not be a goal. Microsoft shows us what it can lead to. I would instead say it should be a motivation. Do not cut compatibility where it's not necessary, and if you do, try to provide a way for other developers to easily make it compatible.
Julien
The available demo is out to date. One of the guys at Peroxyde told me to forget the demo, the current engine being far better than the old one we can play with. For instance, the day cycle with light shifts etc is implemented in the current development version.
They have NO idea why there is a "war" going on in the Middle East.
That's the problem with american people: they've no idea at all.
Jeez, it remembers me those guys, on TV, who thought France was somewhere in Canada...
Julien.
-- "One day, American people will get smart!" (The American Dream)
I just was wondering who are the guys to fear.
On one side, we have a bunch of well organized terrorists (maybe one thousand guys).
On the other side, we have millions of "partiot-at-all-price" brainless american guys, ready to iitiate WW3 and to "eradicate all the countries who helped those bastards" (this is a quote from a previous comment), killing one million times more innocent people that the previously mentionned bastards actually killed.
Sad to say, but if the DEFCON goes one more level up, I would more likely fight against the US than on their side...
Julien.
-- Warning: drinking too much beer while watching Jerry Springer's shows can make you as idiot as an american patriot.
I use both command line tools and GUI equivalents. I think GUI can help a lot, providing graphical tools for designing (UML modeler, inheritance charts, etc.) and for coding (words completion, syntax highlightning, inline error checking, etc.).
However, command line tools are often more flexible.
I would say it depends on the target's environment: if you're developping a server for some *n*x platform then command line tools may be more efficient (vim has syntax highlightning, after all:p), but if you're targeting a Win32 platform with hundreds of resources files, etc. Visual Studio stays the best choice.
I've never used KDevelop nor CodeWarrior (except on BeOS, but it was just a poor shitty IDE with not so much features), but I don't see anything valuable currently in unix IDEs that are more source editors than true design platforms.
Julien.
As far as I remember, there is an open source package computing diffs between two images, given a certain difference tolerance. It could be used to compare the webcams shots, with a tolerance high enough to get rid of false alerts due to changes in sun illumination.
Sorry, I do not remember the package name!
What about re-packaging the STL (and the extensions) with nested namespaces, like packages in Java (eg, std::sql::oracle::query) ? It may make the reading clearer, and could lead to better programming.
Slashdot Mirror
- And that brings us to my point: making software compatible with older hardware shouldn't be a goal in and of itself. Why? One need only to venture over to Pricewatch to see that an AMD 1800+ mobo/CPU combo sells for under $300. Systems faster than what anyone could ever need are commodities now.
While this is true in most western countries, it is absolutely wrong in many regions of the world.The CS school I've been in has many students organizations. One of them, EFREI Aide Humanitaire (french link, use the fish) helps schools in Africa to get CS related stuff, by collecting "old" mobos, towers, harddisks, and so on.
What's old for us is quite new for them. Hey, the whole "desktop computer" thingy is quite new (it's only 25 years old) ! One may argue that old hardware may run old software well, so why don't they use linux 1.x ? I would answer I don't see the reason why they should use buggy software when bugs have been corrected for years. Linux is a modern OS these people can use.
I agree that making backward compatible software should not be a goal. Microsoft shows us what it can lead to. I would instead say it should be a motivation. Do not cut compatibility where it's not necessary, and if you do, try to provide a way for other developers to easily make it compatible.
Julien
Will OpenAL be continued ?
Give also credit to XEROX, who invented among other things the mouse and the GUI...
Give credit to Microsoft when they deserve it: they forced Intel to not get rid of real mode in x86 design back to the 386 ages...
Julien.
This leads to fitness propagation, which may lead to neural nets, neural nets lead to anger, anger leads to... you know
Julien.
Damn it! How do you know ??? :)
Julien.
I've been studying Scheme in my CS class. However, it was a fork called "SWIT" (Scheme WIth Types, somewhat useless).
:)
I didn't learn anything but enabling vim's bracket matching option
The previous year (and the next year as well), this class was dedicated to Prolog, which I think as being more useful and less boring.
Julien.
I've been studying Scheme at schools, but I've still no girlfriend. Isn't that a design flaw of the language ?
Julien.
Actually this is old news.
The available demo is out to date. One of the guys at Peroxyde told me to forget the demo, the current engine being far better than the old one we can play with. For instance, the day cycle with light shifts etc is implemented in the current development version.
Julien.
Jeez, it remembers me those guys, on TV, who thought France was somewhere in Canada...
Julien.
--
"One day, American people will get smart!" (The American Dream)
I just was wondering who are the guys to fear. On one side, we have a bunch of well organized terrorists (maybe one thousand guys).
On the other side, we have millions of "partiot-at-all-price" brainless american guys, ready to iitiate WW3 and to "eradicate all the countries who helped those bastards" (this is a quote from a previous comment), killing one million times more innocent people that the previously mentionned bastards actually killed.
Sad to say, but if the DEFCON goes one more level up, I would more likely fight against the US than on their side...
Julien.
--
Warning: drinking too much beer while watching Jerry Springer's shows can make you as idiot as an american patriot.
How is this system supposed to work on my PC, since I primaryly use my keyboard for web browsing ? :)
Gotta track space, enter, tab and arrow keys hits too !
Julien.
I use both command line tools and GUI equivalents. I think GUI can help a lot, providing graphical tools for designing (UML modeler, inheritance charts, etc.) and for coding (words completion, syntax highlightning, inline error checking, etc.). However, command line tools are often more flexible. I would say it depends on the target's environment: if you're developping a server for some *n*x platform then command line tools may be more efficient (vim has syntax highlightning, after all :p), but if you're targeting a Win32 platform with hundreds of resources files, etc. Visual Studio stays the best choice.
I've never used KDevelop nor CodeWarrior (except on BeOS, but it was just a poor shitty IDE with not so much features), but I don't see anything valuable currently in unix IDEs that are more source editors than true design platforms.
Julien.
55-60 hours a week.
As far as I remember, there is an open source package computing diffs between two images, given a certain difference tolerance. It could be used to compare the webcams shots, with a tolerance high enough to get rid of false alerts due to changes in sun illumination. Sorry, I do not remember the package name!
What about re-packaging the STL (and the extensions) with nested namespaces, like packages in Java (eg, std::sql::oracle::query) ? It may make the reading clearer, and could lead to better programming.