You should more clearly mark, what gain can be expected by which measure. Allocating on the stack (with alloca() or something similiar) gains you speed, some convenience, but no security (buffer overflows are more readily exploited to inject harmful code, if the buffer is allocated on the stack).
You failed to describe what's wrong with strncat(), strncpy() etc. IMHO people who can't comprehend the man pages for those functions probably should avoid C altogether, but definitively must be hindered to write security relevant software (as should sleep-deprived coders who try to do it on a Sunday morning;-}.
Said that, I can only appreciate your attempt to raise this issue (once more, maybe for a new generation of C coder).
It already exists: sourceforge.net Thousands of OSS projects are hosted there. And exactly this is a concern for some people: the sheer concentration of projects there. If sourceforge.net would go black and cold, it would be a harsh backlash for the OSS community and probably mean the end for many smaller projects.
SourceForge isn't doing any security screening at this time. The costs would be forbidding. Open Source doesn't mean free of charge, but most people came to expect it to be cheap. RedHat and SuSE could raise the price of their CDs by a couple of bucks to finance this additional service, but Debian can not.
The best answer to this problem IMHO is to have the developers sign their source and binary distributions. But people need to verify those signatures (apparently the attacker didn't bother to adjust the MD5 hash of the infected files -- I wonder how many infected packages are out there, with proper hashs).
From iRobot's website (http://www.irobot.com/corp/p02.asp): "Our Corporate Mission statement is fourfold, but simple: Make money, have fun, build cool stuff, and change the world."
I don't care if the fonts in X11 are ugly. In fact I don't know what he's talking about. I can read them without getting a headache and was able to do so ten years ago.
I dare to suspect that X11 developers think similiar. Since they don't see the need to 'improve' the fonts, it doesn't get done.
The OSS modell works (most of the time) as follows: fix it yourself or find someone you can pursue to do it, by e.g. paying him or her. This are viable options only for a few.
What if companies like RedHat and SuSE would adopt the Transgaming modell (paying customer vote for features)? This way, features overlooked by developers but (for whatever reason) desired by the broad masses, would make it into popular distributions.
What next? People chose Windoze because the colours are more vibrant? The tunes are more transparent? It has a richer bouquet?
Then again, people have sold their souls for less, I guess.
And yes, I'm one of those who 'look under the hood' when buying a car. My '84 240DL Volvo is still runing strong at 184000 miles, albeit not pretty (never was;-)
Don't even think about commenting on my girlfriend! I'm serious about that.
Slashdot Mirror
You should more clearly mark, what gain can be expected by which measure. Allocating on the stack (with alloca() or something similiar) gains you speed, some convenience, but no security (buffer overflows are more readily exploited to inject harmful code, if the buffer is allocated on the stack).
;-} .
You failed to describe what's wrong with strncat(), strncpy() etc. IMHO people who can't comprehend the man pages for those functions probably should avoid C altogether, but definitively must be hindered to write security relevant software (as should sleep-deprived coders who try to do it on a Sunday morning
Said that, I can only appreciate your attempt to raise this issue (once more, maybe for a new generation of C coder).
It already exists: sourceforge.net
Thousands of OSS projects are hosted there. And exactly this is a concern for some people: the sheer concentration of projects there. If sourceforge.net would go black and cold, it would be a harsh backlash for the OSS community and probably mean the end for many smaller projects.
SourceForge isn't doing any security screening at this time. The costs would be forbidding. Open Source doesn't mean free of charge, but most people came to expect it to be cheap. RedHat and SuSE could raise the price of their CDs by a couple of bucks to finance this additional service, but Debian can not.
The best answer to this problem IMHO is to have the developers sign their source and binary distributions. But people need to verify those signatures (apparently the attacker didn't bother to adjust the MD5 hash of the infected files -- I wonder how many infected packages are out there, with proper hashs).
From iRobot's website (http://www.irobot.com/corp/p02.asp):
"Our Corporate Mission statement is fourfold, but simple: Make money, have fun, build cool stuff, and change the world."
I feel so much safer now.
I don't care if the fonts in X11 are ugly. In fact I don't know what he's talking about. I can read them without getting a headache and was able to do so ten years ago.
I dare to suspect that X11 developers think similiar. Since they don't see the need to 'improve' the fonts, it doesn't get done.
The OSS modell works (most of the time) as follows: fix it yourself or find someone you can pursue to do it, by e.g. paying him or her.
This are viable options only for a few.
What if companies like RedHat and SuSE would adopt the Transgaming modell (paying customer vote for features)? This way, features overlooked by developers but (for whatever reason) desired by the broad masses, would make it into popular distributions.
Just my 2c .
Fonts? Because the fonts are ugly in X11?
;-)
What next? People chose Windoze because the colours are more vibrant? The tunes are more transparent? It has a richer bouquet?
Then again, people have sold their souls for less, I guess.
And yes, I'm one of those who 'look under the hood' when buying a car. My '84 240DL Volvo is still runing strong at 184000 miles, albeit not pretty (never was
Don't even think about commenting on my girlfriend! I'm serious about that.