We're doing our own internal time-tracking applications (as mainly an exercise to keep us busy as more work comes down the pipe.)
I decided to fire up the RC2 version of the web app under Firefox.
Worked without a hitch.
Granted we're not using anything really complex... (we're using some 3rd party data grids built off the MS grids, that's it)
But still.
Worked fine in Firefox.
If people actually bothered to TEST their applications, they may find that they work in alternate browsers. Or at least, they could hack around any incompatibilities.
It's useful at stopping buffer overflows--you can put executable code in pages marked as writable, but you can set the NX bit, and it can be executable, but it's not going to be executed.
OpenBSD uses a scheme called W^X (write xor execute) which is basically what I stated above.
However, AFAIK Windows doesn't support usage of the NX instruction (the new SP may...) And Intel's new chips certainly don't as of last I heard (today), so that gives Microsoft less incentive to implement it.
OSes like OpenBSD who absolutely thrive on this sort of instruction (hell, they implemented a decent W^X function on chips that DIDN'T have it built in because they wanted it that badly.) will immediately snatch it up and praise it, and there will be some market share gains.
OTOH, You can build all the greatest functionality in the world but if no one uses it, there isn't much point.
Of course, when people start adopting usage of those instructions, they're glad it's there.
--Benanov
To ensure reliable operation, circuits were constructed of rigidly tested standard components, which were operated at current, voltage, and power levels below their normal ratings.
(And apparently the thing still ran hot.)
Hats off to a pioneer.
--Benanov
Well, it'll make people want to upgrade their machines because they're so slow, when really they're just maintained improperly.
Free as in Beer not too useful an arguement then?
on
Linux in Iraq
·
· Score: 1
Since any Iraqi government has more immediate measures to worry about, piracy is probably fairly attractive and there probably isn't much of a penalty for it in terms of the end user or even middlemen.
So Free As in Beer really doesn't present much alternative to what's currently going on.
Sounds like a good fit for a lot of security-related applications--rackmount DMZ servers, firewalls, central controller for various physical security devices (biometrics controlling electronic locks on doors, etc.)
I knew that the hardware RNG and AES-encrypt/decrypt functions were supported by OpenBSD a while ago (don't ask me when...I think by the time 3.4 came around. Confirmation?) It's no small feat to have them in other kernels, and bodes well for OSS applications.
I'd assume it's also low-power due to the small form factor, which is another plus. Can't check the article, as my manager's about to breathe down my neck.:)
I'm just worried about the potential security vulnerabilities.
I mean, imagine someone running down the street, flailing their arms wildly, screaming "My underwear's been infected by a virus! I can't take it off!"
Slashdot Mirror
We're doing our own internal time-tracking applications (as mainly an exercise to keep us busy as more work comes down the pipe.)
I decided to fire up the RC2 version of the web app under Firefox.
Worked without a hitch.
Granted we're not using anything really complex... (we're using some 3rd party data grids built off the MS grids, that's it)
But still.
Worked fine in Firefox.
If people actually bothered to TEST their applications, they may find that they work in alternate browsers. Or at least, they could hack around any incompatibilities.
It's useful at stopping buffer overflows--you can put executable code in pages marked as writable, but you can set the NX bit, and it can be executable, but it's not going to be executed. OpenBSD uses a scheme called W^X (write xor execute) which is basically what I stated above. However, AFAIK Windows doesn't support usage of the NX instruction (the new SP may...) And Intel's new chips certainly don't as of last I heard (today), so that gives Microsoft less incentive to implement it. OSes like OpenBSD who absolutely thrive on this sort of instruction (hell, they implemented a decent W^X function on chips that DIDN'T have it built in because they wanted it that badly.) will immediately snatch it up and praise it, and there will be some market share gains. OTOH, You can build all the greatest functionality in the world but if no one uses it, there isn't much point. Of course, when people start adopting usage of those instructions, they're glad it's there. --Benanov
Well, it'll make people want to upgrade their machines because they're so slow, when really they're just maintained improperly.
Since any Iraqi government has more immediate measures to worry about, piracy is probably fairly attractive and there probably isn't much of a penalty for it in terms of the end user or even middlemen. So Free As in Beer really doesn't present much alternative to what's currently going on.
Sounds like a good fit for a lot of security-related applications--rackmount DMZ servers, firewalls, central controller for various physical security devices (biometrics controlling electronic locks on doors, etc.) I knew that the hardware RNG and AES-encrypt/decrypt functions were supported by OpenBSD a while ago (don't ask me when...I think by the time 3.4 came around. Confirmation?) It's no small feat to have them in other kernels, and bodes well for OSS applications. I'd assume it's also low-power due to the small form factor, which is another plus. Can't check the article, as my manager's about to breathe down my neck. :)
I'm just worried about the potential security vulnerabilities. I mean, imagine someone running down the street, flailing their arms wildly, screaming "My underwear's been infected by a virus! I can't take it off!"