The fact that windows mail clients recognise some file types and allow easy execution is a feature. Ultimately it is human error that allows a trojan to work.
A bug would be if some kind of mail client feature allowed the script to be executed on just opening of the message.
Development aside. How can perl be easier on servers?
Same work done in less cycles, requiring less memory, the only issue is fork() exec() delay, but that is a server design issue.
You could load the CGIs into the server context as a library, saving the fork() hit. Clearly this is not a good idea from a stability point of view, however it isn't any worse than mod_perl and it would run at full compiled speed.
I am a small ( 1000 user) Australian ISP, but I don't consider myself a 'shonk' or 'cowboy'. I find censorship distasteful, and am not at all pleased with this decision.
It could have been much, much, worse, for that I am thankful.
12 assorted x86 machines, just an ascend router and a tigris remote access server.
We don't have an NT ASP box yet, and probably never will I have anything to do with it.
Last years Australian ISP survey indicated that over half the ISPs in Oz use linux on the majority of their machines, only 19% used NT exclusively, most that had NT machines only used them for ASP serving for virtual clients, some RADIUS servers too (the reason escapes me).
Sorry I can't post the pdf with that in it, as it was given free only to participants of the survey and I am not allowed to leak the info. (oops!)
They make linux's flexibility sound like a disadvantage. Maybe for people that lack the money/brains, but to my mind enterprise sized systems would be maintained by people with a clue. Solaris ain't no cinch to setup either.
Saying that NT is better than Solaris is just plain dishonest. Funny how they attack linux for poor SMP support but then gloss over the huge difference in NT vrs Solaris SMP support.
Try remote administration of NT boxes you turkies, then tell me unices are hard to maintain remotely.
The article isn't that bad I guess, once you realise that it is just another marketing driven review.
> The problem is that Linux is indeed inferior to > most operating systems. Look at BSD/OS and > FreeBSD. Look at OpenVMS. Look at Digital UNIX.
Digital Unix, don't make me laugh! Open VMS, yeah OK, I'll give you that, the facist OS is much more mature than unix. xBSDs, well I have used lots of unices, and the xBSDs aren't that much different, Linux performs better that all the BSDs I have tried.
> I rarely see a Linux uptime greater than a month > whereas with the systems listed above, it is > just expected
Where are you looking? YMMV, but I think that is complete garbage or figures for workstations that are turned of every night:-).
Yeah they are a bit heavy on memory, but they sure work well compared to the xBSD ones. GNU have the wrong ideas about many things (eg. man pages) but they supply great high performance software that tries hard to be POSIX complient and useful to everyone.
> I have never seen a Linux distribution that came > with full source that could be rebuilt with one > command
Slackware: SlackBuild (on every CD in/source)
Not many people need this, but I have used it just for fun a few times (takes a _long_ time). I am sure there are others. It seems to me that your experence of Linux is limited to RedHat et al.
Besides, linux is just a kernel.
> my Linux 2.2.x systems, it fails to properly > unmount its drives.
Running the old glibc? There is a known issue with the ld-linux.so for oldish versions of libc6. Not a kernel issue, RTFM (well Changes) next time.;-)
There are many other userspace reasons why this may happen also; if you are keeping the fs busy it obviously can't umount it.
> simply quits responding to IP packets for a > while
Never seen it happen, sounds like the _old_ SCSI bug, or broken/cheap and nasty hardware. More info please, we can help to correct your issue?
> I never see BSD systems broken into
I have, Solaris, Linux, and even VMS too.
It is a userspace problem in almost all cases. The famous broken IMAPd was the same code used on most unices, this is not Linux specific.
> ship without documentation or manpages
Yeah I have to agree here, lazy programmers. Many GNU morons (sorry) seem to think that man pages are old hat, well I think they are better than the info system (yuck). Many interactive commands are self documenting though and come with other format documentation.
All in all, not a great issue, you can always email the guy or ask a friend if you are completely lost. strace comes in handy too;-)
> makes for a great research system
Being open source is great for students, tinkers, and sysadmins alike. I wouldn't limit its usefullness to just research though, I personally use it in mission critial serving for my ISP business, as my desktop OS, for realtime hacking that I do for fun and profit, and as a system to develop portable network apps on.
Hell I use linux for everything I do with computers. If it can't do something I want, I just add the functionality and get on with business, something I can't do in many other unices.
Yeah I used VSTa at Uni. Even got to talk to the author (damn I forgot the man's name, we got on pretty well too).
VSTa was pretty damn bad, it was written mainly as an OS for educational purposes.
I did some realtime hacking on it, bah it was horrible. It would crash when you exited vi or emacs at the wrong time. Close examination of its source led me to believe that the main problem was the semaphore locking was broken (not atomic).
I mentioned that to my tutor and after about half an hour of logical argument he agreed that is was broken and would require significant effort to fix.
I really liked tfork() though, very nice, easier than pthreads or clone(). Still I'd rather use Linux anyday. Message passing by MM is fast and the microkernel design is interesting, but I think I have been too mentally mutalated by monolythic kernel hacking.
I have throughly abused linux networking code and blasted terabytes of data through fast ether cards with only marginal support (at the time), never crashing a box.
The only times I have ever crashed linux has been either my fault (kernel stack smashing et al) or broken/ill supported hardware like my el cheapo PV-BT848 framegrabber.
I have a RivaTNT and after repeated attempts to get nvidia to even respond to email I am about to take it back and opt for a lesser product that 3dfx drivers exist for.
It is great under X, but with no OpenGL libs, quake et al are stuck in software mode.
Anyone else want to try talking them into giving linux 3D drivers for it?
I don't think too many of the 'HERF gun' guys really have any idea what they are talking about though. The power supply unit for a CW microwave HERF device would be so big and heavy you wouldn't back pack it.
Kinda reminds me of the idea of carrying a rail gun, sure you might just be able to lift it, but the recoil would liquify your skeleton.
EMP generators are practical though, I have a bit of text (hersay basically) on my homepage that talks a little about them. As a matter of fact I recently registered herf.org to put up a site to dispell some of the EM weapon rumours. (it isn't up yet)
I have a 3" round by 10" long device that can kill digital watches and other small unshielded electronics from about 1 metre. It was built from an old photo flash unit reworked to store about 20 Joules. It has an anoying effect of magnetising the shadow mask in VDUs too.
I remember reading about some PET brain scan research that found using a computer in CLI mode uses similar brain centers to human language processing.
Makes sense really, the semantic constructs may be different, but it is still the same fundamental function, parsing or generating a formal structure.
While a lot of the brain is dedicated to processing visual information, I doubt that there is much language learning value from a GUI. The visual centers dump detail as much as possible. The language centers can't do this as that data has a much higher 'entropy' as far as the brains concept processing goes.
Depends upon how non-linear the lasing system is. At low amplitudes near the lasing threshold any laser will act randomly as far as spatial and temporal coherance goes.
Intresting field of study actually, you can see quantum effects in the macroscopic output of the laser. Although it could be argued that a sparking fluro light or dripping tap demonstrates exactly the same thing.
Does this just sound like conventional data spreading to anyone else? Seems like they are scrambling the data stream with a pseudo-random sequence. What is new about that, lots of modem protocols use it?
I gather they have some other protocol development to improve the data throughput?
More info please, that article is lacking any real information.
Indeed, I even go as far as to write CGIs in gawk. Something to which it is very suited but few seem to try, they just opt for a big, slow, and ugly perl script.
If you know the language well you don't need to start (many) other processes like most people seem to think.
Gawk starts and runs faster than perl5 for most programs (that are suitable for it). In fact I only go to perl when I need to use sockets.
When you get to using sockets, you give up on perl after about 5 minutes and go use C like you should have in the first place.
Slashdot Mirror
Optically or some kind of MEMS accelerometer?
What about the actual marking engine?
Every now and then, my karma sucks.
Of late, I have been a lurker more than a poster as the SNR went south.
I 'viewed' the message in outlook at work.
It doesn't run it automatically, it requires a user double click to execute it.
In fact, as I have 250 of them in my Inbox at the time, the first thing I did was save it as a file and right click->edit to view the source.
It is not a bug.
ILOVEYOU is just another trojan.
The fact that windows mail clients recognise some file types and allow easy execution is a feature. Ultimately it is human error that allows a trojan to work.
A bug would be if some kind of mail client feature allowed the script to be executed on just opening of the message.
Really it is no worse than having . in your path.
Development aside. How can perl be easier on servers?
Same work done in less cycles, requiring less memory, the only issue is fork() exec() delay, but that is a server design issue.
You could load the CGIs into the server context as a library, saving the fork() hit. Clearly this is not a good idea from a stability point of view, however it isn't any worse than mod_perl and it would run at full compiled speed.
The fork(), exec() hit is pretty tiny anyway.
The SlackBuild script can do it.
:)
Takes a _long_ time though
It isn't quite FreeBSD's make world, but it a hell of a lot closer to it than anything any other Linux distribution has.
Yes I was rather offended by that.
I am a small ( 1000 user) Australian ISP, but I don't consider myself a 'shonk' or 'cowboy'. I find censorship distasteful, and am not at all pleased with this decision.
It could have been much, much, worse, for that I am thankful.
Well the ISP I run is _fully_ linux.
12 assorted x86 machines, just an ascend router and a tigris remote access server.
We don't have an NT ASP box yet, and probably never will I have anything to do with it.
Last years Australian ISP survey indicated that over half the ISPs in Oz use linux on the majority of their machines, only 19% used NT exclusively, most that had NT machines only used them for ASP serving for virtual clients, some RADIUS servers too (the reason escapes me).
Sorry I can't post the pdf with that in it, as it was given free only to participants of the survey and I am not allowed to leak the info. (oops!)
uid/gids are 16 bits (at least on x86 linux)
They make linux's flexibility sound like a disadvantage. Maybe for people that lack the money/brains, but to my mind enterprise sized systems would be maintained by people with a clue. Solaris ain't no cinch to setup either.
Saying that NT is better than Solaris is just plain dishonest. Funny how they attack linux for poor SMP support but then gloss over the huge difference in NT vrs Solaris SMP support.
Try remote administration of NT boxes you turkies, then tell me unices are hard to maintain remotely.
The article isn't that bad I guess, once you realise that it is just another marketing driven review.
> The problem is that Linux is indeed inferior to
:-).
/source)
;-)
;-)
> most operating systems. Look at BSD/OS and
> FreeBSD. Look at OpenVMS. Look at Digital UNIX.
Digital Unix, don't make me laugh! Open VMS, yeah OK, I'll give you that, the facist OS is much more mature than unix. xBSDs, well I have used lots of unices, and the xBSDs aren't that much different, Linux performs better that all the BSDs I have tried.
> I rarely see a Linux uptime greater than a month
> whereas with the systems listed above, it is
> just expected
Where are you looking? YMMV, but I think that is complete garbage or figures for workstations that are turned of every night
> Linux, specifically GNU, utilities redefine
> bloat.
Yeah they are a bit heavy on memory, but they sure work well compared to the xBSD ones. GNU have the wrong ideas about many things (eg. man pages) but they supply great high performance software that tries hard to be POSIX complient and useful to everyone.
> I have never seen a Linux distribution that came
> with full source that could be rebuilt with one
> command
Slackware: SlackBuild (on every CD in
Not many people need this, but I have used it just for fun a few times (takes a _long_ time). I am sure there are others. It seems to me that your experence of Linux is limited to RedHat et al.
Besides, linux is just a kernel.
> my Linux 2.2.x systems, it fails to properly
> unmount its drives.
Running the old glibc? There is a known issue with the ld-linux.so for oldish versions of libc6. Not a kernel issue, RTFM (well Changes) next time.
There are many other userspace reasons why this may happen also; if you are keeping the fs busy it obviously can't umount it.
> simply quits responding to IP packets for a
> while
Never seen it happen, sounds like the _old_ SCSI bug, or broken/cheap and nasty hardware. More info please, we can help to correct your issue?
> I never see BSD systems broken into
I have, Solaris, Linux, and even VMS too.
It is a userspace problem in almost all cases. The famous broken IMAPd was the same code used on most unices, this is not Linux specific.
> ship without documentation or manpages
Yeah I have to agree here, lazy programmers. Many GNU morons (sorry) seem to think that man pages are old hat, well I think they are better than the info system (yuck). Many interactive commands are self documenting though and come with other format documentation.
All in all, not a great issue, you can always email the guy or ask a friend if you are completely lost. strace comes in handy too
> makes for a great research system
Being open source is great for students, tinkers, and sysadmins alike. I wouldn't limit its usefullness to just research though, I personally use it in mission critial serving for my ISP business, as my desktop OS, for realtime hacking that I do for fun and profit, and as a system to develop portable network apps on.
Hell I use linux for everything I do with computers. If it can't do something I want, I just add the functionality and get on with business, something I can't do in many other unices.
Yeah I used VSTa at Uni. Even got to talk to the author (damn I forgot the man's name, we got on pretty well too).
VSTa was pretty damn bad, it was written mainly as an OS for educational purposes.
I did some realtime hacking on it, bah it was horrible. It would crash when you exited vi or emacs at the wrong time. Close examination of its source led me to believe that the main problem was the semaphore locking was broken (not atomic).
I mentioned that to my tutor and after about half an hour of logical argument he agreed that is was broken and would require significant effort to fix.
I really liked tfork() though, very nice, easier than pthreads or clone(). Still I'd rather use Linux anyday. Message passing by MM is fast and the microkernel design is interesting, but I think I have been too mentally mutalated by monolythic kernel hacking.
OK you got me curious?
What kernel version and card?
I have throughly abused linux networking code and blasted terabytes of data through fast ether cards with only marginal support (at the time), never crashing a box.
The only times I have ever crashed linux has been either my fault (kernel stack smashing et al) or broken/ill supported hardware like my el cheapo PV-BT848 framegrabber.
Yep, I have bothered Zoid and nVidia over TNT support.
Zoid wrote me a nice letter back.
The email(s) to nVidia went unanswered despite the return receipts I got back.
nVidia seems to want me to buy that voodoo2.
Yeah it sucks hey :-/
:(
I am thinking of borrowing a v2 for the '2 weeks' while nVidia gets their butt into gear.
And a Voodoo Banshee I scored for $45 today isn't supported yet
I have a RivaTNT and after repeated attempts to get nvidia to even respond to email I am about to take it back and opt for a lesser product that 3dfx drivers exist for.
It is great under X, but with no OpenGL libs, quake et al are stuck in software mode.
Anyone else want to try talking them into giving linux 3D drivers for it?
HERF and EMP weapons are nothing new.
I don't think too many of the 'HERF gun' guys really have any idea what they are talking about though. The power supply unit for a CW microwave HERF device would be so big and heavy you wouldn't back pack it.
Kinda reminds me of the idea of carrying a rail gun, sure you might just be able to lift it, but the recoil would liquify your skeleton.
EMP generators are practical though, I have a bit of text (hersay basically) on my homepage that talks a little about them. As a matter of fact I recently registered herf.org to put up a site to dispell some of the EM weapon rumours. (it isn't up yet)
I have a 3" round by 10" long device that can kill digital watches and other small unshielded electronics from about 1 metre. It was built from an old photo flash unit reworked to store about 20 Joules. It has an anoying effect of magnetising the shadow mask in VDUs too.
I remember reading about some PET brain scan research that found using a computer in CLI mode uses similar brain centers to human language processing.
Makes sense really, the semantic constructs may be different, but it is still the same fundamental function, parsing or generating a formal structure.
While a lot of the brain is dedicated to processing visual information, I doubt that there is much language learning value from a GUI. The visual centers dump detail as much as possible. The language centers can't do this as that data has a much higher 'entropy' as far as the brains concept processing goes.
Get a sense of humour, I ended up ROFLMAO at some of the comments, particuarly the one about NT being better in bed.
BTW: Humans are just machines too.
Why not just common board and CPU, with 1Gb or so or RAM, and a 100Mb/s ether card, that boots linux of a floppy.
Add a UPS and a harddisk for power-off memory retention if you require it. (don't need the floppy then). Much cheaper than SCSI and not much worse.
I have a degree?
Well I am a Electrical Engineering student, and a Computer Science student, but I have not graduated yet.
If you are looking at the vk2zay, that is my HAM call. I know a fair bit about lasers, I experiment with them a bit at home.
It was just a comment, jeeze... Sounds like envy.
Depends upon how non-linear the lasing system is. At low amplitudes near the lasing threshold any laser will act randomly as far as spatial and temporal coherance goes.
Intresting field of study actually, you can see quantum effects in the macroscopic output of the laser. Although it could be argued that a sparking fluro light or dripping tap demonstrates exactly the same thing.
Does this just sound like conventional data spreading to anyone else? Seems like they are scrambling the data stream with a pseudo-random sequence. What is new about that, lots of modem protocols use it?
I gather they have some other protocol development to improve the data throughput?
More info please, that article is lacking any real information.
Indeed, I even go as far as to write CGIs in gawk. Something to which it is very suited but few seem to try, they just opt for a big, slow, and ugly perl script.
If you know the language well you don't need to start (many) other processes like most people seem to think.
Gawk starts and runs faster than perl5 for most programs (that are suitable for it). In fact I only go to perl when I need to use sockets.
When you get to using sockets, you give up on perl after about 5 minutes and go use C like you should have in the first place.
SPARCs have 'The NSA Instruction'.
Now Intels will have 'The MS Instruction'.
The only possible use of this stupid hardware ID is tracking or software copy 'protection'.
The random number source is nice, but I sure hope it is _random_. Wouldn't suprise me if it is actually pseudo-random and based on the ID.