Maybe I have a different definition of insightful. Addionally I didn't say that this posting wasn't worth the high rating (it was!), I just wouldn't call it insightfull, it just call it a good question (which, as you said sometimes is equally good or even better than a insightful comment).
The difference is that those are static formats that don't run any code (at least if you believe in the difference between code and date).
Additionally there are quite some different gif and jpg parsers out there, but the number of usefull Flash-Players is rather limited (1 comes to my mind). So if you'd be able to make a gif file that runs arbitary code on the machine that views it, it would most probably be targeted only on this gif-reader software (and this version, and this platform, and...).
And I think the checks form alformed GIF and JPEGs are rather strict in most image-loading libraries, 'cause defect GIFs and JPEGs are known to exist.
For Java to do anything bad it has to have explicit permission from the user. In that case, in my opinion, it isn't a virus, just a dangerous program and the user should acuatlly read the warning boxes.
I don't think that this disqualifies it as a virus. The user may accept that the program may "access the local file system", but he certainly doesn't want it to trash his harddisk.
Additionally I'd keep in mind that "Users don't read documentation" which can be gerneralized to "Users don't read.", so Joe Average won't be interested what the message box says that stops him from playing with this cool "web thingy" (which in technical terms could be described as an Java Applet), he just wants to find out which button he must press for the warning dialog to go away.
Right. But a scripting language, that can't get out of its sandbox is rather useless (except for some special cases like Flash). A scripting language without a sandbox is of course much worse.
But there are two ways a script can get out of a sanbox (in some languages there is only one...):
A Bug in the Sandbox. this is the most obvious but can be avoided rather well with some good design (not completly of course, good code and constant security audits are still needed)
The 'official' way.
As I said a scripting language without a official way out of the Sandbox is rather limited. In Java (not strictly a scripting language, but the Sandbox I'm most familiar with) an Applet can escape the Sandbox if it is both signed and gets the permission by the user (the signing part can be skiped, but therefore you have to modify client settings). We all know that the permission of the user is only a problem of social engineering and virus authors are pretty good in this (or at least good enough for Joe Outlook-User out there).
The signing part is actually quite good. A virus author would have to get a valid, certified key from an Certification Aurthority (like Verisign) and sign the Virus with this key... well, this obviously would be stupid, except if he is planing to find out about live in prison pretty fast.
Now the really big problems arise when a [scripting] language allows a script/program to escape the sandbox, when it is not sign (or is sign with a self-signed certificate), even when it does so after a big red flashing DONT-EVER-DO-THIS sign, where the user has to enter a 12-digit prime number he has to calculate from a formular that is printed on page 123 of his handbook... in reverse, using polish translation. Nothing of this would prevent the user from executing harmfull, unkown code.
Actually I just remembered a third method, or rather a combination of the first two: A bug in ther Certification-Check-System. IIRC Netscape had some in their 4.x-releases that allowed any valid Signature to verify the validity of any host and not just that of the host it was made for.
I'm on a Karma-Trip, so I'll burn some of it by asking a offtopic meta-question:
I myself wouldn't really call the parent-posting "Insightful", rather quite the opposite (it even mentioned the lack of insight), but I think it is a "+1, Good Question", don't we need more different Reasons? Especially now that we can grade the Reasons (maybe I'm into questions and I hate all those "+1, Insightful" and "+1, Interesting" posts, giving them -6, but I really dig those with "+1, Good Question" and "-1, Redundant" (Redundancy never harmed anyone was harldy harmed by Redundancy).
And what about "-1, Karma whore", hell even "+1, Karma Whore", or more neutral "+0, Karma whore"?
Man, I'm happy that I don't have to moderate this comment, I wouldn't know what to do... maybe "+1, Offtopic"?
Nope, wouldn't work either. The best you can get on average over all possible inputs is 1:1.
Of course. But noone is actually likely to work with a significant perfentage of all possible inputs. What I want to say, that each usefull data, that is not already compressed is less than random (otherwise it wouldn't be useless). The really interesting average is that over the average/home/foo
I'm going to agree with you here. If there's no pattern in the data, how can you find one and compress it. The reason things like gzip work well on c files (for instance) is because C code is far from random. How many times do you use void or int in a C file? a lot:)
This is a proof ('though I doubt it is a scientificly correct one), that you can't get lossless compression with a constant compression factor! What they claim would be theroretically possible if 100:1 where an average, but I still don't think this is possible.
So where do I enter this password in my old, trustworthy 10" monochrom vt220 (or my PuTTY at work if you're reaction to the former is "yuck! those should've died thousands of years ago").
Start your own company (or get really important in an already established organization) and then refuse to accept any resumes but those in PDF format.
So your company will only employ professional layouters and some geeks who are willing to pipe tex into dvi2ps into ps2pdf. Maybe thats what you want, but maybe not. At least I'd suggest accepting text/html and text/plain as well.
Well not only the TCP/IP-Stack, but when you open IE and look at the about box, you'll see that MSIE is based on NCSA Mosaic, which was at least somewhat open-source (look at the initial announcement and the README).
Much of the internet community has wised up and now only post under psuedonyms.
First of all I'd not confuse Internet with Usenet. Discussions on Internet Forums are generally of much lesser quality than in many Usenet-groups. (with exceptions, as allways). Additionally there are parts of the Usenet where a Real Name is still considered to be an important part in the From:-Header-Field. (like the entire de.* and at.* hierarchies (that is almost the entire german-speaking part of the Usenet)) Of course you can put any name you want in there, but most regulars really use the real name.
Younger folks probably won't find this too interesting as it will be more like history to them rather than us old farts re-living younger days...
Well, I think I'm more into the "younger folks"-category (although, when I look at the age of some dot-com-millionaires I think I might not; at least I was alive, before the first post;-) but I still find this quite interesting. Especially reading the Linus vs. Tanenbaum dispute (which I already read before, but not in the google-view, which I got used to for up-to-date infos;-), or the problems in the First Post (tm). They are quite fascinating. And I allways knew that Usenet is much older than "the Internet" but this is something else, you get to feel that this is an old beast (in web-years only, of course).
Most the spam I get comes through open relays in China, Korea and Japan.
And the phone numbers you should call, or the PO-Boxes you should send the money to, or the incredibbly trustworthy companies you should invest in are located in _______ (insert correct answer here).
> I filtered on [a-z]*[0-9]{3,}@.*...).
> You do realize that this blocks out more valid email addresses than just the example [a-z]@domain you gave, correct?
well... everything that comes from [a-z]*[0-9]{3,}@.* will be moved to my "probably spam" folder, which I got trough once a week and delete. It is not ment to catch valid e-mail-addresses, but computer-generated accounts like somelamename1234@yahoo.com or spammer0815@hotmail.com. Noone I know uses an e-mail-address that matches this regexp (apart from said students, for whom I now have an extra "this-is-not-spam"-rule).
So apart from the specific key combination it has nothing at all to do with pine, right?
And IMHO filtering in the MUA is the most low-level spam-prevention possibility. There are several fine anti-spam systems out there.
'though I myself don't use them, yet. I make a sport out of finding a rule that catches the spam that reaches my inbox and is sensible enough not to filter any other mail (just recently I got a problem, that I couldn't receive mail from any student of a specifiy university, 'cause they got [a-z]@university-domain and I filtered on [a-z]*[0-9]{3,}@.*...). It certainly isn't the most effective way, but... well, at least I learn some regular expressions;-)
If you ever find out how to calculate this one magic number, you'll be quite famous). Don't you think that all the benchmarking is for a reasons? There is no single number that measures the "Fastness" of a PC. Thats why each benchmark that is worth looking at puts multiple Numbers for each candidate out (Not that this stops everybody from takeing one of those and saying "This one is the fast-value")
How about making a GIMP-based ultra-mini photo studio tool?
I don't think that you'll win this content with anything GIMP-based. GIMP is based on GTK (and GDK), which are essentially concurent products to QT (though not yet as strong in the embedded sector AFAIK). Additionally you'll have essentially rewrite the GIMP, 'cause GTK and QT are quite different (both from design and implementation, not to speak of two different languages). A nice port of Konqueror would be nice, but that has already been done.
If you're running NTFS, AND you've been hit, *sigh*..
If you're in a german-speaking country you might want to fetch the most recent issue of c't. They got an article about Virus-Cleaning on NTFS-Platforms (from DOS and Win9x). Take a look at The download Links for the article. I don't think the article itself is available on the net. It's german but I'm sure even those of you, who don't speak this language will find a way through ("NTFS", "DOS" and "Download" are the same;-)
A quick search on vil.nai.com for "Tiny" turns up sever small Virii. The smallest beeing Tiny Di with 94-110 Bytes.
But I think that is only possible because.com (the only files those virii infect) are much simpler in design than.exe (not to speak of.exe-files running in win32) and those virii had no way of spreading over a network on themselfes (they depended on some person to distribute the infected file in some way).
Aliz has the ability to distribute via the network and is much smaller than Goner (just 4098 Bytes).
All those Virii definitley don't come out of a Virus-Construction-Set (yet).
Of course I've seen the missing "Bytes" in the split second between pressing submit in the Preview-Page and the loading of the newly posted comment... Sigh...
Please check the facts! It's _not_ 159! Not even the first self-replicating Virii were this small (AFAIK). It's aprox. 159 kb if unpacked from its PE-compressed format! The File you have to download to enjoy the virus is aprox. 38 kb.
You're right... partly. I wouldn't have given it a second glance, but that doesn't mean, that it wouldn't have been so interesting (or "+1, Insightful").
I might not have read it if Linus weren't involved (I might have nevertheless, 'cause Alan Cox is involved;-), but that doesn't mean that it's not good. It would have been worth beeing postet even if Linus weren't involved. If your and your friends discussion was equally interesting I'd love to read the logs.
Slashdot Mirror
Maybe I have a different definition of insightful. Addionally I didn't say that this posting wasn't worth the high rating (it was!), I just wouldn't call it insightfull, it just call it a good question (which, as you said sometimes is equally good or even better than a insightful comment).
The difference is that those are static formats that don't run any code (at least if you believe in the difference between code and date).
...).
Additionally there are quite some different gif and jpg parsers out there, but the number of usefull Flash-Players is rather limited (1 comes to my mind). So if you'd be able to make a gif file that runs arbitary code on the machine that views it, it would most probably be targeted only on this gif-reader software (and this version, and this platform, and
And I think the checks form alformed GIF and JPEGs are rather strict in most image-loading libraries, 'cause defect GIFs and JPEGs are known to exist.
I don't think that this disqualifies it as a virus. The user may accept that the program may "access the local file system", but he certainly doesn't want it to trash his harddisk.
Additionally I'd keep in mind that "Users don't read documentation" which can be gerneralized to "Users don't read.", so Joe Average won't be interested what the message box says that stops him from playing with this cool "web thingy" (which in technical terms could be described as an Java Applet), he just wants to find out which button he must press for the warning dialog to go away.
Right. But a scripting language, that can't get out of its sandbox is rather useless (except for some special cases like Flash). A scripting language without a sandbox is of course much worse.
But there are two ways a script can get out of a sanbox (in some languages there is only one ...):
As I said a scripting language without a official way out of the Sandbox is rather limited. In Java (not strictly a scripting language, but the Sandbox I'm most familiar with) an Applet can escape the Sandbox if it is both signed and gets the permission by the user (the signing part can be skiped, but therefore you have to modify client settings). We all know that the permission of the user is only a problem of social engineering and virus authors are pretty good in this (or at least good enough for Joe Outlook-User out there).
The signing part is actually quite good. A virus author would have to get a valid, certified key from an Certification Aurthority (like Verisign) and sign the Virus with this key ... well, this obviously would be stupid, except if he is planing to find out about live in prison pretty fast.
Now the really big problems arise when a [scripting] language allows a script/program to escape the sandbox, when it is not sign (or is sign with a self-signed certificate), even when it does so after a big red flashing DONT-EVER-DO-THIS sign, where the user has to enter a 12-digit prime number he has to calculate from a formular that is printed on page 123 of his handbook ... in reverse, using polish translation. Nothing of this would prevent the user from executing harmfull, unkown code.
Actually I just remembered a third method, or rather a combination of the first two: A bug in ther Certification-Check-System. IIRC Netscape had some in their 4.x-releases that allowed any valid Signature to verify the validity of any host and not just that of the host it was made for.
I'm on a Karma-Trip, so I'll burn some of it by asking a offtopic meta-question:
... maybe "+1, Offtopic"?
I myself wouldn't really call the parent-posting "Insightful", rather quite the opposite (it even mentioned the lack of insight), but I think it is a "+1, Good Question", don't we need more different Reasons? Especially now that we can grade the Reasons (maybe I'm into questions and I hate all those "+1, Insightful" and "+1, Interesting" posts, giving them -6, but I really dig those with "+1, Good Question" and "-1, Redundant" (Redundancy never harmed anyone was harldy harmed by Redundancy).
And what about "-1, Karma whore", hell even "+1, Karma Whore", or more neutral "+0, Karma whore"?
Man, I'm happy that I don't have to moderate this comment, I wouldn't know what to do
Of course. But noone is actually likely to work with a significant perfentage of all possible inputs. What I want to say, that each usefull data, that is not already compressed is less than random (otherwise it wouldn't be useless). The really interesting average is that over the average /home/foo
So a perl programm can't be compressed?
This is a proof ('though I doubt it is a scientificly correct one), that you can't get lossless compression with a constant compression factor! What they claim would be theroretically possible if 100:1 where an average, but I still don't think this is possible.
So where do I enter this password in my old, trustworthy 10" monochrom vt220 (or my PuTTY at work if you're reaction to the former is "yuck! those should've died thousands of years ago").
So your company will only employ professional layouters and some geeks who are willing to pipe tex into dvi2ps into ps2pdf. Maybe thats what you want, but maybe not. At least I'd suggest accepting text/html and text/plain as well.
Well not only the TCP/IP-Stack, but when you open IE and look at the about box, you'll see that MSIE is based on NCSA Mosaic, which was at least somewhat open-source (look at the initial announcement and the README).
First of all I'd not confuse Internet with Usenet. Discussions on Internet Forums are generally of much lesser quality than in many Usenet-groups. (with exceptions, as allways). Additionally there are parts of the Usenet where a Real Name is still considered to be an important part in the From:-Header-Field. (like the entire de.* and at.* hierarchies (that is almost the entire german-speaking part of the Usenet)) Of course you can put any name you want in there, but most regulars really use the real name.
Well, I think I'm more into the "younger folks"-category (although, when I look at the age of some dot-com-millionaires I think I might not; at least I was alive, before the first post ;-) but I still find this quite interesting. Especially reading the Linus vs. Tanenbaum dispute (which I already read before, but not in the google-view, which I got used to for up-to-date infos ;-), or the problems in the First Post (tm). They are quite fascinating. And I allways knew that Usenet is much older than "the Internet" but this is something else, you get to feel that this is an old beast (in web-years only, of course).
And the phone numbers you should call, or the PO-Boxes you should send the money to, or the incredibbly trustworthy companies you should invest in are located in _______ (insert correct answer here).
> I filtered on [a-z]*[0-9]{3,}@.* ...).
... everything that comes from [a-z]*[0-9]{3,}@.* will be moved to my "probably spam" folder, which I got trough once a week and delete. It is not ment to catch valid e-mail-addresses, but computer-generated accounts like somelamename1234@yahoo.com or spammer0815@hotmail.com. Noone I know uses an e-mail-address that matches this regexp (apart from said students, for whom I now have an extra "this-is-not-spam"-rule).
> You do realize that this blocks out more valid email addresses than just the example [a-z]@domain you gave, correct?
well
So apart from the specific key combination it has nothing at all to do with pine, right?
...). It certainly isn't the most effective way, but ... well, at least I learn some regular expressions ;-)
And IMHO filtering in the MUA is the most low-level spam-prevention possibility. There are several fine anti-spam systems out there.
'though I myself don't use them, yet. I make a sport out of finding a rule that catches the spam that reaches my inbox and is sensible enough not to filter any other mail (just recently I got a problem, that I couldn't receive mail from any student of a specifiy university, 'cause they got [a-z]@university-domain and I filtered on [a-z]*[0-9]{3,}@.*
If you ever find out how to calculate this one magic number, you'll be quite famous). Don't you think that all the benchmarking is for a reasons? There is no single number that measures the "Fastness" of a PC. Thats why each benchmark that is worth looking at puts multiple Numbers for each candidate out (Not that this stops everybody from takeing one of those and saying "This one is the fast-value")
'Cause a Makefile will at least make the errors reproduceable. Try to teach that to an intern.
I don't think that you'll win this content with anything GIMP-based. GIMP is based on GTK (and GDK), which are essentially concurent products to QT (though not yet as strong in the embedded sector AFAIK). Additionally you'll have essentially rewrite the GIMP, 'cause GTK and QT are quite different (both from design and implementation, not to speak of two different languages). A nice port of Konqueror would be nice, but that has already been done.
If you're in a german-speaking country you might want to fetch the most recent issue of c't. They got an article about Virus-Cleaning on NTFS-Platforms (from DOS and Win9x). Take a look at The download Links for the article. I don't think the article itself is available on the net. It's german but I'm sure even those of you, who don't speak this language will find a way through ("NTFS", "DOS" and "Download" are the same ;-)
Wow! almost ...
A quick search on vil.nai.com for "Tiny" turns up sever small Virii. The smallest beeing Tiny Di with 94-110 Bytes.
But I think that is only possible because .com (the only files those virii infect) are much simpler in design than .exe (not to speak of .exe-files running in win32) and those virii had no way of spreading over a network on themselfes (they depended on some person to distribute the infected file in some way).
Aliz has the ability to distribute via the network and is much smaller than Goner (just 4098 Bytes).
All those Virii definitley don't come out of a Virus-Construction-Set (yet).
Of course I've seen the missing "Bytes" in the split second between pressing submit in the Preview-Page and the loading of the newly posted comment ... Sigh ...
Please check the facts! It's _not_ 159! Not even the first self-replicating Virii were this small (AFAIK). It's aprox. 159 kb if unpacked from its PE-compressed format! The File you have to download to enjoy the virus is aprox. 38 kb.
You're right ... partly. I wouldn't have given it a second glance, but that doesn't mean, that it wouldn't have been so interesting (or "+1, Insightful").
;-), but that doesn't mean that it's not good. It would have been worth beeing postet even if Linus weren't involved. If your and your friends discussion was equally interesting I'd love to read the logs.
I might not have read it if Linus weren't involved (I might have nevertheless, 'cause Alan Cox is involved
They even succeeded in providing a point & click local root exploit (for details take a look at Bugtraq).
I don't know if they are the first to offer this feature, but it's definitley nice.