svante
Newbie
Posts: 1
|
 |
Re: AxCrypt 1.6.4.1
« Reply #6 on: Apr 13th, 2008, 2:59pm » |
|
Hello,
I am the author of AxCrypt, and also the writer of the comment about TrojanHunter, deemed unprofessional in this thread.
I would like to point a few things, before being judged.
The comment was made in a personal and private e-mail addressed to "lolpo kiujk" (here using the name 'enriewill'), which sent me a e-mail with a false positive report from TrojanHunter stating that the AxCrypt installer contained a trojan. I found the e-mail a bit agressive in tone, and choose to respond to the specific question in an ironic manner.
The comment was not directed towards the makers of TrojanHunter in general, but was a personal response made in frustration with both the fact of the false positive, as well as the tone of voice in the e-mail from "lolpo kiujk" (obviously not a real name, but rather a random hammering on the keyboard).
Now for some background to explain a bit of the frustration.
AxCrypt is free and open source. Every last bit of source code is available for download, perusal and recompilation by anyone with the appropriate tools. It also uses other open source components, among them NSIS, a program to make installers for Windows.
NSIS comes with many plug-ins. One of them is called 'nsisdl', and is a simple component that can perform a HTTP GET and retrieve the result. This is used by the AxCrypt installer to send the registration information (if allowed by the user - nothing whatsoever is sent without the users consent) to our server, so we have somewhere to send notifications of upgrades.
Several years ago, Norton Anti-Virus picked up on this plug-in component and flagged AxCrypt and numerous other softwares as malware - all being false positive - just for having the capability of issuing a HTTP GET (that's what any browser does anytime you enter something in the address bar and hit return). Thereafter during the coming year, one after another of the anti-virus software makers did the same thing. Each time I had to defend myself from these false claims, and a general feeling of helplessnes came with it. What can I do? A single developer of free software agains Norton et. al.? Not much - fortunately bigger fish than me were troubled by this and slowly it resolved it-self.
Since then, perhaps once a year, along comes a new maker of scanning software claiming that AxCrypt contains this or that malware. I get the feeling that these scanning signatures are slowly making the global rounds, popping up in new places time after time again.
It is very frustrating having to defend myself against these claims time after time, and even more so when I get angry e-mails from various users around the globe who think that I'm a bad guy trying to take control over their computers or whatever.
I have yet to receive so much as an apology from any anti-virus maker for the trouble caused. Not once have I heard anything, at best the false positive has been removed more or less promptly.
So, when this popped up again today, in the form of an e-mail (this was not the first actually about TrojanHunter, there was another one the same day) that was formulated rather agressive as I found it, I vented my frustration a bit.
I was not expecting the sender of the e-mail to publish this private e-mail in a public forum without asking my permission, which would have been denied.
If there's a 'report abuse' function of this board I will ask to have the post removed - if not, and a moderator reads this, please remove the quotation earlier in this thread by 'enriewill' of the personal e-mail sent by me to him. It was never intended for publication.
I would not use a derogatory remark in a public forum, although I still am of the opinion that makers of anti-virus software takes these issues far too lightly. A single false positive can cost years of reputation-building, and there is absolutely no defense against frivolous signature pattern matching like this.
I have even gone to the length of purchasing a certificate enabling me to digitally sign the installer with Authenticode. One would think that an anti-virus scanner would think twice before reporting an officially signed and verifiable executable as malware... After all, it's not like I'm anonymous if there's a digital signature on the executable.
Best regards,
Svante (my real name by the way).
|
| « Last Edit: Apr 13th, 2008, 3:22pm by svante » |
 IP Logged |
|
|
|
Mischel Internet Security Forum
Notify of replies
Send Topic
Print
Author
Search
Members
Login
Register