Closed Source Reverse Engineering

[billy_gates]

I know that reverse engineering closed source software is probably extremely difficult and time consuming.  However if you do reverse engineer the source of software, is that copyright infringement, to then change it and redistribute it?

[rtgwbmsr]

quote:Originally posted by Billy Gates: Mac Commando:
I know that reverse engineering closed source software is probably extremely difficult and time consuming.  However if you do reverse engineer the source of software, is that copyright infringement, to then change it and redistribute it?

Yes, and yes. In the case of Winblows and many major Winblows software titles you are also violating the EULA.

[ December 10, 2002: Message edited by: The Muffin Man ]

[Pissed_Macman]

That sounds exciting.

[Pantso]

quote:Originally posted by Billy Gates: Mac Commando:
I know that reverse engineering closed source software is probably extremely difficult and time consuming.  However if you do reverse engineer the source of software, is that copyright infringement, to then change it and redistribute it?

Yes it certainly is, since most of that software is released under the EULA, which most of the times (if not all) forbids any kind of reverse engineering attempt to the code of the respective software.

[ December 10, 2002: Message edited by: Panos: Mac Commando ]

[TheQuirk]

quote:Originally posted by Billy Gates: Mac Commando:
I know that reverse engineering closed source software is probably extremely difficult and time consuming.  However if you do reverse engineer the source of software, is that copyright infringement, to then change it and redistribute it?

What are you planning on reverse-engineering? It might be legal under its EULA. . .

[voidmain]

Reverse engineering is allowed to an extent. I would have to do some research so don't take what I say as fact but I believe it to be so. I believe that you are allowed to reverse engineer closed source software that do not document their protocols and interfaces so as to make your software work/communicate with said closed source software.

That is what allows Samba to exist and allows it to communicate with Windows on many levels. However, I believe with Samba most of the reverse engineering is done to the protocol and not necessarily the Windows software itself. That is, network sniffers come in handy to figure out what the closed source software is doing. Again, this is from memory of reading several articles and discussions about it, and if I recall there have been court rulings on such activity to support it.

Someone please correct me if/where I am wrong on this.

[lazygamer]

Just how possible is it to reverse engineer? Do you just put binary code into a program that turns it into incomplete sourcecode?

[flap]

Getting from binary code to high level source code is known as decompilation, and is not easy. There are some decompilers available but I have no idea how well they work. The problem with decompiling code is that high level code -> machine code is not a simple 1:1 translation. That is, one particular piece of object code could be produced from any number of pieces of high level code, so the translation back is ambiguous.

Disassembling, on the other hand (translation from machine code to assembly code) *is* a simple 1:1 translation so is more reliable/easy. But it's obviously harder to get useful information about a program's workings from assembler than it is from C/C++ etc.

[choasforages]

and for legal issues involving coding, after seeing an asm dump, you are no longer a "virgin" and you could comprimise the licensing of an opensource project. now with the attack shark spray used, what are trying to take apart?

[Kintaro]

Its not about law, but weather you are going to get caught or not.

[beltorak0]

from void main:
 

quote:Reverse engineering is allowed to an extent. I would have to do some research so don't take what I say as fact but I believe it to be so. I believe that you are allowed to reverse engineer closed source software that do not document their protocols and interfaces so as to make your software work/communicate with said closed source software.  

You are right as far as I know -- i don't have the time to look it up right now, but I believe the act was passed right along with the "fair use" stuff.  Basically, lazygamer, all you do is plug things into {software}'s input ans see what comes out as output. The same process allowes OO.org and Abiword read M$ Word Docs -- although not perfectly.  I can never get any embedded images to show.  It is also what alows read support for NTFS under linux -- again not perfectly, especially write support.

Just wait until Longhorn and the new Database File System comes out.  I was wondering how M$ was going to bundle thier half-ass SQL server into the OS; now I know.  How bloated will that make the kernel?  Not to mention the damage caused by bugs.  Since the whole point of a database information setup is to eliminate redundancy, much of the information stored on the disk will be nothing more than "links" to other parts of information.  That creates a worse single-point of failure than the registry does.  At least you can back up the registry.  Now eveytime you download your email you will have to "ghost" the harddrive just to be sure.

whatever.

-t.