On Sun, Mar 18, 2007, Maxim Veksler wrote about "Using the code of dead project - Political question": > Hello List, > A political question if I may. > > What to do if you wish to use the code of some GPL'ed project hosted > on sf.net but the project is dead? Do you fork the project for your > own code, or do you start maintaining two different projects?
As someone with a bit of experience with dead projects (I wrote one, sendSMS), I think the answer is yes: you can, and should, fork the project. (by the way, any takers for SendSMS? ;-)) The other alternative you mention - starting from scratch - makes sense if the original project doesn't have much code you can use, or if you think you can do better if you start from scratch. If the original project has a lot of useful code that will take you months or years to rewrite, then the answer should be obvious - reuse the original project, or at least parts of its code. > How should I approach this? Should I contact the developer and ask his > permission to do so? Should I simply inform him that I will be forking > his code? I think you should politely write this guy that you plan to fork his project, but you don't need to ask for his permission. On your new project's site you should prominently explain the situation - that your project is a fork of, or based on, or contains parts of, (or whatever is applicable) of the original XYZ project, but you should be contacted about problems with this version. If you take a big project and only make a small change or add a small part, make sure you admit so. One example of someone who did this properly is Mitz Pettel's Hebrew Spelling Service for macintosh (http://www.mitzpettel.com/software/hspell.php), which fills an important need that the Hspell developers (Dan and myself) could not fill or did not bother to fill, but it clearly admits what part is new to this project, and what part was taken from the other project. Also, try to be polite when you choose a name for the new project. Two prominent examples of offensive fork names come to mind and can help you think how NOT to name your project: The first example is the Emacs fork, which (after several incarnations) was named "XEmacs". The name "XEmacs" unfairly led people to believe that the difference between "Emacs" and "XEmacs" is XEmacs' better, or perhaps exclusive, X-Windows support. This may have been true for a short time, but over a decade after not being true any more, the name persists. The second, example, is Mosix vs. OpenMosix. Mosix is an interesting cluster OS project from HUJI, that started its life as patches to a proprietary operating system (BSDI) but later became free software on top of Linux (see mosix.org). Someone who owned a company that tried to profit from Mosix decided to fork it to have better control over its development, and called the new free fork "OpenMosix", which implied that the original was "closed" in some way. This turned out to be a self-fulfilling prophecy, when the developers of Mosix decided to turn it back to being proprietary :( -- Nadav Har'El | Sunday, Mar 18 2007, 28 Adar 5767 [EMAIL PROTECTED] |----------------------------------------- Phone +972-523-790466, ICQ 13349191 |My typos are intentional copyright traps. http://nadav.harel.org.il | ================================================================= To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
