Slashdot Mirror
On the Ethics of a Code Split?
McWizard asks: "We've recently had a code split at a project I'm leading. (No name given, as this is a question, not an advertisement campaign). While both projects have done some major design decisions in opposing directions, we've been keeping a close eye on the changelog of the spinoff for small changes that could be used. So, whenever we've found an interesting piece of code (mostly GUI stuff, nothing longer than 20 lines of code), we transferred it to our project and gave credit to the spinoff team in the changelog.
What does Slashdot say on that matter? Is this unethical or are such things fair game?"
"Yesterday, I was contacted by the leader of the spinoff project who told me that he's quiet angry at us for doing that and that it's considered unethical and rude to copy code from the spinoff.
As both projects are under the GPL, we have an opposing opinion on that matter and we've more than once invited him to copy code from our project. Nevertheless he's thinking about obfuscating his changelog and only open the source as packages when he's doing a release, which is, as he says, his right under the GPL."
In every case I've ever heard of, the primary reason why a fork has happened is because the lead of the original project was percieved to be an obnoxious megalomaniac. From what I've read this was true of XFree86, and from what I've been reading if it happens to Gnome, it'll be true there as well. (And yes, I'm privately inclined to believe it was at least partially true in the case of the Emacs/XEmacs fork as well ;-))
It sounds like the leader of the project you were involved with is now attempting to verify this truism by preventing you from copying the parent project's code to your new tree. Tell him that unless he either a) is the original copyright holder, and b) therefore wishes to change the code's license, (in which case he fairly obviously never intended to abide by the GPL in the first place) that he can go and perform an anatomically impossible act with a shovel with your blessing, because as long as the GPL is binding on the project, he won't be able to do anything else.
To RMS' credit, while he might not have been happy about the XEmacs fork (and he wasn't...I've read the email archives on this) but other than claiming Richard Gabriel had shown a "bad spirit" he never explicitly tried to stop it from happening to my knowledge, precisely because he would have known that forking is one of the rights that the GPL specifically grants.
The right to fork is crucial, because it protects against that part of human frailty which causes the behaviour of some of us to degenerate into fascism. If the leader of a project that you've devoted considerable time and effort to for whatever reason suddenly decides to start being a control freak, the right to fork ensures that the effort you've already invested will not go to waste. You can simply copy the project and relocate said copy to your own site/machine, and then continue working on it.
I agree completely that credit should be given in the changelog/wherever else to whoever has worked on the parent code, but for the parent project's lead to try and prevent forking of it if it uses the GPL is completely wrong, IMHO. In the XEmacs situation RMS might have tried to dominate people in spirit, but in practice he was able I think to recognise the necessity of abiding by his own rules.