|
From: Barry <we...@i1...> - 2005-09-16 06:06:07
|
Craig White wrote: > >---- >you should have at least looked at the license before you commented. >It's not GPL, it's BSD > >http://www.webmin.com/intro.html > > Yeah, I thought this morning that it would work better if it was BSD instead of GPL - that would be fine. But before I looked, Joe Cooper wrote (and I trust he should know): > Thanks for the encouragement! We felt that way too. It just became > apparent that it was the best way to answer the needs of commercial > hosting providers, without causing any pain to users of the GPL version. Hey wait , even though Joe wrote that, I followed your advice and checked - we are talking about virtual min, not webmin. And on http://www.swelltech.com/virtualmin/ it says: "09/12/2005 Virtualmin had gone Pro! Visit Virtualmin.com to find out more. We're now providing two versions of Virtualmin, one GPL and one proprietary with additional features. The GPL version will trail development of the proprietary version by some amount." and "07/23/2003 Ransom for the first milestone has been raised. Tye Gaddis and Hosting4Less.com were the contributors today who pushed us over the top! The current module has been released under a GPL license today. Kudos to all of the sponsors." So, the code base we are talking about has been GPL for 2+ years. I stand by my comments and questions. On what legal basis can GPL code be productized, released, and source code *not* made available? Not trying to stir up trouble, but with such a high profile project as this, it is important to not make any missteps that will misguide the community in what GPL actually means, and not to put the code at risk. This is interesting: "What's the license? Virtualmin is licensed under the GNU General Public License. Virtualmin is periodically ransomed off under a non-GPL license to fund major new developments. A $200, or higher, sponsorship is required for you to receive the module including the "ransom" features. You may then use the module in any way you see fit for commercial or non-commercial purposes. However, you cannot resell the module, relicense the module, or give away the module, without discussing it with us (and purchasing the rights to do so)." Is there any other history of other projects doing this, and having the general community that guards the GPL approve? Maybe there is, I don't know for sure. But I think you guys should look very closely at this model before jumping and creating a legal mess and expense for yourselves. And if you have it on some authority that it is fine, then can you please cite that authority so the rest of us can consider this method for promoting GPL code too? Here is an opinion from folks at Swelltech: http://www.swelltech.com/virtualmin/ransomware.html that cites no legal basis at all, but is filled with warm goodhearted intentions I wholly agree with. I would maintain that given otherwise, this "ransomware" concept is about as contrary to the spirit and the letter of GPL as you can get. It is quite simple really - you release the product to even one person, you must release the code under the GPL to them (and possibly to everyone?). Now, if it was called "Virtualmin Beta" instead of "Virtualmin Pro" you might be able to make the rhetorical case that the product isn't yet really released per se, but you didn't hear that from me because I believe that any GPL project should have a public anonymous-read source code control system with all available code :) >Furthermore, the sweat and the equity is to those that put it into it >which as far as I know, has to be like at least 98% Jamie if not higher >but only those who contributed code to it are entitled to question it >and I doubt there are many who have earned that right. > > That is a nice warm fuzzy but simply not true. The GPL is the GPL. You can read it at gnu.org. It doesn't say "some people have rights that others don't have". It says what it says. You simply can not fork a non-GPL branch off of a GPL code base. That is precisely the point of GPL. Otherwise, companies like JBOSS would do that in a second and they would have a completely different business model then they do. If you think anyone in the user base can not question what goes on, and make a difference, review the fiasco at the as-popular-as-webmin-and-maybe-more Mambo/Joomla split last month. See mamboserver.com discussion forums, and then opensourcematters.com (or .org not sure which) where the developers moved to first, and then to joomla.com where the developers landed. Try posting in those three places that the users have no say and no clout about the direction of the code and see what happens :) Unless you are RMS, who sees only the GPL and "everything else", you will see that this is the reason why other licenses exist. so, if the product is not GPL then it doesn't matter because the other license will allow what I was asking about and that is fine. Perhaps you think *I* have no standing in the matter, or that I am shooting in the breeze. You would be very wrong. I was one of the early development managers at Cobalt Networks deciding these issues every day. Cobalt was the first wildly successful company to blend open source (mostly but not all GPL) with proprietary code. The code served essentially the same functions that webmin and usermin does now: managing hosted Linux servers. And yes, I had my hands in an awful lot of it at the code level. I can offer quite a bit of insight into what makes people spend money on a product like this, and yes, I have been paid for writing, testing, and marketing this type of code for worldwide installations. >Lastly, if Jamie and Joe can prosper by taking their base and even >devote more time and energy, then the code base would have to benefit so >there's no reason to assume that it wouldn't be a winner all around. > > Agreed that they should be allowed to prosper if they can. But violating the license they chose would not be the right way to do it. I trust they are getting good advice on this matter from Open Country - I know folks there, and I hope they wouldn't blunder on this point. >Clearly Jamie has the benefit of each and every doubt besides the >gratitude for his generosity this far. If he decided to not develop it >any further, I would presume that it could be forked under BSD >restrictions. > > It is not an issue of benefit of the doubt. It is a matter of adhering to the license terms that were freely chosen. Maybe this is allowed, I am jsut asking for citations in the gnu community where they guard this kind of thing jealously, and have seen it all before. I highly doubt the gnu community would ever accept a bsd fork from gpl code being acceptable. I have been in meetings with companies such as Yahoo, Apple, Hotmail/Microsoft where they all raised the issue of the ability to fork BSD *nix without sharing back as the reason why they chose BSD instead of Linux. It could be forked within GPL though regardless of if Jamie or anyone else chooses to develop it or not. This is basically what happened with Mambo a few weeks ago, which you will recall I raised as an issue last week, before this Virtualmin announcement even occurred. As I understood the response, we were told no forks would happen. Yet only days later, we are told there will be, for the first time, a "professional" branch, and a non-professional branch. It was this kind of sloppiness in making announcements that caused the otherwise unnecessary (IMHO) hard feelings in the Mambo community. BTW, for potential developers of open source code: Consider the license you choose *very carefully*. What starts out as a hobby or jsut a few lines of code you give to the community may in fact be restricted form making you the kind of money that might be available to you later, sometimes much later, based on the early decision you made. Seriously. Best, Barry |