<div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><br>If you say "after y2/m2/d2 we will begin using GPLv3 only" the issue<br>is already decided and you have to do it. Bowing to pressure is not my
<br>style. If you're not sure about it it's better to say "we will<br>probably begin...". I recognize that this is a highly subjective decision.</blockquote><div><br>This isn't true, strictly speaking. The current license, or a GPLv2/GPLv3 combined license allow anyone to fork the project with that same license, and the original author has that right no less than anyone else. And that isn't even considering the extra rights held by the copyright holder. If Simon was to say "there will be an update released under GPLv3" that might compel him to do so, but nothing can compel him to remove licensing options.
<br></div></div><br>Now, I'm all for the switch to GPLv3 only for dnsmasq. The only argument I've seen here against it is a rather specious "My company has forbidden use of GPLv3 in our codebase" with strong implication "because we intend to TIVOize". My own reading of GPLv3 is that it leaves quite a few loopholes, but it is better than v2. Forcing those who want to TIVOize to use closed source is a win because it gives a company that chooses to enable user customization a competitive advantage. The anti-GPLv3 argument is that more users = better, but the reality recognized by all versions of the GPL is that leeches aren't valuable users.
<br>