On Tue, Apr 1, 2008 at 5:38 PM, James Tucker <jftucker@gmail.com> wrote:

On 1 Apr 2008, at 00:03, Corey Jewett wrote:
>
> On Mar 31, 2008, at 04:53 , XiaoLiang Liu wrote:
>> Can anyone help me on this?
>>
>> I think Ruby development team should have investigated the license
>> issue
>> before release the package.
>>
>> Thanks a lot.
>
> http://www.catb.org/~esr/jargon/html/F/freeware.html
> http://dictionary.reference.com/browse/freeware
>
> The term "freeware" has no clear accepted definition, but it is
> commonly used for packages which permit redistribution but not
> modification (and their source code is not available). These
> packages are not free software, so please don't use "freeware" to
> refer to free software. -- http://www.gnu.org/philosophy/categories.html

Given that this freeware distribution *is* the source code, I don't
see that there should be any restriction to it. In fact, more than
this, the fact that it was well known to be published as source code
by the author, any case law making this software non-public domain
would be damaging to the industry.

> IANAL.

No, nor am I.

> Freeware is different from public domain. Releasing something into
> the public domain means you abandon your copyright and therefore
> your ownership of the work. Freeware is a somewhat arbitrary
> license, but it generally implies that the author has given you all
> rights to use something. Modification and redistribution may or may
> not be questionable.

Traditionally (I'm talking back before everyone had typed "http"),
freeware was common. From what I remember of the general actions of
people, most freeware apps are considered free to distribute, and free
to use, subject to other restrictions in the license. If there are no
specific additional restrictions, one is effectively granted
permission to use the software for whatever one might please. Reverse
engineering of software is rarely regarded as allowable, however, in
this case the source code *is* the delivered product.

With regard to intellectual property law, anything which has a
copyright on it is well protected in most countries with mature IP
law, and as a consequence I cannot see the argument of public domain
dropping copyright or ownership. This is just the same as a virus
author not being able to release themselves from the law by way of
claiming that they only wrote the code, they did not 'release' it.
Similarly MIT licenses have been known not to provide protection in
these cases, however, as far as I know the license remains sound and
case law from such incidents is not used in copyright dispute cases.

> Unfortunately, given that the term freeware is so vague, your best
> bet is to either track down the original author. Perhaps you should
> contact matz directly and ask why it was committed since it appears
> he did the original checkin: http://svn.ruby-lang.org/cgi-bin/viewvc.cgi/trunk/lib/gserver.rb?revision=4102&view=markup

Finally I would just argue that the GServer implementation is very
largely 'standard' for a multi-threaded server design and
implementation, as such I doubt that any significant copyright is
actually owned here. The code is likely to have been derived from that
which is already in the public domain, and moreover it could be
replicated without reference as MIT licensed code in a short time-
frame. Given this, anyone with concern about the license should
probably just do that.

> Corey
>