[Ardour-Dev] Licensing and enforced payments

[Thomas Vecchione]

On Fri, Jan 23, 2009 at 2:25 PM, John Emmas <johne53 at tiscali.co.uk> wrote:

> ----- Original Message ----- From: "John Rigg"
> But only if your SVN assumption is correct.  Back at the start of this
> thread I did a rough & ready analysis of the bugs reported in December.
> Of the 70 (ish) bugs reported, a very significant proportion were one-off
> reports from people using the SAE or OS-X version.  Therefore, although
> I can't be absolutely certain (there's no way to quantify the data) it's
> very likely that these were found by users running a binary copy.  Believe
> it or not, some of the reporters were even still running Version 0.99 !!
>

John I think your analysis is a bit off actually, but I haven't dug into the
bugs reported over the last month yet due to other things keeping getting in
the way.

For one thing, many of the OS X and SAE version bugs reported are likely
discussed in IRC first, and I point them to report it in Mantis.  These are
not just people that report it and assume it will be magically fixed, but
came seeking help first and could easily turn into more detailed testers.
One or two showed an interest in compiling it themselves on OS X which says
a lot.

For another thing, I would question that ANYONE is still using .9x series,
in fact that ability to report bugs on it is soon going to be obsoleted out
of Mantis and any still relevant bugs will be upgraded to a newer version
they are still relevant on(I have already done this with over half of the
ones I found since I started on it in November, just haven't gotten back to
finishing it up because it requires I test every bug individually).


>
> In fact, I only found around 8 'regular' testers (i.e. people who'd
> reported
> 3 bugs or more) and if paid-for SVN had been in place, no doubt all of them
> would be covered by the exemption.
>

See above.


>
> The bottom line is that any effect on bug testing is just speculation.
> Mantis itself would still be free (to all) so I can't believe there'd be a
> significant drop in one-off reports.  Time would be the only way to judge.
>
>
To be honest, the only way I might see supporting this, and it would still
take a lot on my part, is if binaries were availiable with contributed
patches from Mantis.  It is on my list of things int he distant future I
would like to see is the automatically building these binaries.  But in
doing so we would need to also make the source availiable as well due to GPL
requirements.  Now whether this would be generating patches against tagged
versions of Ardour and leaving it like that would satisfy it(Assuming the
binary was built like that which isn't a bad idea) I am unsure of as I
haven't thought about it long enough yet.  BUt that would actually benefit
testing to be honest but still allow the limit you are referring to.  As I
mentioned before though I am still against it as of right now.

          Seablade
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ardour.org/pipermail/ardour-dev-ardour.org/attachments/20090123/e0df4d19/attachment-0002.htm>