On Sat, 20 Feb 2005 firstname.lastname@example.org wrote:
> Brian Behlendorf <email@example.com> writes:
>> Any particular reason you are using the "SVN" abbreviation in the name
>> of a product that isn't part of the Subversion project, or even open
>> I guess the rest of us will keep using "DumbSVN"...
> Marc, just to clarify:
> I think what Brian's objecting to is not the mere presence of the
> string "SVN" in your product's name. Plenty of Subversion-related
> projects already have that, as a glance at our Links page will show.
> The issue is more the *combination* of that name, "SVN", with the fact
> that you're offering something that takes the place of the 'svn'
> client already offered by the Subversion project.
Well, no, that's not quite accurate.
I think this community should be more protective than it is about the
Subversion name, and its SVN contraction. I think it would be a disaster
if it ended up like the Linux community, where the confusion over
what-is-Linux has opened the door to FUD storms by competitors and
confusion by people outside the circle of Slashdot readers and others in
To illustrate I'll create a fake example far worse than the current one,
to make clear my concern. Imagine a company who noticed the great press
and word-of-mouth Subversion was getting, and decided to release a product
called "SubversionPro". Let's say such a company has never so much as
submitted a bug report or comment on the users list, let alone fixed bugs
or added new features. Let's say they got smart and purchased the
Subversion keyword at Google and other search engines, obtained
subversion.com, paid for a positive review in some tech rag that didn't
make it clear that the "Subversion project" is an open source project but
instead linked to "SubversionPro", etc. How would it make the people here
who contribute to the project feel to have our public image essentially
hijacked out from under us?
I've tried *extremely* hard to fight back the impulses within CollabNet to
pull a JBoss on Subversion, because I've always felt that doing so would
disincent other contributors, and disincent other companies from also
incorporating Subversion into their products. I *want* to see the
community of core developers be much, much more than CollabNet's own
contributions. That's why we chose a very easygoing license, and kept the
CollabNet imprimatuer to a minimum. I want a large number of developers
to feel a sense of ownership over the name "Subversion" because they've
contributed to it; I don't want them to feel like that investment (in code
or street cred) enriches just one particular vendor.
The point made last week about thinking about _Crossing the Chasm_ and
thinking about the "whole product" resonated deeply with me. All these
disparate projects using the SVN name really should be thought of as part
of this "project", speaking broadly. This list has so far limited itself
to core libraries, the server, and the command line client, but it's also
the most logical place to start thinking about coordinating the "whole
product" as Subversion is know to the wider user community, the same way
that other aggregate projects coordinate. If there are lots of deriviate
works that carry the SVN or Subversion name but are outside of the sphere
of the "whole product", that makes managing a positive "whole product"
experience much, much tougher.
This is why, for example, the Apache Software Foundation is pretty fierce
about protecting the use of its name - they don't want to see people
releasing products called "ApachePro" or "Apache++" or "MS Apache",
because of that confusion, but also because it's using someone else's name
without having earned the right to it. Even if "SmartSVN" isn't too far
from other examples we've allowed, and even if I can trust that the
intentions of the SmartSVN developers are to contribute back from time to
time, I worry that we're slipping down a slope in a direction we don't
want to head in.
To be clear, I've got *no* problem with companies building commercial
and/or proprietary derivative works from Subversion source code. The
more, the merrier, because that should lead (in theory) to more developers
and more development across the board.
To date, seeing examples of this have just been somewhat annoying to me,
but I've held my tongue on the basis of "let's see where it goes". I
hoped the community would feel the same kind of urge and eventually
someone would say something. Maybe that conversation has already been
had, and I missed it. Or, maybe people are *expecting* CollabNet to
provide some sort of leadership here. I'm sorry to pick on SmartSVN and
Marc, and do admit that it was the end of a long week and it was more an
accumulation of events rather than specifically SmartSVN. But am I the
only one who feels this way?
 - check it out. turns out not to be such a hypothetical example.
To unsubscribe, e-mail: firstname.lastname@example.org
For additional commands, e-mail: email@example.com
Received on Thu Feb 24 06:36:38 2005