Governance and Release Model updates
kent at caspia.com
Wed Nov 14 19:07:05 UTC 2012
On 11/13/2012 1:15 PM, David Lechner wrote:
> I have been thinking about how I could make a living from open source
> software. Since it relates to this discussion, I figured I would share
> my ideas and apply them to Thunderbird.
"Make a living" is also where I am at.
> I really like the OpenERP philosophy
> (http://www.openerp.com/node/465). Basically, I think that if you want
> to make a living as a software developer, you should be selling your
> services, not a product. My ideas are something along the line of what
> Blake said:
>> I think having a not-Mozilla organization dedicated to enhancing
>> Mozilla Thunderbird could work out well. They wouldn't be responsible
>> for releasing it, but I see no problem with them (say) charging money
>> to support it in enterprises. (Or, maybe we could look into a model
>> like the "Microsoft Solution Providers", where other people build
>> businesses around the Thunderbird codebase?)
"dedicated to enhancing Mozilla Thunderbird" is not how I would describe
the core mission of Swanfox. Rather, Swanfox is a community that is
dedicated to providing income to its members, using means that are
consistent with Mozilla values, and working within the general area of
internet communications clients beginning with the Mozilla codebase.
"Mozilla Thunderbird" as currently envisioned by Mozilla is a desktop
email client whose future development will be done by volunteers. Any
time we try to push beyond those boundaries, we get pushback. At this
point in time, rather than continuing to push, I am ready to accept that
definition of Thunderbird, but claim that I am interested in something
else that I guess is not "Thunderbird".
> So here is my hypothetical future of Thunderbird...
> First you have Mozilla. As has been stated before, Mozilla will
> continue to provide infrastructure for testing (try servers), building
> (build servers), and maintaining (bugzilla, wikis, getsatisfaction,
> etc.) and personnel for security and stability fixes and driving releases.
I am not convinced that Mozilla really wants this job in the long run.
In the short run, yes, but the motivation is that they want to do the
right thing by their existing users, not that they really believe in a
long-term future for Thunderbird. So although I hope that Mozilla
continues to do these things, I am not willing to assume it.
> If an external organization chose to fork the code, it would be an
> enormous task to try to do these things themselves. Not a good idea.
> Rather, Mozilla *needs* to be the keeper of the code.
Nobody is proposing a fork, but "Mozilla *needs* to be the keeper of the
code" does not necessarily follow from "no fork". Again, in the short
run yes, but in the long run I am not positive that Mozilla really wants
Thunderbird, they are too-focused on "the web" with Thunderbird viewed
as an anachronism that no longer fits into their "everything is a web
app" rhetoric. When Axel and I discussed some of these issues with
Mitchell in Warsaw, she said that Mozilla would consider transferring
Thunderbird to another entity, but she would be more comfortable with a
I am talking very long term here, in the short run the assumption is
that Mozilla continues to do the same things, and owns and controls the
Thunderbird brand. Yet it is important when you start something new that
you do not make a mission statement or assumptions that you will regret
in the future.
> Next, you have the volunteer contributors. These are the people who
> just like Thunderbird and want to do something useful to make
> Thunderbird better. These folks are already here and will most likely
> keep doing what they are doing, so no need to explain further.
Actually an important issue for me is that we do not take volunteers for
granted. I specifically reject the idea that because someone has been
willing to do something as a volunteer in the past, that when we decide
to expand an activity for pay that we go outside to hire someone. Even
though taking a volunteer from, say, 10 hours per week for free to 20
hours per week for pay only adds 10 hours per week, while going outside
adds 20 hours per week, the organization needs to explicitly reject that
logic and offer first to honor their dedicated volunteers, should they
> Now, following Kent's lead, we have Swanfox. (As a side note, I might
> choose a name like ThunderbirdPros if you could get away with it).
> Swanfox is a for-profit organization that sells services for
> Thunderbird software. These services would include support,
> development, training, and partner certification.
> Support: As Kent mentioned:
>> the free support pages offered by Mozilla could point to the paid
>> support options provided by Swanfox.
> The 'more sophisticated support' could also include helpdesk type
> services for large businesses.
+1 It should also include not only guaranteed response, but also
higher-tiered support, including involvement of core contributors if needed.
> Development: Swanfox could offer a bug fix service.
> For example, if there is a bug that is not high priority for Mozilla
> or volunteers, but is a real blocker for a business, then that
> business can hire Swanfox to fix the bug and continue to use
> Thunderbird. And of course, the Thunderbird Community as a whole gets
> to benefit from the fix.
> Swanfox could also provide add-on development to integrate Thunderbird
> into a business' specific enterprise needs.
I have done this for clients in the past. Swanfox could provide a
marketing and matchmaking role here.
> I suppose you could even venture into product development. A mobile
> client is a hot topic at the moment, so I will use that as an example.
> Having a mobile client that is "designed to work seamlessly with
> Thunderbird" would defiantly be a big appeal to many businesses.
+10! "designed to work seamlessly with Thunderbird" is exactly the goal
for such a client. It might even be based on the Gaia email app.
> Training: Swanfox could offer training services for a number of
> things. They could offer end-user training for businesses that are
> using Thunderbird as their email client. They could offer more
> technical training for IT staff that have to support Thunderbird. They
> could also offer developer training for those who want to become
> developers and prefer a classroom experience to teaching themselves.
I don't have much experience with this. I think that marketing of this
would be a real challenge.
> Partner certification: This is really the key here. The above tasks
> are really too many and geographically spread out for a single
> organization to take on. ... Hopefully Swanfox would be very well known.
The "well known" part I think is a killer here. For the foreseeable
future, the "Thunderbird" brand name is going to mean more than
Swanfox's brand (and Swanfox is really a project codename now and not a
brand), so this is a case where it would be essential for Mozilla to be
cooperative with their branding. Yet I fully agree that there needs to
be local organizations, and Swanfox is not the right entity to do that,
and we should be careful to be cooperative with rather than competitive
to such local organizations.
> So, in the end, hopefully what you get is Mozilla has an enterprise
> grade mail client that they get to give away to promote a free and
> open web, individuals who get to use enterprise grade software for
> free if they want, businesses who get free software along with the
> level of support that they need and jobs for people who want to make a
> living as software developers.
You used the concept of "free-as-in-beer" three times in that paragraph,
and "free-as-in-freedom" only once ("the free and open web"). We need to
have some really serious discussions about that. While I will defend
"free-as-in-freedom" as a core value, "free-as-in-beer" is the whole
reason that we are in the fix we are in, namely we have 20,000,000 users
and it seems to be an impossible task to get a nickel per year from each
of them so that we can have $1,000,000 to continue with a vibrant
product. The user's need to have a vibrant communications client is much
more important than their need to avoid paying $0.05 per year for that
In our previously mentioned discussions with Mitchell, she said "Mozilla
has no interest in subsidizing a free software product for enterprises".
I'm fully in agreement with her there with respect to my personal
resources. Swanfox as I see it has no interest in promoting
free-as-in-beer as an intrinsic good. We provide something of value to
people, and we need to recoup some of that added value so that we can
earn a living, and continue to develop the product. If free-as-in-beer
is a tactic that we use, fine. But it is not a core value or mission.
Looking forward to your future involvement and discussions!
More information about the tb-planning