[FOSS] On the creation of Rev to Web tools

[David Bovill]

On 15 September 2010 16:10, Malte Pfaff-Brill <revolution at derbrill.de>wrote:

>
> I would be interested in how many people would really think they would be
> willing to invest some effort into various open source projects. I know
> David is a huge advocate of all things OSS. However, as Richmond pointed out
> pretty well, over the last 8 years I´ve spend in this community I have
> rarely seen OSS projects that took up momentum. I have been wondering why
> that is for quite a while now. My main thought is that it is not exactly
> easy to collaborate on rev Projects.


It's mainly due to the economics of cooperating in Rev - too easy to develop
solo, and partly due to the history of the community - it's average age is
pre-open source / more share ware - so the culture is not there, and finally
the community is a little small. For these reasons you need to do a little
bit more than simply than place code up on server and declare it open under
some undetermined license for a project to take off.


> This is partly due to the binary nature of stacks which makes it hard to
> use a version control system on rev projects, partly due to the lack of a
> place where projects like this could be hosted.
>

The version control problem, is effectively solved now that we can create
objects with IDs. It's a red herring anyway, as the majority of useful code
can be shared under version control without problem.

Current state: Everyone that tries to release stuff to the community is
> cooking her own soup. Though most people are very generous with sharing code
> on the lists and forums, there is no central repository where people can go
> to and collaborate on projects. We do have many sites spread all over the
> world with too many gems to dig out.
> Additionally we have revOnline. revOnline is a good place for consumers /
> prosumers though, not suitable for starting a collaborative effort to work
> on code. Especially libraries. Most of the stuff on revOnline is there for
> the visual stuff the stack does, or in a state where the lib is basically
> finished.
>
So the only things an author that uploads to revOnline can gain is
> - giving examples what can be done
> - help someone solve a problem with a complex stuff (requires a lot of
> coordination and is usually easier done by mail)
> - show off what he has done.
>

RevOnline does not work - it is not a collaborative environment, which is
why it is easier for people to post urls to downloadable stacks than
indicate there is a stack on revOnline. It should be replaced.

>
> What an author usually can not hope for is to benefit from changes other
> coders have made once a stack is released into the wild. I have no idea how
> many people here would really willing to dedicate time into OSS projects (my
> last try was rather frustrating, though it has been a few years since I last
> tried.) I might be willing to test the waters again in a couple of weeks.
> More on that later.
>

Not many. They would when it works. The hard part is not the many, its the
first 5. Ever tried to herd cats? Well there aren't any cats in the Rev
community - they are wolves. They growl a lot and are fiercely independent,
but are deep down secret pack animals even though they wont admit it in
public :) The mothership has a lot to answer for.