Shay wrote:
> However, the slow development of POV-Ray leaves me concerned that this 
> quality program (or at least, a good deal of the community around it) 
> may one day disappear. I love POV-Ray, but I feel like an IE user - the 

I share some frustration; the original developers of POV (i.e. ones who
were around in the 1.0 days) have mostly gotten somewhat older, married,
kids, mortgage, in other words 'real life'. So one by one they've become
inactive; I'm the only one left now.

In the process, we haven't picked up as many new developers as those who
have left. And our development model simply does not fit in with the way
things are done in most open software projects these days, which doesn't help.

Rather than change development model mid-stream, we have decided to make
the switch when we go to 4.0. That will have a more typical 'open' plan
with a public code repository and so forth (the exact details have not been
decided, but it will be along those lines).

As for the delay for 4.0, the main issue has really been finding a suitable
license. Up until recently there was no OSI-approved license that we found
to be suitable, and we were very disinclined to attempt to come up with a
new one and submit that for approval. So we decided to wait for the outcome
of the GPL3 process.

Now that process has been completed, as a group we feel the GPL3 is the way
to go and have informally decided that 4.0 will be GPL3-licensed. [While at
this point the GPL3 is not (as far as I know) OSI-approved, we expect that
this will occur in due course, and in any event, the GPL has enough
standing to hold its own.]

The process we intend to follow (note this is not set in stone) is that
once 3.7 is stable enough for an RC, we will make a public release of the
3.7 source. Some time after that (once the source is stable enough) we will
release a modified version of the 3.7 source labeled 4.0, with all portions
of the code that we cannot re-license stubbed out or otherwise removed.

Basically this means gutting much of the functionality. However the basic
infrastructure should survive since it has been extensively re-written for
3.7, and the new code can be re-licensed under the GPL3. This tree will at
least be able to compile and render a sphere-on-plane image, forming a
working framework for the following work.

We will then depend on public participation (along with our own efforts) to
re-implement the removed code using the 3.7 source as a reference.
Basically any code that we cannot bring across has to be re-written and
submitted under the GPL3 and later by the developer concerned.

While this is a big job, having the original source as a reference reduces
the difficulty. While it wouldn't be possible to copy the code, new code
with the same functionality can be written and, moreover, compared with the
original by feeding the same data through it, thus easing testing and
integration.

Of course we would also want to take the opportunity to clean up some older
parts of the code so I am not claiming it will be a 1:1 re-creation of 3.7;
I would expect significant changes in the parser for example, and there may
even be a version of the renderer that can be complied to a library with an
ABI and no parser (if we don't do it, someone else surely will).

Overall you may now have some better idea of what is in our minds and
the process we intend to follow. The matter is we simply have to get the
time to polish off 3.7 and do the requisite modifications to the tree to
make it ready for the above.

We are as always interested in hearing from anyone with reasonable coding
skills who can assist, not only with the future 4.0 work but also with
finishing 3.7.

regards,

-- Chris

Post a reply to this message