POV-Ray : Newsgroups : povray.general : Is Pov-ray development dead ? Server Time
31 Oct 2024 06:14:41 EDT (-0400)
  Is Pov-ray development dead ? (Message 1 to 6 of 6)  
From: PaulHuxe
Subject: Is Pov-ray development dead ?
Date: 28 Sep 2023 04:50:00
Message: <web.65153d54a008833ffa0a99ebd5f99eb2@news.povray.org>
Hi to all.

Today I wanted to reinstall POV-ray after several years of interruption.

I found the 3.8 beta version with your link to GITHUB, and saw that nothing has
moved since 2021.
There are only two people are linked to this project. And both of them have zero
activity the last year.

I searched for the 4.0 repository and saw that there are no new commit since
july 2021.

So my question : Is the project dead or does it have moved elsewhere else ?

I'm not C++ dev (only an amateur of C#...) so I cannot help on it.


Best regards.


Post a reply to this message

From: Le Forgeron
Subject: Re: Is Pov-ray development dead ?
Date: 5 Oct 2023 12:35:58
Message: <651ee5ee$1@news.povray.org>
Le 28/09/2023 à 10:46, PaulHuxe a écrit :
> Hi to all.
> 
> Today I wanted to reinstall POV-ray after several years of interruption.
> 
> I found the 3.8 beta version with your link to GITHUB, and saw that nothing has
> moved since 2021.
> There are only two people are linked to this project. And both of them have zero
> activity the last year.
> 
> I searched for the 4.0 repository and saw that there are no new commit since
> july 2021.
> 
> So my question : Is the project dead or does it have moved elsewhere else ?
> 
> I'm not C++ dev (only an amateur of C#...) so I cannot help on it.
> 
> 
> Best regards.
> 

It's not dead, but some of us (me !) are getting old and busy with other 
things to do.

As far as I am concerned, I made everything I wanted for development in 
my own fork, and it is very calm.

I'm actually wanting ONE change, but I am unable to make it so far: 
dropping all the UCS2 handling for only having UTF-8 everywhere.
My incentive is using font that are past the U+10000 page, as well as 
handling non-ascii paths (Linux with French interface: the Modeles 
directory is Modèles and it is not possible to put it as a search path, 
because the "è" is not ascii and get mangled in UCS2 in non-recoverable 
way (and the filesystem is just using utf-8 for its API, so UCS2 is an 
approach from another time).

Oh, and if someone can find the bug in text handling with fonts which 
have accents (à,é,è,ê,ô,ù;ï): sometime, the accent get just moved not at 
the right horizontal place. Always the same issue for same char & font, 
but not predictable.

I did not want to get power to update the official repository because I 
believe on commite review of features and I would be too much tempted to 
just smash all my changes: not wise.

I also got exhausted after providing a native support of Nurbs in SDL 
(requested by some one, IIRC) only to find later that Blender is hard to 
change to export native Nurbs instead of pre-meshed object.


Post a reply to this message

From: Chris Cason
Subject: Re: Is Pov-ray development dead ?
Date: 6 Oct 2023 09:42:46
Message: <65200ed6$1@news.povray.org>
On 28/09/2023 18:46, PaulHuxe wrote:
> Today I wanted to reinstall POV-ray after several years of interruption.
> 
> I found the 3.8 beta version with your link to GITHUB, and saw that nothing has
moved since 2021.
> There are only two people are linked to this project. And both of them have zero
activity the last year.

I'm one of the two committers you mention. Like Le_Forgeron, I'm getting old (in my
60's now). Plus I haven't used C++ in anger for a long time; I'm a fully C# dev in my
day job.

After around 30 years with the project I see my role as more a steward; a maintainer
of stability and backwards compatibility than a developer of 'new things' (e.g.
changes in 3.8). POV-Ray is a mature program with a long history of stability.
Currently version 3.7 is working as designed and while it doesn't have all of the
bells and whistles of 3.8 I'm not currently ready to replace it myself. That's not to
say I won't merge 3.8's changes in, but if it happens it won't be until I've sorted
out a few other things.

One of the reasons we moved to the AGPL was to encourage other people to fork POV and
experiment, and this has happened to a certain extent. Those forks are the place to go
if you want to try out new ideas.

So, no, POV-Ray development is not "dead"; we're still here but somewhat short on
manpower. For a more than 30-year-old program, not having a new release for many years
is not necessarily a bad thing, as long as everything is working and there is an
active community. There will be a 3.8 release, eventually.

> I searched for the 4.0 repository and saw that there are no new commit since july
2021.

4.0 won't likely have any activity until 3.8 is released, IMO.

-- Chris (current team leader and developer since the Compuserve days).


Post a reply to this message

From: Bald Eagle
Subject: Re: Is Pov-ray development dead ?
Date: 6 Oct 2023 13:45:00
Message: <web.652046e7199d5b231f9dae3025979125@news.povray.org>
Chris Cason <del### [at] deletethistoopovrayorg> wrote:

> After around 30 years with the project I see my role as more a steward; a maintainer
of stability and backwards compa
tibility than a developer of 'new things'
.....
> One of the reasons we moved to the AGPL was to encourage other people to fork POV
and experiment, and this has happen
ed to a certain extent. Those forks are the place to go if you want to try out new
ideas.
.....
> -- Chris (current team leader and developer since the Compuserve days).

I do hope that you have some notes that would give future developers a broad
overview / flowchart about how everything under-the-hood works, and any of those
1000's of tidbits of data concerning why x, y, or z was or was not implemented,
or was implemented in a certain way.

In a sense, there are people "developing" POV-Ray by adapting code for various
algorithms, libraries, and desired features from languages like javascript, c++,
C, python, Fortran, etc. to SDL, so at least the features are usable, even
though they're glacially slow.

I think it would be great if there was some sort of short guidebook that
outlined the necessary and sufficient basics of how to implement new keywords,
primitives, functions, etc. so that anyone considering learning enough to make a
fork or take the time and effort to learn the skills to contribute to 3.8, 3.9,
4.0 would have a definite goal and roadmap to get there.

Also, IIRC, there was a major change from 3.6 to 3.7 in terms of threads or
internal messaging, or both ... idk - but it would probably be worth pointing
out the details of that, so that introductory attempts at monkeying with the
source code were done with the simpler version where there is less to break, and
then once the basic proof of concept is a fait accompli, the code could be
inserted into a newer, more complicated version.

I'm not sure what sorts of materials exist that have been "for developer eyes
only" - but perhaps those could be made available somewhere on the website so
that it's preserved, and accessible to anyone interested in learning how
everything works.

- Bill


Post a reply to this message

From: Chris Cason
Subject: Re: Is Pov-ray development dead ?
Date: 7 Oct 2023 06:03:54
Message: <65212d0a$1@news.povray.org>
On 7/10/2023 04:41, Bald Eagle wrote:
> I do hope that you have some notes that would give future developers a broad
> overview / flowchart about how everything under-the-hood works, and any of those
> 1000's of tidbits of data concerning why x, y, or z was or was not implemented,
> or was implemented in a certain way.

One thing that I suspect could answer a lot of questions about "why is this like so"
and "how do I add such and such a feature" would be the full commit and branch history
we keep in our previous version control system (Perforce). That goes back to sometime
in the late 90's I think. At one point I tried exporting that to git and the result
was not very good, but that was using a very early version of their conversion tool.
One day I will have to take a second look at this.

Converting to Git isn't essential, but helpful as the native web interface for
Perforce wasn't great (but again, it's been a long time since I looked. I ought to
check again).

> I think it would be great if there was some sort of short guidebook that
> outlined the necessary and sufficient basics of how to implement new keywords,
> primitives, functions, etc. so that anyone considering learning enough to make a
> fork or take the time and effort to learn the skills to contribute to 3.8, 3.9,
> 4.0 would have a definite goal and roadmap to get there.

I think that would be great also. But it doesn't currently exist, as such.
Implementing new things in the current codebase can be an involved process, depending
on what it is. POV-Ray's codebase is well over 30 years old and its structure has
evolved organically over the years. 4.0 will be more modular.

> I'm not sure what sorts of materials exist that have been "for developer eyes
> only" - but perhaps those could be made available somewhere on the website so
> that it's preserved, and accessible to anyone interested in learning how
> everything works.

There's very little "developer eyes only" content, really. Apart from the
above-mentioned commit history POV-Ray is pretty much "what you see is what you get".
Not much is "hidden". A few files are, e.g. the original, unmodified bitmap I use in
the windows splash screen, because we don't want forks to display "official" POV-Ray
branding.

-- Chris


Post a reply to this message

From: David Kraics
Subject: Re: Is Pov-ray development dead ?
Date: 21 Dec 2023 10:20:00
Message: <web.65845788199d5b2387b92d44a57e0@news.povray.org>
I knew POV-ray from America Online. Didn't know it was on Compuserve as well.
Now I wish I checked it out then.


Post a reply to this message

Copyright 2003-2023 Persistence of Vision Raytracer Pty. Ltd.