the confusion about Disney/CMU being active devs

General Discussion
1

Anyone new to the Panda3D engine is told in the first few pages of the manual that the engine’s main developers are CMU and Disney, they see “Carnegie Mellon” on the top-right of the website, “© Carnegie Mellon University” on the bottom-left, in the contact page the email and postal mail are CMU’s.

I’ve been using the engine since 2009 and I don’t see CMU developing anything since then to be honest. I’ve seen students contribute few features two times and the last time that was few years ago. And even the tools they make are not maintained after their semester and unfinished projects (like their scene editor) are abandoned.
I’ve also not seen teachers from CMU develop anything for the engine since my arrival in 2009.

In short, I wouldn’t call CMU an “active developer”, actually I wouldn’t call CMU a developer at all after the few years of silence.
Don’t get me wrong, I have nothing against CMU and they have probably contributed a lot to the engine, but that was many-many moons ago.
Right now the only thing they seem to do is allow the Panda3d website to be hosted on their servers.

Now let’s talk about Disney. By my impression David Rose has done tremendous work on the engine and from what I know he has written most of the code for Panda3D when it was first created by Disney. But he seems to be still developing the engine for himself rather than being paid by Disney to do so. I’ve seen few other people from Disney make few contributions (few years ago), but that’s still not enough to call Disney a “main developer” too.

Let’s get back to my first point: new developers are told in the docs and the site as a whole that the active developers are CMU and Disney.

  1. This is very bad as they later realize that CMU and Disney haven’t really done much for years, which gives the impression that this is an abandonware, which makes them less likely to choose the engine/contribute.
    That was my impression too at one point.

  2. I also imagine if more people will know that the engine is currently being developed by an internet community they will be more willing to contribute themselves then if they knew it’s mainly done by a company or a university which might (according to some of them) be less needing/wanting help from freelancers.

What I suggest to do is update the texts on this site to clearly explain that “Panda3D has been created and is still being used by Disney VR, has been actively developed at some point by CMU ETC and is currently developed by contributors around the world”.
This is exactly the case with other engines like Irrlicht or OGRE, who’s main developers have “passed on” and they are currently being developed by an online community.

Many will probably find this idea crazy, but I’d go as far to suggest to move the site to another hosting service like Sourceforge (that’s where the Irrlicht engine’s website has been all along) to prevent every confusion that CMU still has an active role in the development.

Another not-so-crazy suggestion will be to add a page named “Developers” or “Team” on the “information” section of the website which will have the names of the current active developers (and maybe their images… and bio), like the one the OGRE engine website has: ogre3d.org/about/team
If there was a page like that when I was was starting with Panda3D I would be… “more secure” about sticking with this particular library.

Again, I don’t have anything against CMU or Disney and without them Panda3D wouldn’t exist or wouldn’t be as mature, but giving them “false” credit today seems to do more harm than good.
And sorry for the really long post.

2

Edit: rdb told me in irc that launchpad is actively used.

Overall I am in agreement, but this should also be part of a larger effort in increasing the engine dev community.

I only use Panda, I don’t develop the engine. The only thing I can speak to is what I look for when checking out an engine or a library to see if I want to use it. From my experience as a teacher in building communities in a classroom I see three ways to do that.

The first is the move to DVCS. It will be a big help in lowering the barrier of entry for contributing code.

In addition I think the repository should be moved to github, bitbucket, or a similar service and make a good effort at using their issue tracker – even if the number of engine devs is small. IIRC panda currently uses launchpad (launchpad.net/panda3d) and it feels very neglected.

The last part would be improving the website. One of the things I do when I look at a library or project I want to use is look at their issue tracker, last commits, etc to see if the project is under active development. The launchpad page is a very hidden feature. If I were supreme ruler I would use github/bitbucket/etc’s api to feature the latest commits and bug closings. I would also move to a continuous integration system instead of a nightly buildbot.

3

David could probably reply to this better than I could, but I’ll try to address some of your concerns.

It is true that CMU has not been using Panda3D very much of late, sadly. Since the leave of Josh, CMU has not been actively maintaining Panda3D. There was the occasional student project, but the communication surrounding them was very poor.
(However, I don’t see this as a permanent trend. As far as I know, CMU is still invested in Panda3D.)

CMU currently does a great job at hosting the servers. They provide excellent hosting, with a reliable dedicated server for the website and related services (buildbot). Not only that, but they also provide build servers where all of the public releases are produced. These services could not be provided (or at least not as reliably) by free hosters such as SourceForge.

It is not true that David was the only one involved in the initial development of the engine, although it is true that David wrote a lot of code. As far as I’m aware, he is still being paid by Disney to work on the engine. People at Disney still develop on Panda3D, and Disney still financially supports the maintenance of the project.

Let’s keep in mind how much the project owes to CMU, also. It was CMU who did all the work to make the project available to such a wide audience in the first place. Let’s be grateful of that.

It sounds to me like your primary concern is that CMU and Disney are receiving credit that they do not deserve? If so, I would strongly disagree. Why does it matter so much which names appear on top of the source files, isn’t the project itself more important? Is the lack of recent involvement on the development side by CMU reason to scratch their names from some list?

I’m not sure if I like the idea of a “Developers” page. I think that the focus should be on the product, not about the developers who were involved. On top of that, I’d like to emphasise Panda3D as a community-driven project; Panda3D regularly receives great patches from various community members and it is partly due to this community that Panda3D is still alive and kicking.

I agree with croxis that switching to a DVCS is imperative to encouraging community contributions. I am actively pursuing this goal, it is a top priority for me.

4

I would hazard a guess that Anon is still more worried that Panda3D would end up a dead project rather than credit going to the wrong people. I worry about this occasionally too - especially considering how much time I’ve been sinking into it lately.

5

Given the active community, and the active contributions by Disney, I doubt that Panda3D will end up dead. But if you’re worried about that, I implore you to contribute - we are a bit short on developers, after all.

6

Well, looks like rdb didn’t get where I was getting at. I’ll try again.

I’m not as optimistic, even though they haven’t made an official announcement on this, they seem to have silently (for us) switched to the Unity engine.

Which I haven’t said.

Which I’ve clearly said in my post. I think the problem with giving the people the impression that the main devs of the engine now are Disney+CMU doesn’t just have the negative consequence of giving credit to the wrong people , which isn’t really the case and isn’t my concern, there are other consequences which I have already mentioned and which can have a direct impact on the future of the project. I’ll repeat my points again.

It will serve the product in the end. Here, I’ll repeat my points:

Newcomers are given the impression that CMU and Disney are the active and main developers.

This can result in:

  1. Some people thinking “Oh look, the project is done by two large companies, guess they don’t need my help so much then”.

  2. Some people thinking “Oh look, this projects is done by two large companies who haven’t done much in the last few years. Guess the project is dead”.

Updating the website/manual and page listing the “developers” (or “contributors” if you wish, same thing) will help issue 1 by showing who and when has developed the engine and will help issue 2 by showing that just because some devs (CMU+Disney) haven’t done much lately, doesn’t mean nothing has been done for the engine lately.

I don’t think I can explain what I meant any simpler.

7

You’re cherry-picking from my post. You overlooked the most important part, in which I explained that Disney is still involved with Panda3D development and maintenance. The point I was trying to make about CMU is that although in the past few months we haven’t heard much from them, I see no reason to assume there won’t be another student project involving Panda3D within the next year or so.

You did seem to give off the impression that that was your concern in the last sentence in your first post, so forgive me if I addressed that.

Anyway, to address your other concerns:

I find the notion of that to be silly. Many open-source projects are backed by companies, and I haven’t encountered anyone who would find that to be a reason not to contribute.

Why would they think that? I explained to you that Disney is still actively involved. What reason would they have for wrongly assuming the opposite?

And if I’m wrong, do you have any evidence to suggest that people do indeed think this? Any forum posts you can link?

8

Are student projects every year or even less commonly a reason to call CMU active devs?

Well sorry for the confusion, I thought it was obvious.

If you are told a project is actively developed by a large company you have less reasons to assume the project is lacking in developers, which as you seem to agree is the case now.

And if you aren’t clearly told that there is a community involved in the development as well and see the current status of contributions by the company you are more likely to assume the project to be dead or in a very slow developing pace.

By looking at facts? You alone seem to have done at least as much for the library as Disney alone for the last release, by looking at the features list.
And should I say you are cherry picking now? You seem to mainly talk about Disney and very little about CMU.

9

Well, I can’t speak for CMU, but Disney contributes by far more than the community. Removing Disney from the list would definitely make it seem like a dead project.

I still have yet to see people who think the way you suggest.

10

Fine, if that’s really the case then I agree too.
But then you might agree that at least having CMU in the list of active developers can cause confusion.
And when you say you disagree with me, do you also disagree with the idea that “contributors from around the world” shouldn’t be included in the list of developers in the manual?

Can you imagine people who have the impression that this is a dead project telling you or anyone else in the forum that they aren’t going to use this library because it seems dead to them? That wouldn’t make sense.

Or if people who had the impression that the project isn’t lacking in developers told that they aren’t going to contribute. That wouldn’t make sense either.

11

The way I see it, having CMU’s name everywhere may actually be helping the project’s community outreach. CMU is a big university name, and university projects tend to feel more open and welcoming to community efforts than the names of huge megacorporations. If we weigh “Disney’s Panda3D” with “CMU’s Panda3D”, the latter is the clear winner in terms of public developer image.

Anon’s concern of the association of Panda3D with Disney is of a valid line of reasoning in that a direct association between Disney and Panda3D in the public’s eyes would probably yield images of hordes of developers making the engine the best it can be. It would be seen as a product offered by a multi-billion-dollar media empire rather than an open project. For that reason, I’m quite glad it’s a university’s name we’re seeing all over with Disney being painted as simply a user of and contributor to the engine.

There are multiple hurdles to getting the public invested into this project. Seeing the names “CMU” and “Disney” aren’t scaring new developers off. So much more about this entire project’s public image needs serious improvement if developers out there are going to take it seriously.

Right now, Panda3D is a dated game engine powering some old Disney games and a couple other titles with not so impressive hype. Its codebase is housed in a CVS repository on Sourceforge. Its website looks like it was designed in the early 2000s, and its forum software is from the early 2000s. It has a tiny box on the front page with a news blog that was last updated in 2011. There’s absolutely no social connectivity in either the forums or the site at large. The developer collaboration areas are scattered everywhere from this forum to Launchpad to Sourceforge and beyond. There’s no clean, structured hierarchy of code maintenance.

The move to a DVCS and a modern platform like Bitbucket or GitHub would be the first step along with cleaning up the repository to be what it is meant to be: the codebase of the Panda3D game engine library. Repository ACLs need to be cleaned up so that there’s a clear maintainer or two taking pull requests from branch owners. This forum needs to be either changed or updated (currently underway), and also needs to be modernised with some sexy new features enabling among other conveniences social connectivity. The news needs to be much more emphasised and maintained actively. A much broader social presence is critical to attracting new developers. Lack of awareness is probably worse in this project than difficulty of contributing.

In conclusion, there are plenty of problems with pitching this project to the public in the year 2013, and the name “CMU” or the lack of an unsustainable “list of all the names of all the people who contributed any little thing because we love each and every one of them” is not one of them. Let’s work on getting this project even remotely interesting to modern developers first.

12

Well, if the problem is that the project seems dead, then Panda3D’s image is what we should focus on fixing. For instance, having more content on the blog would help with that. I think that will make much more of an impression than whatever name is in a short paragraph on a manual page. Upgrading from CVS to Mercurial would also help.

Reading the manual page again, it is true that the manual page is a bit misleading, though. Hmm.

The problem I have with adding a list of specific community developers is that with any open-source project it changes all the time. It raises all sorts of questions, such as “how many patches should someone submit to make it to the list” and “how long should someone be inactive for him to be removed from the list”.

The manual page could certainly be reworded a bit to highlight the community’s involvement in general, though.

13

I’ve only recently started using Panda3D after looking at a few different game engines.

I was a bit concerned when I first looked at the website, thinking that the project was no longer maintained.

But, after looking at the forums (I’ve been lurking for a while, this is my first post), I realised that the project is still actively being worked on.

So, I agree that it’s maybe a perception thing.

I did not even look at, or consider that the list of active developers might not be active anymore. I just saw CMU and Disney, and thought, wow, this project has a proven track record. Those names were what made me look at Panda3D a bit deeper.

For me, these were the things that made me think that the project was no longer maintained when I first visited the webpage:

  • The latest version considered stable according to the download page is 1.7.2, released two years ago
  • The latest unstable version, 1.8 was released a year ago
  • The latest blog message, dated almost a year and a half ago, talks about the ‘upcoming’ 1.8 version

I just want to say again that I don’t think the project is dead. The above points were just my first impressions.

The forums are what changed my opinion. So perhaps, to change perception, some updates could be done on the website? Even if it’s just to highlight some new projects or cool things people are doing with Panda3D, something to show activity.

14

I plan on releasing 1.8.1 within a matter of days, and I have a major update to the forums coming up soon as well (update to phpbb3 with a complete facelift). So those are two steps in the right direction. :slight_smile:

I would encourage people from the community to help keeping the blog up-to-date as well. If you have a topic that you’d like to write about, e-mail me and we can discuss it.

15

If I may, one thing that might help would be a list of recent builds: I get the impression that those happen fairly frequently, so the recent dates on those might help to demonstrate that the project is still active.
[edit]
For the sake of clarity, I mean something like the list on the right-hand side of this page.
[/edit]

In order to show that community support is welcomed, perhaps a weekly or monthly “user contribution highlight”: mention in the blog some particularly interesting, impressive or useful user contribution, such as the scene editor. If none are found for that month, either give an update on an ongoing contribution (again, such as the scene editor) or pick a project (perhaps from the Showcase sub-forum) instead.

[edit]

I’m glad to read that! I look forward to seeing – and trying – the new version, I believe. :slight_smile:

16

about “who’s name should be put on the list”, think about the big names in IT industry, those with the fame are not the only people who worked, but many more people who also worked are not known to the public. I would think of the big names as symbols and representatives of what can be done, not that they are the only great people who contribute.

the panda website is in deed lacking, maybe that’s because panda needs no advertising. it’s free and open source and it’s backed by big company. advertising just adds more cost.
panda engine has fewer new features than commercial engines out of the box, but it’s free and open source, people can develop new features on it. would you blame Microsoft bundling MSPaint in Windows instead of bundling Photoshop in it?
people choose panda for some reason, and choose Unity or CryEngine for some other reason. panda doesn’t compare to commercial engines.

17

To put my opinion in, I had no idea Panda3D existed until about a year ago. The Digital Media Academy hosted at Swarthmore College in Swarthmore,Pa used the C++ version of Panda3D in the course “Game Progamming in 3D”. It was a 40 hour class that lasted a week and I learned a lot and I created a pretty cool game so I kept programming at home.
To me anyway, it doesn’t really matter if “the project is dead” or not. Panda3D was and still is a great way for beginners like me to create 3D games and applications and without it I would have never decided to major in computer science.

18

I Have to agree that the from the first look project may seem to be dead.

In additional to the above, as far as I can see on the Russian community, people often confused by rather old-looking screenshots and very unpretentious demo included. While I personally believe that the Panda is fully capable to show a picture comparable to the Unreal Engine or even CryEngine, although - it’s mostly contribute of the shaders and artists, but not all newcomers understand it.
It would be nice if in the Panda have had something like “Angry Bots” or other demo-projects from Unity engine to demonstrate the features.

19

A nice, flashy demo – as opposed to tutorial – game might not be a bad idea, actually. It needn’t be anything terribly complicated: perhaps a simple one-level “survive endless waves of enemies” game, with a single weapon, some basic physics objects (crates that can be knocked over, etc.), one or two enemy types, two or three powerups and a simple but three-dimensional level layout, all displayed using pretty shaders and the like.

20

I’m glad to read that! I look forward to seeing – and trying – the new version, I believe. :slight_smile:
[/quote]

I am an odd one , I prefer to keep on using an old software unless the new one has big improvements that I miss. such as updating from win98 to winxp is necessary, but from xp to win7 is not yet.
and I don’t like to run installers which may change my system setting. I prefer to manually insert addon files than installing the whole package. new package is good for a PC which hasn’t got a previous version running.