PCGen sellout

[bushfire]

What I found odd about the announcement was that Wizards is hiring a team with 1) a lot of qualifications in managing D&D source data, 2) qualifications in Java programming, 3) no qualifications in C or C++? (Does CMP have C experience from separate projects? If so, why that horrible Java language?)

Actually I think the reason PCGen is in java is that the original programmer (merton_monk??) used writting PCGen as a way to teach himself java. I think several of the CMP programmers have experience in C/C++

 

[Hardhead]

What I found odd about the announcement was that Wizards is hiring a team with 1) a lot of qualifications in managing D&D source data, 2) qualifications in Java programming, 3) no qualifications in C or C++? (Does CMP have C experience from separate projects? If so, why that horrible Java language?)

Because it's a well-rounded, very fast (for being cross-platform) platform-independant lanaguage? Java is a very popular, very sucessful programing language.

 

[Wanderer]

Calm down Negative Nancy

Just because you have to pay for something now, does not mean anyone sold out. Your probably just angry cause you didn't get included

I guess you live in a country that does not have copyright laws or your rather Chaotic in alignment and therefore don't care. Do you also kick cats on the way to work, because you can?

I never stated I had a problem with paying for anything.
I also respect Wizard's copyright and their right to insist on a charge of some kind.
I certainly don't condone cruelty to animals.

What I find amazing is that key personel of an open source project have incorporated to make a business that is legally bound by agreement to save and promote a rival product.
Any way you cut it, ETools and PCGen are both character generators, and hence there is a clear conflict of interest.

PCGen was built on the contributions and efforts of a large community that have been completely excluded from any say in this process.
I actually think that having the ability to buy WotC IP datasets is a good thing - if its the only way that Wizards will allow it. Everything else about the deal smells pretty bad to me. From a PCGen point of view, nothing else in the deal was necessary, and nothing else was revealed to the community at large until it was all signed and delivered.

I think that the top PCgen hierarchy deserve something rewarding from their efforts. I'd like to think that there were more creative solutions than entering into a commercialized conflict of interest with a rival program.
I have nothing against ETools, and don't begrudge it an opportunity to resurrect itself, but I think that resurrecting it by diverting the key personnel of its big open source community rival is very distasteful.
Open source community projects imply certain ideals that are clearly at odds with what has happened here, and I have read PCGen yahoo posts by disappointed people who have donated their time to enter in PCGen data in the past.

What I would like to know is if Wizard's forced the PCGen leadership to take on ETools and fix up ETools as part of the deal. At least then I could understand that it came down to a question of how badly desirable the IP datasets are.

Some open community discussion would not have gone astray. In fact, it would have been very appropriate. I also wouldn't buy a "Wizard's made us keep it secret" line. Wizard's were coming from nowhere with their ETools PR disaster (shoddy program, no funds to fix, no option for support), and PCGen held all the cards.

Wizards played this one very, very well...

 

[NPC]

Wanderer, you have summed up exactly how I feel better than I ever could have.

This move has left a bad taste in my mouth. And I'm sure it's not just my bad breath.

 

[Gizzard]

Because it's a well-rounded, very fast (for being cross-platform) platform-independant lanaguage?

Maybe my C bias is showing here, but I thought the Java App bandwagon came and went long before the year 2003. When people say "its fast for a cross-platform language" it highlights the basic problem - it is slow and it is always going to be slow. Not to mention quirky little issues like Swing which show that the dream of 100% compatibility is going to remain exactly a dream. IMHO.

Anyway, I say that as an explanation of why I was dismissive of Java rather than as an invitation to a Holy Programming Language War. ;-)

...most skilled coders nowadays have proficiency with several languages.

Proficiency or experience? I could put a ton of languages that I've dabbled in on a resume, but I would truthfully only say that I am proficient in C. And maybe some obsolete assembly languages. ;-) I can't think of many people that I work with that I would hire for a Java gig even though several of us have at least looked at the language - I'd prefer to hire people who were doing Java day-in, day-out.

I mention that because as a project manager I'd want to see some ability to manage a C project from CMP before I hired them. A 90 day schedule doesnt leave you a lot of time to config SourceSafe and VC6 if you have to start from ground zero. And thats just the tools issues; not to mention some time to shake the rust off if you've been away from C for a couple years.

Anyway, I dont know what the CMP guys do in their work day; they could all be C programmers by day and Java hobbyists by night. In which case its a great idea to hire them to fix up e-Tools. If there is an implicit vote of "no confidence" in my even bringing this subject up, its more in Wizards ability to understand software development and hire an appropriate dev house to do the work it needs done. I mean, CMP has shown that they can go from zero to product in a reasonable amount of time, so I really dont intend to slam them.

 

[Duncan Haldane]

Re: Re: PCGen sellout

Not to change the subject, but what is this new product? Where has it been officially mentioned?

From the front page of http://www.codemonkeypublishing.com:

The future is indeed very bright folks. We have more than just e-Tools and data sets going on here. Remember, we are also a publisher ourselves, with an entire line of products slated to be released this year, as well as software support for those products, and _our_ flagship program CampaignGen, we are aiming for a GenCon release. There are a lot of things planned in the future for a lot of different things, so stick about and join in the monkey mayhem!

----

As for PCGen charging, etc - I'm not too bothered by a smallish, one-off charge. What bothers me is the claims that the fee will be annual. But I can't find anything confirming this.

I'm also worried about other publishers charging through CMP for their material to be available in PCGen. This could start to cost me a lot of money.

Also, on the PCGen yahoogroup people have stated that the money generated is likely to be used to pay bandwidth charges, yet PCGen is currently hosted on sourceforge.Net, which states on http://sourceforge.net/docman/display_doc.php?docid=6025&group_id=1:

SourceForge.net is the world's largest Open Source software development web site, providing free hosting to tens of thousands of projects.

*Italics mine...

From what I read there, CMP/PCGen don't have to pay for downloads from SourceForge, so if the data remained open they could distribute it there for free.

Another point: will CMP be re-coding all of WotC's non-SRC data sets, or will they use the data already typed up in the pre-OGL days of PCGen? Because, I don't think they should be able to use the old files - they were typed up by many different people, for free, and I don't think it's right for CMP to benefit from that work, nor do I think they "own" it.

I actually wonder if the reins of PCGen should now be handed over. CMP have the responsibility for WotC's datasets, etools patchs, and their new product - what will happen to PCGen when they find their time too stretched?

Duncan

 

[Hardhead]

Maybe my C bias is showing here, but I thought the Java App bandwagon came and went long before the year 2003. When people say "its fast for a cross-platform language" it highlights the basic problem - it is slow and it is always going to be slow. Not to mention quirky little issues like Swing which show that the dream of 100% compatibility is going to remain exactly a dream. IMHO.

Exactly what do you need a Java program like PCGen to do that it needs to run as fast as C, though?

PCGen runs plenty fast for me both on my Windows computer, and my old iMac running OS X. There's about a three second delay while it loads up the data files, but there's no delay after that doing anything. So, yeah, you wouldn't want to code Quake III in java, but a simple character generation program works just fine. Plus, it's cross-platform. I first started using PCGen when all I had was a Mac. Two of my players still only use Macs. It would be much less useful to me if it wasn't cross-platform.


- Z a c h

 

[Mynex]

I mention that because as a project manager I'd want to see some ability to manage a C project from CMP before I hired them. A 90 day schedule doesnt leave you a lot of time to config SourceSafe and VC6 if you have to start from ground zero. And thats just the tools issues; not to mention some time to shake the rust off if you've been away from C for a couple years.

Anyway, I dont know what the CMP guys do in their work day; they could all be C programmers by day and Java hobbyists by night. In which case its a great idea to hire them to fix up e-Tools. If there is an implicit vote of "no confidence" in my even bringing this subject up, its more in Wizards ability to understand software development and hire an appropriate dev house to do the work it needs done. I mean, CMP has shown that they can go from zero to product in a reasonable amount of time, so I really dont intend to slam them.

All the primary coders do exactly what you describe above... they code in c by day and java 'for fun' by night... someone else pointed out that they thought that Bryan McRoberts (Merton Monk) started PCGen as a way to learn Java, that is 100% accurate.

Between all the Code Monkeys at CMP there is a total ov 90+ years worth of C language coding, for companies ranging from bank one, to lear corp... so rest assured, the experience is there.

 

[Henry]

I have nothing against ETools, and don't begrudge it an opportunity to resurrect itself, but I think that resurrecting it by diverting the key personnel of its big open source community rival is very distasteful.
Open source community projects imply certain ideals that are clearly at odds with what has happened here, and I have read PCGen yahoo posts by disappointed people who have donated their time to enter in PCGen data in the past.

Please show me how the design process, addition of new features, or the design goals of PCGen has changed in any way, shape or form with the coming of this announcement. PCGen will not slow down, since there is more than just Bryan doing the coding here.

Both E-tools and PCGen benefit from the arrangement. PCGen gets the option of datasets it didn't have before, legally. E-Tools gets the option of those same data sets, legally.

E-Tools users get support that they were entitled to, but lost due to business issues. PCGen was not, and will most likely not be, affected by the process; in addition to the fact that there are more code monkeys than are CMP employees, we have been given assurances that everything will still go on schedule.

What I would like to know is if Wizard's forced the PCGen leadership to take on ETools and fix up ETools as part of the deal. At least then I could understand that it came down to a question of how badly desirable the IP datasets are.

The official statement is "mutually beneficial; from outside appearances, however, CMP still holds the most important cards. PCGen is still free. E-Tools is not. Both have to pay for data sets - same price, I'm willing to bet. E-tools has more windows users, PCGen has users in other OS'es.

PCGen has lost nothing.

Is your apprehension that PCGen has been damaged based on a belief that the two programs cannot compete on a level playing field? They have the most level of playing fields now; heck, they're supported by some of the same people! (though not all of same; see above.)

Some open community discussion would not have gone astray. In fact, it would have been very appropriate. I also wouldn't buy a "Wizard's made us keep it secret" line. Wizard's were coming from nowhere with their ETools PR disaster (shoddy program, no funds to fix, no option for support), and PCGen held all the cards.

Wizards played this one very, very well...

I really don't see the game, here. If anything, CMP got the good end of the stick. I don't know for certain, but I am guessing that there is involved some monetary compensation in addition to getting a license to sell data sets - I certainly hope so, because I hate to think Bryan, Greg, Robert, et. al. are doing it all for just selling data sets.

Originally posted by Duncan Haldane
I actually wonder if the reins of PCGen should now be handed over. CMP have the responsibility for WotC's datasets, etools patchs, and their new product - what will happen to PCGen when they find their time too stretched?

PCGen is directed by Bryan McRoberts; I'm guessing he kept ownership of the name, and therefore he'll decide when and if it ever needs to be handed over.

But I will re-iterate a point another poster made: If a person feels that PCGen has gone astray, they are certainly within the rights of the Public License to take the code base for PCGen and develop it themselves. If a person feels the program is going wrong, then fork it. They owe it to themselves, and to their beliefs.

Otherwise, trust Merton Monk and Mynex and the other Head Bananas just a bit longer. They've stewarded this project for almost three years now; they're not homicidal parents.

I for one am almost celebratory: The poor orphan-child known as E-Tools has finally found adoptive parents; PCGen is getting a new relative, and both of them have had the papers drawn up for their long-overdue inheritance.

"Options, not restrictions." It's the mantra for 3E, and it applies to the new deal between WotC and Code Monkey Publishing, too.

 

[Hollywood]

What I found odd about the announcement was that Wizards is hiring a team with 1) a lot of qualifications in managing D&D source data, 2) qualifications in Java programming, 3) no qualifications in C or C++? (Does CMP have C experience from separate projects? If so, why that horrible Java language?)

a) Java is NOT a horrible language. Its actually a very good language by itself. The biggest problem is that its supposed to be cross-platform and doing cross-platform UI is not an easy task and the Swing API for doing UI is really dodgy and slow... IBM's version is actually a bit better, but it also relies on taking advantage UI APIs found on the target machines in which its run on.

b) Java is based quite heavily on the C/C++ syntax anyways. Its what is called a "managed" language, which means that it manages memory allocation/deallocation and some other things that can really trip up even the most experienced programmers.

c) More important to the issue is not whether the team can program in C, or why they chose Java, but whether they can actually DESIGN and DEVELOP software with proven SOFTWARE ENGINEERING concepts. Since none of their resumes are available, I can't say whether any of them have this experience or not. Having seen the Java code, since its open-source, I really suspect not, but may be wrong. If they don't, then it was a horrid move on WotC's part.

Open Source gives everyone an opportunity to make a contribution. If you feel that PCGen will die without some help, then DL the source and do some pro bono work on your own.

But I will re-iterate a point another poster made: If a person feels that PCGen has gone astray, they are certainly within the rights of the Public License to take the code base for PCGen and develop it themselves. If a person feels the program is going wrong, then fork it. They owe it to themselves, and to their beliefs.

Why? Start fresh, don't use bad code thats full of a lot of proprietary and badly designed code and data storage/manipulation.

Please show me how the design process.

How about we see some design and architecture documents for eTools and PCGen to prove that legitimate software development processes are being followed. That won't ensure that broken software won't get turned out, but will prove that some time and effort has been spent on thinking through the project before any coding begins and its not just a "code-n-go" project.

E-Tools users get support that they were entitled to, but lost due to business issues.

ETools users should have gotten their money back, with an apology, from WotC. Fluid should have bailed on the software long before it got to the point it got, when it was apparent that WotC had no true goals that could and would be set in concrete.

Otherwise, trust Merton Monk and Mynex and the other Head Bananas just a bit longer. They've stewarded this project for almost three years now; they're not homicidal parents.

I'm sorry, but I've not seen 3 years of value out of that work. Luke's done far more with his RPM software.

"Options, not restrictions." It's the mantra for 3E, and it applies to the new deal between WotC and Code Monkey Publishing, too.

I disagree. This "deal" doesn't address anything. It only allows TWO tools, one controlled by WotC and the other basically controlled by WotC, to use additional options that are not available to other tools like RPM or Twin Rose's software. Thats not "options, not restrictions", thats "restricted to only our tools" of which both are mishappened hammers being used to pound in wood screws.