PC Gen data sets

[Nylanfs]

Three letters Paul... XML

Until PCGen can handle that and stop using those temperamental custom text files from Hell I for one am not touching it.

Actually coverting to XML would most likely make the data files easier to read, but much, MUCH more verbose, and believe me that matters when supporting many datasets. Just ask Barak over at CMP.

 

[2WS-Steve]

Actually coverting to XML would most likely make the data files easier to read, but much, MUCH more verbose, and believe me that matters when supporting many datasets. Just ask Barak over at CMP.

I would think that the verbosity would only be a slowdown on initial loading of the dataset -- and that after it was parsed into whatever internal format they use there wouldn't be an issue.

In fact, that's the way Hero Lab apparently handles it -- except what it does is you put together an XML dataset, then have HL append it to the database, then, in the future, it loads directly from the database. Which, I take it, makes loading faster and also allows it to retain the speed and ability to handle large quantities of data like pure database programs such as E-Tools.

 

[Graf]

This post contains so many factual errors and misinterpretations that I just can't be bothered to point them out.

Actually everything I posted is a fact.

You can spin the facts however you like (as KingPaul does) but you can't point out the "factual errors" because they don't exist.

There was once an open source project for DnD players, the people running it took it private for a payout from WotC, and the community as a whole lost out.

Now the project, which was over priced and insisted on using a weird custom data format (which they changed every couple of months) has colapsed.

OP should probably be made aware that PCGen isn't really a program for creating DnD characters (despite appearances to the contrary), it's a vehicle for Code Monkey to generate cashflow by selling datasets.

It -is- possible that WotC, finally having put the whole thing out of its misery, will come up with some other option. But until then you're really just out of luck.

 

[Skaven_13]

Actually everything I posted is a fact.

You can spin the facts however you like (as KingPaul does) but you can't point out the "factual errors" because they don't exist.

There was once an open source project for DnD players, the people running it took it private for a payout from WotC, and the community as a whole lost out.

Now the project, which was over priced and insisted on using a weird custom data format (which they changed every couple of months) has colapsed.

OP should probably be made aware that PCGen isn't really a program for creating DnD characters (despite appearances to the contrary), it's a vehicle for Code Monkey to generate cashflow by selling datasets.

It -is- possible that WotC, finally having put the whole thing out of its misery, will come up with some other option. But until then you're really just out of luck.

Um....wow. Talk about a spin. Did someone burn a copy of PCGen to cd and beat you with it?

Look, if you don't like it, you don't like it. But get your facts straight. From pcgen.sourceforge.net:

PCGen is free, open-source software, and is distributed under the GNU Lesser General Public License (LGPL).

Since PCGen is free, it is hard to be overpriced. It is still an open source project. If you don't like paying $6 on average for someone else's time to do data entry and support it, do it yourself. The lst files are not that difficult to learn. And yes, I have edited the lst files before. I have used it to create characters.

The only project that is truly overpriced and probably out of its misery is eTools, and that's WOTC's and Fluid's fault, not CMP. The data sets were still a good deal, as I didn't have to worry about entering the data myself.

Skaven13

 

[karianna]

I would think that the verbosity would only be a slowdown on initial loading of the dataset -- and that after it was parsed into whatever internal format they use there wouldn't be an issue.

In fact, that's the way Hero Lab apparently handles it -- except what it does is you put together an XML dataset, then have HL append it to the database, then, in the future, it loads directly from the database. Which, I take it, makes loading faster and also allows it to retain the speed and ability to handle large quantities of data like pure database programs such as E-Tools.

We've looked at a conversion to XML, but there were a myriad of reasons why we couldn't go there. The main reason being that our current LST format isn't just data, it's actually a programming language in its own right and that's not easy to convert to a language which should be treated as purely representing data . I'd love to see PCGen go to XML or some sort of embedded database, but that won't happen until we rewrite the core (version 6.0).

 

[karianna]

Actually everything I posted is a fact.

I'm afraid that some of your facts are incorrect, hopefully I can explain things clearly, Oh I'm the Chair Person of PCGen BTW and not affiliated with CMP in any way shape or form .

There was once an open source project for DnD players, the people running it took it private for a payout from WotC, and the community as a whole lost out.

PCGen is an open source project that is still _very_ healthy and _very_ active (a team of 60 contributors and over 6000 community members can attest to that). As Paul stated above PCGen and WotC talked at GenCon '02 to discuss our assumption that we were operating under fair use. WotC informed PCGen that they were infringing upon their IP and PCGen realised that we'd have to remove those datasets.

Some PCGen members wanted to still be able to supply WotC datasets to the community and so they discussed that with WotC and CMP was born. I don't think the community didn't lose out because CMP was formed, in fact CMP (or a company like CMP) was the only legal way to get WotC datasets to PCGen users. PCGen _had_ to remove those datasets regardless of CMP forming or not.

Now the project, which was over priced and insisted on using a weird custom data format (which they changed every couple of months) has collapsed.

"The project" I assume you mean CMP (as PCGen is a separate open source entity to CMP). CMP simply sold WotC datasets for PCGen and was forced to use the LST data format since that is what our project uses!

I can't comment on the pricing, I code datasets for myself .

I totally agree that LST is not an ideal format for beginners, it is not just a data language, but a programming language in its own right. Yes we will change this in the future, but that is several versions away. Unfortunately PCGen has evolved with having to support increasingly complex rules so LST has continued to evolve quickly in order to support these complex rules.

We do provide a complete documentation set and tutorials etc to assist users but I can totally understand that it can be difficult to produce these sets (ironically this is the reason why CMP was able to sell sets, people could appreciate the time and effort to code up a complete book and were happy to pay for that).

OP should probably be made aware that PCGen isn't really a program for creating DnD characters (despite appearances to the contrary), it's a vehicle for Code Monkey to generate cashflow by selling datasets.

Actually we have a web community of over 6000 members, so goodness knows how many more thousands of actual users actually still us PCGen for creating D&D characters (and d20 Modern characters and sidewinder characters and....). We can legally supply the RSRD (which is the basis of the core books) which the thousands of users I've mentioned above are more than happy to use. I could give you detailed stats on daily downloads and daily additions to our community, but I suspect I'm blathering here . I certainly know that a majority are not CMP customers otherwise the CMP owners would have retired in the Cayman islands by now

Hope that helps clear things up!

 

[kingpaul]

You can spin the facts however you like (as KingPaul does)

Excuse me?

There was once an open source project for DnD players, the people running it took it private for a payout from WotC, and the community as a whole lost out.

PCGen still exists, so I don't know where you get "there was once". It still is open. A private business was created by some of the people who were involved on PCGen's BoD, that is true.

Now the project, which was over priced and insisted on using a weird custom data format (which they changed every couple of months) has colapsed.

Are you referring to PCGen's *.lst format or the Access Database format that eTools uses?

OP should probably be made aware that PCGen isn't really a program for creating DnD characters (despite appearances to the contrary), it's a vehicle for Code Monkey to generate cashflow by selling datasets.

What?!?!?l PCGen existed before CMP was born, and it still exists. We have numerous users and volunteers. Where are you getting the info on which you are stating these "facts"?

It -is- possible that WotC, finally having put the whole thing out of its misery, will come up with some other option. But until then you're really just out of luck.

Both CMP and PCGen are still around.

 

[jaerdaph]

Once a long time ago there were datasets for PCGen and they were availible under the fair use clause and everyone was happy.

Can someone explain to me how distributing copyrighted material for anyone to download, whether they owned the original or not, is considered fair use?

 

[DMFTodd]

Can someone explain to me how distributing copyrighted material for anyone to download, whether they owned the original or not, is considered fair use?

I'm not speaking for the reasoning of PCGen in the early days, but there are a couple possibilities:

1) Game rules are not copyrightable in US law. If the datasets are purely game rules, then anyone can distribute them. That's a bit tricky though: Where do game rules end and IP, which is copyrightable, begin?

2) Copyright law allows you to use someone's copyrighted material if the use is "transformative" and/or "limited". Is the amount of IP as opposed to game rules within the dataset "limited"? Is a dataset, in a text format with no other use than to power a character generator, a transformative use of the limited IP?

And just to comment a little bit on the above, while it was Wizard's opinion that the PCGen datasets were infringing, it was never legally challenged. PCGen certainly did not "have to" remove those datasets. They could have made the two arguments above and seen whether Wizard's was willing to serve a cease & desist.

 

[Gnal]

And just to comment a little bit on the above, while it was Wizard's opinion that the PCGen datasets were infringing, it was never legally challenged. PCGen certainly did not "have to" remove those datasets. They could have made the two arguments above and seen whether Wizard's was willing to serve a cease & desist.

Correct, PCGen voluntarily removed the content that WotC said was IP. Yes, what you say is correct, maybe it was fair use, maybe it was game rules, etc.

*However* (and you knew there would be!) -
a) If WotC had complained to Yahoo that their IP was being infringed, *most* likely, the groups would have been deleted.
b) Same with Sourceforge, probably. They would have pulled the project, or at least the data portion of the project since at that time the code & data were separate (I think).
c) Merton Monk (PCGen originator) might have been taken to court if the data had not been removed, and at least in the US, being right does not always mean you win. You have to have lots of money in order to win if your opponent wants to make it difficult for you. And *nobody* on the Yahoo group, especially the most vocal complainers, ever volunteered to put their money where their mouths were.

-- david