Reply to thread

Message: <blockquote data-quote="thpr" data-source="post: 3394303" data-attributes="member: 48911">I believe this has been addressed, but as one of the most active committers I would note I haven't worked for CMP and have no intention of doing so.  More importantly, I am one of the critical people driving the direction of PCGen, and if CMP or anyone else tried to hijack the project, I'd fork it so fast their head would spin... just as I would expect the Board of Directors to revoke my commit access if I became a liability to the project.  The board of directors _can't_ stop a fork of the code, nor can CMP.  Even _I_ can't stop a fork of the code unless the fork attempts to violate the license the code is under (LGPL).  The only thing limiting a fork of the data is permission from the publishers.  With good cause and a well written description of the reason for a fork, I doubt that issue would last more than a few weeks.The PCGen formats have changed.  That was recognized as a problem some time ago.  Since that time, effort has been made to deprecate items across revisions, so that older data does not break.  We are making efforts to ensure a smooth transition from 5.10 to 5.12 (and to future versions going forward).I highly doubt anyone has been denied a bug fix.  Given that the code is open source, I don't even comprehend how a fix could be denied to someone.  The production versions are maintained and criticial items are fixed (we just released a 5.10.2).  You're seeing a conspiracy where there is none.  What happens on open source projects is that developers turn over.  Some of them have had children, others have encountered new jobs, others have simply disappeared.  This is not evil, but I think it's a testament to PCGen that there are other developers who have picked up the project and continue developing.As someone who started coding on the PCGen project in late 2005, I think there is a major problem of knowledge concentration.  There is a bunch of undocumented code that performs what can only be described as 'deep magic'.  There are undocumented features in the LST files as well, and that has made it difficult for us to actually make a conversion to a different data format, because we can't explicitly define the new format using our documentation.Yes, this is a problem... we know that.  We have slowly been tightening the rules on the LST syntax to eliminate some items that are not used or are old syntax (Go look at my code checkins and the posts I make to the pcgen_experimental Yahoo! group).  We have been improving our documentation (go look at the number of documentation bugs I have opened in the last 6 months).  This helps us reduce the quantity of code we are dealing with and the complexity of the LST format... this opens the possibility of another data formats.These issues make changing some parts of the current code an extremely difficult proposition.  It's fragile and highly tangled (those are both technical terms, fragile code being code where small changes tend to produce many problems, and tangled code being code that has circular dependencies, which leads to huge complexity to make simple changes).This has led to challenges in doing major conversions like you describe.  It is HARD (I resist stronger language because enWorld tries to produce a "family show").  I am one of the most active committers to PCGen, and I wouldn't go NEAR an XML conversion today, and that has NOTHING to do with my like or dislike for XML.  (I WILL note here that my other project, RPG-MapGen, entirely uses XML for its data storage).  I would LIKE to see a data format that is usable across tools, I would LIKE to see a saved character format that could be opened by other tools, and I think XML is a great candidate.Nor would anyone have had access to WotC data.The problem is that WotC didn't feel it worked under the fair use clause.  The resources to line up a defense in federal court do NOT exist for PCGen, and never have.  The ONLY option as a response to WotC was exactly what happened (ceasing shipment of the closed content data)I would note fair use is a general concept, and the realm of data for role playing games is NOT clearly defined.  The fact that the datasets were free (as in beer) did NOT mean they are not "commercial in nature" (the exact words in 17 USC 107(a)).  What would have happened is that WotC would have sent a cease and desist letter, and the project would have become history.[edited to add clarity to the fork comment]</blockquote>

Verification