Loading ...
Sorry, an error occurred while loading the content.

Re: [BUG!!!!]Hit points for non-humanoids (maybe others) totally FUBARed

Expand Messages
  • Reed Thornton
    ... work ... not ... do. If ... me. ... Knowing ... correct the ... which ... debugging ... It s ... it off by ... experience, the ... set of ... other ... dev
    Message 1 of 36 , Sep 1, 2004
    • 0 Attachment
      --- In pcgen@yahoogroups.com, Lizard <lizard@m...> wrote:
      > On 06:42 PM 8/31/2004, Devon Jones said....
      > >I know I for one had a hard
      > >time not being offended by the characterization of my last year of
      work
      > >(GMGen) as a waste of time.
      >
      > I apologize. I didn't mean to imply it was a waste of time. It is
      not
      > something *I* need, but you know your customer base better than I
      do. If
      > most users are clamoring for it, then, you would be right to ignore
      me.
      >
      > I'm glad I was able to help pin the bug down to default monsters.
      Knowing
      > this helps, as I can ignore all 1 hit dice humanoids and just
      correct the
      > others; this SERIOUSLY reduces the error checking I have to do.
      >
      > As a programmer, let me note something like default monsters --
      which
      > creates a seperate 'mode' for character creation -- smells like a
      debugging
      > nightmare, as it means almost everything needs to be checked twice.
      It's
      > obvious from the early responses that most of the dev team leaves
      it off by
      > default. I can very much attest, from 15 years of personal
      experience, the
      > problem with 'programmer testing'. You use a system with a specific
      set of
      > options, preferences, and so on, and everything works fine. Bugs in
      other
      > modes can linger for months or *years* because no programmer on the
      dev
      > team ever uses a specific option.
      >
      > Testing -- REAL testing, mind-numbing precise, ritualistic,
      testing --
      > isn't fun. And if it's not fun, no one will do it for free. This is
      a major
      > weakness of open source - it requires users to report bugs, because
      the
      > developers are not going to find any but the most egregious. No one
      who is
      > a skilled programmer is going to be satisfied or happy doing the
      tedious
      > job of QAing, and those people who are skilled QAers will do it for
      money
      > only, 'cause it ain't fun. (Of course, there are exceptions, but
      that's the
      > general rule.)
      >
      > If I get a chance, I will try to write up a formal list of
      suggestions and
      > issues brought about by using PCGen for 8 hours a day for a month
      solid.

      If you do get a chance to write up that list, I would be very happy
      to take a look at what you have. I am in charge of the QA team, and
      we are working on making the tests that we have better, and easier to
      run (ie as automatic as possible.) Any information on failures is
      useful, assuming that we can extract what the failure is. It is easy
      for us to miss a problem in the volume of traffic that is generated
      on the various PCGen lists (I watch seven lists, and I'm sure that
      there are some that I am missing.)

      Unforunately, it is all too easy for us to miss the same thing over
      and over. We (the QA team) are supposed to be testing the interface,
      testing the output sheets, testing the core functionality, verifying
      the data sets accuracy,... it would take a large team of experienced
      testers a while just to work out what to test, and how. Throw on to
      the top, the myriad ambiguities within a source (RSRD for example),
      compound that with all of the other sources out there, and then try
      and adjudicate what the *proper* answer is.

      Reed, QA SB, BoD
    • merton_monk
      There has been an awful lot of activity lately on both the data and code teams (including a fix for this and many other items!). Hopefully we ll have a release
      Message 36 of 36 , Sep 5, 2004
      • 0 Attachment
        There has been an awful lot of activity lately on both the data and
        code teams (including a fix for this and many other items!).
        Hopefully we'll have a release date soon for 5.7.5 (which would still
        be an alpha). I imagine it will be sometime this week (maybe next
        weekend).

        -Bryan

        --- In pcgen@yahoogroups.com, James Dempsey <jdempsey@a...> wrote:
        > Hi Joe,
        >
        > Joe Medica wrote:
        > > OK, I should have read the entire thread first.
        > >
        > > Do you have an idea of when this will make it to a beta or production
        > > release? 5.7.5 maybe?
        > >
        > > Thanks.
        > >
        > > Joe Medica
        >
        > It's in CVS, so 5.7.5 will have the fix.
        >
        > Cheers,
        > James Dempsey
        > PCGen Code Monkey
      Your message has been successfully submitted and would be delivered to recipients shortly.