[DotGNU]Reminder - Weekly DotGNU Meeting

[Gopal V]

DotGNU holds weekly meetings to discuss any issues
of relevance; to allow the developers to catch up
with each other; and to allow new members to ask
questions and find out how to join in.

The meetings are conducted via IRC in the channel
#dotgnu on irc.openprojects.net at the following
times:

    Saturday 1000 UTC
    Saturday 2200 UTC

The first is best for DotGNU members east of UTC
and the second is best for DotGNU members west
of UTC.  Some of the key DotGNU members attend
both meetings.

The logs of #dotgnu are always available here:

http://ajmitch.dhis.org/dotgnu/

Note: the times are UTC, not GMT.  That is, they
do not include daylight savings adjustments during
the northern summer.  If you have a Unix system,
then the command "date; date -u" will give you
your current time in both local and UTC, allowing
you do determine when the next meeting will occur.

(Rhys)
--------------------------------------------------

This week's meeting is special . This is less of a
bugs & features discussion. This week I have a couple
of solid agenda items to suggest. This is the longest
agenda list for an IRC meeting ever !!. But this is
intended to put the ideas in perspective.

*) Parrot support for Cscc ?

      I have read the Cola compiler code and other 
      details, to be relatively sure that the transition
      is possible. The so-called Stack to Register 
      transition is relatively easy for Cscc (compared
      to gcc) as we can hack the ILNode_GenValue to
      work here as well. Because ILNode_GenValue does
      not assume stack or register model ... (or so
      I understand).

*) Why Parrot ?

      It would seem that the Parrot suggestion came out
      of the blue sky.. If you think so, you inherently 
      mistake one of the original tenets of DotGNU and
      GNU itself. Avoid Patents . Microsoft has patents for
      .NET and maybe even for IL and C# . So if we restrict
      ourselves to only C# and IL we stand the risk of
      being a sitting duck ... To put it in another way,
      we can't dodge the patent bullet . We've got out of
      the C# issue by supporting C . But still we're not
      out of the woods for the bytecode issue.

*) What's so great about Parrot ?

      AFAIK Parrot is the only GPL'd and patent free universal
      virtual machine available. And one which uses a good &
      flexible bytecode as well. Also we have hands into the
      spec as well (like the #parrot guys said, "if you want
      a feature, send in a patch"). They have a nice JIT as 
      well.

*) System.Xml development

      How to proceed with this .. I've been pondering about developing
      the XML parser using a scanner & parser generator . Something like
      Coco-Csharp. Does anyone know if there are any license , or other
      issues with using it. I know someone is going to say that there are
      better was of doing it ... But I just am trying a new approach.
      (I'll atleast be able to use the lexer ;-)

*) XmlRpc support

      I thought about doing it with dotgnu.xml . It is fairly straight
      forward to implement it with that. I could be able to parse XmlRpc
      data easily. Would some other WW hack out this stuff ?
      (Refer: python's xmlrpclib.py for an idea of what I'm planning).

*) Server frameworks & threading

      I'm thinking about porting one of my old Java WebServers into C#.
      But that seems to need threads for the worker pool. How soon can
          we get the Threads support into the C# part ... test_thread passes
          all tests for me :-)

*) Treecc , JITs and gcc

          I'll reveal this when I come along at the meeting ... If you can't
          guess, you're unaware of or underestimating the abilities of treecc.

Gopal
-- 
The difference between insanity and genius is measured by success