[Pipet Devel] Pied TODO

[Brad Chapman]

> Here is an updated TODO list for Pied.  I need one for each part of 
Piper. 

Okay dokee, at the bottom of this is a TODO list off the top of my 
head for the definition layer.

> They will of course be included in the code, but I also want to put 
them on
> the web (Brad's suggestion) so that potential volunteers will know 
what needs
> to be done.  

I also need to improve the documentation for how to setup and install 
Piper, and I'll work on this (and probably convert it to LaTeX or 
something).

> Hurry, hurry!  Then we can
> announce the project on Gnotices and Freshmeat, right Brad?

Sure :-). You can post without it too if you all want to. You just 
asked for opinions, so I gave one...

Brad

Definition layer project list
------------------------------

* Short term stuff to finish:
-- Finish integration of Overflow into Piper
-- Finish implementing MVC. Debug the code and make sure it works 
properly.
-- Improve and finish implementing the connection with the BL. Modify 
the
code in the DL to deal with the kind of information the BL will give 
us,
and also the kind that it needs.
-- Finish saving and loading of work flow diagrams as XML.

* Update existing documentation to be consistent with what we've got 
in the
code. Add more documentation about how to run Piper and deal with 
different
things.

* Improve the system for storing passwords safely. The current setup 
should
be reasonably secure, but the only way to deal with it is through a 
hard to
use command line program. which probably doesn't even work properly 
with
the current setup. So the basic code exists for this, but it needs some
serious work to fix up and make it easier to use.

* Dealing with databases.  The basic code to connect with three kinds 
of
databases (MySQL, PostgreSQL and Gadfly) already exists, but doesn't 
work
properly with the new CORBA framework since I haven't done this yet.  
So
need to get this working, and also expand it to include support for 
making
SQL queries on the databases.  Then this needs to be integrated with 
some
user interfaces to give a way to manipulate the databases through the 
user
interface.

* Dealing with the filesystem.  Again, the basic code to do this 
already
exists, but it needs to be fixed to work with the new setup of the
definition layer, and fix the user interfaces to deal with it.

* Remote Communication with other Piper instances.  Since we aren't
positive how remote communication will occur between nodes, this should
probably mostly be focused on connecting with remote Piper programs, 
and
then querying them for available nodes, and then work on a way to 
integrate
these remote nodes with local nodes so they work together seemlessly.

* Implementing stuff that make it easier to write XML files for 
programs
that will work with Piper. Need to work on automating the system of 
plugins
so it is easier to deal with.