[Zope-PTK] Workflow states wiki

Kent Polk kent@goathill.org
26 Jul 2000 17:37:37 GMT

On 20 Jul 2000 06:35:02 -0500, Steve Alexander wrote:
>Hi Folks,
>I've set up a Wiki to discuss Workflow states in the PTK, and also as a
>separate concern.
>  http://www.zope.org/Members/stevea/WorkflowStates
>If the results of this get merged into the mainstream PTK, I'll probably
>merge the important points in the Wiki into the PTK wiki, or something
>like that.
>There's not much there yet.
>I'll summarise the discussion. I'd also like to look for some consensus
>on what workflow means, and what its scope its.

Will you be looking for workflow examples that people will be
attempting to implement? I have a bit of a more unusual one that
I am working on (genetics research labs), starting from the edges
that push the bulk of the workflow. It's a little unique because
it uses daisy-chained workflow states that are tied to creating
and refining the core genope records. I.e. the core data stays in
one place but it's metadata and state data moves through the
workflow. The initial client software (functions that can't be
supported by a browser) I'm developing uses xml-rpc.

I'm writing the client/support stuff, waiting for Philip/Ty to get
a little further along with ZPatterns for workflows (and because
I haven't quite figured out how to do a lot of things yet :^).

Speaking of xml-rpc, is anyone working on updating xmlrpclib to
use one of the c-based parsers? I find I need to turn my record
sets into strings before calling the server or incur a 50x performance
penalty over the string representation or a custom parser I wrote
outside of Zope.