[kplato] Hello

[Dag Andersen]

Mandag 13 november 2006 16:47 skrev Nicolas MICAS:
> Hello!
>
> We haven't see the question asked by Dag about our criteria for selecting
> things to do.
> To answer this question it is easy! (or ISI as you wants :P)
> We have studied the TODO list of KPlato on SVN and after clear it from all
> things which were done we have selected those which seems to be most
> important and easily realisable during our project.
Yes, it's nice to be able to finish something :)
I was also thinking of whether you preferred many small fixes, or one or fewer 
larger tasks, gui, non-gui...
I thought maybe something that is distinct which you can point your finger at 
and say "We Made That". Should make it easier for you to document what you 
have done, also (important, no?).

The things mentioned by Jim in his other answer to you could be canditates, I 
think. In general, I have almost not touched functions needed for project 
execution yet, I haven't gotten further than the planning stage.

Another example; I would like to be able to get data from project(s) into 
kspread for further processing. The method would be to make a plugin for 
kspread that fetches data from kplato files. There exists such a plugin for 
kexi which uses SQL to extract data from a database. I was thinking of 
something similar.

> We are studiing a another items to do, relative to the PERT method, it will
> consist to find a simple way to key PERT data and so to calculate
> particulary things of the PERT method without a graph.
The solution now (in the task dialog) is to enter three estimates: Expected 
(E), Optimistic (O) and Pessimistic (P) along with Risk.  O and P is entered 
as a persentage of E. They are stored as durations though, so changing that 
is easy. Risk controlles the PERT distribution used to calculated the 
estimates used during scheduling.
There is no help to analyze the results, I haven't gotten that far yet ;)
>
> Best regards.
>
> 2006/11/10, Gaël de Chalendar <Gael.de-Chalendar at cea.fr>:
> > Hello,
> >
> > Le vendredi 10 novembre 2006 08:30, Dag Andersen a écrit:
> > > Torsdag 09 november 2006 18:34 skrev Kevin Ottens:
> > > > I'm wondering about this one. I'm not 100% about the semantic of this
> > > > dependency. Could someone enlighten me?
> > >
> > > task1 <--- task2.
> > > means task2 has to start before task1 can finish, so I suppose task2
> >
> > should
> >
> > > be calculated first.
> > > I don't think there are many use cases for it, I can't pull one from
> > > the top of my head just now ;)
> >
> > Maybe some specifications (like a file format) of task2 (that could be
> > splitted in more detailed subtasks) should be finished to allow the final
> > implementation of task1 ?
> > This allows to have some tasks not too much detailed.
> >
> > Regards,
> >
> > Gaël
> > _______________________________________________
> > kplato mailing list
> > kplato at kde.org
> > https://mail.kde.org/mailman/listinfo/kplato

-- 
Mvh
Dag Andersen