How do you plan before starting to code ?

[Richi Dubey]

Hi,

Thanks for your suggestions. You've used lots of new keywords that I need
to learn, I'll look them up online.

I like your quote at the end!

On Wed, Jul 8, 2020 at 8:04 PM Per Dalgas Jakobsen <pdj at knaldgas.dk> wrote:

> Hi,
>
> I believe that differs a lot...
>
> If it's something I can't completely get my head around, I start
> modeling the problem/solution using UML or BPMN or ...
>
> In this case I would probably start with a simple Actor-Context model
> (stick-men), a domain model (class model), and some sequence models and
> a state model to give an idea of how the scheduler should act to stimuli
> (timer, blocking, request, etc.). Be careful not to fall into the
> Analysis-paralysis trap - Just do enough modeling til you understand how
> it would work. Documentation should not be redundant with the code itself.
>
> A quick prototype can sometimes be of great value too...
>
> ~Per
>
> "Weeks of programming can save you hours of planning.
> -- Unknown"
>
>
> On 7/8/20 2:56 PM, Richi Dubey wrote:
> > Hi,
> >
> > I am working to implement a Strong APA scheduler (a new SMP scheduler)
> > in RTEMS. I need to develop a high-level description of my
> > implementation so that I can get it reviewed by the maintainers before
> > I start coding.
> >
> > How do you make a plan before coding? Do you use UML diagrams or do
> > you document every function (with argument, prototype and the aim of
> > the function) in a pdf file and get that looked at?
> >
>
> _______________________________________________
> users mailing list
> users at rtems.org
> http://lists.rtems.org/mailman/listinfo/users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/users/attachments/20200709/e4e4c3f7/attachment.html>