[Gambas-user] Restructuring the official Gambas examples

Tobias Boege taboege at ...626...
Tue Apr 30 22:20:43 CEST 2013


On Tue, 30 Apr 2013, Sebastian Kulesz wrote:
> Maybe there is a way to do this already without changing anything. Consider
> this:
> 
> The complexity of an example (project?) *may* be given by the amount of
> components it uses (just thinking out loud). But, if not accurate, it
> should be a really good approximation.
> 
> So, sorting the examples by the amount of components it uses may be a short
> therm solution for this, if not a permanent one, until a consensus is
> reached.
> 
> A slight change to the previous approach: assign each component a numerical
> difficulty and then compute the final score by adding the value of each
> component used. Then sort them using that value. We would then only need to
> assess the difficulty of using each component.
> 
> I know this is not the most accurate solution, but it's much less
> intrusive.

Wait, how do we assign each component a numerical difficulty level? OK, we
have the *.component files but that seems even more intrusive. IMHO more
difficult to rate either. Having "Beginner" and "Advanced" in examples makes
it feasible that everybody classifies his examples himself.

I can only precisely speak about the example I wrote: look at Games/Pong. It
only requires gb.ncurses and is complex as hell. OK, admittedly gb.ncurses
wouldn't climb too high on the easy-to-use ladder but that can't compensate
using three to four graphical components like some other games do.

You see: "Games" tend to always be more complex than "Basic" examples and
that's why I want to conserve the topic grouping.

On the other side, of course, the biggest example, the IDE itself, does
somewhat support your idea.

If we go and rate the project structure, maybe the number of classes is
better, but IMHO still the explicit, human-chosen grouping is best.

Regards,
Tobi




More information about the User mailing list