Re: serialization of dynamically sized arrays

[Benjamin F. <ben...@he...>]

Sorry Matej, I meant to keep the discussion on the list, hope this is ok with
you.

Am Montag, 5. September 2011, um 20:50:45 schrieben Sie:
> > So, does that mean arrays are always sent either wholesale or not at all?
>
> There is an ArrayRequest that allows you to get/set parts of the array.
>
> > This is strange, given the otherwise strong emphasis on performance. Take
> > the example from the spec: what if most of the time the (x,y) positions
> > remain the same, only the values change? Do I have to pay 3 times the
> > price for every monitor event? And what if I have a very large array,
> > but only a few of the elements change at a time?
>
> For sure this is doable and a nice feature. I think even me and Marty
> talked about it.
> However, currently we try to have a rock-solid core and then we can
> add features.
> This is better than non-stable SW with lots of features.
> What do you think?

I don't know if I understand what you are trying to say here. Does rock-solid
core mean that the core protocol is something you do not wish to change,
because changes to the protocol definition would make your software (the client
and server library) unstable? If this is the case, I'd rather know sooner than
later; I am not planning to spend time on reviewing the protocol if the design
is already set in stone. Note that the protocol itself has not been discussed
in any detail on the mailing list. Is it your opinion that it is already as
good as it gets and no further improvements can be made?

Regarding change bitsets for arrays: you seem to think that this is something
that can be added in later versions w/o making any disruptive change to the
protocol definition or the software. If you could explain how this can be done
I might be convinced to let the issue go.

(IME, the road to reliable software is to get the design right from the start,
not to add feature after feature afterwards, when it turns out users need
them.)

Cheers
Ben

________________________________

Helmholtz-Zentrum Berlin für Materialien und Energie GmbH

Mitglied der Hermann von Helmholtz-Gemeinschaft Deutscher Forschungszentren e.V.

Aufsichtsrat: Vorsitzender Prof. Dr. Dr. h.c. mult. Joachim Treusch, stv. Vorsitzende Dr. Beatrix Vierkorn-Rudolph
Geschäftsführer: Prof. Dr. Anke Rita Kaysser-Pyzalla, Dr. Ulrich Breuer

Sitz Berlin, AG Charlottenburg, 89 HRB 5583

Postadresse:
Hahn-Meitner-Platz 1
D-14109 Berlin

http://www.helmholtz-berlin.de