Re: oaf async activation

["Diego Sevilla Ruiz (dsevilla um es)" <dsevilla ditec um es>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Miguel de Icaza wrote:

> > Those times are way too long for anything. If you can reproduce times
> > like that with recent code, please file a Nautilus bug report.
> >
> > In any case, sucky slow to start components will always exist, no
> > matter how hard we try; I think it's only reasonable to let oaf
> > clients avoid blocking their UI in such a case.
>
> Agreed.  I like Mathieu's proposal.
>

I agree too. Ideally, synchrony/asynchrony isn't an "interface" concern,
so it would be better to define "poller" and "callback" objects as the new
AMI (Asinchronous Messaging) proposes, and let the client to choose the
QoS model of each invocation (which includes if it wants sync/async calls,
timeouts, etc.), but... in the meantime, it seems to be a good design: it
maintains the sync interface, so it could be used when ORBit implements
AMI...

    This raises another question... Is ORBit supposed to eventually
implement AMI?

    Regards.
    diego.

--
Diego Sevilla Ruiz -- http://ditec.um.es/~dsevilla/ -- dsevilla um es
       Departamento de Ingeniería y Tecnología de Computadores
           Facultad de Informática. Universidad de Murcia
Campus de Espinardo  -  30080 Murcia (SPAIN).  -  Tel. +34-968-367570
PGP:  http://pgp.rediris.es:11371/pks/lookup?op=get&search=0xC9B964B7
\huge d\em\kern-.36em\lower-.2ex\hbox{\small sevilla}\kern- 1em um es
perl -e'$_="\x4\ FLe\x2&B";for(/../g){print unpack("b*",$_),"\n"}'|tr 01 " #"




-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 5.0i for non-commercial use
MessageID: VmL5skHJ4DrJKAW3MVPIWlXSiKLhl2UZ

iQA/AwUBOdzJX9oq0AfJuWS3EQID0QCff7Lt+2hKVaoql3zPVXgx1PDcqfYAoPQc
WcwhwMq/V6uNV5/kLp2Dtn6o
=GgyN
-----END PGP SIGNATURE-----