Peter, thanks for raising these important questions. You really deserve some
answers, although they don't come lightly, especially since they probably
imply significant work;-)
(I am writing thesis now, so I cannot presently allocate much time for this.)
I think that work on connecting realtime with non-realtime should be given the
highest priority. LXRT sounds promising. That I favor CORBA and ACE/TAO is
probably no surprise (I agree there are drawbacks). But I think we should
keep interoperability with other orbs and programming languages. To give an
example, that excludes the use of the TAO real-time event-channel.
There is also the option that all of orocos converge into using real-time.
That would of course mean a lot of work(?) for people like us. Question is
what we would loose in terms of functionality, library support, ease of use
etc? (My knowledge about real-time programming is almost zero.)
I actually think we should investigate these matters quite thoroughly,
preferably coding some quick implementations to get a clearer view. Unless
there is someone skilled in both real-time and corba who is willing to do
this, people from both camps should probably sit down together for a week of
intense hacking. Again, for my part this will have to wait a while.
I don't regard the issue of 'Use common libraries' as being very urgent to
fix. The 'global repository of idl interfaces' will only have real meaning
after we have achieved interoperability within orocos.
./Anders Oreback
--
Anders Oreback Phone: +46 (0)8 790 6204
CAS/CVAP/NADA Cellular: +46 (0)70 281 9664
Royal Institute of Technology Mailto:[email protected]
SE-100 44 Stockholm,SWEDEN http://www.nada.kth.se/~oreback
lmpx.com only provides a reader for public news (NNTP) servers. It is not
affiliated with the servers or forums shown here and is not responsible for
the content of articles, which is written by their respective authors.