A Dimarts 11 Novembre 2008, Jan Kiszka va escriure: > Leopold Palomo Avellaneda wrote: > > Sorry for the delay ... > > > > A Dimecres 05 Novembre 2008, Jan Kiszka va escriure: > >> Leopold Palomo Avellaneda wrote: > > > > [...] > > > >>> but someone is working on this? > >>> do you know if someone of fsf is looking on it? > >> > >> FMTC and Beckhoff people are said to work on resolving the issue. I > >> don't know any details nor news, but someone just asked again for an > >> update on ethercatmaster-users. > > > > well, to me it's and strange issue. I will ask to some people that know > > more than me about this kind of issues > > > >>>>> Powerlink, but I have understood that someone in the list > >>>>> have no good opinion. > >>>> > >>>> ??? > >>>> > >>> :-) > >>>> > >>>> There is no [L]GPL'ed stacked for Powerlink yet, but at least some > >>>> BSD-licensed code. > >>> > >>> well, there are some the code, bsd licensed. Are you saying that there > >>> is a functional code? Well, probably I didn't understand the answer. > >> > >> I haven't looked into details, but they claim its functional. And as > >> long as the stack can live in user land, the otherwise unfortunate > >> license should also be no problem. But I think to recall that they also > >> had kernel bits. > > > > And, a part of that the license is too permissive, what's the problem? > > Besides that there can be compatibility issues with all the (L)GPL code > around it -- the major issue is that is it gives a counterproductive > idea about open source to people, specifically to those who come from > the land of "cathedral" software development. > > I personally like the "didactic" aspect of quid pro quo in the GPL a > lot. It urges people to do clean(er) separation between their actual > core know-how and common infrastructure and drives them to give > improvements on the latter back. If you have code with which you can do > what you want AND you are not forced to show it to someone else... I've > seen quite a few examples.
Ok, I agree with you. MS (and others) loves the bsd licence... > >>>>> So, someone knows if exists industrial devices that could be > >>>>> controlled by rtnet? or someone could give some opinion about all > >>>>> this mess of "open" protocols? > >>>> > >>>> RTnet is not directly comparable to "full-blown" industrial RT > >>>> Ethernet approaches. RTnet is an open stack that can even be used to > >>>> implement some of those protocols. > >>> > >>> So, someone could implement this kind of protocols "over" rtnet, no? > >> > >> As far as no special requirements (e.g. too demanding timing for a > >> GPOS/RTOS system) prevents it, yes. One example is EthercatMaster over > >> RTnet. > > > > Ok, it was the implementation of the ethercat from FMTC. And etherlab? > > As I see it: a "pragmatical" approach, specifically when it comes to > real-time. It certainly has its use, but you should carefully check if > it can fit your requirements, also on a longer term. Ok, I begin to see ... > >>>> But it does not come with its own abstraction > >>>> of industrial devices (drives, I/O clamps etc.). Ie. there are no > >>>> "RTnet-compliant" industrial device definable due to this undefined > >>>> highest layer (industrial applications). > >>> > >>> Have you think about it? To promote some kind of standard protocol for > >>> industrial using rtnet? > >> > >> Defining and promoting such a protocol takes orders of magnitude more > >> time than a single person can spent. :) > >> > >> Seriously, such things requires a broad user base that wants to drive it > >> or simply contribute to it. > > > > So, please correct me if I'm wrong. RTNet would be the low level part > > (connects to the OS and the devices, driver, etc) and powerlink or would > > use this to implement _over_ its own protocol, no? > > if yes, it would be nice, because they could share the driver part and > > all of us would benefit about it. > > Yes, that's the basic idea. You cannot catch up with the performance of > Powerlink components that use dedicated controllers, for sure. But AFAIK > nothing prevents doing things with the same protocol on a realtime GPOS > with RTnet as low-level interface, just a bit more "relaxed". So, following this, it could be possible to borrow the powerlink code and put it over rtnet, no? uff, too many things to do and only one life ... Regards, Leo ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ RTnet-users mailing list RTnet-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/rtnet-users
