So perhaps this is the right time for the ultimate solution: --lets stop lusting after the idols of Beckhoff.--
I have and I burned badly. Lots of good productive time down the drain, because I liked the IgH stack so much. And of course I still respect that piece of software. But in my experience the Beckhoff attitude, the Beckhoff hardware and Beckhoff service are bad news. The hardware changes version every 5 minutes (read the configuration files) and when you need real service you get the Disperin solution. A while back I spend some very expensive money (i.e. lots of it) sorting out an intermittent problem in a 10 axes machine running on a Beckhoff NT computer. Beckhoff in Verl could not tell me (at any price) how to monitor the Sercos loop in a case like this. I had to find it out myself. In an Opensource environment that is off course quite acceptable, but THIS WAS NOT. (And I dont think it even fair on the local agent to talk about them) In my experience that company has sold its soul to Microsoft, which is good for business, but bad for technological exellence. Perhaps this is a good solution to the communities dilemma: http://www.ethernet-powerlink.org/ The opensource stack might be configured as a master or a slave (good for building your own io) and WAGO and a few others sell io off the shelf. cheers, j. On Fri, Oct 25, 2013 at 2:57 AM, EBo <e...@sandien.com> wrote: > On Oct 24 2013 2:23 PM, John Morris wrote: > > On 10/24/2013 02:06 PM, andy pugh wrote: > >> On 24 October 2013 19:55, John Morris <j...@zultron.com> wrote: > >> > >>> I believe that these restrictions mean that neither the LinuxCNC > >>> project > >>> nor anyone else may distribute EtherCAT drivers in source code form > >>> or > >>> otherwise, since with EtherCAT drivers, the software would > >>> essentially > >>> become an 'EtherCAT device' subject to Beckhoff's licensing > >>> requirements. > >> > >> Does it matter that the HAL driver in question is not itself an > >> EtherCAT master, but simply a glue layer between HAL and a > >> third-party > >> EtherCAT Master? > >> (An inexact analogy would be a GPL filter to save a document in > >> Microsoft Word format) > > > > I'm no authority at all, of course, just trying to interpret these > > matters for myself. I was becoming involved with packaging EtherCAT > > before this topic was raised, so it affects me personally. > > > > I'm focusing on the paragraph of Gerd's email following the title > > 'For > > product /device manufacturers'. > > > > He writes, 'Making, marketing and sale of a product making use of the > > EtherCAT technology requires membership in the ETG and licensing of > > the > > technology'. This sounds like LinuxCNC would qualify as such a > > product, > > and therefore those of us engaged in those activities (most all of us > > here in emc-developers) would be bound by those requirements, were > > EtherCAT to be included. > > > > A bit further down, he writes, 'if a product is a master stack > > software, > > a vendor of the master stack or the master device does require a > > technology license agreement for the master product'. I don't > > understand what 'master device' means or when the term would apply to > > LinuxCNC or systems it runs on. In any case, it does apply to IgH's > > EtherCAT Master for Linux implementation, and is a 'further > > restriction' > > explicitly prohibited by the GPL. > > We are going to run around and around and around wasting time on this > until someone 1) emails both IgH and FSF and ask for a determination on > a) will LinuxCNC require to have such an agreement, and b) if their > added requirement is a violation of the GPL; or 2) someone hires a > copyright lawyer to sort it out. All us arguing about it is a wast of > time unless somone here IS actually a lawyer, hired one, or emailed the > two principal determining parties. > > > ------------------------------------------------------------------------------ > October Webinars: Code for Performance > Free Intel webinars can help you accelerate application performance. > Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most > from > the latest Intel processors and coprocessors. See abstracts and register > > http://pubads.g.doubleclick.net/gampad/clk?id=60135991&iu=/4140/ostg.clktrk > _______________________________________________ > Emc-developers mailing list > Emc-developers@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-developers > ------------------------------------------------------------------------------ October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60135991&iu=/4140/ostg.clktrk _______________________________________________ Emc-developers mailing list Emc-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-developers
