On Thu, Sep 25, 2008 at 10:40 PM, Michael Krufky <mkrufky@xxxxxxxxxxx> wrote: > On Thu, Sep 25, 2008 at 4:18 PM, Jörg Knitter <joerg.knitter@xxxxxx> wrote: >> VDR User wrote: >>> Another option would be to look at both proposals, take the best ideas >>> for each, and marry them into a new hybrid proposal so-to-speak. By >>> that I don't mean something which must be built completely from the >>> ground up.. Most of the work has already been done. I think that is >>> a workable solution that can be found somewhere in the middle. >>> >> Thinking about all the posts within the last days, I did not propose >> something as this as I don´t believe that certain persons are still >> willing to cooperate or change their point of view. To be more precise: >> I think it would be a wonder if Steven Toth and Manu Abraham worked >> together. ;) >> >> With kind regards > > For the record, Manu did a great job with Multiproto. Steve worked > with the multiproto API and added driver support for his devices. > Steve depended on Multiproto in order for his devices to work. Steve > asked Manu when he would merge into the master branch, and waited > patiently for months. Manu wouldn't even merge Steve's drivers into > Manu's own multiproto tree. After repeatedly asking him to merge with > no sign of progress, Steve was left in a tough spot and had to come up > with his own solution. > > There is a saying that goes often for Linux development... "release > early, release often" .. I first heard this saying from Johannes. If > you dont go by this way of thinking, ie: hold on to code for two and > a half years without allowing a merge, there is no guarantee that > somebody else may come up with a better solution. > > There was no sign that Multiproto would _ever_ get merged. All of us > have waited for two and a half years for this. The FIRST pull request > from Manu only appeared after Steve posted his own API proposal. Had > Manu requested merge but one day before that, it would have been > merged and that would have been the end of it. > > As it turns out, Steve's API is more flexible, and allows us to > support more features than Multiproto allows. If Manu would have > allowed Multiproto to have been merged into the master branch, Steve > would never have designed this newer API. Due to the superior design, > that was the chosen extension. This is where we are left today. > > I hope that clears things up. > Julian put together a history (just because Michael keeps writing the same): http://lkml.org/lkml/2008/9/19/67 Markus _______________________________________________ linux-dvb mailing list linux-dvb@xxxxxxxxxxx http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
