The first joint event organized by both Accellera and OSCI was what Stan Krolikoski, moderator of the event, labeled a "Town Hall" meeting held during the lunch time of the first day of DVCon. The food was good but Spartan, the hall full of EDA professional and electronic designers (around 300), and the topic well presented.
The idea was to have UVM experts in the audience answer questions. It worked, at least to the point that I believe everyone left the meeting knowing something more about UVM than what they had known before.
Given that UVM had just been released there were not any reports from actual users. But there were many questions about how the new methodology could be used and many reassuring words about the applicability of UVM to the various modeling languages. This, unfortunately, resurrected the old language wars, something I thought was not only buried, but also quite inconsequential. But you know, it was a town hall meeting, and events like this never fail to give to fringe speakers a ready audience.
Speaking of fringe speakers, I am told that many did not understand my "pigeons" remark. So here is the full explanation. Standards made in the vacuum are no better than statues of unimportant people. The latter are only appreciated by pigeons.
Getting back to UVM. It seems to me that this standard is an important step forward in improving verification quality and efficiency. One of the aspects less understood of UVM is the reference implementation library. The software module that comprise the library are not meant to be used in actual designs, they are example of the use of the methodology. Thus no one should expect Accellera to become a VIP provider.