[openfabrics-ewg] Re: How to generate IBEDreleasesusing 1.0branch

[Matters, Todd]

The way it's structured, you will only have to test with one stack - IBED.  It's just a little confusing how we get there. 

I agree that the marketing group needs some communiqué here.  

Todd


______________________
Todd A. Matters
SilverStorm Technologies
(610)233-4809


-----Original Message-----
From: Bob Woodruff [mailto:robert.j.woodruff at intel.com] 
Sent: Wednesday, April 05, 2006 4:50 PM
To: 'Roland Dreier'; Matters, Todd
Cc: Michael S. Tsirkin; Shawn Hansen (shahanse); Sujal Das; Openfabrics-ewg at openib.org
Subject: RE: [openfabrics-ewg] Re: How to generate IBEDreleasesusing 1.0branch

Roland wrote,

    Todd> But both are sanctioned by OpenFabrics; IBED by the EWG and
    Todd> Release 1.0 directly by the board.  With the establishment
    Todd> of the EWG by the OpenFabrics board the consortium of
    Todd> vendors hardening the stack became an arm of Open Fabrics.
    Todd> If they're both sanctioned by OpenFabrics I don't see why
    Todd> both can't, eventually, be a single release process.

>Yes, in my opinion (emphatically not speaking for Cisco here) it was a
>mistake to make IBED part of OF.  Unfortunately people saw IBED as
>competitive rather than complementary and so politics forced it to
>be embraced by OF.

> - R.

Putting my software ISV hat (Intel MPI)on, we'd really like to only 
have to test with one infiniband stack.