I understand where Matt is coming from. Complete disclosure of the results to the public can create some serious issues for an organization.
Instead of a “Consumer’s Report” style document, it may be better to just create a “Catalogue of Compliant Cables.” That would permit devices that failed to be re-designed and re-tested for compliance and addition to the catalogue. Cheers, Brad Brad Booth Principal Engineer Microsoft Azure Networking brbo...@microsoft.com<mailto:brbo...@microsoft.com> From: opencompute-networking-boun...@lists.opencompute.org [mailto:opencompute-networking-boun...@lists.opencompute.org] On Behalf Of Steve Noble Sent: Tuesday, July 14, 2015 2:51 PM To: Matt Peterson Cc: opencompute-networking@lists.opencompute.org Subject: Re: [Opencompute-networking] Open Networking Consortium Testing - Membership Proposal Doc Hi Matt, See inline Matt Peterson<mailto:m...@peterson.org> July 14, 2015 at 12:44 PM Agree that results need special treatment and don't "go public by default". The obvious end goal of "Consumer Report" is clear, however language is needed to specify that exact scenario you mention. Advertising a failing item isn't doing a positive service for anyone, however members need to be encouraged to remedy problems when they arise. Not sure how to word some equal ground here, maybe a negative test is given a "TTL" or timeout to become compliant with all the parties involved, otherwise it never makes it to the public site. It may also seem a bit weird if all of public matrix data is 100% positive with no defects or failures, maybe some generic reporting is a half-way common ground "X number of pluggables had issues with Y and Z number of switch unique pairs" - without listing the member or organization by name? I believe there should be a waiting period, for the vendor to determine what the failure was and if/when/how it will be fixed, but I still contend that with the "Consumer Reports" type concept, the failure data needs to go out. I don't agree a failing item is negative for everyone, I think the knowledge is useful to anyone who is or might be using the device that failed.
_______________________________________________ opencompute-networking mailing list Unsubscribe: http://lists.opencompute.org/mailman/options/opencompute-networking opencompute-networking@lists.opencompute.org http://lists.opencompute.org/mailman/listinfo/opencompute-networking
