W3C home > Mailing lists > Public > www-qa@w3.org > October 2001

Re: One reason why certification is important

From: Lofton Henderson <lofton@rockynet.com>
Date: Fri, 26 Oct 2001 11:05:11 -0600
Message-Id: <4.3.2.7.2.20011026103925.03093710@terminal.rockynet.com>
To: Alex Rousskov <rousskov@measurement-factory.com>, Rob Lanphier <robla@real.com>
Cc: www-qa@w3.org
At 10:27 AM 10/26/01 -0600, Alex Rousskov wrote:
>On Fri, 26 Oct 2001, Rob Lanphier wrote:
>
>[...]
> > A certification mark would give the rest of the community a tool
> > for making users of MS products feel uncomfortable (at least until
> > MS got certified).

In my past (commercial) experience with product certification, I found that 
one of the main benefits was to simplify the dialog about who's right and 
who's wrong.  "Ask to see their certificate.  Here's ours."  The 
simplification is both good and bad.  It's bad in that it obscures *useful* 
information.  It's good in that it obscures *a lot* of information -- it 
encapsulates some bottom-line information in a PR-oriented statement that 
is quickly and easily grasped by a mass audience.

We used our certificates both pro-actively and defensively.  While some 
competitors and potential clients were immune, it was clear that others did 
pay attention.  In the balance, we were glad for the existence of the 
certification program.

Please that I am not recommending that W3C get into the certification 
business -- I don't believe that it should.  But if others do, and if they 
use the W3C materials (e.g., test suite) and W3C name, then W3C may have an 
advisory and quality-control role to play (accreditation of test labs, 
stake-holder in a Control Board, etc).


>I bet MS products will be ones of the first group to get certified [by
>a MS-sponsored certification company, for example]:
>
>         "In CertificateForTheBuck lab tests, Internet Explorer has
>         passed all our certification criteria, once again
>         demonstrating MS leadership in supporting W3C standards".

This is a danger.  There is some body of experience in how to go about 
setting up certification programs so that a "CertificateForTheBuck" 
operations are clearly marginalized.  Assuming that W3C owns and controls 
the Rec XYZ conformance test suite, for example, it can assure that only 
properly *accredited* test labs can use it for certification, and that the 
W3C name can only be used in association with those labs.

By the way, IMO, the above hypothetical quote starts to border on violation 
of some existing W3C PR rules, doesn't it?  W3C members are required to get 
W3C approval of press statements that invoke the W3C name, aren't 
they?  Once upon a time in the SVG WG, there was actually a bit of 
nastiness when one member was making un-approved (and to some other 
members, false and inflammatory) press statements mentioning SVG and 
W3C.  It was halted from doing so.

-Lofton.
Received on Friday, 26 October 2001 13:05:13 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:40:28 UTC