[Wolves] Certified Open - a potential certification process for Europe?

[Shane M. Coughlan]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

Some business connected to FOSS in Ireland are seriously considering a
certification framework called Certified Open.  The interest in this
matter is mainly centred around the Open Ireland group, though this is
not exclusively their domain.

Certified Open is a framework designed to allow people to purchase ICT
solutions with the confidence of knowing they'll avoid vendor lock-in
and transitions to alternative platforms will be possible.

Certified Open originates in the UK under the auspices of the Open
Source Academy.  It is very new, currently unfinished, and is subject to
review.  The final working of the framework should be ready sometime
around September.

Certified Open is quite vague now, but the current content is a cause
for concern because of misapplications of language and concepts.  The
main document that I have briefly reviewed is the 'Certified Open
Product Framework.'  This document is clearly intended to provide an
assessment method for judging products and services according to Open
Source Initiative definitions, though it's falling short of that now.

Please find some initial thoughts on the Certified Open Product
Framework below:



Section 'Client View: Operating System'

"The product should be tested to demonstrate that it executes
effectively on both Microsoft Windows and at least one Linux desk-top."

Windows is not the only propriety operative system and Linux is not the
only open operating system.  FreeBSD, NetBSD and OpenBSD all represent
strong open alternatives.  MacOS X, QNC and Unix all represent strong
propriety alternatives.

Rewording suggestion:

"The product should be tested to demonstrate that it executes on at
least three operating systems, of which two must be compatible with the
Open Source Definiton published by OSI."



Section 'Client View: Office System'

"The product should be tested to demonstrate that it executes
effectively on both Microsoft Office and at least one out of Star Office
and Open Office, without relying on non-proprietary extensions."

Microsoft Office is neither open nor compliant with the ISO standard for
office documentation (ODF).  In addition, this statement suggests that a
product that runs on two commercial systems (MS Office and StarOffice)
constitutes an open product.  This is incorrect and misleading.

There are other propriety and open office applications like KOffice,
Abiword and WordPerfect Office.  To discount them is potentially serious
in light of the EU anti-trust proceedings against Microsoft.  It could
be understand as deceptive misrepresentation of what constitutes openness.

Rewording suggestion:

"The product should be tested to demonstrate that it executes on at
least two office suites, of which one must be compatible with the Open
Source Definiton published by OSI."



Section 'Application View: Operating System'

"The product should be tested to demonstrate that it executes
effectively on at least one Linux Server distribution together with at
least two other Serv OS's."

Again a bias towards a named product.  It's not necessary, and it's not
fair to other projects.  Also, why two server operating systems that may
be propriety?

Rewording suggestion #1:

"The product should be tested to demonstrate that it executes
effectively on at least three server operating systems, one of which
must be compatible with the Open Source Definition published by OSI."

Rewording suggestion #2:

"The product should be tested to demonstrate that it executes
effectively on at least three server operating systems, two of which
must be compatible with the Open Source Definition published by OSI."



Section 'Database View: Operating System'

"The product should be tested to demonstrate that it executes
effectively on at least one Linux Server distribution together with at
least two other Serv OS's."

As above.  Unnecessary bias towards a named system.

Rewording suggestion #1:

"The product should be tested to demonstrate that it executes
effectively on at least three server operating systems, one of which
must be compatible with the Open Source Definition published by OSI."

Rewording suggestion #2:

"The product should be tested to demonstrate that it executes
effectively on at least three server operating systems, two of which
must be compatible with the Open Source Definition published by OSI."



==

Any certification programme must be clear, well-informed and without
bias towards particular vendors or projects.  If the certification
programme chosen is weak it will be undermined by the proprietary
competition, it will be rejected by the FOSS community at large and it
may create issues regarding legal liability.

Certified Open has the potential to become a pan-European certification
programme because it's being promoted by Open Forum Europe
(http://www.openforumeurope.org).  However, the issues with its
definitions of what constitutes openness need to be reviewed.

I suggest engagement on this matter.  The certification programme is
going ahead and it's under review.  I believe that the director of
Certified Open (abellinger at iitt.org.uk) and people in Open Forum Europe
(bob at openforumeurope.org) need to be told where the product framework is
not working.

For a complete copy of the product framework and for more information on
Certified Open please visit http://www.opensourceacademy.gov.uk/.
Failing that, please email me directly at shane at opendawn.com and I'll
send you a copy of the products framework.

If any really smart people out there can review the frameworks and
reword for Free Software it would make my day :)  To be fair, they are
not aiming for Free Software.  This is an open source thing.  Therefore,
it'd be nice at least to have documents that work a lot better with OSI
open source concepts...

Regards

Shane

- --
Shane Martin Coughlan
e: shane at opendawn.com
m: +447773180107 (UK) +353862262570 (Ire)
w: www.opendawn.com
- ---
OpenPGP: http://www.opendawn.com/shane/publickey.asc
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)

iQCVAwUBRKBLUtwG3M95JPpzAQg5NAP8Dd2UkXnAAyayX8BWUhBaL9T0ubTegFkc
g0qJ5Ae31JXJ4/Bqa7lJZrsKqSp3sx/z2fegH+oa/kX/bYyMKc3c7RtBbhneqYFk
mrTw/D1znJ8Yz9W1ti3tGwA6br57Je9sBYs4wPYD5sgL1n9XtTSDkvsbcWWypTry
8Sfc9u6cOy4=
=8rsU
-----END PGP SIGNATURE-----