Jerry has hit on a good point when he discusses criteria and the value of
the certification. Most IBM certifications simply require that a vendor
fill out some forms on the web and submit them for someone at IBM to
review and approve. This is how you become a member of IBM's
Partnerworld. This is also how you get your software listed in the IBM
software directory and it is also, pretty much, how you get the server
proven moniker.
In order for something like this to work, the certifying agency (IBM?
COMMON?) would need to set up a rather elaborate testing lab and then
actually perform an end to end evaluation test of every package submitted
for certification. That would command quite an investment and is IBM or
COMMON or Midrange.com or any other organization out there prepared to do
it the right way? You could easily just end up in the same boat as having
to choose between a certified solution and a non-certified solution and
have the exact same problem that you have when choosing between a CDP or a
non-CDP developer.
Its a nice idea, but I wonder how practical.
Rich Loeber
Kisco Information Systems
http://www.kisco.com
--------------------------------------------------------------------------
Jerry Adams wrote:
Lukas,
Doesn't this simply mean that they will run on the platform, OS? I doubt that this is what Don was talking about. Heck, we still have applications written in RPG II for the System/34 running on our i5 model 520 running v5r4.
My interpretation of Don's suggestion was that there be some way of
(a) documenting and revealing the functions/features of the package,
(b) how, perhaps, the application achieves the functions, though not anything that would reveal trade secrets, and
(c) how well the application runs on various models and releases (is it v6r1-compliant, for example).
Any certification would have to be objective and repeatable across applications by different vendors. I think that there would have to be different criteria for system utilities; maybe even breaking those down into categories (security, system management, etc.). Back office applications (A/P, G/L, etc.) would have to have another set for evaluation.
I think, personally, that certifications of any nature can be useful, complicated to define, and not an end but a step. Our CFO, for example, is very, very good; she has an accounting degree, but not a CPA. I know some d**n good programmers that do not (to my knowledge) hold a certification, and some CDP's that I wouldn't let IPL our i5.
Jerry C. Adams
IBM System i Programmer/Analyst
B&W Wholesale
office: 615-995-7024
email: jerry@xxxxxxxxxxxxxxx
-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Lukas Beeler
Sent: Wednesday, May 28, 2008 6:21 AM
To: Midrange Systems Technical Discussion
Subject: Re: Vendor/BP application/service/product certification...some thoughts...
On Wed, May 28, 2008 at 4:51 AM, Jim Franz <franz400@xxxxxxxxxxxx> wrote:
Perhaps this certification should not be any one vendor's responsibility. Do
other vendors (MS, Sun, etc) test and certify their partner's software?
Yes, they do. "Certified for Windows Vista", "Certified for Windows
Server", etc.
These certifications ensure a few very basic technical guidelines:
e.G:
* The application works with a standard user account
* The application uses proper MSI files for install/deinstall
But again, it's very basic. Don's proposal goes much further.
--
Read my blog at
http://projectdream.org
--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
To post a message email: MIDRANGE-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit:
http://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at
http://archive.midrange.com/midrange-l.
midrange.com
