Happy New Year to all!
To Bob's comments "
second, this section reeks of ecosystem SOA testing only" - in reality, definition of SOA ecosystem is fuzzy. In my practical work, I have found that SOA ecosystem includes EVERYTHING that provides connectivity to the service interface and EVERYTHING that situates BEHIND the interface. For example, if configuration of 'number of concurrent database connections' is changed - you must re-test all services working with that database, directly or indirectly (if a business service uses a data access service and data access is changed due to re-configuration, your business service might need to reconsider and engage another data access service because the old one does not meet SLA any more). If a policy applied to the service execution is changed, you must re-test all services working with that policy because business functionality and RWE may be affected by that change.
Yes, it is constant and END-to-END re-testing for all IT elements involved in the service interactions (connections and execution). From practical side (not for RA), I know only one tool, which can do such work - it is iTKO's LISA testing suite.
- Michael
----- Original Message -----
From: "Ken Laskey"
To: "Ellinger, Robert"
Cc: "soa-rm-ra@lists.oasis-open.org RA"
Subject: [soa-rm-ra] Re: first complete draft of SOA testing
Date: Tue, 6 Jan 2009 09:11:38 -0500
Bob,
Happy New Years back atcha.
Thanks for your quick comments. Here are some thoughts:
re mockup, I understand your concern and have some of it myself.
Traditionally, mockups were for visual effects and less functional,
although I believe car mockups are used for wind tunnel testing.
I have seen use of mockup in service discussions as something more
on the functional end. This may be a case of evolving terminology
or just sloppy use. It deserves more attention.
The write-up is ecosystem heavy, and I may be overreacting to a lot
of the SOA literature that is sealed in the enterprise like my
swimming pool of sea water. I want an ecosystem emphasis but it
must also be relevant to the enterprise. That will get attention
too.
On the question of continuous testing vs. waterfall, I'll need to
reread because I thought I emphasized this as a major point. It
was certainly intended.
I'll let this sit a couple days, get a few more comments and then
start to revise.
Thanks again,
Ken
On Jan 6, 2009, at 7:48 AM, Ellinger, Robert wrote:
> Ken:
>
> From a quick read I have three concerns. But, first Happy New Year, and
> this is a good start. Now:
>
> First, is your use of the term mock up. In aircraft, cars, and
> industrial products, mockups are used as part of an early evaluation of
> the conceptual design of a product and does not have great fidelity with
> the actual product. And typically, mockups are of the product not the
> context. If I remember correctly, in modeling, a simulation of the
> "test" environment is termed the simulandum (sp?). The term mockup as
> you use it in this section, "jars my eye teeth."
>
> Second, this section reeks of ecosystem SOA testing only. While that is
> unique and new, I suspect that there will be many more business or
> organizational critical enterprise SOAs initially, than ecosystem SOAs
> and there is plenty of testing of services with respect to the business
> processes that will need to be performed, which is also unique and new.
>
> Third, this section implies a waterfall development process. I think
> that in service/composite application lifecycle of the future, testing
> will continue through the entire lifecycle, not just during
> implementation. This didn't jump out at me, though it may be buried in
> the text.
>
> Just my two cents to try to help.
>
> Bob
>
> -----Original Message-----
> From: Ken Laskey [mailto:klaskey@mitre.org]
> Sent: Monday, January 05, 2009 11:21 PM
> To: Francis McCabe; Jeff A Estefan; dellis@sandia.gov Ellis; Danny
> Thornton; Ellinger, Robert; Rex Brooks
> Subject: first complete draft of SOA testing
>
> OK, folks, here it is.
>
> I'm not completely satisfied with the organization and I'll need some
> help in defining models, but I think I have covered most of the
> territory I wanted. I took into account comments from Frank, Rex, and
> Bob, and I believe your comments and concerns have been addressed to
> what I felt was a consistent level of detail appropriate for the RA.
> (There are several mentions of further detail being out of scope.)
>
> I'm sure you'll all let me know if I missed something.
>
> Ken
>
-----------------------------------------------------------------------------
Ken Laskey
MITRE Corporation, M/S H305 phone: 703-983-7934
7515 Colshire Drive fax: 703-983-1379
McLean VA 22102-7508
---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that
generates this mail. Follow this link to all your TCs in OASIS at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php