I have had an amazing amount of commentary to get back to people on… At first, I was traveling and then busy with family stuff… then I looked and there were a lot of good (and some very long) discussions that had arisen. I am only part way through creating detailed answers for these but I thought I would get the partial results posted.
So many of these discussions were VERY long! I especially want to that John for his copious work in creating excellent discussion questions. In many cases, I decided to move the responses to the commentary track of this blog because they were SO long. John’s in particular were hard to give proper response without the trick of keeping his discussion and interleaving my responses. This gets pretty overwhelming after a while so I’m posting these to the side and giving you links to follow and a brief synopsis if you care.
I had a comment from Marcus on 3/18 (see Re: http://blogs.msdn.com/pathelland/archive/2004/03/18/91825.aspx#91911 )
He was asking me to address the apparent conflict between my post on “What’s in a Name?” and Ramkumar’s post. (see http://blogs.msdn.com/ramkoth/archive/2004/03/08/85802.aspx )
I responded that Ram and I really did agree but that I had screwed up the terminology in my post and that made it very confusing. Ram and I had previously used the terminology that Ram had used in his blog entry and I should have followed suit.
Moving forward, I will align with Ram’s usage.
Service — (same as a Foo) for the named endpoint for an interaction, and
Business-Service — (same as a Bar) for the collection of data, code, and Foos that is a disjoint set.
See http://blogs.msdn.com/pathelland/archive/2004/03/22/94000.aspx#107204 for the complete discussion.
I had a comment from JJ on 3/19 at http://blogs.msdn.com/pathelland/archive/2004/03/11/88058.aspx#92937
This is a discussion of the size of “bars” and how an enterprise app every gets to the point of having a disjoint set of data when, over time, you start seeing all the database tables in most enterprises updated and used by lots of app code that spans lots of tables… great discussion.
Next, we had the mother-of-all-comments from John on 3/23
Really, John was commenting on my comments on his comments of my post… I’m now continuing this effort. I’m warning you, this is a LONG post.
See: http://blogs.msdn.com/pathelland/archive/2004/03/22/94000.aspx#107212 for my latest contribution to this thread. My response to this one includes all of John’s commentary with my commentary nesting in between.
The big issue John raises is his belief that we need concurrency control (optimistic or pessimistic) to connect the services. I argue that this is how services are different than components. This is an in-depth discussion about the lack of shared transactions across service boundaries in SOA.
This one took me some energy to respond to… sorry for the delay.
Next, we had a post from John Cavnar-Johnson on 3/22. See http://blogs.msdn.com/pathelland/archive/2004/03/22/94000.aspx#95578
In this, JohnCJ raises concerns about the accuracy of relational data and the challenges in relating it to the real world. JohnCJ talks about business documents (as they flow around in messages) and their relation to back-end relational data. Great discussion!
My comments can be found at http://blogs.msdn.com/pathelland/archive/2004/03/22/94000.aspx#107215
Next, I found another LONG commentary from John on 3/25.
I responded with http://blogs.msdn.com/pathelland/archive/2004/03/22/94000.aspx#107227
(again, to properly respond to this long comment, I included John’s comments and nested my discussion in the midst of it).
This one gets into some fun discussion of types and schema in addition to the locking and transaction issues.
I am still trying to write responses to:
http://blogs.msdn.com/pathelland/archive/2004/03/22/94000.aspx#104760 posted by John on 3/31 – this gets into synchrony and asynchrony of service interaction.
http://blogs.msdn.com/pathelland/archive/2004/03/22/94000.aspx#104960 posted by John (winner of the “most voluminous post award”) also on 3/31 business documents in response to JohnCJ’s post.
http://blogs.msdn.com/pathelland/archive/2004/03/22/94000.aspx#105665 posted by Bart Elia on 4/1 on business function and messaging.
I’m sorry I can’t get responses to every comment submitted written… I will try to keep up.