Kamen, we are talking about 2 things in here.

1. Satisfaction. I do not like this term because we are trying to deal with 
measurable things in SOA. 'Satisfaction' carries more connotation than 'need'. 
The latter represents concrete requirement. I think if a service meets 
requirements, its work is done well, we can control it. But we still do not 
know for sure that the customer is satisfied. This is simple semantic.

2. As of meta-data, I just tried to re-phrase (seems like with no success) 
Rob's explanation why an aggregation of data appears as a service in question - 
is it real SOA service or not. That is, we mostly agree that just retrieval a 
data with SQL expression from a database is not a service's but a data access 
component's work, it does not bring a RWE. At the same time, an aggregation of 
data via the same SQL, plus, aggregation business rules, brings concrete, new 
business value, i.e. the RWE.

This is not about a user and its interest in meta-data. It is about what we 
consider a service in SOA.

- Michael


----- Original Message ----
From: kaburov <[EMAIL PROTECTED]>
To: [email protected]
Sent: Sunday, June 29, 2008 9:30:25 AM
Subject: [service-orientated-architecture] RWE based on data aggregation (was 
ESB/Intermediary in SOA (was Data services (was Re: Definition of SOA)))


I see

I have seen in many cases that most users do not need metadata. 
Probably metadata can be compared with plan or base of a building. 
Most users do not need plan or base without other parts but they need 
the building as a whole

Michael why don't you agree that many people are satisfied when they 
have what they need? I think satisfaction and fulfilling the needs 
easy can merge into one another.

Probably you agree that fulfilling the needs of a customer, amount in 
bank account, number bottles of vodka for free can merge into one 
another. Because if client is satisfied or he has what he needs he 
will pay for it. And these money can buy vodka. 

I agree that there are frequent conflicts of interests between a 
person and organization, between a business unit and the enterprise 
it belongs to. But each company has something in common or common 
style of the company. This style for me is the user's way of 
thinking. It is better anthropologists, sociologists, business 
analysts or other similar specialists to outline this style and they 
to describe the common way of thinking in company. This description 
will show what is more important for company and case - so some users 
will need metadata other users will need the information( actual data) 
described by the meta-data... If the code is aligned to this way of 
thinking in the company most users are satisfied. 

Kamen

--- In service-orientated- architecture@ yahoogroups. com, Michael 
Poulin <[EMAIL PROTECTED] .> wrote:
>
> Kamen, it is Rob who is asking. I just propagated his question.
> 
> >  Michael probably you are asking about alignment of code to 
business.
> 
> No, the question is not about alignment; it is about what we can 
consider as a Real World Effect - only a new meta-data (Rob, please, 
correct me) 
> resulted from an actions or also new information (actual data) 
described by the old meta-data ( i.e. new values aggregated according 
to known rules or meta-data).
> 
> > Well I wrote several weeks ago that alignment to the way of 
thinking 
> > of users provides maximum customer satisfaction. In other words 
for 
> > user is most important to have the functionality in his 
intentions or 
> > way of thinking. 
> 
> Customer satisfaction is a *slicky* ground. A SOA service cannot be 
sure it satisfies a consumer because 1) consumer decides about it by 
itself; 2) satisfaction and fulfilling the needs are not the same 
things. 
> 
> Also, how does a service know if a customer need is objective or 
subjective? Don't we know about frequent conflicts of interests 
between a person and organization, between a business unit and the 
enterprise it belongs to? Who said that customer's way of thinking is 
right for its own organization (i.e. fits into organization business 
model or addresses real problems in that model)?
> 
> Believe it or not, but I lived in the country where customer 
satisfaction (in any business) was measured by a sum in hard currency 
on his/her bank account (desirably abroad) and/or by a number bottles 
of vodka, for free of cause. Did it relate to their business?  - No, 
not at all.
> 
> - Michael

> 
> 
> ----- Original Message ----
> From: kaburov <[EMAIL PROTECTED] >
> To: service-orientated- architecture@ yahoogroups. com
> Sent: Saturday, June 28, 2008 5:14:59 PM
> Subject: [service-orientated -architecture] RWE based on data 
aggregation (was ESB/Intermediary in SOA (was Data services (was Re: 
Definition of SOA)))
> 
> 
> --- In service-orientated- architecture@ yahoogroups. com, Michael 
> Poulin <m3poulin@ .> wrote:
> >
> > > 
> > Is the combining/aggregati ng/merging of data a RWE? Or is it the 
> > > 
> > actions resulting *after* that activity that are important? e.g. 
> > > 
> > direct mail campaign, identification of a correlation between 
data 
> > > 
> > points resulting in special promotions, etc.
> > 
> > Hey folks, anybody?
> > 
> > - Michael
> 
> Michael probably you are asking about alignment of code to 
business. 
> Well I wrote several weeks ago that alignment to the way of 
thinking 
> of users provides maximum customer satisfaction. In other words for 
> user is most important to have the functionality in his intentions 
or 
> way of thinking. So my opinion is that in cases when user expects 
to 
> get aggregation of data this is more important,when user expects to 
> have direct mail campaign this is more important, etc 
> 
> Kamen 
> 
> 
> > ----- Original Message ----
> > From: Rob Eamon <reamon@>
> > To: service-orientated- architecture@ yahoogroups. com
> > Sent: Friday, June 27, 2008 4:14:48 PM
> > Subject: [service-orientated -architecture] ESB/Intermediary in 
SOA 
> (was Data services (was Re: Definition of SOA))
> > 
> > 
> > --- In service-orientated- architecture@ yahoogroups. com, 
Michael 
> > Poulin <m3poulin@ .> wrote:
> > >
> > > Rob, I even capitalised the service action name. A "Client 
> Business 
> > > Data" or "Single Customer View" are not services, they are RWE 
of 
> > > the services.
> > 
> > Those describe the data or message submitted to or returned by a 
> call 
> > to an operation. A service may have operations that simply read 
or 
> > write the data, but there should be more operations in the 
service 
> > than just those (as we've already agreed, I think).
> > 
> > Is the combining/aggregati ng/merging of data a RWE? Or is it the 
> > actions resulting *after* that activity that are important? e.g. 
> > direct mail campaign, identification of a correlation between 
data 
> > points resulting in special promotions, etc.
> > 
> > In other words, is the "Client Business Data" or SCV the end 
point? 
> > Or are they really steps along the path to the "real" activity? 
How 
> > many business folks will say "Ah, now that I have the Client 
> Business 
> > Data, I'm done. My day is complete."?
> > 
> > IMO, noone would ever say that. That's because they intend to do 
> > something with that data after they get it. So we should focus on 
> > that "do something" and not only on the data retrieval.
> > 
> > > I agree with you that just a data retrieval with an execution 
of 
> an 
> > > SQL statement is not a service  - it is pure technical solution 
> > > dictated by particular type of the data storage.
> > 
> > +1
> > 
> > > BTW, do we consider an abstraction like a business data storage 
> > > being a business thing? 
> > 
> > Yes, it is at least implicit in the service defintions. How a 
> > business service stores or accesses its data is of no concern--
it' 
> s 
> > an implementation detail.
> > 
> > > It seems that existance of business data  model (which is very 
> > > important business thing becuase it carries 
> > > business information, knowledge, business treasure) w/o a 
> > > definition of the place where this information materialises is 
a 
> > > bit incomplete.. .
> > 
> > Yes, as I mentioned earlier in the thread, the model is indeed 
> > important. It materializes in the services and operations, 
> > represented by the data/messages/ events exchanged by the 
consumers 
> > and providers.
> > 
> > > At the same time, I think that an aggregation of business data 
> > > according to business rules from different data sources may be 
> > > accepted as a business service because it can provide new RWE 
> > > unavailable via simple data retrieval. What do you think?
> > 
> > Presumably the data aggregation is intended to serve some 
> actionable 
> > purpose, as I mention above. I think we agree that a service can 
do 
> > whatever it needs to perform its work, including aggregating data 
> > from various sources (be they other services, data abstraction 
> > layers, direct database accesses, etc.). 
> > 
> > -Rob
> >
>

    


      

Reply via email to