I read the text of Udi Dahan in the section [Separation of command request and SOA] [1]. Thinking about how I will use this in practice in the system in which I am currently working, some questions arise ...
1.
Consider the following situation when I have a WPF client application that allows a user to edit a list of entries:
The client application starts. At startup, he subscribes to a team service that will process and post changes to the records and requests the WCF service for a complete list of records. After receiving the list of records, any changes (by other clients) that could be published to the client queue when they were busy, waiting for a response from the WCF, are combined into a list.
Now it seems to me (perhaps a very small) chance that the client signed up immediately after the team responsible for the change was processed, and that the WCS service request was launched just before the same change was made to the database, In this situation, I receive incorrect data on my client that is not synchronized with the database (and will not come across) (if the client application does not restart). This problem really exists, and if so, how should it be handled?
2. My second question: how to create / implement a user interface:
; , ok: .
( ),
.
( , , - )
- , ok, , .
- , ...? ? ? ?
- , , , - :
- -. , , ? ( " , ..." ), ? , ? , , - (, ), .
, , , , . , -, . , -, , , , , , 1 . , , - , , , .
?;)
, , . , , , , , ?
.
[1]: http://www.udidahan.com/2008/08/11/command-query-separation-and-soa/ " SOA"