Re: SOA and its real state of the art

On 28 Sep 2006 01:08:47 -0700, "Luca" <nioskidol@xxxxxxxx> wrote:

You're right, Andy
what I mean is that the real "robustness" of SOA
using the actual technologies (Java, .NET, XML, etc. etc.)
seems yet to be proved.

For example, seems like that testing of a simple application made of 2
or 3 SOA services must be very careful because of error detecting
difficulties and errors permutations.

In short,
do you know of any SOA-based mission critical application
really used by a big company/organization ?

Its swings and roundabouts really, I think that if one is looking at a
corporate mission critical application, then one is really dealing
with what would be termed a 'back office' type application and they
tend to be heavy in data crunching and high in uptime requirements.

For a corporate type applications, actually writing the software
really should be a minimal part of the overall project effort. I would
expect the testing to be a much larger part of that effort with a
focus heavily on quality assurance. With good quality process one can
minimalise the chance of producing defective product, so the
architecture itself probably is a minimal contributor (unless the
toolset is rather flaky I guess).

My mind is not really made up at the moment if I think its a good
technology or not. If I was writing a CRM application for a telco,
I'm not sure if it would be a good idea, but if I was developing a
version of that same application to be run by that telco's resellers,
I could certainly see merits in using SOA.

One thing I do like is the ability to be able to put the name of a
function and some parameter data in an xml document and fire it off to
a server for validation and execution with another xml document being
returned containing the result set. To me its not as robust or as
well defined as CORBA is (especially in the telco field), but I can
see benefits. At the end of the day, there isnt that much difference
between writing IDL or writing WSDL.

Be interesting to see what others think.

Andy

.

Relevant Pages

  • Re: Writing technical text
    ... then go spend a few months writing Ant scripts or something ... > can someone tell me what kind of XML we're talking about here? ... > end up writing that format in place of LaTeX...? ... and then writing an improvised tool later to convert ...
    (Debian-User)
  • Re: ultra newbie question (dont laugh)
    ... storing employee information in an XML file. ... but not as fun as writing the logic. ... On reason to use OOP is to insulate your main code from things like storage ... from CSV file. ...
    (comp.lang.python)
  • Re: Obsessive Compulsive XMLing
    ... > I'm still having a hard time seeing the problem he is referring to. ... Writing an Ant script from scratch is not really hard. ... could be somewhat harder is writing XML with stricter DTDs like ... Now, designing Swing GUI apps without a visual designer, that is ebil. ...
    (comp.lang.java.programmer)
  • Re: Gallery Pagination Result Problem
    ... I use Altova XML Spy Home Edition when writing simple XML scripts, and it has the added benefit of validating the XML and pointing out errors against the DTD. ... Is there such a free program for writing PHP that has a good debugging mode or validation? ... And if I'm not using a debugger, I've found strategically placed echo statements do wonders for debugging. ...
    (comp.lang.php)
  • Re: Should I use XML as a database for a standalone app?
    ... PDA, webservices, and more stuff. ... I had the idea of exporting the necessary data from the main database as ... XML files, using these as a mini database and the writing a procedure to ...
    (microsoft.public.dotnet.languages.vb)