Politics is the art of the possible.
Otto Von Bismarck, remark, Aug. 11, 1867
German Prussian politician (1815 - 1898)
I have seen first hand the disastrous effects that can be had by the
belief that technology is the answer to a problem. (see also http://www.avorcor.com/morgenthal/index.php?entry=entry080510-181701)
Alex is somewhat correct in his statement, which is why I included the
quote from Otto Von Bismark (above). We technologists like to forego
the politics in favor of an elegant solution to the problem,
unfortunately, life is not always elegant. Hiring the "right"
architect or buying the "right" software package will never make up
for years of attention to the short term over the long term.
Enterprise integration projects fail because they need to be led by
diplomats, not engineers. Companies that see this need front their
projects with top notch project managers with elegant diplomacy backed
by top notch architects feeding them direction.
REST, WOA, SOA, WS, CORBA, DCOM, DCE....Facebook, Google...???
whatever you want to add on to the end of the list is relatively
irrelevant today. People chose to adopt certain technologies because
of the fit a need and because the learning curves had a relatively low
slope (e.g. Web, email, etc.) These technologies can also only handle
a limited albeit large subset of overall problems. They do not lend
themselves easily to multi-phasic transaction-oriented applications
(e.g. ERP). When they do, the solution will be complex, because the
problem domain is complex.
Simplification of complex problem domains into smaller, simpler
problems that can be understood and bought into by the business
community will succeed even if they are not as elegant.
REST can succeed where SOAP-based WS might fail because the community
that can understand them and use them is much larger. I loved the
mid-90's when CORBA was just coming into being and I was one of the
few that understood the technology and could charge $200+/hr for my
services. However, the world community had realized that for most
tasks, technologies like CORBA are overkill and the problems can be
solved by lighter weight technologies when broken down in small chunks
and, sometimes, by changing the requirements to fit the available
technology domain.
I know that last line is sacrilegious for many, but it's probably the
most important lesson I've learned in the past 10 years, and a very
painful one to learn at that.
Politics are the art of the possible and sometimes we have to change
the requirements to suit the possible means to getting a solution
implemented ... even if it's not the most elegant or agile or
extensible solution that we could have designed.
JP
__________________________________
JP Morgenthal
(703) 649-0829 : Voice mail
(703) 554-5301 : Cell
[EMAIL PROTECTED]
__________________________________
Confidential: The information in this e-mail message (including any
attachments) is intended only for the use of the recipient(s) named
above and as such is privileged and confidential. If you are not an
intended recipient of this message or an agent responsible for
delivering it to the intended recipient(s), be hereby notified that
you have received this message in error. Any review, dissemination,
distribution, printing or copying of this message is strictly
prohibited. If you believe you have received this message in error,
please notify the sender immediately by return e-mail and delete this
message from your system(s).
On May 24, 2008, at 6:16 PM, Alexander Johannesen wrote:
On Sat, May 24, 2008 at 9:25 PM, Steve Jones
<[EMAIL PROTECTED]> wrote:
> I find it interesting that back in 2003 most SOA groups talked about
> Web Services and technical implementation and now 5 years on we are
> talking about the harder problems of organisational and cultural
> change, while REST is back again at the way to build a better wheel.
Actually, I persist with REST exactly because it has a number of
things that makes it easier to adopt to more human problems. Resource
orientation being one (closer to the human perception of "things"),
the limited interface another (humans have limited interfaces, too :),
and finally the state of application through hyperlinks (which is
horrible to explain, terrible to spell out, but has a grounded human
essence in it; the angle of bias and scope directly there with the
"thing" itself).
To do SOA on a organizational / human level I find that since people
can grok the human aspects of REST I get a free shot at actual finding
solutions with them rather than trying so damn hard to separate SOA
and implementation, because, you know, people don't really think like
that. The world is complex, but we humans handle complexity through
adapting to patters, not by breaking it down into bits and
categorizing it.
> The on going problem of IT is that people continue to think that a
> technology will deliver a massive step change in the success,
> maintainability and evolution of IT estates. Don't you find it
> depressing? I do.
Oh, it's a bet one way or the other. Sometimes technology do bring the
goods, other times it does not. The thing is, in the end, it's not the
technology with the best technical design that wins, but the
technology that connects best with the people who will use it.
Alex
--
----------------------------------------------------------
Project Wrangler, SOA, Information Alchemist, UX, RESTafarian, Topic
Maps
------------------------------------------ http://shelter.nu/blog/
--------
