avalon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Pete Carapetyan <p...@datafundamentals.com>
Subject Re: [RT] Avalon == Jakarta Architecture Labs (was: RE: avalon is not a risk dependency)
Date Sat, 22 Jun 2002 14:45:23 GMT
Leo Sutic wrote:

>Instead of trying to be the "do everything", the ultimate
>dependency, the framework for *all* of Jakarta, we would
>release framework 4.0, Excalibur 4.0, and then not consider
>A5 an upgrade, but something entirely new. The A4 architecture
>would still be 100% supported not as a deprecated product,
>but as a 100% alive and kicking architecture.
>
 From the outside, this looks very logical. May I submit Spock's name 
for the role of King Arthur?

May I also propose that of all the attributes being discussed, the one 
that is perhaps most important and not even mentioned yet is KISS ? 
*KeepItSimpleStupid* translates to elegance even if elegance is really a 
crappy word to use. But in a functional Spock like use of the word, 
elegance does the trick. Personally I think that is what Stefano and 
Berin have been talking about, without using the word, but instead in 
terms of how to get there. And, except for the swordfighting and 
container madness, it is my personal observation only that Avalon is 
sooooo close to already being very elegant, hence it's appeal.

Leo, you are dead on the money as far as issues of dependency are 
involved. Which is why Avalon, though amazingly successful as a 
technology, is still only touching 00.25% of the market which could 
really be helped by using it.

For the rest of the 99.75% who are not prone to swashbuckling and 
pursuit of perfection, consistency, elegance, simplicity and 
trustability all tend to be rather lumped together in a single unspoken 
feeling that one gets when one takes a look at jakarta.apache.org/avalon 
(or anything else) for the first time.. You either get that feeling in 
the first ten seconds or not. If you get it, you keep looking until you 
lose it, or until Avalon is implemented in your code base, whichever 
comes first.

Another useful saying floating around  - *Perfection is the enemy of 
excellence*.

Before King Arthur there was tribalism - and the resources of the 
country were consumed in infighting between tribes. What made King 
Arthur nifty-keeno was that he unified the country so that the resources 
were focused on the welfare of the country and it's citizens rather than 
infighting between the tribes. This is very similar to OO, where 
packages are the tribes. What Avalon does is IoC the strength of the 
packages back up to the country where they are most useful.

 It wasn't about King Arthur, and it wasn't about the Round Table. They 
just helped facilitate the move. And so it must be with Avalon. Keep It 
Simple and get out of the way. The country and it's citizens are the 
objectives, not the court, and not the process of gaining IoC, however 
critical.

What made King Arthur cool was that he focused on getting the job done, 
not on king-ing around and swashbuckling and building summer palaces and 
stuff. He had a very simple set up. There was a court and a table and a 
few guys hanging around to help with the IoC. Other than that, quite a 
modest affair, hence it's allure. Keep It Simple, personified.


--
To unsubscribe, e-mail:   <mailto:avalon-dev-unsubscribe@jakarta.apache.org>
For additional commands, e-mail: <mailto:avalon-dev-help@jakarta.apache.org>


Mime
View raw message