Weblogs Forum - Reference Architecture

Articles |
News |
Weblogs |
Books |
Forums

Artima Forums | Articles | Weblogs | Java Answers | News

Sponsored Link •

Weblogs Forum
Reference Architecture

20 replies on 2 pages. Most recent reply: Jun 4, 2007 8:31 AM by nick longinow

Welcome Guest
Sign In

Back to Topic List

Reply to this Topic

Search Forum

Threaded View


Previous Topic		Next Topic

Flat View: This topic has 20 replies on 2 pages [ « | 1 2 ]

J. B. Rainsberger

Posts: 12
Nickname: jbrains
Registered: Jan, 2004

Corporate standards

Posted: Jan 6, 2004 2:33 PM

My comment is 100% anecdotal evidence. Tough.

I have not yet experienced a case in which any corporate-wide standard has resulted in positive ROI (including cost to implement and inform about the standard) due to fungibility of programmers. The net-positive-performing programmers I have worked with are able to switch from project to project effectively, irrespective of whether those projects follow any corporate standards; and conversely, the net-negative-performing programmers perform net-negatively in spite of the corporate standards that are meant to help them.

I have, however, experienced a number of cases -- say at least five -- where corporate-wide standards resulted in additional cost (both at the time and in hindsight) due to not being able to implement a simpler solution.

They call it the lowest common denominator for a reason....

J. B. Rainsberger

Posts: 12
Nickname: jbrains
Registered: Jan, 2004

Re: Reference Architecture

Posted: Jan 6, 2004 2:35 PM

> Maybe reference architecture are more useful in
> organizations that have a lot of less-experienced/talented
> developers that cannot be trusted to make design
> decisions--in other words, managing for the lowest common
> denominator.

If you don't trust them to make decisions, then don't hire them. They will be net-negative performers. More net-negative performers just means a larger loss.

Steve Ropa

Posts: 3
Nickname: steveropa
Registered: Jan, 2004

Reference Architecture

Posted: Jan 9, 2004 3:26 PM

> Excellent, now if only Management types would read this!!!!

Who says we don't! :-)

Steve Ropa

Posts: 3
Nickname: steveropa
Registered: Jan, 2004

Re: Reference Architecture

Posted: Jan 9, 2004 3:30 PM

> > Maybe reference architecture are more useful in
> > organizations that have a lot of
> less-experienced/talented
> > developers that cannot be trusted to make design
> > decisions--in other words, managing for the lowest
> common
> > denominator.
>
> If you don't trust them to make decisions, then don't hire
> them. They will be net-negative performers. More
> net-negative performers just means a larger loss.

I have also always thought that we have a responsibility to the less-experienced developers to impart the benefits of our experience. IOW, if we have already learned that there is a set of disciplines that work better than forcing an RA, we shouldn't cheat them by giving them an RA. They will remain less-experienced\talented developers.

MIchael Dean

Posts: 1
Nickname: twogun
Registered: Aug, 2004

Re: Reference Architecture

Posted: Aug 5, 2004 11:42 AM

We are in the fourth year of a multi-year flagship system replace project. All designed, architected and produced by a enormous, well-known entity. The RA in use is: PanDOORA...

I'm just wondering if anyone currently has, in production, a PanDOORA-architected system and any feedback from the software developer/maintainer?

As an OO developer, on the face, the RA seems bloated and more of a benefit to the system builders (concurrent development) than the system maintainers/users.

Also, what are some of the other RA on scale with PanDOORA?

Any and all feedback is appreciated cuz' we haven't gotten Drop 1 yet...not sure what to expect..

nick longinow

Posts: 1
Nickname: jakeslumba
Registered: Jun, 2007

Re: Reference Architecture

Posted: Jun 4, 2007 8:31 AM

I realize this post is very old now, but wanted to comment in favor of RA. Their greatest, but not only, value is for large companies doing infrastructure-related projects, who need to coordinate short-term developers across different geographical regions. Thus, the "known-quantity" benefit of known-performance profile and deterministic learning curve, has value to schedule predicability. Further, by employing insightful design patterns (ie, go4) you get the ability to evolve over time. RA is being used with proven cost-effectiveness in many companies; in the right application its the right answer.

Flat View: This topic has 20 replies on 2 pages [ « | 1 2 ]

Previous Topic

Next Topic


	Web Artima.com