Monday, February 21, 2005
« awk | Main | Serialization of events in .NET 1.x and ... »

A reader commented on my previous post about patternshare.org:

 

The content on patternshare.org is weak. I don't think the content will ever be as strong as the books that the Authors are trying hock.

 

It seems like every good design choice these days are being labeled a "Pattern". Quite a shame.

 

Don't misunderstand... I'm a huge pattern evangilist.

 

We must keep in mind that patternshare.org can not replicate the patterns from the books. That would violate copyright law. The goal of patternshare.org is to be an index, to make it easier for you to figure out which books to buy and/or where in those books to find interesting patterns. That is an admirable goal and one that I think patternshare.org can accomplish.

 

In 50 years (or whenever it is that copyrights run out) we can put all the book content online and then we won’t have “weak” content. But up to that point I’m afraid we’re kind of stuck with the reality that the patterns are in books, and the books are copyrighted and that’s that…

 

Regarding the comment that "everything is being labeled a pattern" I agree. It is the current over-hyped trend in the architecture/design space - competing only with SOA.

 

But I think that the current pragmatic value of patterns is to provide an abstract language we humans can use to discuss our software designs. A language better than we have without “patterns”. Regardless of whether some of these "patterns" are really patterns or not, the fact is that the collective effort of all these books and articles is providing us with that common language - and thus is allowing us to communicate at a higher level than was possible even 3-5 years ago.

Monday, February 21, 2005 6:03:57 PM (Central Standard Time, UTC-06:00)  #    Disclaimer  |  Comments [7]  | 

Monday, February 21, 2005 7:38:15 PM (Central Standard Time, UTC-06:00)
Sadly, as a Project Gutenberg volunteer. I can tell you that copyrights now essentially never run out. They are set at 75 years from the death of the author. So generally books written today will be off copyright sometime in the 22nd century.

Bryan

Bryan Kramer
Monday, February 21, 2005 8:58:46 PM (Central Standard Time, UTC-06:00)
Cool! At least my grandkids will have full access to these ancient "patterns". I'll encourage them to be techno-archaeologists so they have some interest in our quaint findings. 'Cause I gotta say that given the amount of change during my lifetime thus far, in 75 years there's no one who'll really care a whole lot about Java, .NET, C++ or any of the things we're so wrapped up in today.
Tuesday, February 22, 2005 9:45:25 AM (Central Standard Time, UTC-06:00)
right on Rock - I agree totally with your "common language" point completely. You see this problem at every level of the software development lifecycle. Effective communication is often the biggest hurdle in any problem domain.

...which is why I like the term "software nomenclature" to describe what you're talking about. It's an effective term that immediately draws the attention of the listener and avoid any potential confusion with... verbal languages, programming languages, etc. etc. etc.


-my $0.02

-kb
Tuesday, February 22, 2005 4:51:17 PM (Central Standard Time, UTC-06:00)
Speaking of patterns and the books they are presented in...

I recently bought a book on patterns which seemed at first like it would be a good read.
I'm about half way through it and I'm hating it. Sure I'm getting the patterns out of it but they constantly use these inane analogies like pizza stores to represent OO designs (complete with topping decorators). It drives me crazy. How am I supposed to translate that into the realistic business objects I'm using. Painfully that's how. I understand the concept of using simple to understand everyday objects so people can grasp things but it gets a bit rediculous.

Anyways point is. Any suggestions for good pattern books Rocky?
tory@syndacon.com (Telarian)
Tuesday, February 22, 2005 5:00:04 PM (Central Standard Time, UTC-06:00)
hmm... I see RSS Bandit has some reply posting issues when it comes to identity.
Tory Netherton
Friday, February 25, 2005 9:18:23 AM (Central Standard Time, UTC-06:00)
Tory, I totally agree! That is my number one complaint with a lot of books. I read a book on C# where one of the 15 authors (but that's a whole other issue) used a "jumper" as the example object. (that's a sweater to the non-brits among us) Lame. I don't even remember the concept they were trying to communicate. I think it was to do with inheritance and polymorphism. Lame. Whenever someone asks for feedback about how to improve their books or help docs the only suggestion I have is REAL WORLD EXAMPLES!.

At least Rocky uses a video store. The Video Store example becoming a bit of a cliche these days but I think that's only because it is such a readily understandable, real world scenario. Almost everyone has rented movies and is intimately familiar with the transactions involved. Plus, conceptually, it scales well. It deals with inventory that can be rented and with product that is for outright sale. You could substitute lawnmowers or cars or Karaoke machines for videos and not even have to change much. You can rent em and you can buy em.

Sweet.
Tuesday, December 06, 2005 12:07:34 AM (Central Standard Time, UTC-06:00)
find the best payday site here
Comments are closed.