Loading ...
Sorry, an error occurred while loading the content.

Re: [XP] Re: Software Agility: Seven Design Concerns

Expand Messages
  • Tim Ottinger
    ... Just for the record, my advice to people is that the right time to use a design pattern in code is when _not_ using it makes the code _even_ _worse_. I
    Message 1 of 84 , Apr 3, 2012
    • 0 Attachment
      >From: MarvinToll.com <MarvinToll@...>
      >To: extremeprogramming@yahoogroups.com 
      >The concern is communicating patterns outside the immediate project team and using class naming as evidence of pattern consideration.



      Just for the record, my advice to people is that the right time to use a design pattern in code is when _not_ using it makes the code _even_ _worse_. I don't reach for them as a tool of global communication, because I find that nothing can screw up a code base as much as using too many patterns. I don't know that I can claim a majority in this, but I know that there are a great many others who have told me that they now share my opinion after years of overbuilding.  Maybe the four rules of simple design are really the thing you're looking for, perhaps augmented by the seven code virtues I listed already in separate email (working, unique, simple, clear, easy, developed, brief).

      http://agileinaflash.blogspot.com/2009/02/simple-design.html
      http://pragprog.com/magazines/2011-08/how-virtuous-is-your-code

      A coherent model ("metaphor") is a different story, because it's not a conglomeration of patterns handed down to hapless programmers who must implement the architectural vision. 
    • Lior Friedman
      ... at the risk of opening a tangent thread, I would say not really. Given the right tools,know how and will, (almost) any code can be tested. I believe that
      Message 84 of 84 , Apr 9, 2012
      • 0 Attachment
        On Mon, Apr 9, 2012 at 3:47 PM, Adrian Howard <adrianh@...>wrote:

        > **
        > Can we usefully say things about degrees of testability?
        >

        at the risk of opening a tangent thread, I would say not really.
        Given the right tools,know how and will, (almost) any code can be tested.

        I believe that the term "tesability" is many time used to describe the
        "ease of writing tests" or as a proxy to "properly designed" code. So I
        find it more useful to talk about what makes good code.

        Lior


        [Non-text portions of this message have been removed]
      Your message has been successfully submitted and would be delivered to recipients shortly.