<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>Supple software comments on Indecision and attachment</title>
    <link>http://supplesoftware.com/</link>
    <language>en-us</language>
    <ttl>40</ttl>
    <description>Supple software comments</description>
    <item>
      <title>"Indecision and attachment" by petrovg</title>
      <description>&lt;p&gt;With the multitude of options that modern software technology allows, indecision seems to get more and more widespread. I&amp;#8217;ve felt like this myself not once or twice - completely paralised by the inability to make a design decision. It could be about the simplest thing - how to name a variable or a more complex one like do we go for an EJB or a web service or just good old plain objects.&lt;/p&gt;

&lt;p&gt;People can spend days on that. They&amp;#8217;ll organise meetings, call consultants, the consultants will then argue with one another and the simple decision will become more and more difficult and political.&lt;/p&gt;

&lt;p&gt;A few years ago when entity beans were still in fashion, I knew people who stopped speaking to one another because of a CMP/BMP disagreement. If the developer had just gone for one of them and not made a big fuss of it, that might have been one slightly easier environment to work in.&lt;/p&gt;

&lt;p&gt;As with many problems, this one does not come alone - the longer you invest in a decision, the more people you involve, the longer you argue one way or another, the more attached you become to the final decision. Meaning, you won&amp;#8217;t change it even if it&amp;#8217;s obviously wrong. Some kind of love grows between the proponent and the idea he was not sure about just a minute ago, the kind of love that just won&amp;#8217;t yield to any reason or common sense.&lt;/p&gt;

&lt;p&gt;It need not be that way. People just have to realise that their decisions do not actually have such a big impact as they fear (or may be secretly hope). Say the argument - EJB or no EJB. I have spent hours in meetings over this. What if we just chose one at random? How long does it take one experienced developer to wrap an interface in an EJB? A couple of hours at most. To rip out an EJB when it seems unnecesary? - about the same. And note - you will know then, because you will have tried it&amp;#8230;&lt;/p&gt;

&lt;p&gt;But this is valid at much lower level too - every individual developer has tonnes of decisions to make every minute - Singleton or not? Who the hell cares?? You switch between the two in minutes. Shall I split the line here or not? Shall I keep my lines under 80 characters long in case they ever get printed? Do I import whole packages or individual classes? First carriage or last (train)? Arrrrghh!!! The modern tools can format and reformat and refactor all these faster that you can ask them. Just go fo one please.&lt;/p&gt;

&lt;p&gt;It will soon become obvious if got it wrong. And then you won&amp;#8217;t care - you&amp;#8217;ll just change it.  What&amp;#8217;s even better, you&amp;#8217;ll soon learn to make your software in a way that&amp;#8217;s easy to change.&lt;/p&gt;

</description>
      <pubDate>Sat, 11 Mar 2006 00:43:00 PST</pubDate>
      <guid>&lt;a href="/articles/2006/03/11/indecision-and-attachment"&gt;Indecision and attachment&lt;/a&gt;</guid>
      <link>&lt;a href="/articles/2006/03/11/indecision-and-attachment"&gt;Indecision and attachment&lt;/a&gt;</link>
    </item>
  </channel>
</rss>
