Anyone else “fried” by Nulls?

Just finished another lengthy debugging session, helping resolve a problem with null handling. Nulls can be a struggle at times in DataStage, but that is hardly the issue. It’s painful for every part of our industry. Do a search on “problems with nulls” — you’ll get hits with subjects on just about every database, product offering, language and package that exists. Five minutes reading and you’ll find lengthy diatribes on both sides of the fence (liking or disliking) regarding nulls and null handling. I remember in the early 80’s how “nulls” would simplify many types of processing, improve the quality of COUNTs and averages, and make lives simpler and applications more manageable. I wonder in hindsight how many billions of dollars in man-hours have been spent (wasted!) trying to correct the problems introduced by nulls in the past 25 years, the incompatibility of null handling among tools and processes, and the errors introduced by the disparity of skill sets and expertise that exists across the spectrum on this topic. I at least hope that I’ve been able to clear the smog surrounding nulls in my little world here on this blog (I have an entry on NULLs and their impact on XML)….

Ernie

Advertisements

4 Responses to “Anyone else “fried” by Nulls?”

  1. Stewart Says:

    Nulls = either Dark Matter or Anti Matter

  2. dsrealtime Says:

    Well said. Still, I’d like to go back in time with educated “grey matter”and suggest that we (IT overall) do a better job of supporting the concept or just skip it altogether. The collective problems and design issues introduced into legacy apps for dealing with data “not there” (or the consequences of ignoring it) would have cost less.

  3. Mark Says:

    What do you put in a middlename column in a database table for someone that doesnt have a middlename.

    null wrong with nulls ?

    • dsrealtime Says:

      That’s a good example…….and “if” I was doing an analysis of a population in my application, where I needed to do “COUNT”s on middle names and understand how many people had them vs didn’t have them, then nullibility (one way or the other) would be a useful feature. If I’m just printing addresses and keeping records? Is it in my way?


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: