Comentários (2)

1 IvanM comentou às Link permanente

So I have to confess that I've only skimmed part of this book - but the main editor also says that Bad Data is not just the stuff that you see in data profiling, but also:
- data that eats up your time, pull out your hair and scream in frustration (or something like that)
- data you can't access, had and lost, not the same today as it was yesterday.

This is his preference to his generic statement about Bad Data is data that gets in the way. But I have more reading to do in my copious free time ....

2 smithha comentou às Link permanente

I had a couple good examples of data that ate up my time just yesterday:
- one was a classic cut-and-paste issue where I replicated one small parameter string in an xml file for a data import and generating a rather undecipherable error message.
- the other was an xml file saved via Word which put a ton of extraneous formatting information into the xml file. In this case, the file looked very nice (tabbed, colored, etc.) in Word, but was unreadable by an API.
In both cases the xml files could be read just fine by various editors but they weren't fit for their purpose -- data cleansing, if you will, was required for them to work, so in these contexts both can be seen as Bad Data.

Incluir um Comentário Incluir um Comentário