Are delimited files Structured or Unstructured?

[Adam Kall]

The material seems to imply that delimited files are structured, but on the practice exam there was a question asking which file types are considered unstructured and the only right answer was Delimited. Was that an error on the question? or are delimited files a weird hybrid of both structured and unstructured?

 

[Jarrel]

To what I understand, delimited files are semi-structured.
Examples of semi-structured data are email, HTML pages, XML files and zipped files.

On the other hand, examples of unstructured data are video, audio and images.

 

[Brian Ford]

Since the actual data field is defined by the delimiter there is no suggestion that the contents of the fields are structured the same way. The field defined by delimiters can be either null (two delimiters back to back) or as long as it takes to reach the next delimiter. Thus delimiters are a sign that the data is unstructured.

 

[Rick Butler]

All data, on some level, has "structure". Media files, blob data - there is all a measure of "structure" behind it. But as Brian alluded to - that structure in the case of a CSV is implied, because the data, as a CSV, is being presented in a raw form. Without getting too deep into semantics, delimited data is unstructured by definition, because, if you look at a CSV in notepad, for example, it's just characters - even if you account for the tabs/commas/etc that is used to provide "structure". Notepad doesn't look at the CSV data and see a "structure".

The only time data gains structure by definition is when that structure is applied by an external source. When we take that CSV and plug it into a framework, it gains some structure, making it semi-structured in this case. Though unstructured data has a native, internal structure that is based on the application that created it, there is no data model that organizes it in a predefined way; it is stored in its native format.

If that makes any difference in understanding...my humble 2¢.

/r