I took the precaution to say “Not only Logic” so I would not have to reach back and invent a soothing explanation for saying “NoLogic.”
The marketing reasons for parroting “NoSQL” are obvious and I won’t belabor them here.
There are some less obvious reasons for saying “NoLogic.”
Logic, as in formal logic (description logic for example), is rarely used by human user. Examples mainly exist in textbooks and journal articles. And of late, in semantic web proposals.
Ask anyone in your office to report the number of times they used formal logic to make a decision in the last week. We both know the most likely answer, by a very large margin.
But we rely upon searches everyday that are based upon the use of digital logic.
Searches that are quite useful in assisting non-logical users but we limit ourselves in refining those search results. By more logic. Which we don’t use ourselves.
Isn’t that odd?
Or take the “curse of dimensionality.” Viewed from the perspective of data mining, Baeza-Yates & Ribeiro-Neto point out that “…a large feature space might render document classifiers impractical.” p.320
Those are features that can be identified with the document.
What of the dimensions of a user who is a former lawyer, theology student, markup editor, Ancient Near Easter amateur, etc., all of which have an impact on how they view any particular document and its relevance to a search result? Or to make connections to another document?
Some of those dimensions would be shared by other users, some would not.
But in either case, human users are untroubled by the “curse of dimensionality.” In part I would suggest because “NoLogic” comes easy for the human user. We may not be able to articulate all the dimensions, but we are likely to pick results similar users will find useful.
We should not forgo logic, either as digital logic or formal reasoning systems, when those assist us.
We should be mindful that logic does not represent all views of the world.
In other words, not only logic (NoLogic).