Is there a reason to choose OWL over OWL2?

Legacy questions from answers.semanticweb.com
#1

I wonder if there are any reasons to choose OWL(1) over OWL2. I understand that certain products may not support OWL2 and that one might argue that OWL(1) has potential incompatibilities with OWL2 (if one already has an existing OWL(1) workflow. But are there any further reasons?

#2

I wonder if there are any reasons to choose OWL (1) over OWL 2

  1. Tool support: many legacy tools only support OWL 1 constructs. (OWL 2 tools can cope better with OWL 1 than vice-versa)
  2. Complexity: if you're a DLophile, OWL 2 DL has an exponentially worse worst-case complexity than its predecessor (if property hierarchies are unbounded). If you're a rules kinda guy, OWL 2 has many new features that requires more rules, and more overhead.
  3. Familiarity: some folks that use your stuff might only be familiar with OWL 1 and not OWL 2.
  4. OWL 1 is sufficient: trivial, but you might not need to use anything new from OWL 2.

Now, ignore all that and use OWL 2 if you want.

#3

one might argue that OWL(1) has potential incompatibilities with OWL2

In the introduction of OWL 2 Mapping to RDF Graphs, it is said:

"The mappings presented in this document are backwards-compatible with that of OWL 1 DL: every OWL 1 DL ontology encoded as an RDF graph can be mapped into a valid OWL 2 DL ontology using the mapping from Section 3 such that the resulting OWL 2 DL ontology has exactly the same set of models as the original OWL 1 DL ontology."

So, if you choose OWL 1, you effectively choose OWL 2.

In fact, the question is: is there a reason not to use the OWL 2 specific constructs? In my opinion, if you do need these constructs, use them.

[Remark: the quoted paragraph is in fact wrong because of a subtle difference between OWL 1 and OWL 2 regarding the use of the reserved vocabulary, but this would not affect any tools that I know, except a strict validator.]