suggested mapping of SRAO to OECD terms

[only1chunts]

The OECD are the custodians of something called the Frascati Manual:
https://www.oecd.org/sti/inno/frascati-manual.htm
It includes a controlled vocabulary of terms relating the fields of R&D (FORD) classifications in the main manual (see table 2.2 in this document) or previously as Field of Science and Technology (FOS) classification in the 2007 annex.
These are broad terms, but given the usage of the OECD by governments and economists, I believe there is merit in mapping the SRAO terms to those used by the OECD.
NB- I am no expert in OECD as I only just learned of it myself! and this is just a suggestion, so feel free to ignore it.

[allysonlister]

Thanks Chris,

I am aware of the Frascati Manual, but I do appreciate you pointing it out! The table you are referring to is here.

We are in the process of writing up both SRAO and DRAO, but until we complete that I'm happy to share my thoughts on Frascati. I realise that it is widely used in web forms etc (I recently had to use it myself in relation to a form a research infrastructure asked me to fill out. However, I didn't use it in the initial creation of SRAO for the following reasons:

• they have lots of compound terms, e.g. 2 - Engineering and technology
• it seems rather limited. I can never find exactly what I need - there is no Computer Science, for example.
• there are no IRIs
• there are no definitions
• there is no machine-actionable version I am aware of

This doesn't mean that we couldn't align with Frascati where possible - after all, it is used pretty frequently as you say. For instance, adding some kind of annotation property or equivalence statement. However the lack of IRIs /PIDs means it's hard to know how best to do this (they have a very generic numbering scheme e.g. 2 - Engineering and Technology, but this isn't particularly useful programmatically). What are your thoughts? With re3data's subject hierarchy, which we also align with, we use imported_from to show the source(s) of this term. If you look at Engineering Science, you see that this term is available both from NCIT and re3data. It is only the NCIT IRI that we have, however, because re3data doesn't offer this information.

We also run into the compound term issue with re3data (see Mechanical Engineering for how we are currently dealing with it), and the issue of refactoring means that our imported_from is more of a general acknowledgement than something logically rigorous.

So, what would your suggestion be about how to show any alignment? As we would be aligning with Frascati rather than using it in the creation of the hierarchy, imported_from doesn't seem quite suitable.

[only1chunts]

I've not used it myself, but maybe Skos, the Simple Knowledge Organization System data model for sharing and linking knowledge organization systems, can be used to specify the type of "match" between SRAO terms and those suggested by OECD as "broadmatch" to indicate the link?
The issue of a no unique ID or link is somewhat problematic for anything semantic, so it might be impossible to link without some URI for those OECD terms, maybe they have them somewhere? I guess the only way to find out is to reach out to them to ask.

[allysonlister]

I think it's worth looking into ahead of the next release, which will probably happen in the next few months - I'll reach out to them and leave this ticket open for now while I do so.

Thanks!