Stability of identifiers

[strogonoff]

We want to switch to new IEEE identifiers in relaton-ieee gem (and therefore relaton-data-ieee dataset and IETF BibXML service).

The goal is to replace all docid with primary: true with new identifiers: so that, for example, this

IEEE 802.15.4j-2013

becomes this

IEEE Std 802.15.4j-2013 (Amendment to IEEE Std 802.15.4-2011 as amended by IEEE Std 802.15.4e-2012, IEEE Std 802.15.4f-2012, and IEEE Std 802.15.4g-2012)

The purpose of this is to assign resources strongly normalized and stable identifiers that are unlikely to change in near future.

From what I heard, about 5–10% of IEEE identifiers cannot be parsed by pubid-ieee yet. This may or may not be a barrier that prevents us from this switch, so here are questions that try to determine that. Perhaps someone can either answer them or give some pointers.

• What does it mean that some identifiers cannot be parsed?
  • Does it mean pubid-ieee may not to produce a “strong” ID at all (that would be OK)? Or does it mean it may produce an incorrect/unstable ID (that would be a problem)?
• What documents are affected? From what I heard, ANSI and some cross-published ones are affected.
  • Is it easy to get a percentage of documents that are currently unparseable due to IEEE’s data error (something that cannot be fixed with pubid-ieee parsing)?

(ping @ronaldtse @andrew2net too)

[strogonoff]

According to Ronald:

I think around 5-10% identifiers still don’t parse properly. Around 2-5% of the total cannot be fixed ourselves. We are going to bring this up to IEEE SA directly in our meeting on August 6. (These are my “feeling” numbers)

[ronaldtse]

I have raised the topic officially to IEEE SA, and will keep this issue open until we have a formal strategy of addressing the full range of identifiers.

[ronaldtse]

This task is blocking https://github.com/ietf-tools/bibxml-service/issues/18