search

acl-org / acl-anthology

Data and software for building the ACL Anthology.
https://aclanthology.org
Apache License 2.0
431 stars 288 forks source link

Submit ACL Anthology to OpenDOAR #929

Open logological opened 4 years ago

logological commented 4 years ago

The Open Access policies of some institutions and funding agencies require authors to ensure that their publications are published or archived in an Open Access repository. At least some of these policies specifically direct authors to select from the Open Access repositories listed in OpenDOAR (the Directory of Open Access Repositories). It would be nice if the ACL Anthology were listed in this repository, as this way authors of *ACL and related papers wouldn't need to separately deposit a preprint in a third-party repository.

It seems that the ACL Anthology easily meets OpenDOAR's inclusion criteria (which are concerned only with things such as ensuring that full-text access is free and unrestricted, and doesn't require that the repository accepts submissions from the general public).

I could submit a new OpenDOAR entry myself, though it might be better if someone more involved with managing the ACL Anthology did so in case the OpenDOAR folks have any questions. (The submission form specifically asks if the submitter is the repository manager.)

akoehn commented 4 years ago

Short remark: we do not host all publications in the anthology, e.g. for LREC we have the metadata but link to LREC-hosted PDFs. I don't know whether this is important for OpenDOAR (would probably fall under "This is fine as long as at least some content in the repository is NOT restricted in this way and actually has a full text copy available.")

It is not clear to me what the anthology "policy" is supposed to be; we simply host proceedings at discretion and have therefore no upload policy that I know of.

logological commented 4 years ago

Those two issues are probably not likely to cause any problems. The "URLs for Policies" fields are not mandatory, but if the directory maintainers ask about this, a simple prose explanation along the lines you've already given will probably suffice. I wasn't aware that we don't actually host some of the papers we index; this should probably be mentioned in the "Additional information" field of the submission form. Very likely they won't care, or maybe they would be satisfied if we made sure to include a notice somewhere on the website drawing attention to the fact that not all of the indexed papers are hosted.