summaryrefslogtreecommitdiffstats
path: root/notes/cleanup_tasks.txt
blob: 812d1c2ea10c011dc836d9c10cd51cb636c71a9c (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48

This is a list of relatively simple bibliographic metadata bugs, which have not
been fixed yet. Some of these need fixes in importers, others might be one-time
runs with a simple tool (even `fatcat-cli`).

## Cambridge Chemical Database (NCI)

    doi_prefix:10.3406 release_type:article
    doi_prefix:10.14469 release_type:article

    193,346+ entities

    should be 'dataset' or 'entry' or something, not 'article'

    datacite importer

## Frontiers

    Frontiers non-PDF abstracts, which have DOIs like `10.3389/conf.*`. Should
    crawl these, but `release_type` should be... `abstract`? There are at least
    18,743 of these. Should be fixed in both crossref-bot, then a retro-active
    cleanup.

## Applied Physics Letters

    doi_prefix:10.2172 title:10.2172

    For 700+ entities, the title is the DOI number. They all seem to be
    "deleted" DOIs, and should be marked as stubs.

## Far-Future Release Years

If year is more than 20 years in the future (arbitrary cut-off), both the year
and date should probably be cleared.




--------

The following may be more difficult

## Antarctica: A Keystone in a Changing World

container_adgy773dtra3xmrsynghcednqm
homepage URL is wrong

36k releases of unknown type and unknown publication stage.