Walt at Random

While this week looks good, that’s partly because of a one-day break in the treatment process that has five more weeks to run–and as of the past day or two, it does appear that I’ll be getting less done for those weeks, with fatigue setting in. But we shall see…

Before the counts, yet another reminder: I need feedback about whether/what to do with countries, and specifically whether a Diamond OA book makes sense. See https://walt.lishost.org/2023/02/goa8-new-direction-on-countries/.

So:  1,.3100 more journals checked.

The overall counts at this point:
9,900 journals checked, of which
8,753 published 1,038,507 articles in 2022 and
9,186 published 942,849 articles in 2021.

The rest of the numbers:

• Fee versus diamond/no-fee: 3,730 journals with fees, 6,170 without.
• New vs. continuing: 1,243 newly-added, 8,657 continuing (including all of the “x”status below).
• Status code:
  8,828 “a”–clean.
  232 “bi”– inactive (no articles since at least 2020).
  43 “bx”–done but at a different URL.
  69 “xd”–defunct, no articles since at least 2016.
  147 “xm”–malware (but not last year).
  19 “xn”–not an OA journal (including those removed this year but before I get to them).
  384 “xx”–unreachable or unworkable.
  And the two oddities:
  150 “xm2”–malware,also malware last year
  28 “xx2”–unreachable or unworkable, as was true last year.
• Ease of article counting articles:
  “d” 5,498: easiest, taken directly from DOAJ (sometimes with 2022 count modified)
  “w” 758: easy, journal website provides direct numbers at either volume or issue number
  “f”  2,553: middling; numbers calculated using Find function for constants (e.g. “doi.” or “pdf”)
  “c” 389: slowest; articles counted manually.

As of the last 20 journals checked in this year’s scan, I’m now just over halfway through: 9,400 journals down, 9,370 to go. The 9,400th journal (sorted by publisher, then by journal title) is Stem Cells Translational Medicine from Oxford University Press.

There’s another milestone: at this point, scanned journals show more than a million 2022 articles (1,017,142).

So am I actually halfway done? Yes, in terms of effort for the initial scan. No, for the time required. I’ll lose a week or so to a special US mid-April celebration involving reviewing lots of records; I know that the first two weeks of Daylight Savings Time will be slow for me (it’s complicated). I’m hoping that the first scan might be done by mid-May, which could (cross fingers) the data would be ready for final polishing and preparing the books around the start of June. This is much more optimistic than I’ve been, and not an actual prediction.

I’ll do the usual weekly update on Saturday, but thought the halfway point deserved a mention.

And I still need feedback about Diamond Open Access 2022-2017: is it worth doing?

As anticipated, some additional slowdowns–but that big publisher did indeed help. So, well, this week was not quite as productive as last, but not bad.

Before the counts, another reminder: I need feedback about whether/what to do with countries, and specifically whether a Diamond OA book makes sense. See https://walt.lishost.org/2023/02/goa8-new-direction-on-countries/.

So:  1,100 more journals checked.

The overall counts at this point are 8,600 journals checked, of which 7,589 published 925,671 articles in 2022 and 7,976 published 834,598 articles in 2021. The eagle-eyed among you may note the enormous increase in article counts: Even though I’m less than halfway through, there are more than 900,000 articles. Why? That same Big Publisher, MDPI, which accounted for just under 237,000 articles in 2021 and around 297,500 in 2022 (it’s growth is actually slowing in percentage terms: it had around 160,000 in 2020 and 97,000 in 2019.) As for revenue, well, MDPI appeared to have a possible fees revenue of more than $640 million (USD) in 2022.

Of the 10,169 journals left to do, more than 4,700 are in the “Univ…” category, and usually small.  The remaining journals showed 467,301 articles in 2021, so it’s reasonable to assume that this year’s total will be at around 1.4 million articles, but that’s a really rough estimate.

The rest of the numbers:

• Fee versus diamond/no-fee: 3,302 journals with fees, 5,298 without.
• New vs. continuing: 1,079 newly-added, 7,521 continuing (including all of the “x”status below).
• Status code:
  7,660 “a”–clean.
  193 “bi”– inactive (no articles since at least 2020).
  37 “bx”–done but at a different URL.
  57 “xd”–defunct, no articles since at least 2016.
  127 “xm”–malware (but not last year).
  16 “xn”–not an OA journal (including those removed this year but before I get to them).
  348 “xx”–unreachable or unworkable.
  And the two oddities:
  137 “xm2”–malware,also malware last year
  25 “xx2”–unreachable or unworkable, as was true last year.
• Ease of article counting articles:
  “d” 4,842: easiest, taken directly from DOAJ (sometimes with 2022 count modified)
  “w” 583: easy, journal website provides direct numbers at either volume or issue number
  “f”  2,237: middling; numbers calculated using Find function for constants (e.g. “doi.” or “pdf”)
  “c” 327: slowest; articles counted manually.

The first week of medical interruptions was not too bad: Despite losing a chunk of each weekday, side-effects haven’t set in yet (they say 2-3 weeks out). This coming week will have more disruptions (one afternoon shot)–but I see one very large publisher coming up that may (or may not) be easy to handle. So, well, this week was not quite as productive as last, but not bad.

Before the counts, a reminder: I could really use feedback about whether/what to do with countries, and specifically whether a Diamond OA book makes sense. See https://walt.lishost.org/2023/02/goa8-new-direction-on-countries/.

So:  1,200 more journals checked.

The overall counts at this point are 7,500 journals checked, of which 6,604 published 599,507 articles in 2022 and 6,955 published 566,675 articles in 2021.

• Fee versus diamond/no-fee: 2,783 journals with fees, 4,717 without.
• New vs. continuing: 958 newly-added, 6,542cluding all of the “x”status below).
• Status code:
  6,679 “a”–clean.
  165 “bi”– inactive (no articles since at least 2020).
  35 “bx”–done but at a different URL.
  51″xd”–defunct, no articles since at least 2016.
  106 “xm”–malware (but not last year).
  14 “xn”–not an OA journal (including those removed this year but before I get to them).
  309 “xx”–unreachable or unworkable. (Last week’s number was a typo.)
  And the two oddities:
  117 “xm2”–malware,also malware last year
  24 “xx2”–unreachable or unworkable, as was true last year.
• Ease of article counting articles:
  “d” 4,132: easiest, taken directly from DOAJ (sometimes with 2022 count modified)
  “w”541: easy, journal website provides direct numbers at either volume or issue number
  “f” 1,999: middling; numbers calculated using Find function for constants (e.g. “doi.” or “pdf”)
  “c” 291: slowest; articles counted manually…including one charmer that only shows dates when you open PDFs, preferring “about x months ago.” Arggh…One way to promote open source software, I guess!

A fleshed-out version of this appears here.

The “country book” portion of the GOA series has been relatively little-used. For GOA7, I changed it significantly, making it a guide to “the long tail” by region and country, eliminating the Big Eleven OA publishers.

That hasn’t helped. Nobody’s purchased a book and there have been very few free downloads.

The question now is what to do in GOA8. My agreement only calls for the data gathering and analysis, shared dataset, and the free ebook and nominally-priced paperback for GOA in general. I believe the country-by-country rundown can be valuable, but only if it’s being used.

I see three possibilities for GOA8, once the general project is done (still no schedule, but right now very late spring is looking more possible–the next six weeks will tell a lot). Here are the three options:

1. Drop the detailed country analysis completely.
2. Try the “long tail” one more time.
3. Diamond Open Access 2017-2022: Articles in No-Fee Jourrnals: Basically take the subset of the overall dataset that consists of no-fee journals (most journals, but not most articles), and do a new book that includes shorter versions of portions of the general GOA8 (probably not including subject chapters), and add (shorter) chapters for all countries with at least 10 no-fee journals.

That overall analysis would, I believe, show a sharp shift toward eastern and southern regions, and some other sharp shifts. (E.g., toward humanities and social sciences.) And yes, I would use the D-word.

Please forward this to anyone you know who might find the results useful and might use them. I will rely on feedback–here in the short term, as email (waltcrawford@gmail.com) or on Mastodon (guess my handle there!) in the longer run. I might even check the Muskbird. Or maybe not.

[If you’re wondering: the Week Six wrapup will be on Saturday as usual. It will include fewer journals than last week but more than I was expecting.]

The daily interruptions start on Monday; next week should show much difference they’ll make. Meanwhile, yet another productive week. I’d thought of doing posts when certain milestones were reached, but that’s silly–and this was a big week for milestones:

• I’ve now done just over a third of the journals.
• There are already more 2022 articles in counted journals than there were in the entirety of the 2016 count.
• And that means more than half a million have been counted.

So:  1,300 more journals checked, The overall counts at this point are 6,300 journals checked, of which 5,546 published 538,460 articles in 2022 and 5,843 published 506,073 articles in 2021.

Some details–as always, about the full dataset to date, not this week’s portion.

• Fee versus diamond/no-fee: 2,471 journals with fees, 3,829 without.
• New vs. continuing: 808 newly-added, 5,492 continuing.
• Need rechecking: 858 will be rechecked (including all of the “x”status below).
• Status code:
  5.629 “a”–clean.
  141 “bi”– inactive (no articles since at least 2020).
  31 “bx”–done but at a different URL.
  40 “xd”–defunct, no articles since at least 2016.
  87 “xm”–malware (but not last year).
  11 “xn”–not an OA journal.
  232 “xx”–unreachable or unworkable. (Last week’s number was a typo.)
  And the two oddities:
  109 “xm2”–malware,also malware last year
  20 “xx2”–unreachable or unworkable, as was true last year.
• Ease of article counting articles:
  “d” 3,548: easiest, taken directly from DOAJ
  “w”430: easy, journal website provides direct numbers at either volume or issue number
  “f” 1,644: middling; numbers calculated using Find function for constants (e.g. “doi.” or “pdf”)
  “c” 245: slowest; articles counted manually.