commonmeta-ruby 3.2.14 → 3.3
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/Gemfile.lock +2 -2
- data/bin/commonmeta +1 -1
- data/lib/commonmeta/author_utils.rb +1 -1
- data/lib/commonmeta/cli.rb +14 -0
- data/lib/commonmeta/crossref_utils.rb +56 -14
- data/lib/commonmeta/readers/json_feed_reader.rb +30 -2
- data/lib/commonmeta/utils.rb +34 -0
- data/lib/commonmeta/version.rb +1 -1
- data/spec/cli_spec.rb +12 -3
- data/spec/fixtures/vcr_cassettes/Commonmeta_CLI/doi_prefix/doi_prefix_by_blog.yml +997 -0
- data/spec/fixtures/vcr_cassettes/Commonmeta_CLI/doi_prefix/doi_prefix_by_uuid.yml +256 -0
- data/spec/fixtures/vcr_cassettes/Commonmeta_CLI/encode/by_blog.yml +997 -0
- data/spec/fixtures/vcr_cassettes/Commonmeta_CLI/encode/by_uuid.yml +256 -0
- data/spec/fixtures/vcr_cassettes/Commonmeta_Metadata/get_doi_prefix_for_blog/by_blog_id.yml +997 -0
- data/spec/fixtures/vcr_cassettes/Commonmeta_Metadata/get_doi_prefix_for_blog/by_blog_post_uuid.yml +389 -0
- data/spec/fixtures/vcr_cassettes/Commonmeta_Metadata/get_doi_prefix_for_blog/by_blog_post_uuid_specific_prefix.yml +389 -0
- data/spec/fixtures/vcr_cassettes/Commonmeta_Metadata/get_json_feed_item/by_uuid.yml +136 -0
- data/spec/fixtures/vcr_cassettes/Commonmeta_Metadata/get_json_feed_item_metadata/blog_post_with_non-url_id.yml +136 -0
- data/spec/fixtures/vcr_cassettes/Commonmeta_Metadata/get_json_feed_item_metadata/ghost_post_with_organizational_author.yml +91 -0
- data/spec/fixtures/vcr_cassettes/Commonmeta_Metadata/get_json_feed_item_metadata/substack_post_with_broken_reference.yml +1316 -0
- data/spec/fixtures/vcr_cassettes/Commonmeta_Metadata/write_metadata_as_crossref/json_feed_item_from_rogue_scholar_with_organizational_author.yml +91 -0
- data/spec/readers/json_feed_reader_spec.rb +280 -186
- data/spec/utils_spec.rb +8 -0
- data/spec/writers/crossref_xml_writer_spec.rb +28 -0
- metadata +14 -2
data/spec/fixtures/vcr_cassettes/Commonmeta_Metadata/get_doi_prefix_for_blog/by_blog_post_uuid.yml
ADDED
@@ -0,0 +1,389 @@
|
|
1
|
+
---
|
2
|
+
http_interactions:
|
3
|
+
- request:
|
4
|
+
method: get
|
5
|
+
uri: https://rogue-scholar.org/api/posts/1898d2d7-4d87-4487-96c4-3073cf99e9a5
|
6
|
+
body:
|
7
|
+
encoding: UTF-8
|
8
|
+
string: ''
|
9
|
+
headers:
|
10
|
+
Connection:
|
11
|
+
- close
|
12
|
+
Host:
|
13
|
+
- rogue-scholar.org
|
14
|
+
User-Agent:
|
15
|
+
- http.rb/5.1.1
|
16
|
+
response:
|
17
|
+
status:
|
18
|
+
code: 200
|
19
|
+
message: OK
|
20
|
+
headers:
|
21
|
+
Age:
|
22
|
+
- '0'
|
23
|
+
Cache-Control:
|
24
|
+
- public, max-age=0, must-revalidate
|
25
|
+
Content-Length:
|
26
|
+
- '8589'
|
27
|
+
Content-Type:
|
28
|
+
- application/json; charset=utf-8
|
29
|
+
Date:
|
30
|
+
- Sun, 18 Jun 2023 05:50:12 GMT
|
31
|
+
Etag:
|
32
|
+
- '"m63byipmbt6mf"'
|
33
|
+
Server:
|
34
|
+
- Vercel
|
35
|
+
Strict-Transport-Security:
|
36
|
+
- max-age=63072000
|
37
|
+
X-Matched-Path:
|
38
|
+
- "/api/posts/[slug]"
|
39
|
+
X-Vercel-Cache:
|
40
|
+
- MISS
|
41
|
+
X-Vercel-Id:
|
42
|
+
- fra1::iad1::zplgv-1687067411738-3a40cc7c9726
|
43
|
+
Connection:
|
44
|
+
- close
|
45
|
+
body:
|
46
|
+
encoding: UTF-8
|
47
|
+
string: '{"id":"tag:blogger.com,1999:blog-4948885059517209129.post-2525787567927280589","uuid":"1898d2d7-4d87-4487-96c4-3073cf99e9a5","url":"http://sfmatheson.blogspot.com/2023/01/quintessence-of-dust-2023-restart-why.html","title":"Quintessence
|
48
|
+
of Dust 2023 restart: the why","summary":"It''s early January 2023, a little
|
49
|
+
before sunset in Tucson. Live image below, showing the glorious Santa Catalina
|
50
|
+
mountains (the snow on the upper reaches is more apparent earlier in the day)
|
51
|
+
and my dinner preparations (shrimp and veggies on the grill).I''ve decided
|
52
|
+
to start writing here at Quintessence of Dust, after another long hiatus.
|
53
|
+
Here are some of my reasons.1. I like to write, and I have things to say,
|
54
|
+
and I self-identify as an author. For eight years, I have co-organized and
|
55
|
+
taught in...","date_published":"2023-01-09T03:03:00Z","date_modified":"2023-04-02T21:17:07Z","date_indexed":"1970-01-01T00:00:00+00:00","authors":[{"url":null,"name":"Stephen
|
56
|
+
Matheson"}],"image":null,"content_html":"It''s early January 2023, a little
|
57
|
+
before sunset in Tucson. Live image below, showing the glorious Santa Catalina
|
58
|
+
mountains (the snow on the upper reaches is more apparent earlier in the day)
|
59
|
+
and my dinner preparations (shrimp and veggies on the grill).<div class=\"separator\"
|
60
|
+
style=\"clear: both; text-align: center;\"><a href=\"https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEghgBx9BU21qMDFnLxgHmnsl6TQVfb3QCwOXrC1zcXq7WH9gN8E0TcH3pTslRBG6O8mb5gcuF9JVtDlJ2je6dFcfyzKE4OD38-ftr66nBxddo892_NkyuevrrX65ndSbwmXMaLh3F5yiqU1QIj8JtA8FLkKOcHOEVwVafz0rzh7PejbFzp3XT25nQxc/s4032/Jan%207.jpg\"
|
61
|
+
style=\"clear: right; float: right; margin-bottom: 1em; margin-left: 1em;\"><img
|
62
|
+
border=\"0\" data-original-height=\"4032\" data-original-width=\"3024\" height=\"320\"
|
63
|
+
src=\"https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEghgBx9BU21qMDFnLxgHmnsl6TQVfb3QCwOXrC1zcXq7WH9gN8E0TcH3pTslRBG6O8mb5gcuF9JVtDlJ2je6dFcfyzKE4OD38-ftr66nBxddo892_NkyuevrrX65ndSbwmXMaLh3F5yiqU1QIj8JtA8FLkKOcHOEVwVafz0rzh7PejbFzp3XT25nQxc/s320/Jan%207.jpg\"
|
64
|
+
width=\"240\" /></a></div><br /><div><div>I''ve decided to start writing here
|
65
|
+
at <i>Quintessence of Dust</i>, after another long hiatus. Here are some of
|
66
|
+
my reasons.</div><div><br /></div><div>1. I like to write, and I have things
|
67
|
+
to say, and I self-identify as an author. For eight years, I have co-organized
|
68
|
+
and taught in the <a href=\"https://meetings.cshl.edu/courses.aspx?course=C-WRITE&year=23\"
|
69
|
+
target=\"_blank\">Scientific Writing Retreat at Cold Spring Harbor Laboratory</a>.
|
70
|
+
I''m a writer and I need to write, if only for myself.</div><div><br /></div><div>2.
|
71
|
+
I have an idea for a book, along with some introductory work (but no sample
|
72
|
+
chapters yet) and writing here will help me develop those thoughts. The idea
|
73
|
+
is over twelve years old and has never faded away, which I take to mean that
|
74
|
+
I need to get it out of my system somehow.</div><div><br /></div><div>3. I
|
75
|
+
have other ideas kicking around in my head and most of them are worth writing
|
76
|
+
about. I have one new intellectual passion that is totally worth writing about:
|
77
|
+
<a href=\"https://skyislandalliance.org/the-sky-islands/\" target=\"_blank\">the
|
78
|
+
Sky Islands</a> that nearly surround us here in Tucson.</div><div><br
|
79
|
+
/></div><div>4. I have an exciting new job with great new people at <a href=\"https://plos.org/\"
|
80
|
+
target=\"_blank\">an organization</a> that''s all in for <a href=\"https://plos.org/open-science/\"
|
81
|
+
target=\"_blank\">open science</a>. I recently turned over the tens digit
|
82
|
+
on my age-o-meter. My kids will very soon be all out of college. (One is about
|
83
|
+
to start a postdoc!) All of this led, predictably, to a spasm of reflection
|
84
|
+
on projects and vision. One clear result is that I''m feeling more inspired.<span><a
|
85
|
+
name=''more''></a></span></div><div><br /></div><div>5. The demise of Twitter
|
86
|
+
has led to a lot of <a href=\"https://www.theatlantic.com/technology/archive/2022/11/twitter-facebook-social-media-decline/672074/\"
|
87
|
+
target=\"_blank\">useful commentary</a> about the nature of social media.
|
88
|
+
I left Facebook two years ago and all but left Twitter two months ago. I''m
|
89
|
+
on a <a href=\"https://fediscience.org/explore\" target=\"_blank\">good
|
90
|
+
server at Mastodon</a> and I like it; maybe microblogging there can satisfy
|
91
|
+
my desire for conversation and connection. But <a href=\"https://www.theverge.com/23513418/bring-back-personal-blogging\"
|
92
|
+
target=\"_blank\">this recent piece</a> at <i>The Verge</i> by <a href=\"https://moniquejudge.com/\"
|
93
|
+
target=\"_blank\">Monique Judge</a> convinced me that blogging could (again)
|
94
|
+
have a place in the future.</div><div><br /><div>Here''s her summary, emphasis
|
95
|
+
mine:</div><div><blockquote>At the end of the day, we don’t know what is going
|
96
|
+
to happen next with Twitter or any of these platforms. We don’t know what
|
97
|
+
changes <a href=\"https://www.techtarget.com/whatis/definition/Web-30\" target=\"_blank\">Web
|
98
|
+
3.0</a> is going to bring to the internet. We do know that <b>we will all
|
99
|
+
still be here, wanting to share our thoughts, talk about anything and everything,
|
100
|
+
and commune with our people</b>. Personal blogging is the simplest and fastest
|
101
|
+
way to do all of that. </blockquote></div><div>That''s what I want! A
|
102
|
+
place to write, and (if I''m lucky) a place to discuss and \"commune\" with
|
103
|
+
people. That used to happen a lot at <i>Quintessence of Dust</i>.<span></span></div></div></div><div><br
|
104
|
+
/></div><div>Okay, but why here? This blog is over 15 years old and was started
|
105
|
+
when I was a Christian believer. <a href=\"https://sfmatheson.blogspot.com/2007/08/kicking-off-my-blog.html\"
|
106
|
+
target=\"_blank\">Its founding themes</a> were anchored in a desire to help
|
107
|
+
Christians understand and enjoy biology, to help them shake free of misinformation
|
108
|
+
and dishonesty. It has twice languished through long hiatuses and was <a href=\"https://sfmatheson.blogspot.com/p/about.html\"
|
109
|
+
target=\"_blank\">remodeled</a> back in 2017 a few years after I deconverted.
|
110
|
+
Maybe it''s time to start anew? I think not, for many of those same reasons:
|
111
|
+
I''m still a biologist who loves science, still worried about misinformation,
|
112
|
+
and still rooted in the power of scientific explanation. I''m still a bardolator
|
113
|
+
and a Red Sox fan. <a href=\"https://sfmatheson.blogspot.com/p/about.html\">I''m
|
114
|
+
still me</a>, and <i>Quintessence of Dust</i> is still my blog. The new <a
|
115
|
+
href=\"https://sfmatheson.blogspot.com/p/about.html\" target=\"_blank\">About
|
116
|
+
page</a> is slightly remodeled from 2017, and hints at the next post, which
|
117
|
+
will outline some new goals and ongoing projects.</div><div><br /></div><div>A
|
118
|
+
final note on inspiration, from Imani Perry, writing in her newsletter (<a
|
119
|
+
href=\"https://newsletters.theatlantic.com/unsettled-territory/\" target=\"_blank\">Unsettled
|
120
|
+
Territory</a>) at <i>The Atlantic</i>. The piece is \"<a href=\"https://newsletters.theatlantic.com/unsettled-territory/63af3facace609003751bacc/becoming-writer-habit-journalism/\"
|
121
|
+
target=\"_blank\">Writing is a Democratic Art</a>.\" It''s for subscribers
|
122
|
+
only but here is her challenge:</div><div><blockquote>Thinking and writing
|
123
|
+
at a faster pace stretched me and gave me greater confidence. Sometimes I
|
124
|
+
changed my mind about what I’d written a day later. But that was okay; a newsletter
|
125
|
+
is a moment in time. Writing it each week gave me a lovely rhythm in what
|
126
|
+
has otherwise been a difficult and disorienting season in history. Feeling
|
127
|
+
stuck can get you stuck. But writing can, and should, inspire deeds.</blockquote></div>","tags":["Introduction"],"language":"en","references":[],"blog_id":"5764g49","blog":{"id":"5764g49","title":"Quintessence
|
128
|
+
of Dust","description":"<i>Quintessence of Dust</i> explores science, society,
|
129
|
+
and human nature, focusing on genetics, development, evolution, neuroscience,
|
130
|
+
systems biology, and topics related to scientific literacy. I occasionally
|
131
|
+
discuss intelligent design, creationism, science denial, and other political/social
|
132
|
+
influences on scientific literacy. Additional topics: philosophy, baseball,
|
133
|
+
scientific culture, and Shakespeare. My main theme is <b><u>scientific explanation</u></b>.","language":"en","favicon":null,"feed_url":"http://sfmatheson.blogspot.com/feeds/posts/default","home_page_url":"http://sfmatheson.blogspot.com/","user_id":"8498eaf6-8c58-4b58-bc15-27eda292b1aa","created_at":"2023-05-31T14:21:27+00:00","indexed_at":"2023-04-18","feed_format":"application/atom+xml","license":"https://creativecommons.org/licenses/by/4.0/legalcode","generator":"Blogger
|
134
|
+
7.00","category":"Social Sciences","prefix":"10.59350","modified_at":"2023-06-06T20:22:00+00:00","version":"https://jsonfeed.org/version/1.1","backlog":true,"authors":null}}'
|
135
|
+
recorded_at: Sun, 18 Jun 2023 05:50:12 GMT
|
136
|
+
- request:
|
137
|
+
method: get
|
138
|
+
uri: https://rogue-scholar.org/api/posts/2b22bbba-bcba-4072-94cc-3f88442fff88
|
139
|
+
body:
|
140
|
+
encoding: UTF-8
|
141
|
+
string: ''
|
142
|
+
headers:
|
143
|
+
Connection:
|
144
|
+
- close
|
145
|
+
Host:
|
146
|
+
- rogue-scholar.org
|
147
|
+
User-Agent:
|
148
|
+
- http.rb/5.1.1
|
149
|
+
response:
|
150
|
+
status:
|
151
|
+
code: 200
|
152
|
+
message: OK
|
153
|
+
headers:
|
154
|
+
Age:
|
155
|
+
- '0'
|
156
|
+
Cache-Control:
|
157
|
+
- public, max-age=0, must-revalidate
|
158
|
+
Content-Length:
|
159
|
+
- '17762'
|
160
|
+
Content-Type:
|
161
|
+
- application/json; charset=utf-8
|
162
|
+
Date:
|
163
|
+
- Sun, 18 Jun 2023 05:52:11 GMT
|
164
|
+
Etag:
|
165
|
+
- '"rm8wu4t2aydoe"'
|
166
|
+
Server:
|
167
|
+
- Vercel
|
168
|
+
Strict-Transport-Security:
|
169
|
+
- max-age=63072000
|
170
|
+
X-Matched-Path:
|
171
|
+
- "/api/posts/[slug]"
|
172
|
+
X-Vercel-Cache:
|
173
|
+
- MISS
|
174
|
+
X-Vercel-Id:
|
175
|
+
- fra1::iad1::wcrgj-1687067530550-fefe168832b8
|
176
|
+
Connection:
|
177
|
+
- close
|
178
|
+
body:
|
179
|
+
encoding: UTF-8
|
180
|
+
string: '{"id":"https://doi.org/10.54900/6p6re-xyj61","uuid":"2b22bbba-bcba-4072-94cc-3f88442fff88","url":"https://upstream.force11.org/an-initial-scholarly-ai-taxonomy/","title":"An
|
181
|
+
Initial Scholarly AI Taxonomy","summary":"Although advances in artificial
|
182
|
+
intelligence (AI)1 have been unfolding for over decades, the progress in the
|
183
|
+
last six months has come faster than anyone expected. The public release of
|
184
|
+
ChatGPT in November 2022, in particular, has opened up new possibilities and
|
185
|
+
heightened awareness of AI''s potential role in various aspects of our work
|
186
|
+
and life.It follows that in the context of the publishing industry, AI also
|
187
|
+
holds the promise of transforming multiple facets of the publishing process2.
|
188
|
+
In this...","date_published":"2023-04-11T08:00:34Z","date_modified":"2023-04-11T15:29:38Z","date_indexed":"1970-01-01T00:00:00+00:00","authors":[{"url":null,"name":"Adam
|
189
|
+
Hyde"},{"url":"https://orcid.org/0000-0002-7378-2408","name":"John Chodacki"},{"url":null,"name":"Paul
|
190
|
+
Shannon"}],"image":"https://upstream.force11.org/content/images/2023/04/1-1.png","content_html":"
|
191
|
+
<!--kg-card-begin: html--><p class=''u-drop-cap-small''>Although advances
|
192
|
+
in artificial intelligence (AI)<sup>1</sup> have been unfolding for over decades,
|
193
|
+
the progress in the last six months has come faster than anyone expected.
|
194
|
+
The public release of ChatGPT in November 2022, in particular, has opened
|
195
|
+
up new possibilities and heightened awareness of AI''s potential role in various
|
196
|
+
aspects of our work and life.</p><!--kg-card-end: html--><!--kg-card-begin:
|
197
|
+
html--><p>It follows that in the context of the publishing industry, AI also
|
198
|
+
holds the promise of transforming multiple facets of the publishing process<sup>2</sup>.
|
199
|
+
In this blog post, we begin the development of a rough taxonomy for understanding
|
200
|
+
how and where AI can and/or should play a role in a publisher’s workflow.</p><!--kg-card-end:
|
201
|
+
html--><p>We intend to iterate on this taxonomy (for now, we will use the
|
202
|
+
working title ‘Scholarly AI Taxonomy’).</p><h2 id=\"scholarly-ai-taxonomy\">Scholarly
|
203
|
+
AI Taxonomy</h2><p>To kickstart discussions on AI''s potential impact on publishing
|
204
|
+
workflows, we present our initial categorization of the \"Scholarly AI Taxonomy.\"
|
205
|
+
This taxonomy outlines seven key roles that AI could potentially play in a
|
206
|
+
scholarly publishing workflow:</p><ol><li><strong>Extract</strong>: Identify
|
207
|
+
and isolate specific entities or data points within the content.</li><li><strong>Validate</strong>:
|
208
|
+
Verify the accuracy and reliability of the information.</li><li><strong>Generate</strong>:
|
209
|
+
Produce new content or ideas, such as text or images.</li><li><strong>Analyse</strong>:
|
210
|
+
Examine patterns, relationships, or trends within the information.</li><li><strong>Reformat</strong>:
|
211
|
+
Modify and adjust information to fit specific formats or presentation styles.</li><li><strong>Discover</strong>:
|
212
|
+
Search for and locate relevant information or connections.</li><li><strong>Translate</strong>:
|
213
|
+
Convert information from one language or form to another.</li></ol><p>The
|
214
|
+
above is the first pass at a taxonomy. To flesh out these further, we have
|
215
|
+
provided examples to illustrate each category further. </p><p>We thoroughly
|
216
|
+
recognise that some of the examples below, when further examined, may be miscategorized.
|
217
|
+
Further, we recognise that some examples could be illustrations of several
|
218
|
+
of these categories at play at once and don’t sit easily within just one of
|
219
|
+
the items listed. We also acknowledge that the categories themselves will
|
220
|
+
need thorough discussion and revision going forward. However, we hope that
|
221
|
+
this initial taxonomy can play a role in helping the community understand
|
222
|
+
what AI could mean for publishing processes.</p><p>Also note, in the examples
|
223
|
+
we are not making any assertions about the accuracy of AI when performing
|
224
|
+
these tasks. There are a lot of discussions already on whether the current
|
225
|
+
state of AI tools can do the following activities <em>well</em>. We are not
|
226
|
+
debating that aspect of the community discussion; that is for publishers and
|
227
|
+
technologists to explore further as the technology progresses and as we all
|
228
|
+
gain experience using these tools. </p><p>These categories are only proposed
|
229
|
+
as a way of understanding the <em>types of contributions</em> AI tools can
|
230
|
+
make. That being said, some of the below examples are more provocative than
|
231
|
+
others in an attempt to help the reader examine what they think <em>and feel</em>
|
232
|
+
about these possibilities.</p><h2 id=\"initial-categorization\">Initial categorization</h2><p>Our
|
233
|
+
initial seven categories are detailed further below.</p><h3 id=\"1-extractidentify-and-isolate-specific-entities-or-data-points-within-the-content\">1.
|
234
|
+
Extract - <em>Identify and isolate specific entities or data points within
|
235
|
+
the content</em></h3><p>In the extraction stage, AI-powered tools can significantly
|
236
|
+
streamline the process of identifying and extracting relevant information
|
237
|
+
from content and datasets. However, an over-reliance on AI for this task can
|
238
|
+
lead to errors if the models are not well-tuned or lack the necessary context
|
239
|
+
to identify entities accurately. Some speculative examples:</p><ol><li>Identifying
|
240
|
+
author names and affiliations from a submitted manuscript to pre-fill forms
|
241
|
+
and save time during submission while increasing the accuracy of the input.</li><li>Extracting
|
242
|
+
key terms and phrases for indexing purposes.</li><li>Isolating figures and
|
243
|
+
tables from a research article for separate processing.</li><li>Extracting
|
244
|
+
metadata, such as title, abstract, and keywords, from a document.</li><li>Identifying
|
245
|
+
citations within a text for reference management.</li></ol><h3 id=\"2-validateverify-the-accuracy-and-reliability-of-the-information\">2.
|
246
|
+
Validate - <em>Verify the accuracy and reliability of the information</em></h3><p>AI-based
|
247
|
+
systems can validate information by cross-referencing data against reliable
|
248
|
+
sources or expected structures, ensuring content conformity, accuracy and/or
|
249
|
+
credibility. While this can reduce human error, it is essential to maintain
|
250
|
+
a level of human oversight, as AI models may not always detect nuances in
|
251
|
+
language or identify reliable sources. Some examples:</p><ol><li>Cross-referencing
|
252
|
+
citations to ensure accuracy and proper formatting.</li><li>Verifying author
|
253
|
+
affiliations against an established database.</li><li>Ensuring proper image
|
254
|
+
attribution and permissions.</li><li>Checking factual information in an article
|
255
|
+
against trusted sources.</li><li>Validating claims made in a scientific paper
|
256
|
+
against previous studies.</li></ol><h3 id=\"3-generateproduce-new-content-or-ideas-such-as-text-or-images\">3.
|
257
|
+
Generate - <em>Produce new content or ideas, such as text or images</em></h3><p>AI
|
258
|
+
can create high-quality text and images, saving time and effort for authors
|
259
|
+
and editors. However, the content generated by AI may contain factual inaccuracies,
|
260
|
+
lack creativity, or inadvertently reproduce biases present in the training
|
261
|
+
data, necessitating human intervention to ensure accuracy, quality, originality,
|
262
|
+
and adherence to ethical guidelines. Some examples:</p><ol><li>Generating
|
263
|
+
social media content (e.g., summarising longer text to a tweetable length)
|
264
|
+
or promotional content for a new publication.</li><li>Creating keyword lists
|
265
|
+
for search engine optimization (SEO).</li><li>Automatically generating an
|
266
|
+
abstract or summary of a manuscript, particularly a plain language summary
|
267
|
+
pitched at a certain audience.</li><li>Creating a list of suggested article
|
268
|
+
titles based on the content and target audience.</li><li>Producing visually
|
269
|
+
engaging charts or graphs from raw data.</li></ol><h3 id=\"4-analyseexamine-patterns-relationships-or-trends-within-the-information\">4.
|
270
|
+
Analyse - <em>Examine patterns, relationships, or trends within the information</em></h3><p>AI-driven
|
271
|
+
data analytics tools can help publishers extract valuable insights from their
|
272
|
+
content, identifying patterns and trends to optimize content strategy. While
|
273
|
+
AI can provide essential information, over-reliance on AI analytics may lead
|
274
|
+
to overlooking important context or misinterpreting data, requiring human
|
275
|
+
analysts to interpret findings accurately. Some examples:</p><ol><li>Analyse
|
276
|
+
an image to create accessible text descriptions.</li><li>Determining the sentiment
|
277
|
+
of reviews.</li><li>Identifying trending topics in a specific field to guide
|
278
|
+
editorial direction.</li><li>Analyzing the readability level of a manuscript.</li><li>Discovering
|
279
|
+
patterns in citation networks to identify influential articles and authors.</li></ol><h3
|
280
|
+
id=\"5-reformatmodify-and-adjust-information-to-fit-specific-formats-or-presentation-styles\">5.
|
281
|
+
Reformat - <em>Modify and adjust information to fit specific formats or presentation
|
282
|
+
styles</em></h3><p>AI can reformat content for specific media channels or
|
283
|
+
alternative structures, enhancing user experience and accessibility. However,
|
284
|
+
AI-generated formatting may not always be ideal or adhere to specific style
|
285
|
+
guidelines, requiring human editors to fine-tune the formatting. Some examples:</p><ol><li>Formatting
|
286
|
+
content to comply with a specific style guide.</li><li>Adapting a long-form
|
287
|
+
article for a shorter, mobile-friendly version.</li><li>Converting a manuscript
|
288
|
+
into XML or converting datasets to open formats.</li><li>Rearranging content
|
289
|
+
to fit different print and digital formats.</li><li>Adjusting images and graphics
|
290
|
+
for optimal display across various devices.</li></ol><h3 id=\"6-discoversearch-for-and-locate-relevant-information-or-connections\">6.
|
291
|
+
Discover - <em>Search for and locate relevant information or connections</em></h3><p>AI
|
292
|
+
can efficiently find and link information about a subject, streamlining the
|
293
|
+
research process. However, AI-driven information discovery may yield irrelevant,
|
294
|
+
incorrect, or outdated results, necessitating human verification and filtering
|
295
|
+
to ensure accuracy and usefulness. Some examples:</p><ol><li>Finding relevant
|
296
|
+
articles within a publisher’s corpus to recommend for further reading.</li><li>Identifying
|
297
|
+
potential reviewers for a submitted manuscript based on their expertise.</li><li>Discovering
|
298
|
+
trending topics for a call for papers.</li><li>Locating similar works to provide
|
299
|
+
context for a piece of content.</li><li>Searching for related images or multimedia
|
300
|
+
to accompany a text.</li></ol><h3 id=\"7-translateconvert-information-from-one-language-or-form-to-another\">7.
|
301
|
+
Translate - <em>Convert information from one language or form to another</em></h3><p>AI
|
302
|
+
can quickly translate languages and sentiments, making content more accessible
|
303
|
+
and understandable to diverse audiences. However, AI translations can sometimes
|
304
|
+
be inaccurate or lose nuances in meaning, especially when dealing with idiomatic
|
305
|
+
expressions or cultural context, necessitating the involvement of human translators
|
306
|
+
for sensitive or complex content. Some examples:</p><ol><li>Translating a
|
307
|
+
research article or book into another language.</li><li>Converting scientific
|
308
|
+
jargon into more accessible language for a popular science article.</li><li>Adapting
|
309
|
+
a text''s cultural references to be more understandable for a global readership.</li><li>Translating
|
310
|
+
the sentiment of a text.</li><li>Converting spoken language into written transcripts
|
311
|
+
(or vice versa) for interviews or podcasts.</li></ol><h2 id=\"balancing-ai-and-human-intervention-in-publishing-workflows\">Balancing
|
312
|
+
AI and Human Intervention in Publishing Workflows</h2><p>There is potential
|
313
|
+
for AI to benefit publishing workflows. Still, it''s crucial to identify where
|
314
|
+
AI should play a role and when human intervention is required to check and
|
315
|
+
validate outcomes of assisted technology. In many ways, this is no different
|
316
|
+
to how publishing works today. If there is one thing publishers do well, and
|
317
|
+
sometimes to exaggerated fidelity, it is quality assurance.</p><p>However,
|
318
|
+
AI tools offer several new dimensions which can bring machine assistance into
|
319
|
+
many more parts of the process at a much larger scale. This, together with
|
320
|
+
the feeling we have that AI is, in fact, in some ways ‘doing work previously
|
321
|
+
considered to be the sole realm of the sentient’ and the need for people and
|
322
|
+
AI machines to ‘learn together’ so those outcomes can improve, means there
|
323
|
+
is both factual and emotional requirements to scope, monitor, and check these
|
324
|
+
outcomes.</p><p>Consequently, workflow platforms must be designed with interfaces
|
325
|
+
allowing seamless ‘Human QA’ at appropriate points in the process. These interfaces
|
326
|
+
should enable publishers to review, edit, and approve AI-generated content
|
327
|
+
or insights, ensuring that the final product meets the required standards
|
328
|
+
and ethical guidelines. Where possible, the ‘Human QA’ should feed back into
|
329
|
+
the AI processes to improve future outcomes; this also needs to be considered
|
330
|
+
by tool builders.</p><p>To accommodate this ''Human QA'', new types of interfaces
|
331
|
+
will need to be developed in publishing tools. These interfaces should facilitate
|
332
|
+
easy interaction between human users and AI-generated content, allowing for
|
333
|
+
necessary reviews and modifications. For instance, a journal workflow platform
|
334
|
+
might offer a feature where users are asked to ''greenlight'' a pre-selected
|
335
|
+
option from a drop-down menu (e.g., institutional affiliation), generated
|
336
|
+
by AI. This way, researchers and editors can quickly validate AI-generated
|
337
|
+
suggestions while providing feedback to improve the AI''s performance over
|
338
|
+
time. Integrating such interfaces not only ensures that the content adheres
|
339
|
+
to the desired quality standards and ethical principles but also expedites
|
340
|
+
the publishing process, making it more efficient.</p><h2 id=\"the-speed-of-trust\">The
|
341
|
+
Speed of Trust</h2><p>Trust plays a large role in this process. As we learn
|
342
|
+
more about the fidelity and accuracy of these systems and confront what AI
|
343
|
+
processes can and can’t do well to date, we will need to move forward with
|
344
|
+
building AI into workflows ''at the speed of trust.''</p><p>Adopting a \"speed
|
345
|
+
of trust\" approach means being cautious yet open to AI''s potential in transforming
|
346
|
+
publishing workflows. It involves engaging in honest conversations about AI''s
|
347
|
+
capabilities and addressing concerns, all while striking a balance between
|
348
|
+
innovation and desirable community standards. As we navigate this delicate
|
349
|
+
balance, we create an environment where AI technology can grow and adapt to
|
350
|
+
better serve the publishing community.</p><p>For example, as a start, when
|
351
|
+
integrating AI into publishing workflows, we believe it is essential to provide
|
352
|
+
an ‘opt-in’ and transparent approach to AI contributions. Publishers and authors
|
353
|
+
should be informed about the extent of AI involvement and its limitations,
|
354
|
+
and presented with interfaces allowing them to make informed decisions about
|
355
|
+
when and how AI will be used. This transparent ‘opt-in’ approach helps build
|
356
|
+
trust, allows us to iterate forward as we gain more experience, and sets the
|
357
|
+
stage for discussions and practices regarding ethical AI integration in publishing
|
358
|
+
workflows.</p><h2 id=\"conclusion\">Conclusion</h2><p>The potential of AI
|
359
|
+
in publishing workflows is immense, and we find ourselves at a time when the
|
360
|
+
technology has taken a significant step forward. But it''s essential to approach
|
361
|
+
its integration with a balanced perspective. We can harness the power of AI
|
362
|
+
while adhering to ethical standards and delivering high-quality content by
|
363
|
+
considering both the benefits and drawbacks of AI, identifying areas for human
|
364
|
+
intervention, maintaining transparency, and evolving our understanding of
|
365
|
+
AI contributions.</p><p>This initial taxonomy outlined in this article can
|
366
|
+
serve as a starting point for understanding how AI can contribute to publishing
|
367
|
+
workflows. By quantifying AI contributions in this way, we can also discuss
|
368
|
+
the ethical boundaries of AI-assisted workflows more clearly and help publishers
|
369
|
+
make informed decisions about AI integration.</p><p>By adopting a thoughtful
|
370
|
+
strategy, the combined strengths of AI and human expertise can drive significant
|
371
|
+
advancements and innovation within the publishing industry.</p><hr><!--kg-card-begin:
|
372
|
+
html--><p class=''u-drop-cap-small''><sup>1</sup> It''s worth noting that
|
373
|
+
we use the term AI here, but we are actually referring to large language models
|
374
|
+
(LLMs); AI serves as useful shorthand since it''s the common term used in
|
375
|
+
our community. As we all gain more experience, being more accurate about how
|
376
|
+
we use terms like AI and LLM will become increasingly important. A Large Language
|
377
|
+
Model (LLM) can be described as a sophisticated text processor. It''s an advanced
|
378
|
+
machine learning model designed to process, generate, and understand natural
|
379
|
+
language text.</p><!--kg-card-end: html--><!--kg-card-begin: html--><p class=''u-drop-cap-small''><sup>2</sup>
|
380
|
+
By publishing, we are referring to both traditional journal-focused publishing
|
381
|
+
models as well as emergent publishing models such as preprints, protocols/methods,
|
382
|
+
micropubs, data, etc.</p>\n<!--kg-card-end: html--><p><em>Many thanks to Ben
|
383
|
+
Whitmore, Ryan Dix-Peek, and Nokome Bentley for the discussions that lead
|
384
|
+
to this taxonomy at our recent Coko Summit. This article was written with
|
385
|
+
the assistance of GPT4.</em></p> ","tags":["Thought Pieces"],"language":"en","references":[],"blog_id":"pm0p222","blog":{"id":"pm0p222","title":"Upstream","description":"The
|
386
|
+
community blog for all things Open Research.","language":"en","favicon":"https://upstream.force11.org/favicon.png","feed_url":"https://upstream.force11.org/atom/","home_page_url":"https://upstream.force11.org","user_id":"8498eaf6-8c58-4b58-bc15-27eda292b1aa","created_at":"2023-05-31T07:23:49+00:00","indexed_at":"2023-01-13","feed_format":"application/atom+xml","license":"https://creativecommons.org/licenses/by/4.0/legalcode","generator":"Ghost
|
387
|
+
5.25","category":"Humanities","prefix":"10.54900","modified_at":"2023-06-06T08:00:49+00:00","version":"https://jsonfeed.org/version/1.1","backlog":true,"authors":null}}'
|
388
|
+
recorded_at: Sun, 18 Jun 2023 05:52:11 GMT
|
389
|
+
recorded_with: VCR 6.1.0
|