-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathIFAR.html
429 lines (378 loc) · 32.5 KB
/
IFAR.html
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
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
<i><small>
Tijdens presentatie continu een dataviz in beeld dat aangeeft HOE VER je in WELK HOOFDSTUK zit.<br>
Afhankelijk van de beschikbare tijd onderdelen skippen of sneller doorheen gaan<br>
Liefst een filmpje, met de hoofdstukken aangegeven.</small>
</i>
<span style="background-color:orange;">orange</span> means: needs work
<br><br><br>
tl;dr : interop-usability for systemically working at big problems<br><br><br>
<span style="font-size:200%;"><b>FAIR-ly REASONable collectivity?<br>
(Or FA<big>I</big>R data in a broader context)</b><br>
(Or making information work better for humanity)</span><br>
<table border=1><tr><td>
<i>chapters, or steps of a <b>reasoning:</b><br>
(so read this again if you lost thread)</i>
<ul>
<li> <b>WHY ?</b> WE’RE IN <i>[CENSORED]</i> CRISES! </li>
<li> WHAT ARE WE <b>THINKING</b> ? ARE WE? HOW?</li>
<li> LESS DETOUR, LESS CHAOS TO BEGIN WITH ? ATTENTION, <b>NOT FRAGMENTATION</b> !</li>
<li> MORE CONNECTED ? <b>FA<small>i</small>R</b></li>
<li> MORE FITTING ? <b><big>I</big>NTEROPerability</b>!</li>
<li> MAKING IT EASIER ? <b>USABILITY</b> !</li>
<li> <b>END ?</b></li>
</ul>
</td></tr></table>
<br><br><br><br><br><br>
<b>WHY ? WE’RE IN <i>[CENSORED]</i> CRISES!</b>
<table border=1>
<tr><td>
Something’s off …<br>
<img src="https://upload.wikimedia.org/wikipedia/commons/thumb/d/d1/20181204_Warming_stripes_%28global%2C_WMO%2C_1850-2018%29_-_Climate_Lab_Book_%28Ed_Hawkins%29.png/495px-20181204_Warming_stripes_%28global%2C_WMO%2C_1850-2018%29_-_Climate_Lab_Book_%28Ed_Hawkins%29.png" width="400">
</td></tr><tr><td>
Ratrace habit$<br>
<img src="https://www.ft.com/__origami/service/image/v2/images/raw/http%3A%2F%2Fcom.ft.imagepublish.upp-prod-eu.s3.amazonaws.com%2Ff1ffa452-6cd8-11df-91c8-00144feab49a?fit=scale-down&source=next&width=470" width="400">
</td></tr><tr><td>
need changing …<br>
<img src="https://storage.googleapis.com/climate-science.appspot.com/en-v5/C1_9-Details-5-Figure_8_Principles_Of_Degrowth-01.png" width="400">
</td></tr><tr><td>
… MUCH!!! faster<br>
<img src="https://www.ncei.noaa.gov/sites/default/files/styles/max_1300x1300/public/sites/default/files/billion-dollar-disasters-collage_1200x480.png?itok=dawxWoPH" width="400">
</td></tr></table>
<br><br><br><br><br><br>
<b>WHAT ARE WE THINKING ? ARE WE? HOW?</b>
<table border=1><tr><td>
<a href="https://en.wikipedia.org/wiki/Vulcan_(Star_Trek)#Mind_melds"><img src="https://blog.developer.atlassian.com/wp-content/uploads/dac-import/mind-meld.jpg" width="400" align="left" valign="top"></a>
<b>Are we paying attention? Listening?<br>
Do we think the same language?<br>
Can we improve our communication (of thoughts)?</b><br>
Some inspiration below<br>
</td></tr><tr><td>
<b>„Philosophische Untersuchungen“ (Philosophical Investigations), Wittgenstein</b><br>
Stated language only works as part of the context-specific interplay.<br>
<a href="https://en.wikipedia.org/wiki/Philosophical_Investigations"><img src="https://images.slideplayer.com/36/10550752/slides/slide_3.jpg" width="200"></a><br>
</td></tr><tr><td>
<a href="https://vimeo.com/115154289"><img src="https://i.vimeocdn.com/filter/overlay?src0=https%3A%2F%2Fi.vimeocdn.com%2Fvideo%2F501083941-8d7ad359986c4cc094ae8bfa457f17a624fe0933a52e689d1f2e13c5ab13164d-d_1280x720&src1=https%3A%2F%2Ff.vimeocdn.com%2Fimages_v6%2Fshare%2Fplay_icon_overlay.png" width="400" align="left" valign="top"></a>
<br><br>Says something like: (tiny rectangles & tiny movements) ==> (tiny thoughts & tiny actions).<br>
And shows directions to instead be full humans working with information and powerful tools.<br>
<i style="background-color:orange;">especially 13:45-14:50 & 16:05-16:25</i><br>
</td></tr><tr><td>
<b>“Thinking beyond the brain”, Annie Murphy Paul</b><br>
Mentions how too often the model of a knowledge worker is a brain stuck too close to a computer screen
and how that is highly ineffective as it ignores powerful capabilities that evolved over human history, to think with our body, surroundings and community.<br>
<img src="https://media.sciencephoto.com/image/c0544094/800wm/C0544094-Computer_screen_with_a_brain,_illustration.jpg" height="200">
<img src="https://image.similarpng.com/very-thumbnail/2021/06/Cross-mark-icon-in-red-color-on-transparent-background-PNG.png" height="200">
<a href="https://anniemurphypaul.com/books/the-extended-mind/"><img src="https://anniemurphypaul.com/wp-content/uploads/2020/11/3dbookcover02-200x300.png" height="200"></a>
<img src="https://raw.githubusercontent.com/steltenpower/PowerGists/main/GreenCheckmark.png" height="200">
</td></tr><tr><td>
<table><tr><td colspan="2">
<b> "Ons werk is stuk" (our work is broken), Martijn Aslander</b> explains</td></tr><tr><td>
<a href="https://pixiecreations.nl/product/ons-werk-is-stuk/"><img src="https://pixiecreations.nl/wp-content/uploads/2022/02/Cover-boek.jpeg" height="200"></a>
</td><td>
<ul>
<li>how we’re still mostly stuck in the dogma of the document<br>
<li> how organisational hierarchies need a fundamental update into the current web era of highly educated people<br>
<li> how far the effects of serious neglect of usability reach; we postpone/procrastinate, won’t use the tool, rush it, the resulting data is off and so are the decisions based on it.<br>
<li> how we need to stop the fragmentation of data over a wide and always changing application landscape.<br>
<li> how the GDPR (or “AVG” in Dutch) brings a lot of good (though it needs serious improvements), but also scared many into limiting more than demanded.
</ul>
</td></tr></table>
</td></tr><tr><td>
<img src="https://learning.oreilly.com/library/cover/9781491916902/250w/" align="left" valign="top">
Yes, this book is also for you. In doubt? Remember Forbes' article: "Why Every Company Is A Data Company".
<ul>The book states:
<li>Everybody is responsible for data quality
<li>Data is about content, not technology. Don't put IT in charge of it
<li>Broadly use data visualisation to explore and convince
<li>You need diversity in tools too
<li>Use a data inventory to show what is available
<li>Have a metadata catalogue to prevent misunderstanding
<li>Have data stewards, data analysts and data scientists, with domain knowledge too.
<li>Know what you're not measuring and that measuring influences the measured
<li>Don't jump to conclusions</ul></td></tr>
<tr><td>
"Flow in Organisaties" (Dutch): There's so much keeping us from actually working productively that most not seem to know.
If you work somewhere that uses open office plans; seriously? Nobody there ever read some research on the topic?
</td></tr>
<tr><td>
"Combifuncties": how having a significant bit of your personnel in 2 related but still quite different functions can be an advantage for your organisation (flexibility, communication, learning, retaining employees) and your employees.
</td></tr>
<tr><td>
"Als alle breinen werken": Bekijk neurodivergentie primair als een kans in de strijd om talent voor diversiteit waar je niet zonder kunt in de diversiteit
die nodig is voor hoog presenterende organisaties. De hoge uitschieters in hun spiky profiles bieden nieuwe inzichten, maar voor de lage uitschieters geldt:
een vis zal nooit goed worden in het inklimmen van bomen en niet alleen de 'vissen' hebben er profijt van als het bomenklimmen geen verplicht onderdeel is.
</td></tr>
</table>
<br><br><br><br><br><br>
<b>LESS DETOUR, LESS CHAOS TO BEGIN WITH ? ATTENTION, NOT FRAGMENTATION !</b><br>
<table border=1><tr><td>
<b>Use the power of open …</b><br>
In case you think your data is not fit for different uses, remember “The best use of your data is often created by someone else”. A colleague or fellow earthling will surprise you, if only with a more insightful datavisualisation.
Having your sheets open might save you from typos before presenting them.
Popular open source software has a community bringing quality and being a welcome force against expensive vendor lock-in.
</td></tr><tr><td>
<b>… and flavours of partly open …</b><br>
It's a great idea to have a general data inventory, but probably with limitations for parts:<br>
- leaving out some datasets for those not having special clearance<br>
- showing not the data itself, but only the structure, along with a <button>synthesize data!</button> button
</td></tr><tr><td>
<b>… and the power of hyper to adapt web applications to your liking</b><br>
With "hyper" I of course mean web technologies and the structuring they allow, but specifically:<br>
- browser extensions, that bring adaptability without needing to adapt code on the server.
Basically they are little JavaScript programs that change web pages between loading and displaying. <a href="https://chrome.google.com/webstore/category/extensions?hl=en">Many browser extensions</a> exist, <a href="https://alternativeto.net/software/greasemonkey/">some generic using so-called widely available userscripts</a>.<br>
- APIs, if they're provided, as they allow to automate things that otherwise would need to be done manually. If you have a user-friendly form to fill in a 1000 times, how useful is it still?
</td></tr><tr><td>
<b>More DOING directly together instead of only communicating ABOUT it, especially pairing.</b><br>
<img src="https://martinfowler.com/articles/on-pair-programming/benefits_overview.jpg" width="400" align="left" valign="top">
All those important things deemed too little for regular meetings like “what button did you touch there?”, or “Actually I didn’t understand, can you explain?” or “how are the kids?” or
“if only it could have been magically documented while doing”, especially since we’ll never go back to the office every day and are social creatures still. And as a bonus you’ll get an intelligently filtered context-specific implicit messaging service: “Oh, in that case, I have a handy trick for you that Trudy showed me the other day …”<br>
Also screens are great, but not for everything:<img src="https://en.wikipedia.org/wiki/Pair_programming#/media/File:Pair_programming_1.jpg">
</td></tr><tr><td>
<b>Less “copy-paste waste“</b><br>
<img src="https://raw.githubusercontent.com/steltenpower/PowerGists/main/DeadPapers.png" height="200" align="left" valign="top">
Much of our ‘digital work’ is actually mostly simulating paper.
<a href="https://wiki.surfnet.nl/pages/viewpageattachments.action?pageId=59803702&preview=/59803702/65242021/ELN%20-%20V2-1080p-220420%20edit2.mp4">“Rock, Paper, Science notebooks” (3:12)</a> explains how that’s error-prone and very unlikely to be reproducible.<br>
Reproducibility needs more than 'only' literature references you manage with e.g. Zotero
</td></tr><tr><td>
<b>Everybody edit</b><br>
Learn from the success of Wikipedia and Github. Do your access restrictions mainly limit the introduction of errors (easily rolled back with proper version control),
or limit quick improvements and updates from your community and beyond? Demand that access too when working with external parties.
And when you really (really?) can't have others edit the production version, use this method sometimes named Pull Request:
Anybody can <button>EDIT</button> and <button>Send in Suggested edits</button>
so the person responsible for code or content can <button>Highlight DIFFerences side by side</button> and <button>Accept Suggestions</button>
</td></tr><tr><td>
<b>Write on the open web!</b><br>
Whatever you seem to be repeating over and over, people could have found earlier on your blog, Wikipedia, etc.
And as we tend to forget things; your future self will thank you too.<br>
<a href="https://github.com/steltenpower/tabs-as-sheets"><img src="https://repository-images.githubusercontent.com/191587320/4e4fc780-583e-11eb-9d86-98cba0ebad61" width="400" align="left" valign="top"></a>
Oh, and instead of pages of Powerpoint, why not try browsertabs of content to go through? Also avoids some licensing hassle (though better ask a copyright expert, maybe at your library).
</td></tr><tr><td>
<b>Give inbox zero, worship agendas</b><br>
<a href="https://gripboek.nl/preorder"><img src="https://gripbook.com/img/grip-book-cover-uk.png" height="200" align="left"></a>
Other people like inbox zero too, so send them less. Agenda, Tasks, Messages is your order of priority. If people keep sending you “SAVE THE DATE” messages without an invite, or 5 reminders for an event you already declined or registered for, maybe start replying “SPAM-BLOCKED: repetitive”?<br>
Team schedules in other places are fine, but usually much better when tightly coupled with personal agenda’s.
And when “life happens” with an emergency or killer headache it’s hard to remember who to call this time, inform about things to cancel/postpone,
let alone register days. Maybe it can be all automated by "Agenda: book me sick till tuesday" ? Matching your tasks with your energy levels and moods is an important part of working with your agenda too.
</td></tr><tr><td>
<b>“Do disturb” scheduling</b><br>
Next to “do not disturb” have “please DO disturb” blocks in your agenda for communication of things that work better through immediate interaction where planning a meeting feels as “too much”.
</td></tr><tr><td>
<b>NO double manual entry</b><br>
For example: You keep your agenda really well and as a reward you can manually enter all again in project and overall time registration systems?
And in the mean time there’s no warning that everybody of a 24/7 team is saving up free days till the end of the year?
This can be done smarter, and not by some imaginary 'silver bullet' 'total solution' application, but by standardizing on data formats and APIs, and allowing choice in software.
Your agenda already figures out you need traveling between meetings and gives suggestions, right?
</td></tr><tr><td>
<b>Less context-switching saves even more capacity</b><br>
Trimming time spent on non-primary things also trims time context-switching; getting into something, concentrated, again, costs capacity big time. Times how many people?
</td></tr><tr><td>
<b>Everybody responsible for data quality and rewarding</b><br>
Where many years ago it became unacceptable to not be able to type, now the minimum IT-skillset for non-IT people includes making your data friendly for further processing.
Learn some basic rules with <a href="https://datacarpentry.org/spreadsheet-ecology-lesson/">Data organization in Spreadsheets</a> from carpentries.org,
a website that offers many more "foundational coding and data science skills" for those interested.<br>
If you think you can completely leave all of that ‘to the IT folks’ also, you’re setting things up for the ump-teenth over-expensive IT-project that should be named a disaster.
</td></tr><tr><td>
<b>Explain value, demand value</b><br>
If you want certain data entered, explain why, make it quick and easy, and reward by quickly giving value created from it. If you don’t, you will be highly overestimating the actual value you get.
Also, ask what apps people love to use outside of work and why? A combined payment&expense app perhaps, with a little budget included for everybody?
</td></tr><tr><td>
<b>Recognize paper era origins</b><br>
What we see simulated on screens a lot is US Letter paper, graph paper and presentation transparents. Though there's load of advanced funtionality, is what we actually use really much conceptually different?<br>
<br>
<a href="https://github.com/steltenpower/CitationStyleOnReading"><img src="https://repository-images.githubusercontent.com/151168459/62a2cd80-789c-11eb-8083-56f4b2fadc66" width="200" align="left"></a>
An slightly less obvious example: For findability, consistency and readability reasons scholarly communication uses 'citation styles'.
When sticking to 1 subject field you probably only really get to use a few of the thousands of available styles,
but especially now science is more and more disciplinary, would it be nice if the reader can choose, instead of the author deciding in a permanent way?
To allow future developments, please unambiguously write down the style used. I'm not aware of a machine-readable standard for it yet.
</td></tr><tr><td>
<b>Think beyond staring at Microsoft Teams at home or office.</b><br>
Next to lightboards, telepresence robots, VR, acting in front of a 2-way video wall,
and zooming in on hands instead of face there’s also just going for a long walk through old forests together, resulting in deeper insights than those aha-moments we get in the shower/bathroom. Less tech can be smarter.
A bit more thoughts in <a href="https://github.com/steltenpower/combined-presence">"combined presence"</a>
</td></tr><tr><td>
<b>Apply datavisualisation inspiration everywhere</b><br>
For example: Most search user interfaces still look a lot like the original ’10 blue text links’.
<a href="https://openknowledgemaps.org/">Open Knowledge Maps</a> (also in use by the Dutch national library) show how things are related and helps you figure out the right search keywords at the same time.
</td></tr><tr><td>
<b>Try something new, or crazy</b><br>
<a href="https://github.com/steltenpower/walk-a-desk"><img src="https://repository-images.githubusercontent.com/69238742/0d63bd80-7896-11eb-8ccb-dc2bf192d962" width="400" align="left"></a>
Just 1 example: Nowadays you can tell your device, by voice, what you want in an image and get a reasonable one in seconds.
Possibly saving you hours of more manual image creation.
Or … prototype a new shape of interaction, putting human interaction and hands-free first, starting with buying a broom stick. ☺
</td></tr><tr><td>
What seems less, can be more: see DaylightComputer
</td></tr></table>
<br><br><br><br><br><br>
<b>MORE DATA CONNECTED ? FA<small>i</small>R</b><br>
<table border=1><tr><td>
<b>The FAIR (data quality) principles</b> are here to help you (classify and reap the benefits, quickly) for your (research) data.
There’s a giant and growing body of interesting expert theoretical FAIR work and practical experience (for example Virus Outbreak Data Network, VODAN) but here I focus on only some awareness, misunderstandings and suggestions.
</td></tr><tr><td>
<b>FAIR Oversimplified</b><span style="background-color:orange;">to be translated</span><br>
<img src="https://raw.githubusercontent.com/steltenpower/PowerGists/main/FAIRsimplified.png" width="400" align="left" valign="top"><br>
Or <i>“tying together (research) data worldwide for as far as allowed and reasonably possible”.</i><br>
In which everything as machine-actionable as possible, to not only make current type of research easier, but also allowing <b>a new scale of multi-disciplinary research</b> through machine-assistants.<br>
<br>
The mentioned <b>metadata</b> is a description, where the described can also be data.
A great example:<br>
<img src="https://raw.githubusercontent.com/steltenpower/PowerGists/main/soepblik.png" width="400"><br><br>
Though of course some people …<br>
<i style="background-color:orange;">3:30-3:37 https://www.dailymotion.com/video/x3w1pog</i></span><br>
<br>
But now first some remarks about FA.R that I feel are not mentioned enough:
</td></tr><tr><td>
On <b>F(indable)</b>:
<img src="https://upload.wikimedia.org/wikipedia/commons/thumb/7/7e/Schlagwortkatalog.jpg/1418px-Schlagwortkatalog.jpg" width="200" align="left">
A <a href="https://direct.mit.edu/dint/article/doi/10.1162/dint_a_00160/112599/fair-data-point-a-fair-oriented-approach-for">FAIR Data Point (FDP)</a> is like
a library card catalog, where the cards are digital and called “Metadata Record” containing properties and relations, both written in Linked Data vocabularies.
Both the object described and its Metadata Record have their own URL, often a type of Persistent IDentifier (PID).<br>
<br>
It can well be that some of the metadata is already available in the described file itself, but with several possible notations and possibly huge files,
doubling a little metadata is efficient.
Of course an API is built-in to let search engines harvest.
For the administration systems in PID minting agencies I imagine those ‘library cards’ to be be partially pre-filled already depending their type (DOI, ORCID, ROR, PIDINST, etc.)
</td></tr><tr><td>
<b>A(ccessible)</b>: is a bit broader than how to be able to download files from a certain server: Files can be distributed for reasons of capacity (e.g. BitTorrent)
or control (e.g. SoLiD PODs).
And for data that is either giant (CERN) or highly sensitive (Personal Health Train) the analysis script is moved towards the data,
and only the results are returned (after a possible manual check). And even having no access to actual data (yet) can be helpful,
cause with the structure available, dummy or synthesized data can be generated to already start working with.
<br><br>
Sidenote: Not “accessibility”, aka “a11y”<br>
<img src="https://assets-global.website-files.com/583347ca8f6c7ee058111b55/5c119a2e1c06a0b428d6b052_keuhQlsNfQYW-besjf8oWesFJiHVYPJDggHZsTu766B3Q8JnKwt5-73q617zdDXhtG8Scrf0TpI0QS_GIN9PhACLd9e5nBxhP690yUWjdi9Al4Mqt8MDPU2tWEfdNGVOK8zxyIvI.png" width="200">
</td></tr><tr><td>
<b>R(e-usable)</b> is about being allowed by law. As closed as necessary, as open as possible.
An open license boosts re-use according to both traditional citation count and altmetrics.
<a href="https://github.com/steltenpower/CopyPaste-copyright"><img src="https://repository-images.githubusercontent.com/151335850/75306700-bc3f-11eb-873c-a20b77423bbe" width="200" align="left" valign="top"></a>
Just as <a href="https://help.libreoffice.org/6.1/he/text/swriter/classificationbar.html">some Office suites can limit cross-document copy-paste based on security classifications</a>,
wouldn't something similar be very welcome for faster working with licenses, as too often people ignore those?<br>
</td></tr></table>
<br><br><br><br><br><br>
<b>MORE FITTING ? <b>I</b>NTEROPerability!</b>
<table border=1><tr><td>
“best for last” of <small>FA</small><b>I</b><small>R</small>, though certainly not the last to attend to for success)
</td></tr><tr><td>
<b>Meaning/semantics/ambiguity</b><br>
If you process information you must know and tell what it means, what it’s about.
How often do people seem to understand what the others mean, to find out in the next meeting weeks later,
it’s slightly or completely different, if not forgotten at all? And diversity, though usually improving potential,
can make misunderstanding a little more likely.
To help that, below we’ll focus on FAIR's character I, the one that goes beyond machine-readable into allowing machine-actionable.
</td></tr><tr><td>
<b>Artificial Intelligence (AI) learns, our limitations</b><br>
Though very powerful, AI is still ‘just calculation’ applied to ‘pattern recognition’, not magic:<br>
<img src="https://cdn-media-1.freecodecamp.org/images/1*bt-E2YcPafjiPbZFDMMmNQ.jpeg" width="200"><br>
And it needs training data, somewhere based on A LOT of manual classification:<br>
<img src="https://blog.apify.com/content/images/2021/08/bike-captcha---Copy-1.jpg" width="200"><br>
That together with the ambiguity of natural language can easily go wrong:<br>
Like what an AI engine came up with for "a salmon swimming down a river"<br>
<a href="https://www.reddit.com/r/Damnthatsinteresting/comments/yijd3c/they_asked_an_ai_engine_to_recreate_a_salmon/"><img src="https://i.redd.it/omthqo8yj6x91.jpg" width="200"></a><br>
</td></tr><tr><td>
<b>Being exact is hard, even for simple daily tasks</b><br>
<i style="background-color:orange;">0:09-0:14 & 0:25-0:32</i>
<iframe width="560" height="315" src="https://www.youtube.com/watch?v=cDA3_5982h8" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
</td></tr><tr><td>
<b>We at least can improve and must</b><br>
To prevent things like again crashing an expensive mars lander by ASSUMING the wrong units (metric (cm, etc.) versus imperial (inch, etc.)):
<a href="https://www.simscale.com/blog/nasa-mars-climate-orbiter-metric">"When NASA Lost a Spacecraft Due to a Metric Math Mistake"</a>
or get people killed daily by ElectronicPatientDossiers (EPD) not being interoperable: <a href="https://www.npostart.nl/2doc/15-09-2022/KN_1727593">"Dodelijke zorg"</a> (Dutch)
<i style="background-color:orange;">8:19-10:31 12:35-12:53 13:15-13:25 15:52-16:19 26:00-26:15 35:34-36:01 39:58-40:15 41:05-41:18 42:11-42:24 45:47-45:56 58:26-58:48 1:08:05-1:08:21 1:16:11-1:16:31</i>
</td></tr><tr><td>
<b>Start Data-centric</b><br>
<img src="https://raw.githubusercontent.com/steltenpower/PowerGists/main/DataCentric.png" height="200" align="left">
To make meaning explicit, it shouldn’t be something you start thinking about only after having acquired or created an application.
Applications implicitly put meaning to values, data standards makes it explicit.
Also a way to heavily save on constructing app2app connections (possibly number of apps squared, n²), e.g. in Enterprise Service Bus (ESB).
Whenever someone utters “application landscape” ask them to read the <a href="https://datacentricmanifesto.org">DataCentricManifesto</a> (again) to prevent trouble.
</td></tr><tr><td>
<b>Be consistent</b><br>
Similar to how spelling checkers use a list of valid words for quality control, that also helps search engines index better, there’s something similar for tabular data:
<a href="https://frictionlessdata.io/">Frictionless data</a> helps you, among other things,
to define what values are valid for each column in so-called schema’s and check data against them.
Part of that is fixing for the many differences introduced by many different versions, localizations and preferences of Excel and others,
e.g.: Ever seen □ or � where you expected different characters,
or asked your colleague with a diacritic in their name how often systems seem unable to work with them properly?
IT is many layers of agreements, with usually multiple options for each, but so-called validators can help find your mistakes in most of them.
</td></tr><tr><td>
<b>Be specific</b><br>
What does “1/2” mean? Jan 2nd, Feb 1st, or 0.5 or 0,5?
Is “string” about musical instruments or underwear?
Is salmon an animal or a food?
And of course meaning may at this moment seem obvious to you, but you better specify from the start …
</td></tr><tr><td>
<b>Use, specify, combine and co-create standards</b>
<table><tr><td>
<img src="https://imgs.xkcd.com/comics/standards.png" align="left" height="200">
</td><td>
<ul>
<li>Use ISO8601 to write dates like YYYY-MM-DD to have the extra advantage of chronological being alphabetical.
<li>In some notations (e.g. XML) you can mix standards, where sometimes an extra identifier named “namespace” is used to make sure there's no misunderstanding when multiple standards use the same term.
<li>If there are no standards yet for what you try to describe, you can make your own definitions.
A clearly readable unambiguous practically implementable specification can be quite the undertaking, depending the scope.
Doing it together with other experts from the domain might turn into something as ubiquitous as HTML.
<li>For data there’s also a (web)technology building on URLs for addressing and typing the links,
where the types are defined in domain-specific vocabulaires 👇
</ul></td></tr></table>
</td></tr><tr><td>
<b>Linked Data (LD)</b><br>
<img src="https://raw.githubusercontent.com/steltenpower/PowerGists/main/LinkedData.png" height="300" align="left">
aka semantic web aka RDF/SPARQL aka knowledge graph, it's especially fit to implement a significant part of what the FAIR principles demand, especially with more research being multi-disciplinary nowadays.
Often the primary research data files are not of one of the linked data formats, but the metadata or descriptions are.
With the Linked Data Wizard it’s easy to batch convert from tabular data like CSV.
Relational SQL databases can also have a live translation layer, RML, so it can be SPARQL-queried like a triplestore (linked data database). And don't forget you can also have selections in documents/graphics/applications as data to refer to or from and processed by search engines (see RDFa and schema.org).
You can create your own vocabulary for the relations if you wish,
but for the most matches you first want to look into vocabularies already in use, e.g. <a href="https://lov.linkeddata.es/dataset/lov/">Linked Open Vocabularies</a>.
If you think your data quality is not up to it yet, I'll add "Not in production, but as it increases the chance of detecting those problems …".
</td></tr><tr><td>
<b>LD+AI = semantic web</b>. Giving validity <a href="https://www.youtube.com/watch?v=WqYBx2gB6vA">constraints and enormous energy savings</a> compared to only Large Language Models.</td></tr><tr><td>
<img src="https://pbs.twimg.com/media/FoUTExMWAAE9lP2?format=png&name=small" width="400px"><br>
<b>networks of data need more than tech</b><br>
<img src="https://pbs.twimg.com/media/FnV2DB-XgAAjUKj?format=png&name=small" width="400px"></td></tr>
</table>
Used with paper textbooks too: <a href="https://www.itc.nl/about-itc/organization/resources-facilities/living-textbook/">living textbook</a>
<br><br><br><br><br><br>
<b>MAKING IT EASIER ? USABILITY !</b>
<table border=1><tr><td>
<b>existing: <a href="https://github.com/ITC-CRIB/JupyterFAIR">JupyterFAIR</a></b><br>
A jupyterLab extension for seamless integration of Jupyter-based research environments and research data repositories.
</td></tr><tr><td>
<b>existing: <a href="https://maastrichtu-library.github.io/the-FAIR-extension/">FAIR browser add-on</a></b><br>
</td></tr><tr><td>
<b>prototyped: <a href="https://github.com/steltenpower/FAIRfilesListing">FAIRfilesListing</a> / FAIRwarning / FAIRsuggestion</b><br>
<img src="https://repository-images.githubusercontent.com/149428210/f8bcf200-77cf-11eb-93f8-452cf9db3dcb" align="left" width="400">
As prevention is better than cure, and even so called loss-less conversions introduce change by definition, you don’t want to find out at the moment of publication or archival that something’s wrong with your files.
What if things like Windows File Explorer or the file menu of any (web)application could give you useful warnings and suggestions?
</td></tr><tr><td>
<b>designed: <a href="https://github.com/steltenpower/Train-Of-Thought">TrainOfThought</a> aka IdeationGraph</b><br>
<img src="https://raw.githubusercontent.com/steltenpower/Train-Of-Thought/main/RELATE.jpg" align="left" width="400">
Providing good metadata is not a researchers primary concern to make quite the understatement.
But what if you make it really easy&quick and valuable as early as possible?
What if during brainstorming about what to research next you rapidly draw a mindmap that is surrounded by related auto-suggestions?
A digital multi-dimensional variant of not just picking the book you want from the shelf,
but also taking a peek at what more is on that shelf, for serendipity reasons.
Using linked data because on one side it's ideal to connect data from different contexts fragmented over application silos
and on the other side our brains work associatively too.
Also, there’s a rich ecosystem of Linked Open data sources with billions of statements already available,
the LOD cloud (OpenAlex for example connects wikidata concepts with ORCIDs and DOIs for scholarly metrics and more.
SURF’s edusources.nl uses it to relate open educational resources to domain-specific vocabulaires).
Could this help humanity? If we keep an eye on energy use, possibly.
Let’s find out! Get in contact.
<i style="background-color:orange;">SHOW a short video to fake it</i>
</td></tr><tr><td>
<b>Keep improving</b><br>
Praise colleagues, supporters, everybody, and then explain what you think could be even significantly easier, faster, more valuable, maybe how, design, prototype, or even implement. Invest time and money to take a step back and see the higher summit behind the closer one.
</td></tr></table>
<br><br><br><br><br><br>
<b>END ?</b><br>
<table border=1><tr><td>
So …. , not futile, right?<br>
Do your thing!<br>
Together!<br><br>
2 sides of the same thing, if done right:<br><br><br><br><br>
<img src="https://raw.githubusercontent.com/steltenpower/PowerGists/main/IMG_20221213_232848237.jpg" width="400" style="transform:rotate(-90deg);">
<img src="https://raw.githubusercontent.com/steltenpower/PowerGists/main/IMG_20221213_232831392.jpg" width="400" style="transform:rotate(-90deg);">
</td></tr></table>
<br><br><br><br><br><br>
<i><small>By <a href="https://github.com/topics/steltenrepo">systems thinker Ruud Steltenpool</a> trying to be a catalyst for insight, to have people make more impact, by improving their information skills and tools.<br>
Suggestions to improve welcome.</small></i>