https://web.archive.org/web/20241117122125/https://support.signal.org/hc/en-us/articles/6829998083994-Phone-Number-Privacy-and-Usernames-Deeper-Dive#:~:text=A%20username%20is%20a%20way,are%20chatting%20with%20in%20Signal

Signal allows you to set usernames. They are unique but temporary (and you can have only 1 at a time). User names can be used to connect to you without sharing your phone number. Set an optional username in Settings Profile. They have two numbers at the end (you can set them).

User names can be shared in three ways: - tell someone (and then change it so they cannot communicate it further) - share a QR code - share a unique URL (which does not contain your username in clear text)

Signal can't 'easily' see which phone number has which username. But given a username it can find the associated phonenumber. 'easily' means it can be done though, and thus both ways.

An old username will become available to others after a week, meaning imo they should not contain any identifiable or associative information.

Found this through someone suggesting that sharing your Signal username through Mastodon would allow private msgs. Yes, but the world will know your username, so you're open to all people who might think it fun to msg you.

the purpose of a system is a what it does, POSIWID, Stafford Beer 2001. Used a starting point for understanding a system as opposed to intention, bias in expectations, moral judgment, and lacking context knowledge.

HCI and HCD as fields have failed to respond to LLM tools and chatbot interfaces a generic solution to everything forcefully.

author calls LLMs and similar AI tools hegemonic, worsened by capital influx

Author moved from mitigating harm of algo systems to the moral standpoint that actively resisting, sabotaging, ending AI with attached political projects are valid reaction to harm. So he's moving from monster adaptation / cultural category adaptation to monster slaying cf [[Monstertheorie 20030725114320]]. I empathise but also wonder, bc of the mention of the political projects / structures attached, about polarisation in response to monster embracers (there are plenty) shifting the [[Overton window 20201024155353]] towards them.

https://web.archive.org/web/20241116074149/https://ali-alkhatib.com/blog/fuck-up-ai Ali Alkhatib (anthropology/informatics academic, QS and HCI) #2024/06/24 call for active resistance against AI / ML

https://web.archive.org/web/20241115135937/https://workforcefuturist.substack.com/p/ai-agents-building-your-digital-workforce

On AI agents, and the engineering to get one going. A few things stand out at first glance: frames it as the next hype (Vgl plateau in model dev), says it's for personal tools (doesn't square w hype which vc-fuelled, personal tools not of interest to them), and mentions a few personal use cases. e.g. automation, vgl [[Open Geodag 20241107100937]] Ed Parsons of Google AI on the same topic.

https://web.archive.org/web/20241115134320/https://garymarcus.substack.com/p/confirmed-llms-have-indeed-reached?triedRedirect=true Gary Marcus in a told-you-so piece on algogens hitting a development wall, same as the other piece by Erik Hoel on models plateauing.

https://web.archive.org/web/20241115134446/https://www.theintrinsicperspective.com/p/ai-progress-has-plateaued-at-gpt Erik Hoel notices that LLM development is stalling at the GPT-4 level. No big jumps in recent releases, across the various vendors. Additional scaling is not bringing results. Notice the graph, might be interesting to see an update in a few months. Mentions overfitting, to benchmarks as in teaching to a specific test.

Counternarrative to rural-urban divide in US politics, says data suggests it was city dwellers not showing up for Harris that tipped the balance. Main point Trump got about the same popular vote numbers but where Biden got 81M in 2020, Harris got just under 72M losing the popular vote. The diff is in core metropolitan counties in swing states.

This worked for me. I had to switch to Apple Mail a few months ago and it is extremely annoying that if you add small attachments it will preview inline. And at times even resizes them for that, and sends only the smaller version.

https://web.archive.org/web/20241114063424/https://www.zotero.org/groups/4676190/tools_for_thought/items/S2SC37HT/item-list

A great collection of sources on tools for thought, by the looks of it by [[Chris Aldrich]] Books, texts and more.

Like MS Teams is your teammate, like your accounting software is your teammate. Do they call their own Atlassian tools teammates too? Do these people at Atlassian get out much? Or don't they realise that the other handles in their Slack channel represent people not just other bits of software? Remote work led to dehumanizing co-workers? How else to come up with this wording? Nothing makes you sound more human like talking about 'deploying' teammates. My money is on this article was mostly generated. Reverse-Turing says it's up to them to say otherwise.

And as usual in these articles the truth is at the end, it's again just promises.

At the center of an AI application, like what, mechanical Turks?

After a section celebrating examples doing just that!

This sentence preceded by 2 examples where interactions and collaboration were delegated to bots to hand-out generated warm feelings, does not convey much positive about Atlassian. This basically says that a lot of human interaction in the or is seen as meaningless, and please go do that with a bot, not a colleague. Did their branding ai-agent write this?

Like Linked-In wants you to congratulate people on their work-anniversary?

Welcome in our company, you'll meet your first human colleague after you've interacted with our onboarding-robot for a week. No thanks.

anthropomorphing should be in this article's don't list. 'inviting someone on your team' is a highly social thing. Bringing in a software tool is a different thing.

An AI agent to help you speak corporate better, because no one actually writes/reflects/talks that way themselves. How did the receivers of these reports perceive this change in reports? Did they think it was better Q, or did all reflections now read the same?

Low end tasks for agents don't really need AI do they. Vgl Ed Parsons last week wrt automation as AI focus.

This does not fully chime with my own perception of (AI) agents. At least the titles don't. The tails of descriptions 'trained to adhere to brand guidelines' and 'expertise in internal decision-making framework' makes more sense. I suppose I also rail against this being the org's agents, and don't seem to be the team's / pro's agents. Vibes of having an automated political officer in your unit. -[ ] explore nature and examples of AI agents better for within individual pro scope #ontwikkelingspelen #netag #30mins #4hr

https://web.archive.org/web/20241113110726/https://www.experimental-history.com/p/so-you-wanna-de-bog-yourself

from 2024/01 by Adam Mastroianni

Giving repeating neg patterns wrt procrastination / not getting into action, a silly name helps in defeating the pattern (rather than beating yourself up over it I suppose).

Ameera Kawash (artist, researcher) def of decolonizing AI.

https://web.archive.org/web/20241112124556/https://www.heise.de/news/Lokale-KI-Nemotron-70B-auf-vier-Mac-minis-M4-Pro-mit-Thunderbolt-5-Interconnect-10014248.html

Exolabs.net experiment running large LLMs locally on 4 combined Mac Mini's. Links to preview and github shared code. For 6600-9360 you can run a cluster of 4 Minis locally. Affordable for SME outfits.

https://web.archive.org/web/20241112122725/https://lexfridman.com/dario-amodei-transcript

Transcript of 5+ hrs (!) of Dario Amodei (CEO Anthropic) talking about AI, AGI and more. Lots to go through it seems. Vgl [[My Last Five Years of Work]] by Amodei's 'chief of staff' whatever that means wrt a CEO other than sounding grandiose.

Ha, rather than "Please note that the transcript is machine generated and is certain to contain errors".

I'm tempted to run this transcript through Claude for summaries and structure. See how that works.

To [[Matt Webb]] the project using GPT3 extracting data from web pages saved him 4d of work (compared to 20 mins coding up the GPT-3 instructions, and ignoring GPT-3 then ran overnight). Saying that's about 10yrs of Moore's law happening to him all at once. 'computers got 10yrs better' an enticing thought and framing. It depends on the use case probably, others will lose 10 yrs of their time making sense of generated nonsense. (Vgl the #pke24 experiments I did w text generation, none of it was usable bc enough was wrong to not be able to trust anything). Sticking to specific niches probably true : [[Waar AI al redelijk goed in is 20201226155259]], turning the issue into the time needed to spot those niches for yourself.

[[Matt Webb]] used gpt-3 in Feb 23 to extract data from a bunch of webpages. Suggests it's the kernel for programmatic AI idea among SV hackers. Vgl Google AI [[Ed Parsons]] at [[Open Geodag 20241107100937^aiunstructdata]] last week where he mentioned using AI to turn unstructured (geo) data into structured. Page found via [[Frank Meeuwsen]] https://frankmeeuwsen.com/2024/11/11/vertragen-en-verdiepen.html

overview of European alternatives for digital products. Not in all relevant categories (payments e.g.) but some novel providers in there.

Explore wrt [[Infosec ladder van techniek en gedrag 20190530190335]] in context TGL.

On covid and long term impact on cognition

Vgl [[Wayfinding by Michael Bond]]

Good and interesting points by Baldur Bjarnason. With arsonists in our own gov, and full blown fascists making a US clean sweep of all branches of gov, relevant q's. Vgl [[Mijn werk is politiek 20190921114750]] mbt tech en TGL. Wat haal ik hier uit? -[ ] lees dit in detail, maak vergelijking met mijn huidige werkzaamheden langs zelfde lijnen.

Map of the exact location of the Berlin wall, plotted on a current map of streets and buildings. Useful in pinpointing exact locations of images have from my trip in 1987.

Hedgedoc is a collaborative markdown tool. Handig voor team mbt samen Obsidian notes bewerken. Kan op yunohost, al wordt een issue/dependency met chromium genoemd in 2023 wv onduidelijk is of het is opgelost.

-[ ] doe een testinstallatie v Hedgedoc op Yunohost #webbeheer #tgl

Spanish immigration minister diff tone than other EU MS / Meloni.

Knower / observer not separate from the universe, not outside the system boundary Vgl [[Systems convening landscape als macroscope 20230906115130]] where the convener is integral part of it too, not an external change agent.

Hadn't realised Daniel Dennett died last April. I read his The Mind's I (1981), Consciousness Explained (1991) and Darwin's Dangerous Idea (1995) while at university, those last two as they appeared. Have Elbow Room (1984) on the reading stack currently.

Yes. I use lots of small local html/php pages. Also webforms to search websites elsewhere, without going there. I had local pages to browse local image files in the 90s. I started writing html by hand in '93 and still do for local stuff. I do use a local on-device webserver though, as I include php.

https://web.archive.org/web/20241017043750/https://alexwlchan.net/2024/static-websites/

I like this idea of having static html as page to explore folders, I had that in the 90s to better search for image files. Author offers no clues as to how he uses the affordance it provides though, in terms of 'showing the metadata' they care for and the little bits of extra functionality. And I wonder about the effort involved when adding new files. Presumably new files are added manually too, otherwise it's not 'static html'. Stores files by year, type and first letter of file name. That makes no immediate sense to me in terms of finding things back. Then again I never understood why you would have folders for file types. It's like sorting items on the type of box it came in. Good example though of making your computer your own.

CoS to a CEO is what? It all reads like your typical mid-20s arrogance before finding out you don't actually know all, mistaking a hunch/'vision' for reality.

Anthropic is the outfit that made the Claude model.

another apples/oranges. We have software that is good at math. Regular people call them spreadsheets. What we don't have, also not in algogens is software that understands what it is they're doing. My model can do sums is not a useful comparison wrt if it can do cognitive tasks.

another over the horizon precondition for author's premise to happen mentioned here. Notices that robots are bound to laws of nature, and thus develop slower than software environs but doesn't notice same is true for AI. The diff is that those laws of nature show themselves in every robot, but for AI get magicked out of sight in data centers etc, although they still apply.

Weird notion of remote work as only screen interaction. My team works remote, meaning they think independent from any screen tasks.

? young? Author is in their 20s, case of 'my first encounter with something means it is globally new'?

Ah, article based on assumption of future improvement. compute and data are limiting factors, and you will end up making the equation if compute footprint is more efficient than doing it yourself. Data even more limiting, as the most meaningful stuff is qualitative rather than quantitative, and stats on the Q stuff won't give you meaning (LLMs case in point)

Is it though? Wrt GAI that is as far away as before imo. The rainbow never gets nearer, because it is dependent on your position.

True, and that is where this fails outside of bullshit tasks. The unmentioned assumption here is that algogen output can have meaning, rather than just coherence and plausibility.

equates 'content production' w k-work

right. 'content production' where the actual meaning isn't relevant?

cogent content / passably well isn't the quality benchmark for K-work though.

Sightful h1 2025 launch on Win via John Philpin https://john.philpin.com/2024/10/12/spatial-computing-has.html

every image change in the page moves focus from this form to page. Irritating

https://web.archive.org/web/20241012060204/https://www.felicis.com/insight/the-agent-economy

for the image listing various AI 'agent' services. Some seem dubious as example, and not agents but generally 'ai tools'. -[ ] maak lijst v voorbeelden uit illustratie [[241012Felicislijstaiagentscorps.png]] #15mins mbt acties/sectoren die interessant lijken. Now an agent for that task....

so much wrong with that phrase and a tell for how these corps view this tech. let's have your customers talk to machines.

seems that way, and to mostly remain that way. I'd be curious to incorporate agents in my tasks ([[Aazai CL]] list of such tasks)

also burying the lede much, this is the key verdict and it's in the penultimate paragraph?

Real agents would start at the individual level. It all smacks so much of corps automating away their own direct interaction with customers, bc they're a pain to talk to. Blind, see gripes of existing silo customers about the impossibility getting to talk to someone

almost by def asymmetric, leaving customers to talk to a blind wall.

The gap is a constant I suspect. In the tech itself, since my EE days, and in people's expectations. Vgl [[Gap tussen eigen situatie en verwachting is constant 20071121211040]]

Also costly to run at all. If this is to increase efficiency of a corp or individual it needs to be energy efficient too. Otherwise doing it yourself is the more efficient option. AI is bound to the same laws of nature as us. [[AI heeft dezelfde natuurwetten 20190715135542]] Hiding away the inefficiency in a data center's footprint and abstracting into a service fee doesn't change that dynamic ultimately.

Again it's just a promise, which seems to be the AI mantra at every step.

multi step is what they're for no? Automator can do better than agents at this time it seems.

Exactly. All promise no delivery yet. It may work if the other side is equally automated, but if it's human or a dumb web form it won't. It also reveals on the side of the human demonstrator a big lack in reflecting on their own preferences that the AI should attach to its choices.

Vgl Phillips voice automation train tickets in 90s. 'Where do you want to go' 'It's not for me but for my mom' 'Destination not found: mom'

Note this is only 'nice' from the buyer's perspective. The 'phone call' to the shop still means having a human be subject to a computer call. It also probably means you don't care about what's being bought. No back story to e.g. a gift. Beware [[Spammy handelings asymmetrie 20201220072726]]. You automate 10 million things be sent, but need to be deleted by a human e.g.

Indeed. [[AI personal assistants 20201011124147]] https://www.zylstra.org/blog/2020/10/narrow-band-digital-personal-assistants/ We should start with the personal here, wrt automation, not the AI to get to quicker results: [[small band AI personal assistant]] where the personal limits the range of possible inputs for a task and the range of acceptable outputs for a task, leaving a smaller area for an AI agent to do its thing in and thus be more effective.

Still, how much of that is already available to automate on-device? 'routine tasks automated' is not in need of AI. What are examples?

Agents are prompt chains that include fetching info (params!) from elsewhere for their function. vlg [[Standard operating procedures met parameters 20200820202042]] I wonder how you generalise them, other than 'go buy/book', and when you do if they are above what on-device automation can do. In the end individuals need to be able to set the params/boundaries of any agent, make it their own agent, rather than some corps agent. What I see at consumer facing level is not aiding consumers but aiding corps reduce human interaction with consumers. Agents should increase agency, is the lithmus test.

yup

Ben Werdmüller makes an interesting distinction. Internet tech, and thus Silicon Valley, originated in defense (ARPA etc.), whereas the web originated in academia in a spirit of open academic debate (CERN). Now ARPA etc had deep ties w academia too, and it's mostly defense funding at play. Still there may be something to this distinction. You could also say perhaps it's an Atlantic divide, the web originated at CERN in Europe.

https://web.archive.org/web/20241007071434/https://www.dbreunig.com/2024/10/03/we-need-help-with-discovery-more-than-generation.html

Author says generation isn't a problem to solve for AI, there's enough 'content' as it is. Posits discovery as a bigger problem to solve. The issue there is, that's way more personal and less suited for VC funded efforts to create a generic tool that they can scale from the center. Discovery is not a thing, it's an individual act. It requires local stuff, tuned to my interests, networks etc. Curation is a personal thing, providing intent to discovery. Same why [[Algemene event discovery is moeilijk 20150926120836]], as [[Event discovery is sociale onderhandeling 20150926120120]] Still it's doable, but more agent like than central tool.

https://web.archive.org/web/20241005072338/https://ruk.ca/content/blog-posts-are-breadcrumbs#comment-28817

[[Peter Rukavina]] on how his blog is something others come across and make connections. Commented that [[Hoe emergence tot stand komt 20040513173612]] is from longer traces. My PKM system is leaving those traces for me, my blog for me and others. My blog is the longest, due to it being 22+ yrs old, trace I'm leaving publicly for others to connect around.

Friction can forestall writing. I've moved to blogging from inside my Obsidian notes through micropub, rather than using the WP back-end. Gives me two things: the back-end pushed me to only write when I had time to finish, the notes allow multiple writings in parallel, and publishing is one key click.

Vgl [[Sturgeons Law most is crap 20190328205135]] https://www.zylstra.org/blog/2019/03/90-of-everything-is-crap-a-source-of-imposter-syndrome/ Provides agency and ratchet too.

https://web.archive.org/web/20241002133454/https://tracydurnell.com/2024/09/29/the-secret-power-of-a-blog/

Tracy Durnell on the personal affordances of keeping a blog. Haven't read just glanced, but usually she makes interesting points, marked to read -[ ] Read this wrt #pkm #15mins

via [[Euan Semple]]

https://web.archive.org/web/20241002103957/https://whyy.org/segments/is-giftedness-a-form-of-neurodivergence/ mentioned in Mensa Heurekasig mailinglist. Art from #2024/05/20 on seeing hiq as neurodivergence, wrt 'gifted burnout'

my comment, in response to someone getting lost in up front organising of notes, and ending up in a 'mess'. Embrace the mess, lengthen traces to stumble upon, earn structure (they're a personal tool not an outside standard or demand). Organise during note usage rather than during note taking, except for curation when saving something external with a remark (tags sometimes) and an internal link.

https://web.archive.org/web/20241002072359/https://www.axios.com/local/charlotte/2024/10/01/hurricane-helene-tech-chip-shortage-spruce-pine-quartz-supply

Spruce Pine is a key resource for high grade quartz. There's only a few places this is available. It has been hit by a tropical storm in North Carolina and now temp closed. The lenght of closure may impact semiconductor production.

Het EU dataportaal heeft een ERPD data flag voor Chapter II v.d. DGA. Vu NL moet iov met CBS data.overheid.nl dit verzorgen. 1 jr na van toepassing worden zijn er ~1400 datasets veel CZ, maar ook NL zie ik van CBS zelf en rivm. 1045

-[ ] vraag #bzkdga en #cbs om hoe dit tot stand komt wel/niet via data.overheid.nl? #10mins #prio

iAwriter pointing out that from their perspective G and A appstores don't compete but divvy things up between them. A 20B USD / 17,5% revenue deal makes it tangible. Say they appreciate the DMA because of it.

https://web.archive.org/web/20240929075044/https://pivot-to-ai.com/2024/09/28/routledge-nags-academics-to-finish-books-asap-to-feed-microsofts-ai/

Academic publishers are pushing authors to speed up delivering manuscripts and articles (incl suggesting peer review be done in 15d) to meet the quota they promised the AI companies they sold their soul to. Taylor&Francis/Routledge 75M USD/yr, Wiley 44M USD. No opt-outs etc. What if you ask those #algogens if this is a good idea?

well yes that is standardisation. If you exclude intended users from its creation it won't be one. A standard gets created by the intended field of users, who commit to adopting it once created. It's not idealism or altruism, it's industry.

This entire piece gave me nothing to understand 'what's the problem' other han a personal beef with a key figure, a dislike for organisation and not understanding standardisation as an industry effort. So I see the author's problems, but still don't know anything about Social Web Foundation, other than that many people seem to feel left out.

This entire thing indeed seems to be about the author's personal perspective on Evan Prodromou

organising is suspect by def then?

a standard is only a standard if it is adopted by those in the user group. Creating your 'own' by definition isn't a standard, at most it's a method or protocol.

The Social Web by naming itself thus reduces social web to fediverse and then to AP only.

And 'Fediverse' is what? Some trekkie couldn't decide between universe and The Federation. Idk what the problem is yet w the foundation, but this is the lamest critique thinkable.

https://web.archive.org/web/20240924182745/https://www.theguardian.com/global-development/2024/sep/24/drug-resistant-typhoid-disease-pakistan-antibiotics-superbugs-children

Typhoid of a antibiotics resistant variety has taken hold in Pakistan. International action required in the face of superbugs.

https://web.archive.org/web/20240924124652/https://edition.cnn.com/2024/09/20/business/23andme-board-resigns-nightcap/index.html

23andme postorder DNA profiling is slowly collapsing. CEO wants to take it private, and board has resigned in protest. This is one of the corps that went public using a SPAC to capitalise on the height of hype. Founded 2006, SPAC in 2021. Revenue is down, and money should run out soon. Seems there's no business model on top of the 1 time purchase of a DNA test. Key asset obv is the data, so I think we can wait for it to be sold to whoever bids most.

Charlie Stross provides a personal anecdotal data point to [[Changes in memory and cognition during the SARS-CoV-2 human challenge study]] that revising his writing after covid showed him cognitive issues he didn't realise as he was writing. Comments / responses add to it. Personally I use a type of puzzle and a timer to gauge my concentration and have done for several years. Since my last Covid and in my burnout I now at times don't finish or make mistakes in a puzzle, where I would consistently be in the 5% fastest solvers before.

https://web.archive.org/web/20240923184711/https://nlnet.nl/project/MaemoLeste-Telepathy/interview.html

Maemo Leste is intended to be a mobile OS independent and completely separate from Android and iOS. e/OS in comparison I think is more a degoogled Android. Its origins are in a Nokia project called Maemo (never heard of it).

https://web.archive.org/web/20240923175503/https://www.baldurbjarnason.com/2024/the-promise-and-the-distraction/

glorious rant by Baldur Bjarnason, but not much surprisal here. As with other stuff, albeit agile scrum, getting things done, and any of the pitched perfect ways to make notes, whenever the process becomes the thing rather than a tool in the hand of an knowledge artisan stuff is useless and boring.

It's about output, not in units or volume, but in quality. Needing to know why your are making these notes, and weaving your network of meaning.

The people who do things with their system usually don't talk about it much. I've done it on occasion and am happy to share and show how/why I do things, but never with the intention to convince another to do the same or similar.

https://web.archive.org/web/20240923073444/https://www.trend-mill.com/p/i-kinda-hate-the-internet-now

Stephen Moore rants about the internet he does not like. Incl the deterioration of search, the adtech, growth hacking for engagement etc. Now no longer human created even but generated slop from bots. Calls it a addictive dopamine machine without joy.

What was that old news editor adagio again? Never use a question mark in the title bc it signals the answer is 'No'. (If it is demonstrably yes, then the title would be affirmative. Iow a question means you're hedging and nevertheless choose the uncertain sensational for the eyeballs.)

Where is this different from the prompt-chaining thing I run locally? Which you prompt for a bunch of steps and then self prompts each one and goes online to further detail or do them.

yeah, externalising the cost of getting it wrong at scale. Testing it in real world circumstances is extremely useful and needed, yet OpenAI's general public customers will mostly not show their own CT and assume any result is true (seen it happen a lot) and thus moving the cost externalisation further down the chain, where it is more likely to have negative real world consequences.

https://web.archive.org/web/20240916044350/https://donaldclarkplanb.blogspot.com/2024/09/critical-thinking-was-famous-21st.html

Donald Clark on the latest closed OpenAI iteration. I can see how it may do the rule based bits of CT (although OpenAI's stuff until now still does very basic stuff wrong even before getting to CT here and at unpredictable times making every output suspect and in need of checking) Anything rule based is codable or will stand out as pattern for prediction.

Says AGI is now more visible. Seems rather Chomsky-esque, assuming language is thinking. Sociocentric (CT!) too, wrt English, while the rest of linguistics world has a century of saying language is communication and thinking a different thing.

https://web.archive.org/web/20240923070540/https://www.thelancet.com/journals/eclinm/article/PIIS2589-5370(24)00421-8/fulltext

(Even mild) covid cases are associated with persistent cognitive damage. Empirical data from 2021/2022. Largest diff between measured groups wrt memory and executive functions. No volunteers self-reported cognitive symptoms. Iow covid is associated with cognitive damage, but you won't notice yourself.

Trender et al 2024.

Paper TrenderChangesMemoryCognition2024 in Zotero

https://web.archive.org/web/20240923064617/https://wordpress.org/news/2024/09/wp-engine/

Matt Mullenweg calls WP Engine 'not WordPress' bc it disables key features to save hosting costs. Calls it a VC funded cancer putting the century goal of WP at risk.

Kara-Murza's take on Russia is that collapse will be swift, much like twice before, 1917 and 1991.

Again lineair and productivity/effectiveness overtones. 'learning something irrelevant' as 'wasting time and energy'? ugh. Curiosity and interestingness/surprisal can be directed with intention without being goal oriented, which seems to be the premise here.

#openvraag I wonder if Wu put relevance in the eye of the learner or not. Vgl Feynman's [[Twaalf favoriete vraagstukken 20201006163045]] vs 'society's' relevance.

This is a highly linear sketch, figure out what to learn, gather information, done. In complexity figuring out what to learn does not then give you a clear path to the 'right' information, as it doesn't exist in that form. You iterate your way forward based on pattern recog. Fractals of figuring out what to learn repeatedly along the way

http://www.rachelwu.com/Wu_2019.pdf

proposes ...adaptation is relevant for all age groups because the environment is dynamic, suggesting that learning what to learn is a problem relevant across the lifespan

reviews new research demonstrating the importance and ways of learning what to learn across the lifespan, from objects to real-world skills 2018/2019pub

Robyn Speer will no update longer Wordfreq States that n:: there is no reliable post-2021 language usage data! Wordfreq was using open web sources, but it getting pollutted by #algogens output

Robyn Speer says in his view natural language processing as a field has been taken over by #algogens And most NLP research now depends on closed data from the #algogens providers.

Reddit was another key data source for wordfreq but they too no longer provide public archives, and sell it at high prices (to the likes of the #algogens)

Twitter was a key resource for wordfreq for colloquial use of words. No longer as API shut down and the population of X is skewed to hatemongering in a way that makes it lose utility as data source.

Example of how #algogens slop pollutes corpus data: ChatGPT uses the word 'delve' a lot, an order of magnitude above human usage. #openvraag Is this to do with the 'need' for #algogens to sound more human by switching words around (dial down the randomness, and it will give the same stuff every time, but will stand out immediately as computer generated too)?

paywalled article.

Wordfreq is shutting down because LLM output on the web is polluting its data to the point of uselessness. It would track longitudinally the change in use of words across a variety of languages. Vgl human centipede epistomology in [[Talk The Expanding Dark Forest and Generative AI]] by [[Maggie Appleton]]

Obsidian is a viewer, each tab essentially a diff one. Putting the two representations each in their own tab makes it possible to work both in the text and visual version of the same file. A combination of the 'flip' that Zsolt's plugin allows plus the tabs in Obsidian.

https://web.archive.org/web/20240920194505/https://www.zylstra.org/blog/2024/09/enjoying-seamless-visual-and-text-work/

nn: visueel en text naadloos naast elkaar maakt visueel makkelijker adopteerbaar.

https://web.archive.org/web/20240919071804/https://www.myrasecurity.com/

Myra CDN, based in Germany. CDNs need to temp decrypt https traffic, potentially creating a gap in GDPR (and DSA?) compliance. Unless the CDN provider can be shown to be part of the compliance chain. Thus select one that ensures this / is based inside the EU.

Revisiting this after 2yrs, now as board member of mastodon.nl, need to read relevant DSA sections again, and think about if/how this applies. The list here as such is almost completely covered by default, except for the reports, which will come as we're heading towards ANBI status anyway.

-[ ] lees DSA met mastodon.nl bril tav kleine platforms #activityclub #30mins

Matrix seeing DMA as supportive towards their goal of open web's communication layer. Actively demo'ng Android interoperability while preserving E2EE, and participating in IETF / MIMI ( https://datatracker.ietf.org/group/mimi/about/ )

The ability to do deepwork is related to create less distraction because of discomfort during it. To keep at it, where a more gratifying small task is just a click away.

it is easy to pretend interruption is the work. Matuschak descibes focus work as a trained thing he improved upon.

This rings true. The friction, the struggle is the work, at least when it comes to my knowledge work. Interesting is that when the jump happens I tend to phrase it as an escape, a way of fleeing forward. When I got stuck in a major research project in 2020, the key insight to unlock it was a gasp of desperation more than a bolt of lightning. Colleagues immediately told me that was the key, but to me it felt like using a cheat code. Now in hindsight, I think it was the best possible outcome but that oroginal sense of escape remains.

Matuschak through focusing on specific people in a specific context, entering into a deeper emotional connection would provide meaning both in the now and over the long term. This certainly applies to my work #hazp08 and perhaps all my project that stand out in hindsight either have that or are singular efforts where the doing held such a link to myself.

I find the concepts behind my work meaningful, and enjoyable, but usualll not the work. The most enjoyable work usually is disconnected from anything else.

[[Andy Matischak]] on the value and quality of his work. Over the long haul, he found his work (at Apple) meaningless, even if it felt good at the time. The statement 'if someonee alse had done the work' the results would have been similar chimes. My work may be seen by others as meaningful in the moment, but I only see that it doesn't matter in the long run. A million others for any of us. M writes he felt confident, I never had any answer to the question what I'm good at. I just get total internal silence in response, and always have gotten.

vgl [[Sturgeons Law most is crap 20190328205135]]

this paragraph reads like making commonplacing navigable in a new way. Also turns 'snippets' into potenital entry points without them being separate notes, and pivots like tags. Note the clear spatial overtones (landmarks, being lost, navigate, ways, stumble upon, point).

Again the important distinction, here in the context which itch Latticework scratches, between 'evidence' and 'kindle' perspectives. The latter is an emotional thing, where knowledge is not an external thing, but a internal network of meaning.

I recognise this. We spent an hour in Feb. Which was fun, and useful because it was a real effort on actual notes and for an actual purpose (for me a workshop design). Then afterwards I didn't use it much due to tech hurdles, so I didn't get to experience ongoing value. Reinstalled it now because of this article. (Which [[Maarten den Braber]] pointed me to.

adjustable snippet ranges, letting your emergent insight impact your original highlighting/annotation sounds like a very interesting idea. Not because you're pinpointing the info at source more accurately, but because the emergent purpose of your sensemaking reflects back on your source material. It shifts the exact point where your Surprisal originates around.

Naming clusters can be incorporated into sensemaking efforts though, when not used as result but as intermediate step. As in [[2 step archetype extraction 20121130152904]]

exactly. Tags are emergent structure, and are not per se to describe the information stated nor to be used as a taxonomy. Vgl [[%On Tagging 20200818120917]] as associative emergence, as search/find history, as pivots in an exploratory path etc.

This is good way to make the distinction with qualitative research tools, incl. those where the narrators of qualitative bits do their own signification which then serves as filters. Tagging like those two types serve a different purpose, spotting patterns 'out there' rather than provoking thinking 'in here'. Both useful and not unrelated, but different activities. The 'evidence' vs 'kindling' metaphors make sense to me. Diff points of application.

This is where I derive value from Excalidraw as Obsidian plugin: text and image are joined in the same note. And it can have hyperlinks to othern notes, drawings, as well as embed.

The discontinuity between visual and textual has been a main issue for me for decades

this is problematic even if you ignore the obvious hurdle of having to wear overheavy ski goggles.

5 reasons L's bidirectionality is different to pkm tools / Xanadu. The first two are most key imo: they can be initiated from both sides of a link, and result in the same thing. (e.g. diff from [[Webmention 20200926203019]], and two the links are standard URLs (in standard Markdown), whereas Roam / Notion abstract them away in a db, moving outside their role as viewer, and Obs maintains its viewer role, but adds things to the notes not obviously interpretable outside of it.

I'd add that the linked snippets are a different unit of linking from what Obs et al support. Much closer to the granularity where the knowledge work is done. Surprisal I rarely find in a paragraph, mostly in part of a sentence. Questions latch onto a single word sometimes.

Bidirectionality here explicitly tied to Ted Nelson's Xanadu. Call block transclusion in pkm tool transclusion "primitives" which sound like a right charcterisation. Say Latticework differs from both

I did #2024/02/01 https://www.zylstra.org/blog/2024/02/matthew-and-andy-watched-me-test-the-obsidian-reference-plugin/ where my own similar observations are captured. - shifting from highlighting to linking as emergence occurs - structure is earned over time - keeping the copy commands straight was hard (I kept forgetting too) - preview and multiple panes is what I did too, but it seemed to clash with the plugin then - selection of snippets is not blocks, but phrases inside par's and sentences. Not whole blocks (and this is why I hardly use block links inside my notes) - after paraphrasing collapsing snippets keeps overview possible, or to collect examples, and treat list as tasks, collapsing when done.

Using Latticework does not break the 'Obsidian is only a viewer' principle cf [[3 Distributed Eigenschappen 20180703150724]]. It adds markdown style links according to W3C selector url standard. Nice, because it maintains readable plain text files.

The processing document can get as unwieldy as the source material for which it is a solution. Latticework lets you collapse stuff therefore.

This pane awareness is what seemd to clash with some other plugin I run. At least it did in feb. I notice their code repo still warns about clashes with other plugins and to run it in a separate vault with no other plugins.

adding small observations to either the foraging or sensemaking side of things is reflected in the other. Another bi-directionality. Nifty, also because this is exactly what happened to me when I tried out an early version with Matt/Andy watching. Working with material leads to new thoughts/observations which I threw in for later follow-up/expansion. It allows me to capture my conversation with a text both as annotation at source, and as refinement in the working doc.

Bi-directionality is a key feature in Latticework, which is great. At the same time, on the source end it is also ephemeral. If your remove a linked snippet from the sensemaking document, the highlight, which is just a styling element, gets removed from the source. The source is not modified to produce the highlight. (Any permanent link to source should be made consciously, which is right) Bidirectionality, other than linking seems to me a key affordance in #pkm, something that not now exists in either my annotations / reading / processing flows. #openvraag: where in my workflows would bidirectional trace leaving be useful.

Latticework sees foraging as tied to source, and sensemaking as tied to working document, and aims to make the movement between the two fluid, so you can shift focus between the two docs and thus the two activities. It leaves traces of your work in both (vgl [[Hoe emergence tot stand komt 20040513173612]] for the role of such longer traces to more easily stumble across in emergence.)

Making sense of material is not a linear process of ever more refinement, as e.g. Tiago Forte suggests with [[Progressive summarising 20200922080651]]. Siu/Matischak embrace the non-linear, recognising you go from 'foraging' (their term, great, K-garden style) to annotating, rearranging, noting an idea, back to foraging, back to rearranging etc. This is a key thing imo.

The video demo also shows adding comments to a snippet either in source or in the sensemaking document. I wonder if they put that in the released plugin (the say it doesn't do everything they did in the research project)

Intended effect of Latticework plugin (great name btw) Latticework allows you to fetch snippets in one note, and paste untalterable into another, with a link added in the original and copy. In the new note you can rearrange, paraphrase etc, purposefully add a link to source material, and otherwise do away with the snippet over time. Allows a better overview of what comes from where etc, preventing getting lost in the source material, which often happens.

As annotations already flow into my notes this helps reinforce their use.

This is the 'final' result of [[Matthew Siu]] [[Andy Matischak]] research into an Obsidian plugin for making sense of several sources into one, emerging an outline. I tested an earlier beta on #2024/02/1 [[Andy Matthew Obsidian plugin]] https://www.zylstra.org/blog/2024/02/matthew-and-andy-watched-me-test-the-obsidian-reference-plugin/ I stopped using it after a few weeks due to clashes with other plugins I could not pin down. At first glance this is a good description of the process and intended purpose. Re-installed this version of the plugin.

https://web.archive.org/web/20240918094904/https://ibestuur.nl/artikel/miljoenennota-toewerken-naar-een-digitale-overheid/

iBestuur brengt samen wat in K-plannen over digitalisering/data staat.

Via [[Lee Bryant]]

corporation from corpus. Medieval roots of corporation were people brought together in a single purpose/economic entity. Guilds, cities. Based on Roman law roots, where a corpus could have legal personhood status. Overtones of collective identity, governance. Pointer suggests a difference with how we see corporations as does the first paragraph here, but the piece itself sees mostly parallels actually. Note that Roman/medieval corpora were about property, (royal) privileges. That is a diff e.g. in US where corporates seek to both be a legal person (wrt politics/finance) and seek distance from accountability a person would have (pollution, externalising negative impacts). I treat a legal entity also as a trade: it bestows certain protections and privileges on me as entrepreneur, but also certain conditions and obligations (public transparancy, financial reporting etc.)

A contrast with ME corpus is seeing [[Corporations as Slow AI 20180201210258]] (anonymous processes, mindlessly wandering to a financial goal)

[[Difference and Repetition - Wikipedia]] v Gilles Deleuze refs The Open Work by Umberto Eco.

https://web.archive.org/web/20240916134240/https://climbtothestars.org/archives/2024/09/15/things-i-use-and-things-i-dont/

[[Stephanie Booth]] continues from an April posting about seeing spaces as there so serve me, making cleaning/tidieing a service to future-me, and talks about things you use and don't (from her ADHD perspective.) Living space as user interface.

https://web.archive.org/web/20240916134154/https://climbtothestars.org/archives/2024/04/28/my-space-is-there-to-serve-me/

[[Stephanie Booth]] on keeping things tidier around the house. from the perspective of seeing space as something that serves me. Reframes cleaning not as reactive but as pro-active in aiding my future self. I once, mid 90s, wrote a column on cleaning/decluttering as a personal battle against entropy, reducing entropy and thus postpone the heat death of the universe. Heck, the very def of life is decreasing entropy around itself, as per James Lovelock in Novacene (2019).

https://web.archive.org/web/20200730082905/https://www.chronicle.com/article/googles-book-search-a-disaster-for-scholars/ https://www.chronicle.com/article/googles-book-search-a-disaster-for-scholars/ paywalled though. The older archived link above has it in full text. Full text [[Googles Book Search A Disaster for Scholars 20240916073335]]

Nunberg on the issues with Google Books

A point worth repeating. I think there is a strong parallel here with algogens. The way 'progress' in released models is celebrated by e.g. Donald Clark. It beats a PhD exam, it does CT etc. What does comparison with their deep roots yield though? Keep history short so you may be the biggest giant of all time.

https://web.archive.org/web/20240916051503/https://newrepublic.com/article/94175/ann-blair-managing-scholarly-information

via Chris Aldrich. Ann Blair of [[Information edited by Ann Blair]]

https://web.archive.org/web/20240916043530/https://euansemple.blog/2024/09/13/bending-the-truth/

Euan Semple describes how he has to fill in an online appointment form for medical care because his doctor asked him to make an appointment, but none of the pre listed answer options match that case. It reads like #prompting #promptengineering as we do in #algogens You change the input because of a desired output, but the input itself is just that means and becomes meaningless in the process. Yet in this case that input is kept as 'truth' in a database, impacting #dataquality

https://web.archive.org/web/20240916042814/https://www.lisihocke.com/2024/05/newcrafts-paris-2024-a-memorable-conference-to-rave-about.html

Lisi Hocke on NewCrafts 2024 conf in Paris. Great images of handwritten notes too! #pkm

Saul Justin Newman's 2018 paper criticising some papers wrt longevity and aging. Says the results can be generated by having a few randomly distributed age-misreporting errors. Barbi et al's models turn out to be sensitive to that. Barbi posit their data means there's an evolutionary dimension to their aging data, whereas Newman says it's just faulty data that causes the effect. Won an 2024 Ig Nobel for this topic.

https://doi.org/10.1371/journal.pbio.3000048

Ha!

Longevity data from the UN says researcher is highly suspect, given where they report highest rate of centenarians. Those reaching are often, poor, lacking in administrative systems, or even without government. Says data is fully untrustworthy from the start.

certainly been that way through the ages, no reason it shouldn't be like that now.

Relying on documentation for age measurement is highly problematic. Yet it determines a lot in terms of pension rates, insurance and health care cost planning.

Researcher proposes developing a way to measure human age independent of documentation. (what would that be? telomeres? x-rays (like they do to determine where refugees are rightfully claiming to be underage?))

High registered age caused more likely by bad admin and fraud pressures. Worst places in terms of aging, and listing the lowest number of 90 yr olds in the UK have the highest of 100yr olds.

Greek and Italian centenarians are regularly dead / fraud cases.

Largest predictor of having many 100+ yr olds in Okinawa is the records being destroyed in WWII.

Researcher says Eurostat data in 1990 https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Mortality_and_life_expectancy_statistics#Life_expectancy_at_age_65 (note that link does not seem to give that data) list EU blue zones actually quite low. Sardinia 51st, Ikaria (Greece) 109th. Research estimates 72% of Greek old age cases are untrue or fraudulent.

Japanese so-called blue zone Okinawa actually has a very bad health and dito eating/drinking habits.

Supposedly centenarians in Japan turn out to be mostly unregistered deaths.

https://web.archive.org/web/20240915125021/https://theconversation.com/the-data-on-extreme-human-ageing-is-rotten-from-the-inside-out-ig-nobel-winner-saul-justin-newman-239023

Saul Justin Newman won an Ig Nobel for finding most claims about people living over 105 are wrong / faulty.

Blue zones wrt human aging are actually bad data zones. Either because actual birthdata is missing (war, bad administrative quality) or pension fraud is rife.

Jumped out at me as I just yday saw snippets of a docu about increasing personal longevity which visited Sardinia, one of the blue zones.

https://web.archive.org/web/20240912102433/https://www.anildash.com//2024/09/11/23-what-was-911/

Responded with a blogpost https://www.zylstra.org/blog/2024/09/24937/

I was at Penn when the news of the first strikes on Afghanistan rolled visibly through the main hall. People fell to their knees, people cried. I didn't know what caused that, until I saw a screen in a sports bar showing the news a minute later. Suddenly there were news crews all over too.

Vuurwerkramp Enschede 2000 similarly brought people together. It made me feel really rooted, a local, for the first time, although I had been in the city for 12yrs. Brought this up in my tech phil paper comparing NYC (I was there 3 weeks after) and Enschede and the coping strategies people deploy in such moments. Down to the exact same rumours going round in both situations.

Similar to Enschede Vuurwerkramp. Even though it was an accident not deliberate. Highways into town were closed off to keep gawkers away. I cycled to work 2 days after and came across lorries with soldiers being unloaded, to prevent looting after the fires went out. That never happened because it pulled the town together, and looting wasn't a response.

Still there when I was in NYC 3 weeks later.

Dash is right here. Smell, and in Enschede's case the shockwave too. The shockwave going through your body was the dividing line between those who were there that day and those who weren't. It was and is a clear tell 24 yrs on.

When I stood at ground zero a few weeks after, the smell is what made me cry then. It catapulted me suddenly back to the explosions in my home town a year before. That a smell could so abruptly and vividly surface those emotions took me by suprise.

Same in Enschede 2000 we just had to go out. https://www.zylstra.org/blog/2020/05/enschede-13-5-2000/

Anil Dash in a #2001/09/11 remembrance post describing how online 'imaginary' friends became a core part of his social life over time. This is true for me/us too, and I think something tied to that time of early blogging and the meet-ups and events that emerged from it. Vgl [[Building community out of strangers – Tracy Durnell]] and [[Nancy White]]'s birthday party in Seatlle in '08 when her neighbours greeted me as 'ooh you're one of her imaginary friends'

https://web.archive.org/web/20240911121048/https://www.baldurbjarnason.com/2024/facing-reality-in-the-eu-and-tech/

Bookmarking it belatedly, bc I find I regularly point to it when discussing the mischaracterisation by big tech of the DMA scope, regulations and purpose, esp Apple this year.

AI to teach us complexity and sensemaking / sense of wonder in viewing the world. It might, given who builds the AIs I don't think so though. Can we build sensemaking tools that seem AI to the rest of us? genAI is statistical probabilities all around, with a hint of randomness to prevent the same outcome for the same questions each time. That is not complexity just mimicry though. Can sensemaking mimic AI to, might be a more useful way?

Yes, this I can see, but that's not making AI into K, but embracing complexity and being able to adapt fluidly in the face of it. To increase agency, my working def of K. This is what sensemaking is for, not AI as such.

the jury is still very much out that they might.

AI or rather our understanding of complexity that needs to step in? The examples [[David Weinberger]] gives of general things that can't do particularised events are examples of linear generalisations failing at (a higher level of) complexity. Also I would say 'prediction' which is assumed to here be the point of K is not what it is about. Probabilities, uncertainties (which is what linear approaches do: reduce uncertainties on a few things at the cost of making others unknowable within the same model, Heisenberg style), that in complexity you can nudge, attenuate etc. I'd rather involve complexity more deeply in K than AI.

[[David Weinberger]] on K in the age of AI. AI has no outside framework of reference or context as David says is inherent in K (next to Socrates notions of what episteme takes). Says AI may change our notion of K, where AI is better at including particulars, whereas human K is centered on limited generalisations.

Companies, half of an anecdotal sample of some 20 US CDOs, have turned Copilot off / restricting it strongly. This as it surfaces info in summaries etc that employees would not have direct access to. No access security connection between Copilot and results. So data governance is blocking its roll-out.

Key observation, ecology surrounds everything. Vgl [[De Europese dataspace als eenheidsmarkt 20200120144254]] waar mensen niet snappen dat je er per def in zit.

Comes from [[On the Emergence of an Ecological Class by Bruno Latour and Nikolaj Schultz]] january 2023. Schultz is a Danish sociologist, Uni CPH and Aarhus School of Achitecture.

https://web.archive.org/web/20240802092537/http://mcld.co.uk/blog/2023/politics-in-the-new-climate-regime-a-review-of-latour-2013.html

Dan Stowell on [[Down to Earth by Bruno Latour]] I really should read more from Latour. wrt [[Latours Actor Network Theory ANT 20201129164732]]

For over 2 decades I struggle with this. Because my paraphrasing is mostly unsuited for my blog, regularly because it is mixed language and often bc it contains words that serve as shorthand. A blog is more performance, written for not-me, while annotation is for me, and after editing might be publishable for not-me. Annotating publicly here in .h, even if the readership is highly limited, introduces a level of performance-awareness for me. At times I've done annotated link blogging, but it never became a practice as with [[Stephen Downes]].

https://web.archive.org/web/20240725080148/https://fossacademic.tech/2024/02/11/Move-Slowy-Preview.html [[Move Slowly and Build Bridges by Robert Gehl]] is a forthcoming book on 'Mastodon, the Fediverse, and the Struggle for Ethical Social Media'. This post gives summaries per chapter of the draft. Ch1 focuses on Xodus after Musk only. Odd, there are many examples where costs of leaving socmed platforms played a role, which may well be more informative than just n=1. Ch 2 on AP as protocol Ch 3 CoC as a social layer on networked tech (no regard here it seems for the fact that human networks exist outside of tech and span multiple tech platforms simultaneously, and themselves have social norms that guid behaviour regardless whether codified in CoC or expressed in federation choices) Ch 4 on blocking and defederation as a needed safety tool. Socially I think the default might need to be the other way around, federating is the choice, defed the default, as it is how we do it socially irl. We are not unwelcoming to newcomers in a group but we are wary. Ch 5. Who pays for the fediverse infra. Short answer is we all do/many of us do. I pay my own instance, and also contribute hours to the governance of the largest Dutch instance. Good point about people forgetting there are other bizz models for digital media than what centralised adtech kraken do. Ch 6. on eco impact of socmed, and need of awareness what running this stuff costs ecologically. Seems to then pivot to how degrowth and solarpunk people using fediverse tech to interact, which is not the same thing. (It says mitigate, but compared to what, X? ) Ch 7. Threads , or the corp reaction to a growing fediverse. Conclusion, this is where the ethics will be discussed finally.

Forthcoming w Oxford Univ Press. Not sure this is for me, reads like a snapshot with a limited time window in which it might be informative. Perhaps of interest for [[Stichting ActivityClub Bestuur Hoofdnote]].