3,272 Matching Annotations
  1. Dec 2025
    1. Horizontal EU data legislation regulates the access to, reuse, interoperability, and governanceof public sector data in a coherent and technologically advanced manner. This includesDirective (EU) 2019/1024 (Open Data Directive)7 and Implementing Regulation (EU)2023/138 (High-Value Datasets)8, Regulation (EU) 2022/868 (Data Governance Act)9, andRegulation (EU) 2024/903 (Interoperable Europe Act)10. Horizontal EU data legislationintroduces open-by-default principles, structured metadata, mandatory ApplicationProgramming Interfaces (APIs) and where relevant as bulk download formats for high-valuedatasets, as well as a streamlined, common governance model for cross-border data us

      Tied to DA (incl DGA,ODD), HVD, and Interoperable Europe Act here. Note: #dgdigit assumes no connection w Interop act until its review.

    2. This proposal seeks to modernise and simplify the INSPIRE Directive by removing technicalrequirements for data and data sharing and aligning its obligations with more recent horizontalEU datal legislation.

      removes tech reqs --> harmonisation?

      connect to horizontal legislation --> ODD(DA) / HVD

    1. Today's simplification package is composed of six legislative proposals.

      6 legislative proposals (but press release lists 5)

      1. Environmental assessments wrt permits
      2. industrial emissions directive
      3. SCIP database (substances of concern, in the Waste Framework directive) to be replaced with DPP ( #openvraag DPP is not in effect yet, so repeal of SCIP early / protection erosion?)
      4. Extended Producer Responsibility req changed for EU producers.
      5. INSPIRE
    2. The current technical requirements for geospatial data under the INSPIRE Directive will be fully aligned with the horizontal legislation governing public sector high value geospatial data. This simplification will lower compliance costs for public authorities and facilitate access to high value geospatial data sets for all public and private users.

      INSPIRE to be aligned with HVD (as expected)

    1. https://web.archive.org/web/20251210080852/https://www.nature.com/articles/s41467-025-66634-7

      Cool! Hotmixing turns out to be the secret behind the durable Roman 'concrete'. A 2023 theory now confirmed by a Pompei construction site find.

      Researchers think this new insight may be applied in current building, and impact the climate footprint of modern day concrete.

      An unfinished Pompeian construction site reveals ancient Roman building technology in Zotero

    Tags

    Annotators

    URL

    1. On 24 November 2023, the EU and Canada launched a Digital Partnership to reinforce cooperation on digital issues. The partnership reflects a shared vision for a positive and human-centric digital economy and society. The EU and Canada agreed to work together in crucial areas such as AI, secure international connectivity, cyber security, online platforms, digital identity and digital skills. This Digital Partnership and its importance were highlighted in the New EU-Canada Strategic Partnership of the Future, adopted at the Canada-EU Summit on 23 June 2025.

      Digital Partnership started in 2023. June 2025 incorporated in strategic partnership. Now a first physical meeting at high level.

    2. EU and Canada are committed to working together on secure international connectivity, for example in 5G and subsea cables, and have agreed to explore new cable routes to strengthen global network resilience, including in the Arctic region. The EU and Canada will deepen the collaboration in priority topics such as quantum technologies, semiconductors, and high-performance computing. They also reaffirmed their commitment to resilient semiconductor supply chains and secure and sovereign cloud infrastructure and data centres.

      Covering deeper layers of the stack. So it's a digital sovereignty / autonomy pact.

    3. Both partners plan to establish a forum to facilitate joint testing of digital credential technologies, drive pilot projects, and share information. They will develop joint use cases and pilot projects towards interoperability of digital identity wallets and digital credentials and trust services.

      joint testing facility for wallets / data connectors

    4. the cooperation on digital credentials and trust services, including on technical interoperability and solutions based on digital identity wallets, the EU and Canada signed a Memorandum of Understanding on Digital Credentials and Trust Services.

      Relevant to EU wallett efforts and to trusted data transactions. This is a 2nd MoU

    5. the partners will share  best practices to accelerate AI adoption in strategic sectors such as healthcare, manufacturing, energy, culture, science and public services, and support SMEs. They committed to work together on large AI infrastructures and support industry and academia's access to AI compute capacity. They will also explore scientific cooperation on fundamental AI research, and the development of advanced AI models for the public good, including in areas such as extreme weather monitoring and climate change. In addition, the EU and Canada will set up a structured dialogue on data spaces, of particular relevance to the development of large AI models.

      Elements in the MoU: - share good practices to support adoption - collab on large AI infrastructure (Apply AI strat EU, HPC network) - collab on access to HPC (in line w AI factories in EU) - explore coop in fundamental ai research (weak) - development of AI for public good (in line w EU AI goals) - structured dialogue on data spaces (as data source for AI models) Only the last one is not immediately obviously connected to existing EU efforts and actions.

    1. The labelled solutions are:   Core Vocabularies – common semantic building blocks that help public administrations describe key concepts (such as persons, organisations or locations) in a consistent way across borders and sectors.  DCAT-AP – a common specification that makes data catalogues interoperable and helps make public sector data easier to discover and reuse.

      Two things labeled as interoperability solutions now. DCAT-AP and core vocabs. Odd as they're not 'solutions' but DCAT-AP is a standard. Risk of board repeating existing standardisation efforts?

    1. Companies will not be obliged to use the European Business Wallets. The Regulation places obligations solely on public sector bodies to accept its core functions, while companies remain free to decide whether to adopt the wallets for their commercial operations or interactions with public authorities

      wallets are voluntary, but PSBs must accept them. Asymmetry may impact adoption? ('costly to prepare but noone uses them')

    2. In parallel, the Commission will work closely with Member States and the private sector to define the technical standards and requirements for European Business Wallets through ongoing efforts under the European Digital Identity Framework and in large-scale pilot projects funded under the Digital Europe Programme such as the WeBuild consortium.

      technical standards and reqs not yet defined.

    3. It is estimated that simplification efforts will save up to €5 billion in administrative costs between now and 2029. Moreover, the European Business Wallets should unlock at least EUR 160 billion in savings for businesses each year.

      Savings projects between 2025 and 2029, odd bc it is only now in proposal. May well be 2027 before law is there, implementation another 2 yrs, so no real change until 2029/2030?

    4. The technical architecture and features for the European Business Wallets will build on the one for the EU Digital Identity Wallets.

      Business wallets will build on existing personal wallets tech and infra. What about existing entity identifiers (Digidentity e.g.)?

    5. Designed to be accessible for businesses of all sizes, including SMEs and microenterprises, as well as public administrations

      EU wallets for all types of orgs. SME's / micro entities and PSBs, are 99% of all EU entities.

    6. The European Business Wallets are one of the key deliverables of the Digital Package. The Commission proposes a regulation to establish those Wallets as a harmonised digital solution to ease the administrative burden and to enable companies and public sector bodies to identify, to authenticate and to exchange data in a secure and user-friendly way, with full legal effect across the European Union.

      EU Business Wallets proposed as regulation in the 2025 work programma. So in first step of legislative process.

    1. this type of thing sounds like what I thought wrt annotation of [[AI agents als virtueel team]]. The example prompts of questions make me think of [[Filosofische stromingen als gereedschap 20030212105451]] die al per stroming een vraagstramien bevat. Making persona's of diff thinking styles, lines of questioning. Idem for reviews, or starting a project etc.

    1. For browsers on mobile devices (including iPads) To install the Hypothesis Bookmarklet you’ll make a bookmark in your mobile browser out of any page, and then edit the bookmark, replacing the URL with some javascript we specify below. To install the Bookmarklet

      this, suggested by the team should work.

    1. Bovendien geldt er nu een adequaatheidsbesluit voor de VS (Data Privacy Framework), waardoor doorgifte naar de VS, voor zover er aan het DPF wordt voldaan, ook voldoet aan de eisen van de AVG die gelden voor doorgifte.

      DPF is ingegaan op 10 juli 2025. Daarvoor was er geen adequacy, nadat Schrems II er in juli 2020 een einde aan maakte.

    2. Minister gaat uit van papieren werkelijkheid ipv daadwerkelijke. - Ook voor geaggregeerde statistieken worden de details wel vastgelegd en doorgegeven aan Google, alleen niet getoond aan de gebruiker van Analytics. - Je kunt Google Analytics ook gewoon uitzetten, ipv zeggen dat verbieden niet kan. - Al is de groep geinteresseerden niet hetzelfde als de groep sollicitanten, is het niettemin een kleinere en dus traceerbare groep dan 'random'

    1. GoVolta heeft een optie op extra NMBS-rijtuigen en onderzoekt een uitbreiding naar Parijs in 2027. De beoogde Nederlandse stops zijn Amsterdam, Haarlem, Den Haag, Rotterdam, Lage Zwaluwe en Roosendaal. Lage Zwaluwe is bewust gekozen vanwege de gratis parkeermogelijkheden. In België hoopt GoVolta via Gent te kunnen rijden. De strategische samenwerking met het Franse Keolis – een dochteronderneming van de SNCF – moet de toegang tot de Franse markt vergemakkelijken.

      Researching an Ams-Paris route for 2027. Through Gent in B, not Brussels. Keolis is French which should help in getting the space on F rail network.

    2. Keolis verzorgt de tractie in zowel Nederland als Duitsland. GoVolta richt zich op commercie en pakketreizen; Brouwer op onderhoud. Eerder wilde GoVolta zelf spoorvervoerder worden, maar die rol wordt nu door Keolis ingevuld.

      GoVolta no longer a rail provider itself , Keolis is the transporter.

    1. GPT-OSS OpenAI's open-weight models designed for powerful reasoning, agentic tasks, and versatile developer use cases.

      GPT-OSS is by OpenAI. It is available locally in Ollama it seems in various versions.

    1. Het bedrijf ontwikkelt al een AI-naar-FPGA-platform waarmee elk AI-model kan draaien op goedkope, in de EU geproduceerde herconfigureerbare chips. Als ze hierin slagen, zou dit de afhankelijkheid van Europa van buitenlandse GPU-fabrieken volledig kunnen wegnemen, een terugkerend thema in de strategie van Vydar.

      A potential path away from NVIDIA it seems, but not at the moment, the text suggests.

    2. We maken geen eigen AI-chips”, merkte Crijnen op, “maar omdat de hardware speciaal voor dit doel is gebouwd, kunnen we toekomstige in Europa gemaakte chips gemakkelijk integreren. Die flexibiliteit is cruciaal.”

      This suggests they do use NVIDIA Jetson now, but don't need to if alternatives are available?

    3. Hun systeem wordt nu voor 100% in Europa geproduceerd en weegt 30 gram, tegenover 176 gram voor concurrenten die op Jetson zijn gebaseerd. Het heeft een stroomverbruik van 3 watt, een efficiëntieverbetering van 88%.

      a sixth in weight, power reduction from 15 to 3 Watt range.

    1. If you’re constantly just making notes without structure

      Not having an upfront determined structure, is not the same as 'making notes without structure'. Structure is emergent and earned working with the notes (linking, grouping, adding to an index / MoC etc.)

      Also: not all structure needs to be in your note taking system. Many of my structures are in my own head. I know how I phrase things, and do things, I'm pretty predictable to myself. That is a structure to work with.

    2. I don’t see obsidian as my second brain, I see it as my second subconscious.”

      I think the Second Brain metaphor is getting to the point of creating damage, not promise.

    3. a decent use case for some kind of personal LLM

      yes, vgl [[Communicating with Slip Boxes by Niklas Luhmann]] as convo partner. Using a local LLM makes total sense to me.

    4. the best thing to do is just write the note. It’s a bit like throwing the text into a big bucket, and that’s okay.

      That's all there is to it indeed. Jot a note, perhaps add a link / choose a single physical location.

    1. The Apertus models also expand multilingual coverage, training on 15T tokens from over 1800 languages, with ~40% of pretraining data allocated to non-English content. Released at 8B and 70B scales, Apertus approaches state-of-the-art results among fully open models on multilingual benchmarks, rivalling or surpassing open-weight counterparts

      Apertus is trained on over 1800 languages (!?) with 40% non English content, meaning many of them can only have had 1/100 or 1/1000 of a procent (1/10k, 1/100k) 60/1799 is 0,033%

    1. Gettier claims to have shown that the JTB account is inadequate because it does not account for all of the necessary and sufficient conditions for knowledge.

      Edmund Gettier, Gettier problem, epistemology. 1963 paper.

      holds that the K def of 'justified true belief' is inadequate. Encountered in Is Understanding A Species Of Knowledge? in Zotero which posits that understanding is a species of knowledge that 'can be Gettiered'

    1. Binnen MeshCore is het niet alleen het apparaat, maar vooral de software die een belangrijke rol speelt. Bij het inrichten van een repeater dienen we dan ook bij de flasher te kiezen voor de rol repeater.

      A repeater is only different in software from a terminal.

    2. Eind 2025 is er een flinke verhuizing gaande van Meshtastic naar MeshCore. Dat heeft er o.a. mee te maken dat MeshCore een betere ontvangst lijkt te hebben en zijn de eerste signalen dus ook positief. Steeds meer eigenaren van repeaters zetten er dan ook MeshCore-firmware op en stappen daarmee over naar MeshCore.

      Implies there's a migration wave from Mehstastic to Meshcore

    1. The model is named Apertus – Latin for “open” – highlighting its distinctive feature: the entire development process, including its architecture, model weights, and training data and recipes, is openly accessible and fully documented.

      Apertus committed to openness wrt all its aspects. Is it in the overview yet?

    1. However, I also believe that therapy should be about the client; there’s a line between self-disclosure and self-indulgent

      when is it a conversation, when are you making it about yourself only?

    2. Dr Yalom talks about this as well in the book and argues that self-disclosure can be useful to building connection, the thing that accounts for 80% of success in therapy.

      Sessions also contain self-disclosure by Yalom, against therapeutic condition, but here positioned as way to connect. (and I think regular interaction between ppl.)

    3. focusing on the ‘here and now’. Often, he picks up on the dynamic between him and the client and raises this with the client, linking it with the issue that brought them to him. Most of the time, this is effective.

      Yalom makes the meeting between the original request and the current setting / dynamic in the conversational pair the thing to explore.

    4. The book is a collection of stories of these one-off sessions, and they are fascinating to me as a psychologist. How effective can one hour of conversation be? It is true that Dr Yalom (or Irv, as he asks his clients to call him) does not promise to resolve his clients’ needs in one hour; in fact, every consultation is ended with him providing names of other clinicians and urging the client to continue with longer-term therapy.

      Yalom during Covid engaged in 1 hour sessions. Not as self-contained therapy, but still self-contained one-off conversations.

    1. It was, however, the usage of Franz Brentano (and, as he later acknowledged, Ernst Mach[6]) that would prove definitive for Husserl.[14] From Brentano, Husserl took the conviction that philosophy must commit itself to description of what is "given in direct 'self-evidence'."[15]

      Husserl wrt phenomenology based himself on Franz Brentano and Ernst Mach . p7 Moran, Dermot (2000). Introduction to Phenomenology. Routledge.

      Introduction to Phenomenology in Zotero

    1. Support for jumping to and displaying paper edition page numbers [7.17] If an e-book contains embedded information about the pages from the paper edition of the book, the calibre E-book viewer can now jump to a page number via the Go to button in the viewer controls.

      Calibre, if it can detect it, can use paper edition page numbers. It's #openvraag how I want to deal with location numbers in ebooks as ref.

    2. [8.11.1] Supports hundreds of AI models via Providers such as Google, OpenRouter, GitHub and locally running models via Ollama.

      Calibre supported Ollama since 8.11, for Ask AI tab in dictionary panel.

    3. New features Allow asking AI questions about any book in your calibre library. Right click the "View" button and choose "Discuss selected book(s) with AI" AI: Allow asking AI what book to read next by right clicking on a book and using the "Similar books" menu AI: Add a new backend for "LM Studio" which allows running various AI models locally

      AI features in Calibre. discuss book w C, book suggestions, and LMStudio back-end. I set up Calibre w the LM Studio back-end, so things remain loca.

      A posting elsewhere suggested it woud also suggest better metadata through AI. But that article seems generated itself, so disregarded.

    1. Het zijn markdown bestanden met een persoonlijkheid, frameworks, en output templates. Die heb ik niet zelf geschreven - ik heb Claude gevraagd om ze te maken. “Maak een Product Owner agent die goed is in prioriteren en impact/effort analyses kan doen.” Claude schrijft dan het volledige bestand, inclusief werkwijze en voorbeelden.Als ik vervolgens zeg “vraag dit aan Tessa”, laadt Claude dat bestand en wordt Tessa.

      Seems like these agent .md files contain description of a role that is then included in a prompt.

    1. In mijn werkmap heb ik een verzameling “agents” - tekstbestanden die Claude vertellen hoe hij zich moet gedragen. Tessa is er één van. Als ik haar “laad”, denkt Claude vanuit het perspectief van een product owner.

      Author has .md files that describe separate 'agents' she involves in her coding work, for each of the roles in a dev team. Would something like that work for K-work? #openvraag E.g. for project management roles, or for facets you're less fond of yourself?

    Tags

    Annotators

    URL

    1. Luhmann’s card structure:Front: Complete bibliographic information (author, title, publication year, etc.)Back: Page index = “Page x has this concept, page y has that argument”

      States that Luhmann's bibliographic cards had his own indexing on the back. Vgl with my own immediate annotations I keep in a book note.

    1. individuals constantly project into the world around them the interpretive frames that allow them to make sense of it; we only shift frames (or realize that we have habitually applied a frame) when incongruity calls for a frame-shift.

      Vgl brain continuously predicting/projecting, not observing out of efficiency. Predictive coding.

    2. When we want to explain an event, our understanding is often based on our interpretation (frame). If someone rapidly closes and opens an eye, we react differently based on if we interpret this as a "physical frame" (they blinked) or a "social frame" (they winked). The blink may be due to a speck of dust (resulting in an involuntary and not particularly meaningful reaction). The wink may imply a voluntary and meaningful action (to convey humor to an accomplice, for example). Observers will read events seen as purely physical or within a frame of "nature" differently from those seen as occurring with social frames

      Applying different frames yield diff interpretations. Humor often based on frame switch.

    3. Framing is an integral part of conveying and processing data daily. Successful framing techniques can be used to reduce the ambiguity of intangible topics by contextualizing the information in such a way that recipients can connect to what they already know.

      Vgl [[Social netwerk als filter 20060930194648]] , [[Chunking 20210312215715]

    1. Guardrail: Strict adherence to the Tagging Workflow. Everything defaults to #private. Only notes explicitly tagged #share pass through the publishing filter. Use clear folder structures to separate sensitive data that should never touch the publishing pipeline.

      Have this for blog posts, separate folder and explicit status for publishing. A published garden first as a separate folder of cloned notes, then as tagged for publication? Needs a 'touching' process to see diverging note/clone.

    2. Publishing is Pausing: You push a rough idea out to pause your thinking and invite external input.

      Like the phrase. Vgl [[Kunst-artefact is (tussen)uitkomst proces 20140505070232]] any artefact is a (intermediate) output of ongoing process. Vgl [[Artefacten uit kennis met methoden 20220215202011]], artefacts brought forth from K through methods.

    3. What Is a Federated Garden? A federated digital garden is a personal knowledge space that: Starts private (your vault, your chaos, your unfinished thoughts) Grows selectively public (some notes bloom into garden pages) Connects across platforms (one source → many destinations) Links bidirectionally (gardens can discover and reference each other) Key principles: Gradual disclosure: Not everything needs to be public—choose what to share Version multiplicity: A note can exist as a draft, a garden page, a newsletter piece, and a toot Decentralized ownership: You control the source, not a platform Interoperability: Uses open standards (Markdown, RSS, ActivityPub, HTML)

      The page confuses tools and content imo, but this is a workable list of attributes. 'Grows selectively public' is the key imo. #openvraag #webbeheer What can I do to improve the note to webpage pipeline I have but seldom use?

    1. Ontdek hoe wij bijdragen aan een veiliger Nederland en hoe u deel kunt uitmaken van ons team van gepassioneerde radioamateurs.Door het Ministerie van Binnenlandse Zaken is aan de veiligheidsregio’s geadviseerd om de dienstverlening van DARES op te nemen in de rampenplannen.

      The DARES volunteers are actively preparing for emergency situations. Veiligheidsregio's sign convenants with them.

    Tags

    Annotators

    URL

    1. ESP32​ The ESP32 chip is older and consumes more power than the nRF52 chip, but is equipped with both WiFi and Bluetooth. Supported ESP32 devices include:

      ESP32 MCU supports both wifi and BT. Higher power consumption though.

      In general, as this network is supposed to provide alternative comms when regular networks break down: how reliant on battery / solar powered devices is the network?

    1. 530-B HARKLE ROAD, STE 100, Santa Fe, NM, 87505, USA

      Meshtastic LLC is registered in Sante Fe, New Mexico, USA. That information is not contained in the meshtastic website.

      The head office address given is that of a registration agent though: New Mexico Registered Agent by High Desert Corporate Filings LLC

    1. When you send a message on your Meshtastic companion app, it is relayed to the radio using Bluetooth, Wi-Fi/Ethernet or serial connection. That message is then broadcasted by the radio. If it hasn't received a confirmation from any other device after a certain timeout, it will retransmit the message up to three times. When a receiving radio captures a packet, it checks to see if it has heard that message before. If it has it ignores the message. If it hasn't heard the message, it will rebroadcast it. For each message a radio rebroadcasts, it marks the "hop limit" down by one. When a radio receives a packet with a hop limit of zero, it will not rebroadcast the message. The radio will store a small amount of packets (around 30) in its memory for when it's not connected to a client app. If it's full, it will replace the oldest packets with newly incoming text messages only.

      You use your phone or 'companion app'(?) to send a msg to a radio (over BT, wifi or wire). The radio broadcasts incoming messages, including the one you provide through the app.

      Msgs that are not acknowledged by another radion will be send at most 3 times. (will you be able to see it has not propagated?)

      A radio that receives msgs already received will not rebroadcast it. Any broadcasted msg has a 'hop limit' and if it hits 0 it will not be rebroadcast. This limits the spread of a message, no? What is the default hop limit? Otoh the hoplimit does not limit the initial number of paths for broadcasting. So it's an attenuation over paths.

      Theoretically in a dense network, my msg may reach Y number of other radios that all start out with the same hop limit.

      I do not see here yet how you could intentionally set and reach a specific recipient. This description provides attenuated propagation of messages but no direction/addressee?

    1. Additionally, Meshtastic radios can be paired with a single phone, allowing friends and family to send messages directly to your specific radio. It's important to note that each device is capable of supporting a connection from only one user at a time.

      Any meshtastic radio can connect a single phone for receiving / sending (?) messages on. Apart from repeater function 1 single connection for actual comms input?

    2. Meshtastic utilizes LoRa, a long-range radio protocol, which is widely accessible in most regions without the need for additional licenses or certifications, unlike ham radio operations.

      Meshtastic positions itself as a radio protocol without need for licensing such as in ham radio. Meaning it's fully in parallel to ham emergency networks like the Dutch DARES.

    1. Hieronder de lijst van AI-boeken die ik gelezen heb en je aan kan raden. Klik meteen door naar de langere omschrijving of scroll verder. Ze staan op de volgorde waarin ik ze uitgelezen heb: Weapons of Math Destruction: over desastreuze algoritmes Code Dependent: over de achterkant van AI Onze kunstmatige toekomst: over de etische kant van AI Empire of AI: over de opkomst van OpenAI Your face belongs to us: over de opkomst van ClearView AI Atlas van de digitale wereld: over de geo-politiek van AI The Digital Republic: over het reguleren van technologie Toezicht houden in het tijdperk van AI: over de juiste vragen stellen over AI

      [[Elja Daae]] recommended reading list wrt AI [[Weapons of Math Destruction by Cathy O Neil]] (have it since 2017) [[Code Dependent by Madhumita Murgia]] bought it in August in ramsj [[The Digital Republic by Jamie Susskind]] I noted in 2024 as possible reading. [[Atlas van de digitale wereld by Haroon Sheikh]] I have too Other's are unknown to me. Interesting list, as it shaped their view on their role in AI public policy I presume

    2. The Digital Republic van Jamie Susskind (2023) is nogal een boekwerk. Niet iets dat je in een avondje uitleest. Maar het is wel een heel belangrijk boek, want het gaat over de vraag hoe we als samenleving technologie kunnen reguleren.

      [[The Digital Republic by Jamie Susskind]], already jotted down the title [[Daglog 29-10-2024]] at Dussmann's in Berlin, 2023 book. [[Elja Daae]] recommends it in this list. - [ ] check for review / summary [[The Digital Republic by Jamie Susskind]] #digitalpolicy #reading

    1. https://web.archive.org/web/20251202103136/https://www.marketplace.org/story/2024/10/24/private-equity-buying-up-businesses-in-the-skilled-trades-hvac-plumbing-electrician

      #2024/10 article describing how private equity firms are buying up skilled trades companies. Plumbers, solar panel / heat pump installers etc. A weak signal imo. Would you want private equity in real world entities? It's one thing for 'new fields' or scale-ups etc. Another when extraction hits trades we rely on (although easy enough for other entrants to help circumvent)

      (via [[Chris Aldrich]])

  2. Nov 2025
    1. American cloud giants Amazon, Microsoft and Google, as well as Chinese tech giants Huawei and Alibaba, are all members of Gaia-X. In 2021, the annual summit in Milan was sponsored by Huawei and Alibaba, prompting backlash.

      Same is happening wrt IDSA, who have a global orientation, but are treated as EU grouping, which they're not.

    2. Those firms “steered the entire roadmap,” Lechelle said, throwing money and people at it. “The committees were drowning. They [global players] had the capacity, the bandwidth, but we were already underwater ... Americans have full-time lobbyists and massive budgets. Their job is basically to derail any initiative they don’t like.”

      Key friction. One cannot rule out non-EU parties mostly, esp their EU entities. But presence often obstructive / malicious compliance. You'd need much better governance / rulebooks upfront to flag and remove.

    3. of a data space based on Gaia-X standards that French energy company EDF will use to securely coordinate the construction of new nuclear sites

      example of DS project, here led by EDF for nuclear plant construction. Link w energy DS program? -[ ] zoek naar EDF DS voorbeeld voor bouw kerncentrales #geonovumtb/ds

    4. That’s how the mission to create a “federated cloud infrastructure” came to life. But that “staggering complexity” would soon turn into an “unmanageable mess,” said Lechelle.

      federated cloud concept originates in Gaia-X article says. Don't know if that's right. In itself that does not create 'staggering complexity' though. It is a different design path.

    5. “I joined Gaia-X because I believed in the original mission. I left Gaia-X because I didn’t believe it was going in the original direction,” said its former CEO, Francesco Bonfiglio.

      Gaia-X moved the goalposts original member says.

    6. The results we’re providing and the real business benefits these interoperable data spaces are creating are more and more visible,” he said, highlighting the example of a data space based on Gaia-X standards that French energy company EDF will use to securely coordinate the construction of new nuclear sites.

      again an impact example that is still in the future, a planned result, not an achieved one.

    7. Current CEO Ulrich Ahle, who joined in 2023, pushed back — saying Gaia-X is far from a “failure.” It has united the industry — both large and small players — around tangible deliverables, such as federated data spaces and compliance labels, he said.

      federated data spaces are not an outcome of gaiax, are they? EC induced. Do note that deliverables after all those years are still in the future.

    1. On the erosion of middle class America. Poverty line is around 140k if actual costs taken into account. The 1960s benchmark assumed cost of food to be 1/3 of overall costs. Now it's 7% or so, meaning 1/15 of overall costs. This pushes up the poverty line to 5 times the level used, or some 150k USD pa

      Example of a proxy being used as 'measurement' and the assumptions in a proxy never re-evaluated.

    1. This page lifts quite a bit out of my descriptions of networked agency, a term I coined 2016, without attribution. It reads like a generated text. Unclear too what the point of the site is at all.

    1. Resumability is not a word, but it’s an important concept to me. When I say Resumability, I’m talking about the ability to quickly interrupt and later resume a task.

      [[Steven Garrity]] coining 'resumability' , where a device just continues on where you were previously. Mentions e-readers as such devices (and paper books with a bookmark), a smartphone and apps like Slack. The latter is not true I think, it scrolls fwd to newest.

      Thinking about resumability in terms of notes / pkm. Obsidian is always where I was previously e.g. A type of ratchet for task execution.

    1. For example, you may need a working shop or a working painting studio. You may need a working music studio. Or a computer room where you can write something. It’s crucial to have a setup, so that, at any given moment, when you get an idea, you have the place and the tools to make it happen.

      Quote David Lynch https://en.wikipedia.org/wiki/Catching_the_Big_Fish continued.

      Mentions examples of set-ups for different activities. Vgl spaces on laptop. Vgl my home office vs attic space etc.

      Vgl [[%index coachingboekje]] wrt set-ups I need/want/woud like to have or list of set-ups currently available to me.

    2. If you don’t have a setup, there are many times when you get the inspiration, the idea, but you have no tools, no place to put it together. And the idea just sits there and festers. Over time, it will go away. You didn’t fulfill it—and that’s just a heartache.

      Quote David Lynch https://en.wikipedia.org/wiki/Catching_the_Big_Fish continued.

      Not having a set-up ready when an idea hits means no agency. Mentions festering a type of powerlessness, vgl [[%I Networked Agency]] wrt methods and tools for various things. Also vgl making note of any idea in 3Ideeenkweekkas

    1. Lynch's idea that "ideas are like fish. If you want to catch little fish, you can stay in the shallow water. But if you want to catch the big fish, you've got to go deeper". To Lynch, going deeper means experiencing a deeper, more expanded state of consciousness, a transcendental or fourth state of consciousness,[2] an experience he has during meditation but believes is rare in ordinary daily life.[6] According to Lynch, this experience expands artistic capacity.[4]

      Source of title Catching the Big Fish. Aside from the meditation angle, this points to practice / reflection, ratchets, and [[Holding questions 20091015123253]] etc.

    1. Funny. Cryptographers club held elections, but cannot access results as one of the three keys needed to see results was lost by one of the people involved. They did not design for human failure / lapses in operational precision. Typical. Vgl the cartoon with laptop w unbreakable login and person taking a bat to the one who knows the login.

    1. Jarche shares 14 ways to acquire knowledge from the quintessential PKM practicer, Maria Popova at The Marginalian, and her review _You Can Do Anything_ by James Mangan, written in 1936. He then categorizes the methods in terms of how they align with PKM in this graphic from Jarche:

      Maria Popova https://en.wikipedia.org/wiki/Maria_Popova http://themarginalian.org/

      [[You Can Do Anything by James Mangan]] 1936 review: 14 ways to acquire K. https://www.themarginalian.org/2013/04/22/14-ways-to-acquire-knowledge-james-mangan-1936/ "prolific self-help guru and famous eccentric" https://archive.org/details/bwb_W8-ANG-369/ can be borrowed.

    1. Jorge Arango talks about his writing process. He actually co-authored a book on Information Architecture in 2015. This vid more about his latest. uses lots of paper notebooks Mentions 3 (obvious) phases in his non-fiction writing research structuring writing Says all three req diff tools. (he uses Obsidian f research)

      Stage 1 research, vague idea about what topic, (s paper notebooks, index/sequence them) His book duly noted is on this phase.

      Stage 2 structuring. mistake to go directly to writing. (vgl [[A System for Writing by Bob Doto]] ) Mentions author Robert Caro who had book structure on his wall story book like, vgl what I do when writing reports w wall of post its for structure. Limitation is size of wall. But physcial movement helpt. He does this type of wall in [[Tinderbox]] !!!! canvas, with the sticky notes having metadata. His chapter and heading structure emerges from it! Prevents getting lost in the details. Shows him what is missing too (looping back to research phase sometimes). It's not writing but getting a sense of the narrative flow of the entire thing. Seems like outlining of talks, but a book is bigger and thus the outlining is more filled out with links to material

      What question would a reader have entering chapter 2 after reading chapter 1, and go from there. Has opening story/anecdote for each chapter. Mwah. Uses the story board view like how the periodic table suggested the existence of atoms that had not been discovered yet. Dmitri Mendleev.

      Stage 3 writing After structuring the writing tools come into play. He used Scrivener, avoiding having a monolithic text. Now Ulysses, in markdown. Uses word count targets in these tools for sections/chapters

      set up supportive environments for each pahse structure first but keep flexible, don't make it a linear process, but structure ulnocks the writing swith modalities - walking, free form in [[Tinderbox]] or another tool, to avoid getting stuck.

    1. This transition is signaled by focused efforts from several major scientists and technology entities. Meta Chief AI Scientist Yann LeCun has emphasized his intent to pursue world models, while Fei-Fei Li’s World Labs has released its Marble model publicly. Concurrently, Google is testing its Genie models, and Nvidia is developing its Omniverse and Cosmos platforms for physical AI.

      Various examples of world model work: Nex to Yann LeCun. Fei-Fei Li World Labs w Marble model, Google has Genie models, Nvidia Omniverse and Cosmos.

    1. On Tuesday, news broke that he may soon be leaving Meta to pursue a startup focused on so-called world models, technology that LeCun thinks is more likely to advance the state of AI than Meta’s current language models.

      Yann LeCun says world models more promising. What are world models?

    1. De bank maakte woensdag ook cijfers over het derde kwartaal bekend. Daarin boekte ABN een winst van 617 miljoen euro, op een omzet van 2,16 miljard. Beide zijn wat lager dan in dezelfde periode het jaar ervoor.

      Wut? 617/2160 is 28,5% winst in het 3e kwartaal. 'lager dan vorig jaar'. Het lijkt me een heel hoog winstpercentage.

    1. [[Links]] like a wiki (except they are [[Linked Mentions|bi-directional]]).

      The bidirectional takes the form of linked mentions underneath a linked to note. That's common even in wiki I think these days. Any tools out there that link closer to like Ted Nelson proposed? Within a pkm tool bidirectional linking into a specific spot in a note should be somehow doable. Transclusion is a step further even, but if I can link to a paragraph, that paragraph should be able to link to where it is cited, no? In Obsidian the ^ link is still one way, while origin and source are known for it to work (the reference number)

    Tags

    Annotators

    URL

    1. George Gilder. Reagan era supply side economist, 85yo, Called the opportunities of internet tech relatively early, and wrote a book about Carver Mead and CMOS chiptech. Founder of the Discovery Institute and loud on 'intelligent design'. Not sure any of that qualifies him to talk about wafer sized chips systems.

    1. George Gilder indeed calls for wafer scale compute. Article mentions the issues of one flaw meaning having to toss the wafer (as opposed to 1 chip cut from the wafer). Also says that bc of that wafer solutions are for highly specialised set-ups only (bc it's costly to end up with one flawless wafer).

    1. WSJ article contends that chips will soon be obsolete, and replaced by wafers that function as a fully integrated system. I have questions (and most of the article is missing). Wafers are error prone, and while you can toss out individual chips cut from wafers, having to throw out an entire wafer is costly. Other factor is that whatever size you will have surrounding electronics too, making wafers unwieldy / brittle parts of a system. Also wondering about heat. Talks about 'suitcase sized datacenters'. I wonder what the technology jump is for that to be viable

    1. n our latest findings, the share of respondents reporting mitigation efforts for risks such as personal and individual privacy, explainability, organizational reputation, and regulatory compliance has grown since we last asked about risks associated with AI overall in 2022.

      did they also ask whether those mitigation efforts negate gains in efficiency / innovation reported for AI?

    1. AI checking AI inherits vulnerabilities, Hays warned. "Transparency gaps, prompt injection vulnerabilities and a decision-making chain becomes harder to trace with each layer you add." Her research at Salesforce revealed that 55% of IT security leaders lack confidence that they have appropriate guardrails to deploy agents safely.

      abstracting away responsibilities is a dead-end. Over half of IT security think now no way to deploy agentic AI safely.

    2. When two models share similar data foundations or training biases, one may simply validate the other's errors faster and more convincingly. The result is what McDonagh-Smith describes as "an echo chamber, machines confidently agreeing on the same mistake." This is fundamentally epistemic rather than technical, he said, undermining our ability to know whether oversight mechanisms work at all.

      Similarity between models / training data creates an epistemic issue. Using them to control each other creates an echo chamber. Vgl [[Deontologische provenance 20240318113250]]

    3. Yet most organizations remain unprepared. When Bertini talks with product and design teams, she said she finds that "almost none have actually built it into their systems or workflows yet," treating human oversight as nice-to-have rather than foundational.

      Suggested that no AI using companies are actively prepping for AI Act's rules wrt human oversight.

    4. We're seeing the rise of a 'human on the loop' paradigm where people still define intent, context and accountability, whilst co-ordinating the machines' management of scale and speed," he explained.

      Human on the loop vs in