+13
_journals/2025-06-17_0820.md
+13
_journals/2025-06-17_0820.md
···+"…because the avocado slicer is narrowly focused on one task, it’s useless at anything else. If you used a specialized gadget for every single task, you’d end up with a mountain of plastic."
+11
_journals/2025-06-17_0845.md
+11
_journals/2025-06-17_0845.md
···+[[Why Centralized AI Is Not Our Inevitable Future]] is a response to [Sam Altman's "gentle singularity"](https://blog.samaltman.com/the-gentle-singularity), written by [[Alex Komoroske]], founder of [[Common Tools]]+> "between hyper-centralized systems that inevitably tend toward extraction and manipulation, versus distributed systems that enhance human agency and preserve choice."
+9
_journals/2025-06-17_1119.md
+9
_journals/2025-06-17_1119.md
···+[[Devine Lu Linvega|Devine]] wrote notes on [[Malleable Computing XXIIVV|malleable computing]] as a reflection on the [[Malleable Software Essay|Ink & Switch Malleable Software essay]].
+8
_notes/Alex Komoroske.md
+8
_notes/Alex Komoroske.md
···+CEO and co-founder of [[Common Tools]]. He was previously Head of Corporate Strategy at Stripe and a Director of Product Management at Google.+He posts [weekly reflections](https://docs.google.com/document/d/1GrEFrdF_IzRVXbGH1lG0aQMlvsB71XihPPqQN-ONTuo/edit?tab=t.0#heading=h.2f6vox7vwou3).
+15
_notes/Apps Are Avocado Slicers.md
+15
_notes/Apps Are Avocado Slicers.md
···+> …because the avocado slicer is narrowly focused on one task, it’s useless at anything else. If you used a specialized gadget for every single task, you’d end up with a mountain of plastic.+> How does this analogy apply to software? **Many applications are avocado slicers.** They’re a bundle of functionality targeted at some specific use case: planning a trip, tracking workouts, organizing recipes. Because an app needs to handle many tasks associated with a use case, it sometimes doesn’t handle any of them particularly well. You may have come across situations where an app is missing some functionality that’s important to you, while simultaneously including extra bits you don’t need.+> On top of that, solving a larger task using multiple applications often requires manual coordination. We can put windows next to each other and copy-paste data, but not much more. If we want more knife-like software tools, we’ll need better ways for smaller software tools to work together.+> **How might we reorient software around more general, composable tools**—that feels more like a knife and less like an avocado slicer? There are two sub-problems to address: sharing data between tools, and combining tools within the user interface.
+13
_notes/Blaine.md
+13
_notes/Blaine.md
···
+8
_notes/Common Tools.md
+8
_notes/Common Tools.md
+8
_notes/Devine Lu Linvega.md
+8
_notes/Devine Lu Linvega.md
···+Vagabond, lifelong student and co-founder of [Hundred Rabbits](https://wiki.xxiivv.com/site/hundred_rabbits.html). Currently drifting somewhere along the foggy coast of the [Pacific North West](https://100r.co/). I enjoy [living on the water](https://wiki.xxiivv.com/site/pino.html), moving with the seasons, growing food in far flung places, and [reading](https://wiki.xxiivv.com/site/reading.html) by the woodstove when it rains.
+85
_notes/Longform Lexicon for Knowledge Graph Notes.md
+85
_notes/Longform Lexicon for Knowledge Graph Notes.md
···+Bluesky is the biggest user of the AT Protocol, and their definition of a shortform, microblog Lexicon limits each post to 300 characters[^chars].+The [WhiteWind](https://whtwnd.com/) blog was one of the first third-party lexicons to exist on AT Protocol, defining a longform Lexicon storing Markdown posts. There is continued interest in a variety of longform lexicons, and my personal interest is in notes that can form a knowledge graph.+My practice of writing linked notes rather than more article-like content goes back to 2020 for my own website. There has been a rise in awareness and usage of grandly named approaches like tools for thought, second brains and so on, that can be more simply be called notes. Everybody takes notes in one way or another!+We can name [[Notion]] as an extremely commercially successful version of fancy notes, and a long tale of fan favourites that are more local, indie, and/or open source like [[TiddlyWiki]], [[Roam]], [[Obsidian]], and [[LogSeq]].+The challenge with these systems is that ultimately they are all single-user systems. I have my pile of notes, which I do funky things to [turn into a website](https://bmannconsulting.com). Obsidian has Obsidian Publish, where they do funky things to turn your notes into a website.+But what if we skipped the part where we turned things into "web" pages, and created a network of linked notes forming a knowledge graph, directly? And that graph included the user accounts of note creators, as well as note likers and blockers and other social and other metadata.+The first step of putting Knowledge Graph Notes on ATProto would be to define a longform format that can be used, which is called a Lexicon in ATProto.+Ideally, this would be a very minimal format that works for many uses cases. Here are some initial notes on field definitions:+- plain text? markdown flavoured? or support / extend bsky’s rich text facets for links and @-mentions? (how ugly would this be in plaintext?)+- the logic is something like “look for other notes owned by this user with this name and link to it”+- people have strong feelings about tags! eg no spaces vs spaces, no tags only mentions of other notes, folksonomy vs tag management, etc.+Storing publicly published notes where you've published them to your own [[PDS]] is a nice start, but it's not very interesting without search and browse.++This image is a screenshot of my Obsidian notes, highlighting the ATProtocol node and what all is linked to it, and you can see 100s of other notes in the background.+Being able to browse an individual's notes, see notes linked across to other user or topical graphs, and generally to explore data in new ones would be+I also admit that I think that "cool visualization of your data" and "publish data in this format and you'll show up in this interface" are motivators that will make people want to experiment with this format.+Obsidian is a widely adopted tool for making knowledge graph notes. It also has a wide array of community plugins (including one that can [post to Bluesky](https://bsky.app/profile/bmann.ca/post/3lrnpl25kpc2j)).+Obsidian has its [commercial Publish service](https://obsidian.md/publish), as well as many other workflows in order to publish to the web.+Being able to directly publish to the KGNotes ATProtocol format means that each user's notes could be directly connected to the graph of everyone else's public notes.+I previously kicked off discussion by reaching out directly to a number of people & projects interested in integrating with ATProto and wanting something vaguely longform notes shaped.+[^chars]: yes yes, not exactly characters, rather [graphemes because of Bluesky rich text and facets](https://docs.bsky.app/docs/advanced-guides/post-richtext)
+12
_notes/Malleable Computing XXIIVV.md
+12
_notes/Malleable Computing XXIIVV.md
···+> Every layer of a malleable computing system is designed to support arbitrary recombination, reused across environments and freely sharable with others. Modifying such a system happens in the context of use, rather than through some separate tool and skill set.+> In contrast with monolithic applications, these amorphous computers are finely tuned and thoughtfully crafted by their operator over the course of a practice.+By [[Devine Lu Linvega]], who also has entries for [permacomputing](https://wiki.xxiivv.com/site/permacomputing.html), [salvage computing](https://wiki.xxiivv.com/site/salvage_computing.html), and [collapse computing](https://wiki.xxiivv.com/site/collapse_computing.html).
+1
-7
_notes/Malleable Software Essay.md
+1
-7
_notes/Malleable Software Essay.md
···#### [Apps are avocado slicers](https://www.inkandswitch.com/essay/malleable-software/#apps-are-avocado-slicers)-> …because the avocado slicer is narrowly focused on one task, it’s useless at anything else. If you used a specialized gadget for every single task, you’d end up with a mountain of plastic.-> How does this analogy apply to software? **Many applications are avocado slicers.** They’re a bundle of functionality targeted at some specific use case: planning a trip, tracking workouts, organizing recipes. Because an app needs to handle many tasks associated with a use case, it sometimes doesn’t handle any of them particularly well. You may have come across situations where an app is missing some functionality that’s important to you, while simultaneously including extra bits you don’t need.-> On top of that, solving a larger task using multiple applications often requires manual coordination. We can put windows next to each other and copy-paste data, but not much more. If we want more knife-like software tools, we’ll need better ways for smaller software tools to work together.-> **How might we reorient software around more general, composable tools**—that feels more like a knife and less like an avocado slicer? There are two sub-problems to address: sharing data between tools, and combining tools within the user interface.+A lot of people love this section, broke it out into it's own page: [[Apps Are Avocado Slicers]]
+1
_notes/OAuth.md
+1
_notes/OAuth.md
···> OAuth 2.0 is the industry-standard protocol for authorization. OAuth 2.0 focuses on client developer simplicity while providing specific authorization flows for web applications, desktop applications, mobile phones, and living room devices. This specification and its extensions are being developed within the [IETF OAuth Working Group](https://datatracker.ietf.org/wg/oauth/about/).
+14
_notes/WebFinger.md
+14
_notes/WebFinger.md
···+> WebFinger is used to discover information about people or other entities on the Internet that are identified by a URI using standard Hypertext Transfer Protocol (HTTP) methods over a secure transport. A WebFinger resource returns a JavaScript Object Notation (JSON) object describing the entity that is queried. The JSON object is referred to as the JSON Resource Descriptor (JRD).+> For a person, the type of information that might be discoverable via WebFinger includes a personal profile address, identity service, telephone number, or preferred avatar. For other entities on the Internet, a WebFinger resource might return JRDs containing link relations that enable a client to discover, for example, that a printer can print in color on A4 paper, the physical location of a server, or other static information.
+12
_notes/Why Centralized AI Is Not Our Inevitable Future.md
+12
_notes/Why Centralized AI Is Not Our Inevitable Future.md
···+<blockquote class="bluesky-embed" data-bluesky-uri="at://did:plc:qxeeegphuasuwv2oeyqe6byd/app.bsky.feed.post/3lrqcdoprn22y" data-bluesky-cid="bafyreigefsscnyvtipr3zg377dp6xzvcsic7qbqqvkqhk5ixs4qy53rr2a" data-bluesky-embed-color-mode="light"><p lang="en">AI has the potential to have as much positive impact for humanity as the printing press, electricity, or the internet. Whether it will be like electricity (universal infrastructure) or like social media (engagement-maxing aggregators) depends on who owns your context.</p>— Alex Komoroske (<a href="https://bsky.app/profile/did:plc:qxeeegphuasuwv2oeyqe6byd?ref_src=embed">@komorama.bsky.social</a>) <a href="https://bsky.app/profile/did:plc:qxeeegphuasuwv2oeyqe6byd/post/3lrqcdoprn22y?ref_src=embed">June 16, 2025 at 8:30 AM</a></blockquote><script async src="https://embed.bsky.app/static/embed.js" charset="utf-8"></script>