Daily Journal 📓

• Vanilla JavaScript Repository

• A listing of single purpose scripts and widgets that use Vanilla JavaScript rather than being part of a larger framework.

• Microsoft is proposing a privacy-preserving advertising approach called PARAKEET.

• via @johnwilander - who works for Apple on the Safari tracking prevention team.

• From discussing with Claire, also related to Google’s FLoC, and this article Building a privacy-first future for web advertising.

• Just bought There Is No Antimemetics Division:

• An antimeme is an idea with self-censoring properties; an idea which, by its intrinsic nature, discourages or prevents people from spreading it.

Welcome to the Antimemetics Division.

No, this is not your first day.

• via @jonathoda

• Digital Favela and High Tech Gothic is my short hand for two future states, as described by Bruce Sterling.

• TODO I’ll flesh out those notes later, here’s one transcription of the original talk which is actually labeled Gothic Chic in the Future Favela.

• More agora thoughts: what about pulling in openly licensed general purpose Git content?

• For instance, terms.dev is a repository of software industry terms and definitions. It uses the Zola engine to build the live site, but the source is Markdown in git.

• Well, this @ctbeiser Twitter thread will help point me towards lots of other notes to flesh out:

• if you were making a syllabus on tools for thought what would be your key inclusions?

especially interested in ones beyond the “obvious list”:

• Englebart Augmenting Human Intellect
• As We May Think
• Plans and Situated Action
• Inventing on Principle
• Mindstorms

• (wikilinks added by me)

• The video by @edelwax is pretty moving.

• @Dylan_Steck mentioned @edelwax recently as having coined Assisted Introspection.

• I don’t know how much deep study I’m going to do around these things. I’ve always been more about tinkerability and trying these things out in practice. And then maybe falling into the weeds of building out tech building blocks and standards and protocols.

• Codex ¹ is one of a number of note taking tools for thought that I’ve been following on Twitter @codexeditor.

• I hadn’t realized that the Codex Patreon was the main web presence other than the Twitter account. It’s actually great to see so many digital creators getting supported directly to build.

• I keep looking at Microlink so I guess I should at least post a quick journal log before I write it up more:

• From the home page:

• Browser as API

Fast, scalable, and reliable browser automation built for businesses and developers. Proudly open source.

• You can turn websites into structured data, take screenshots, turn pages into PDFs, and more. Power embeds, etc

• The list of open source components is super interesting.

• Brave Browser bought a search engine, “promises no tracking, no profiling – and may even offer a paid-for, no-ad version”.

• Good timing for clearing my tabs of other Brave related stuff. This PDF white paper describes the Brave Search Team’s plans — Goggles:

• This paper proposes an open and collaborative system by which a community, or a single user, can create sets of rules and filters, called Goggles, to define the space which a search engine can pull results from. Instead of a single ranking algorithm, we could have as many as needed, overcoming the biases that a single actor (the search engine) embeds into the results.

• via Daring Fireball

• I have to do some more Webfinger research.

• I added a file to this site at .well-known/webfinger to try out delegating a remotestorage account, and it worked!

• More details on the Fission forum, working with @rosano.

• What else speaks Webfinger and can be delegated like this? It’s very IndieWeb feeling.

Nothing on fire that can’t just burn until morning.

• The UX of selecting time zones in Apple & Google is horrible. Mostly, filtered search as you type in country or city name seems the most workable.
• Particularly relevant as we are living in a more online, more distributed world.
• via @derrickreimer

• Chatting with @flancian who was wondering if he should support the Simply Jekyll margin note syntax in Anagora.
  • Specifically “what do you call this freak of nature???” when he saw the markup 😅
  • I’ll do a search and replace at some point to convert everything to Littlefoot which is probably worth a write up TO DO Implementing Littlefoot for footnotes.
  • The install was completed right at the end of my birthday Feb 22nd, 2021
• Gurlic looks interesting:

• Gurlic is an experiment in building a new/old kind of internet community. It is primarily a place for discovering and sharing technology, science, culture and the arts.

• That’s from the about page, which continues with principles that make for an important and inspiring read. The one about domains stands out:

• Every resource (user, community, publication etc.) must be mappable to custom user-owned domain names.

• Basically: even though this is a shared platform, you can bring your own domain and truly own it.
• Via @liaizon, whose post also links to a Show HN. And, Gurlic may be becoming federated via Matrix protocol to be called Hummingbird?

• I’m having one of those days. Reserved a car from Avis. Show up at 8am. I have a paper temporary driver’s license because I just renewed it. “Policy” is to have picture ID, even though I’m on file, I always rent from this location, and the desk clerk and manager both recognize me.
  • So: do I complete this rental and then never rent from Avis again, after 20 years? Probably. It’s policy after all.
• via @kemitchell’s blog, Jerome Saltzer’s “The Origin of the MIT License” in the IEEE Annals of Computer History.
  • PDF
• Catching up on @kemitchell’s Artless Devices forum where he linked to duallicensing.com. Yup, that’s a pretty fantastic resource for dual licensing.

• I have been sharing a number of @genmon’s posts from his blog. Mostly, I just say, you should go read this: Let’s invent new interfaces, not new products.
  • Yes I will read anything which links to Bret Victor and his Dynamicland project:

  • a physical environment of smart light, human writeable code, and a social interaction with the computer.

• I can already tell that having this box be able to teleport into my journal log — without having to cross post it publicly — is going to make me both post more, and tweet less.
  • Maybe I’ll consider cross posting these Journal posts to Mastodon.
• 7 hours of video calls is too many hours of video calls.
  • There were fun humans on the other side of a number of those hours, but it’s just a lot of screen time.

• Woo hoo! I am so stoked!
• I now have this site auto-publishing using Github Actions for Jekyll, with a Fission/Publish Github Action at the end.
  • I used the nicely commented limjh16/jekyll-action-ts and then just added fission publish at the end.
• Main thing I need to do to complete the rollout of these new journal logs — other than pagination — is to truncate long posts and strip HTML from them.
  • For short posts, I should not truncate, and actually render markdown.
  • Probably something like, if the content is bigger than 500 characters, strip_html and show 250 chars with a read more link. If less than 500, display the content.
  • For reference, this post is about 480 characters without HTML tags.

• The metaphors around apps vs windows vs tabs — especially in a mobile context vs a desktop context — are becoming increasingly blurry.
  • via Michael Tsai pointing at Nick Heer on unified MS Office App for iPad
• Via Roland, Tom MacWright wonders if the future of programming is more common data types:

  • What if a simpler programming language had first-class representations of a lot more than strings and arrays?

  • But if the aim is ease of use and giving power to people who otherwise wouldn’t be doing programming, type-rich systems with lots of assumptions seem like a logical first step. And one that doesn’t need a visual editor or a new dialect of a rare programming language.

• Tinkering with the site, getting the logs section working. This page on group_by in Liquid is useful. Covers posts grouped by year and month, among other expressions.

• It’s my birthday. I’m 46 today. Up late tinkering with my blog. And now I’m going to bed. Good night!

• HuJSON

  • https://github.com/tailscale/hujson — JSON for Humans (comments and trailing commas)
  • Built by Tailscale. Found via nigeltao’s write up on JSON with Commas and Comments

• Like @vgr, I created a Clubhouse account so that I didn’t have to keep turning down invites.

• Create a self-hosted chat service with your own Matrix server, Victoria Drake

  • Dendrite is a Matrix server written in GoLang. This article walks through setting up on an inexpensive VPS.

• High Lotek - Gemini and the Small Internet

  • A good overview of Gemini by Andrew Roach.

  • It’s a protocol partway between gopher and the web, without tracking, with encryption by default, respectful of users above all else.

  • I like a lot of the principles, and I understand the push back against, shall we say, the “complicated web”. There’s probably some content web vs app web feelings going on here too.

• Why I Still Use RSS — > However it wasn’t until I began working from home and everything in my life moved online that I really began to notice how beneficial RSS could be with relation to Digital Wellbeing. By selecting only the sites, blogs, creators etc. that I had a serious interest in, I could effectively remove the negative effects of social media and excessive online usage from my life.

  • via Robin Rendle who in turn has some very interesting thoughts in their post.

• Major surgery on the site, nicely celebrated from almost 24 hours ago :) Yes, working on my blog on my birthday has got to be tradition by now. #BMC

• What was Journals all got dumped into logs. Which is a collection in Jekyll but I’m going to call them all Journals: more granular daily notes and bookmark postings via IndieKit. The Journal page groups things by day.

• Somewhere in there, implemented Littlefoot for footnotes, and will mostly remove the Simply Jekyll specific syntax for margin notes, except for design reasons like the home page.

• Another test of the Logs functionality. A small x should give us Unix timestamps. You can check the current timestamp.
  • And this. This is the note that will go forth and timestamp appropriately in the Logs directory.
  • A final note. To bid farewell. At night. When it is late. And the timestamps are flowing into a git repo.
• I’m spending some time working on Moa Party today.
  • I created a new Mastodon account for the project to complete the setup of all of the social accounts for the project. We were thinking that we might run our own Mastodon server at some point, but accounts can be moved pretty easily within Mastodon, so I went ahead and chose the Fosstodon server to create an account on.
  • From the Fosstodon about page, “a community of like-minded people who enjoy Free & Open Source Software (FOSS)”, plus they have a sustainability model of passing donations on to other open source projects after costs are covered.
  • You can follow Moa Party at @[email protected].
• Git Bug
  • I get excited when I have a chance to try new software and…it just works!¹ That was the case with git-bug. It’s a bug tracker that is fully embedded in git. You use the same git repo that you are using for code development to track and update bugs and issues.
  • On MacOS, brew install git-bug and you’re up and running.
  • git bug user create will setup your local user. It will read from your local git settings for full name and email address. The only other thing you’ll need is a full link to an avatar image — the picture that represents you.
  • Next up is setting up bridges. git-bug supports GitLab, GitHub, JIRA, and minor support for Launchpad, as well as custom implementations.
  • The Moa Party project is what I’m going to experiment using git-bug on, which we host on GitLab. You run git bug bridge configure and walk through a terminal interface to fill out your GitLab server details (yes, you can use it with self-hosted GitLab instances), as well as your personal login and an access token.
  • When you create a personal access token, you’ll need to give it api access permission, which pretty much can do everything on your behalf. I initially created a token without that access, and had a heck of a time figuring out how to fix that. Turns out, go into your local ~/.gitconfig and delete the token and identity with the wrong permissions. And, git-bug bridge rm the original bridge you created with the wrong token, then you can git-bug bridge configure a new default.
  • And now, unfortunately, I nuked identities using the Makefile (and found @agentofuser in the issues from a couple of years ago) and am currently in a state where I can’t create a new identity.
  • OK, I got it working through the age old trick of … downloading a new copy of the repo and setting it up again.
  • There is a recent (7 days ago) issue thread that seems to indicate the GitLab API has been improved and this can be working again. git-bug itself works, and I managed to import the existing GitLab issues, but I can’t push changes to the GitLab issues.
  • Since the Agora itself works with git a lot, we may actually be able to link git-bugs with the Agora itself in some way, but that’s pretty deep git magic for me, since I don’t fully understand how git-bug works / where it stores stuff in git.
• IDGAF Takeout Roulette

  • Can’t decide what to eat and you just don’t give a…? We’ll place a $20 takeout order and have it delivered to you!

  • Seen via @IanColdWater

• Well, everything is working, but need to figure a few things out. I used iA Writer to post to the site and it works! I’ve made Article post types into what I call Notes here, the main post type.
• Posting from Quill as an Article isn’t going to work¹, since it turns everything into HTML and escapes the square brackets in wikilinks.
• For the (Micropub) Notes post type, I’ve created a new kind called Logs. They use the current date stamp as their filename, and I made a temporary logs page to display them.
• So, I can post arbitarily long Notes posts with Markdown formatting and wikilinks. This could replace my manual Journal entries. Instead, I would auto-generate journal pages per day (not sure if this is possible), or maybe just show the last N days of logs on a main journal page.²
• OpenFaaS
  • I have been meaning to tinker with OpenFaas, specifically the non-Kubernetes faasd. This article — Bring a lightweight Serverless experience to DigitalOcean with Terraform and faasd — walks you through using Terraform and Digital Ocean to set things up.
  • I’ve been looking for an open source serverless function runner to work with Fission. I think this could then also run things like IndieKit for me personally.
  • I ended up buying the book Serverless for everyone else, which is an ebook by the creator of OpenFaas, Alex Ellis.
  • I got as far as Terraform automating the creation of a Digital Ocean droplet, which is pretty great, but have some errors that may be DNS related. As part of getting things setup, also installed doctl, the command line tool for Digital Ocean.
• IndieKit
  • I decided that remote publishing of notes would be useful. So, back to setting up IndieKit. It has changed quite a bit since I last looked at it — for the better!
  • The actual server you deploy into Heroku is one index.js file. Here’s mine bmann/indiekit-bmcgarden, which in turn is really just a copy of paulrobertlloyd/paulrobertlloyd-indiekit.
  • I deleted everything except for Articles (which should turn into blog posts) and Notes (which will be notes, as long as notes don’t have an arbitrary character limit coming in). I need to figure out templating to do things like turn Bookmarks into notes with the link: front matter filled out.

• As I was finishing up my Drop in Audio blog post, Jam, an open source Clubhouse-like system debuted on ProductHunt. I took a quick look at the code and it requires a few moving pieces and Docker, but it’s a great little starting point.
• After a good discussion with @flancian, I added a section to the Git Siphon for Moa Party to focus on post-per-file. We discovered that we need to think about Moa Party as being standalone with multiple clients. Anagora is a client, people’s individual gardens are clients, eg. my Jekyll site.
• This may mean that making Moa Party become a whole Micropub server that publishes to Git may make sense. IndieKit has a whole template system so that people can run a variety of git-based static site generators with it.

• Starting last night, I seriously started using my new reMarkable paper tablet.
• I read @arcalinea’s Decentralized Social Ecosystem Review and highlighted sections and scribbled notes on it.
• Then, this morning, as I posted, I wrote out 6 pages of a blog post at the breakfast table, my Drop in Audio post.
• Lots of questions in the Twitter thread, which I’ll capture on the reMarkable notes page, also where I’ll add other things I find out about over time.
• Did a bunch of Moa Party writing to capture some discussions happening in Matrix. I’m excited for the Git Siphon for Moa Party feature I wrote up.
• I also bought moaparty.com. Nothing there right now, I guess I should at least do a redirect or something. It might be the root domain for running an experimental Mastodon / ActivityPub server of some kind for FedStoa experiments with @flancian and friends.

• Unfortunately, I have to tag this with February 10th, 2021, since it just rolled past midnight, but “today” is mainly February 9th, 2021 when I talk about it in this post!

• Learned a very relatable term today: “報復性熬夜” (revenge bedtime procrastination), a phenomenon in which people who don’t have much control over their daytime life refuse to sleep early in order to regain some sense of freedom during late night hours.

  Daphne K. Lee @daphnekylee https://twitter.com/daphnekylee/status/1277101831693275136

• Not quite the case for me…but also some truth there.
• Last night (morning?) was the final piece of the Mac Mini: memory install.
• The whole office here is a mess of boxes and pulled out mess, as I now have to sort through various electronics and recycle them, and so on.
• I also got two final deliveries that I had been waiting on for now. I decided to buy a reMarkable tablet. My paper notebooks ran out, and I really should have at least one device for quiet reading and note taking that isn’t a click away from Twitter. There is some retail therapy there, too.
• The other delivery was Ubiquiti networking gear for my parents’ new Foxglove Terrace house. I’ve now got Ubiquiti here at home, at the Fission HQ on East 6th, and will install it at my parents.
• Other notable events, I was part of a live show put on by Startup Vancouver. Here’s the video from Facebook. I start at around the 17 minute mark, talking about Fission, what we’re building, but a lot of focus on vision.
• I’ll look to get the source video and cut down a clip to just my segment.
• I was down on Clubhouse https://www.joinclubhouse.com/ at the beginning of the video show, but I ended up joining today. I’m thinking of BBB as a pretty good open source equivalent, and am supposed to be helping Social.Coop document it.
• I joined Clubhouse because I started getting multiple people in the last day or two inviting me to it — or asking me if I should be invited or what I thought about it. For the record, @allbombs was the first person to invite me several months back when it was still super exclusive :)
• I also have Soapbox https://soapbox.social/ on my phone that I haven’t used yet. How many social audio apps can I have before I use the first one???
• Had a nice Twitter thread with @tombielecki, where he found my announce tweet for the Digital Garden version of this site, when I first put it on Fission. Wow, Sept 9th, 2020 is already 6 months ago!

• Just saw this. Your site is so fast! There’s a Digital Garden telegram group sharing examples etc they might want to take a look at @FissionCodes? @Mappletons @houshuang

  Tom Bielecki https://twitter.com/tombielecki/status/1359257555462049793

• The site was a bit slow because I am using Cloudflare/IPFS Gateway plus Fission hosting, which occasionally has some caching issues.
• Cryptic notes for today: Claire screenshot, Courtney yay, Lance boo.

• I would apologize for some of the jargon in this, but in fact that is partially the point of this post.
• I am writing up notes from a discussion with @Flancian. We have been “chatting” on Twitter and Mastodon about various topics, and ended up booking a time to speak in person last Wednesday, January 20th, 2021.
• Very quickly on the call it was clear we were “pushing” context into each others brains so that we could get to deeper topics and explore shared areas of interest.
• Querying things like “have you heard of X?” If not, quick context as well as key terms to note and search for later. If yes, then anchor that context and go deeper.
• One of the specific areas we talked about and were exploring together is Digital Gardens, Second Brains, and, as @Flancian calls them, Agoras. His specific implementation is Anagora.
• With Fission, we have interest from a number of different people and software apps that want to have user owned files and notes, as well as thinking about how to link them together between people.
• @Flancian has written a Python server and is interested in connecting Agora at the level of a git repo filled with Markdown files.
• I had mentioned to him that I was interested in experimenting with this site, but after all the recent tinkering I had done, I had kept the git repo private and closed. To prep for the call, it was a good excuse to bring the repo for this site public. It is available at https://github.com/bmann/bmcgarden. Actually connecting it into the Agora will be documented in the future.¹
• Will connecting second brains at the Markdown layer work?
  • My site is “a bunch of Markdown files in a git repo” as its “source code”, but it gets processed by Jekyll, and actually by the double bracket backlink and other features of the Simply Jekyll theme.
  • In fact, the completely normal Jekyll behaviour of specifying what the resulting web permalinks look like from source Markdown files is completely custom per site.
  • I already know that some of the syntax of Simply Jekyll is going to be problematic for portability. I like the side notes and margin notes features, but they are totally custom.
  • I have a note to move many of them to regular Markdown footnotes, and then add BigfootJS to have a nicer display. Knowing myself, I will want to switch around the theme and engine behind this site in the future, so using more standard “source code” is important.
  • I’ve said before a couple of times now that if the various note taking initiatives / engines / apps want to interlink, there are going to need to be some conventions or specifications or something.
  • I stopped using LogSeq in part because of their Markdown formatting.² which isn’t compatible with regular posts in any other system.
  • How is the extended Note-verse going to come to some consensus? How do we layer backlinks or other supersets of Markdown for note-specific purposes together? And, to do this at a text-only layer, which in turn has to work with a build process and/or live hosting software, desktop software, and display layers across multiple different programming languages and software architectures.
  • At the very least, we need a protocol or standard at some layer, and a way for these disparate groups to form consensus.
  • Related: I wrote about a vNotes Format some time ago, but that is in fact at another “layer” — more of a wrapper or sync format.
• Connecting over the Internet
  • We can, of course, also connect over the Internet. Mostly, I mean linking to other pages, but there are some other protocols and standards that may prove useful or be the base layer on which backlinks or extended references could be built.
  • The Web Annotation Standard and Webmention are two that I have in mind.
  • ActivityPub and having someone’s agora participate directly in the Fediverse is interesting.
  • We ended up talking about convention of being able to do double brackets anywhere one can put text, like on Twitter or Mastodon, which could then be processed and sucked back into an agora. No spec, no permission, just a convention that starts being used, which is exactly how @ChrisMessina invented the hashtag.
  • Somewhere in here is the concept I was introduced to by @agentofuser, which is.
  • Micropub is another standard that I see being adopted and working relatively well. When it makes it into mobile / desktop apps like iA Writer, that’s a good sign.
• On Biases
  • Flancian is an SRE at Google. Many Google only tools from software stack to hosting.
  • What’s the simplest that will work without having to dive into more learning?
  • For me, I have a bias towards Heroku³ and 12 Factor App. Both as a way of architecting, but also on how to host to maximize user agency.
  • Making things explicit can help people and communities find connections as well as understanding.
  • Specifically, three things:
    • Goals
    • Values
    • Biases
• On Social Networks and Community
  • I mentioned the Orbit Model of looking at community engagement.
  • Both of us have found the Loomio style community management / engagement in Social Coop to be hard to get started with / put time into.
  • Flancian sees the agora as an integration hub. For Fission, we’re using the label constellation provider. Some equivalencies?
  • Can a social network do more to connect people and have them learn from each other? Similar as the Viznut Eternal September post: social networks beyond consumption or entertainment.
  • If you take two people and most of their opinions are the same, can we dig down into the areas where they are different in a way to learn why they differ? Understanding and learning.
  • I mentioned chicken fingers vs. tentacles: you have to understand the preferences and biases of another person if you hope to give them useful recommendations, because your recommendations are based on YOUR preferences.
  • An agora is a tool for building context.
  • New wording. A Stoa is shared context with explicit overlap defined by its members.
  • A data structure and a social graph that is designed toward problem solving.
  • I introduced the concept of a Squad: a collection of people with a set of skills coming together to jointly work on some tasks or goals.
  • Lightbulb moment: we are in a Stoa together — we have overlapping interests and context. The next step is to form a Squad initially of the two of us, aimed at some tasks / goals.
• Goals
  • What goals can we work on?
    • Cross post between social networks
      • Use Flancian’s agora as an integration hub, leaning into the server aspects. Like Social.Coop, form a Stoa around it to meet common goals.
      • Moa Party is what Flancian currently uses to sync between Twitter and Mastodon. It has an Instagram API, which currently isn’t working.
      • Create an Open Collective to see if we can pool funds to pay someone to enable the Instagram API again, so we can cross post our photos into that social network.
      • Make some blog posts, announce the idea, get in touch with moa.party maintainers, see who else is interested in joining this Squad.
    • Agora Interlinks from Social Posts
      • We’re already experimenting with Twitter / Mastodon double bracket linking.
      • Can we use the agora integration hub to do more of this?
      • What tools do we need to do to ingest these posts into agoras?
      • Also need to explore common patterns and conventions. How do we inter-work with e.g. hashtags?
    • Experiment with linked agoras
    • As mentioned above, Boris’ garden is now a public git repo.
    • What to do next? How do we connect?
    • Make a call for other git repo based agoras to link.
• Other Ideas & Resources
  • I used to run my own Micropub server. IndieKit is the one I recommend.
  • Making it easier for other people to run their own agora should be a goal. I created a Simply Jekyll Template repo on Github, and hooked it up to Forestry and published it to Netlify.
  • Add agora features? Add instructions on how to link into Anagora, and make the template linked in “by default”, or at least following some simple steps.

• I’ve been blogging a lot lately. Daily posts about doing a walk outside. Basically “Instagram-style” photo posts with a few words.
  • Having Mb with Gluon on my phone and it “just works” is nice. I can write something short, or just keep typing and it ends up as a blog post, all written and posted direct from my phone, like this post about @kemitchell’s StrictEq project. Yes, it’s being renamed, head over to the Artless Devices Forum for longer discussion.
  • Longer posts like this music one were done sitting at my desktop, composed and posted with Mars Edit. This is good for short posts, too, like documenting some tide / calendar research I did for my mom. A side effect of spending lots of time WFH — having default desktop app tools.
  • I’ve turned on cross posting to Mastodon and Twitter automatically. People seem to enjoy my more “social” posts on Twitter, which is great. And I definitely reach different people on Mastodon, in a nice way.
  • Using the StrictEq post as an example again, yeah, I hate the “title”:

    • “The code you depend on depends on you” @kemitchell’s commercial license sales for public software

  • But I have to craft it knowing that it will get cross-posted to Twitter and to Mastodon. I guess that’s another wishlist item for Micro.blog: for non-micro-blog posts that have a title, allow for a “cross posting excerpt”.
  • Otherwise, your only other option is to:
    • 1. turn off cross-posting for the initial publish of the blog post and then
    • 2. make a micro blog length post that links to your blog post.
  • Hmm. Now that I think about it, that’s actually not bad, but I’ve found toggling cross posting to be really confusing. I need to experiment with the Mb mobile app and run some experiments to see if I can figure it out, the UI is just not good for this.
• With Trump being kicked off Twitter, the discussion about Parler being kicked off AWS, there’s lots of discussion about federated and censorship proof social networks.
  • RTd @QuinnyPig who has a good thread on AWS terms of service, and other cloud providers more broadly. Aimed at explaining even to non-technical people, a good one to share.
• I got drawn into a long thread kicked off by Stefan George of Gnosis, who wanted to post a bounty to save/store Twitter stuff to IPFS. I pointed out that Twitter archives already exist.
• From there I got into a whole back and forth, including people suggesting that Twitter data should be mass exfiltrated to setup an alt social network. Maybe at one point I would have reached for a purely technical solution like this, but here’s my final take:

• Technical tricks to try extract Twitter data is probably orders of magnitude harder than …just doing the work in building out critical mass communities.

  The difference between the two approaches? One is technical challenges, the other is marketing and community building. @bmann

• I am… somewhat distraught that a lot of decentral type people think about this issue as censorship, rather than the “let’s deplatform fascists issue” that it actually is. Twitter is in no way a public utility. Own your “distribution” — whether it’s direct mailing list subscriptions or your own community social network: run your own Discourse or Mastodon server. Twitter is today’s mass media: use it for distribution, but know that you’re there at the company’s sufferance.
• And of course, when I see lots of crypto people happily using Medium and Substack, I’m not going to take the rest of their opinions seriously. Your words need to get backed up by actions, my dudes.
• Jeff Henshaw popped back up on Mastodon — and he’s hosting his own Mastodon server. He was asking me why I’m on Mastodon, here’s my answer:

• the concept of collective ownership / maintenance.

  Nothing wrong with “multi user” systems, but we have to clearly understand that it costs people time and some server resources to run things.

  I’m working on a model of awareness that people should pay for apps — that behind that app are real people, not just corporations.

  “Exit to community” is a related theme. Plus governance, etc etc @[email protected]

• Yes, I’m posting my own content back to my own site. Which means it becomes part of my forever content, and also I can link to and include it in search.
• I guess I should document my BMC/Twitter Archive, which lives at https://tweets.bmannconsulting.com/. In a week or so, when Fission has Github/Actions support, I can start publishing this to IPFS directly, instead of hosting it on Github Pages. Doing this direct from Google Sheets to Fission would be ideal, as a little one off Twitter Archive app.
• It was really easy to setup. Michael Hawksey’s @mhawksey post explains it all: Keeping your Twitter Archive fresh and freely hosted on Github Pages.
• I just emailed Michael to see if I can send him some thank you money, or donate to some project on his behalf.
• Doing some work on the Marfa Theme for my Micro.blog site. Running Hugo locally is really fast! I still have about 200 orphaned posts that I’d like to import into Mb at some point, which I use for testing, and with Hugo it’s pretty instantaneous.
  • I’m stuck in the dark with Hugo a little bit, as I can’t prototype locally because how to structure this for Mb isn’t really documented. I guess I’ll file another issue in the help documentation, as the only public place you can kind of file this stuff. I really wish Manton Reece would just open source the whole thing or more properly make it non-commercial / source available.