@prologic@twtxt.net so maybe you’re not so sure about the standardized search API then? Or are you saying something else?
@prologic@twtxt.net that a twtxt search engine is super important and necessary.
@prologic@twtxt.net Now that I see your new comment, it sounds like we’re in strong agreement.
@prologic@twtxt.net It’s really easy to publish a flat twtxt file from just about any client you can imagine. The hard part is being notified of mentions and discovering new feeds. There is some degree of centralization/always on needed for that piece. By standardizing the hard part to an interchangeable API, it would allow a diversity of clients to be built with little effort. Just plug in your preferred search engine(s) and you’re good to go.
@prologic@twtxt.net I see registries/search engines as a way of removing the complicated operational piece of twtxt. Just put a static twtxt up on the web and rely on search to notify you of non-follower mentions and to discover others talking about what you care about.
@tkanos@netbros.com I’d be happy to put in a PR. I wanted to see if this had already been considered and decided against for some reason.
@tkanos@netbros.com That’s why I was thinking it might be easy to implement. I am looking at implementing my own client/server, and it would be nice to have as few protocols as possible to deal with. Now that I’m saying this, I don’t actually know how many people are using twtxt registries vs yarn.social json protocol.
@prologic@twtxt.net Any plans/thoughts on implementing the twtxt registry protocol as part of yarn.social?
@prologic@twtxt.net I don’t know this person. But this reeks of the Dunning-Kruger effect.
@prologic@twtxt.net looks like it required alot of hand eye coordination.
pkg/
inside of Go projects. It is just like any other sub-package structure you might otherwise define in your project. It just adds an extra part to your imports. I think it's actually confusing at best and just unnecessary typing and an unnecessary sub-structure. Just keep your packages in the top-level and be done with it 👌
@prologic@twtxt.net I sometimes wonder if this is a breadth of experience problem. I know I did this a lot when I was first starting out. As I’ve learned more programming languages, I have noticed that I’ve gotten better at following each language’s idioms.
pkg/
inside of Go projects. It is just like any other sub-package structure you might otherwise define in your project. It just adds an extra part to your imports. I think it's actually confusing at best and just unnecessary typing and an unnecessary sub-structure. Just keep your packages in the top-level and be done with it 👌
@prologic@twtxt.net Sometimes it feels like a cultural import from the Java world. I work a lot with Kubernetes code and it feels a bit like a bunch of Java devs wrote it. Good Java devs, but Java devs nonetheless.
pkg/
inside of Go projects. It is just like any other sub-package structure you might otherwise define in your project. It just adds an extra part to your imports. I think it's actually confusing at best and just unnecessary typing and an unnecessary sub-structure. Just keep your packages in the top-level and be done with it 👌
@prologic@twtxt.net I think many projects largely started using it because of “the standard go project layout” which is not actually endorsed by the go core dev team. I generally advise against it for similar reasons.
@prologic@twtxt.net Good to know. I thought that .us site looked a bit sketchy (at least on mobile).
@abucci@anthony.buc.ci Wow I agree that’s terrible. Definitely gonna think twice about using Cloudflare in the future. This is potential lost revenue for any company who uses their services!
@prologic@twtxt.net I’m in the states. I wonder if they have show rooms here. I found http://www.marscampers.us but it doesn’t look as legit as the site you linked. This is mostly just idle fantasy for now anyways since I’d need a vehicle to tow it first.
I use bothdaily. But as my primary? Apple.
@prologic@twtxt.net I’m not sure if this is a joke or real. Either way, I kinda want one. 😂
@prologic@twtxt.net @adi@twtxt.net You might have already seen this but local-first software seems related.
@adi@twtxt.net I mean it’d be cooler if fintech was about high tech for finned creatures like dolphins and whales. 😂
@lyse@lyse.isobeef.org haha what do you think?
@markwylde@twtxt.net @prologic@twtxt.net I heard another piece of advice. Don’t pay attention to the news very much. Learn history instead. Then occasionally look at the news for its own larger trends (multi-year narratives). And sadly this “conversation” around privacy is all too consistent with some historical trends and it certainly has been a background narrative for almost a decade—it’s not very entertaining.
@lyse@lyse.isobeef.org While I think some parts of the FP (and Haskell especially) community can get into abstraction for the sake of abstraction. They do have some cool ideas. I think this is one of them.
@prologic@twtxt.net Don’t ya know? Dark web just means anything the governments can’t track your every move with. 😉
@adi@twtxt.net @prologic@twtxt.net tiny bit facetious but not entirely inaccurate. 😂
@prologic@twtxt.net As an aside, this is one thing I admire about Haskell. Its “core” language (sort of like an IR) can fit on a note card. I believe that someone fit it in a tweet. Haskell syntax and type system is more complex though. See this talk by FP God-king Simon Peyton Jones for more about that: https://youtu.be/uR_VzYxvbxg
@prologic@twtxt.net I totally agree. If I ever make a programming language, I really want to find a way to make removing features a regular part of the process. Rust definitely feels like it’s heading in the direction of complexity. However, I did see a ray of hope in a recent talk from rust conf. One of the co-leads for the language suggested removing features.
@lyse@lyse.isobeef.org @movq @prologic@twtxt.net IIRC, the story around rust async was a good example of letting things mature in libraries. They added macros for the keywords but then delegated the implementations to the ecosystem and then formalized stuff with proper language keywords and interfaces when stuff settled. I believe C++ has a similar model in that stuff tends to bake in boost before entering into the standard library.
Thanks! I’ll definitely check it out when I get a chance. Fussy baby makes my time less plentiful as of late.
I feel like most feeds I would want to subscribe to are updated pretty infrequently so once a day would be sufficient. But I’ve seen other feeds that are updated more frequently. Perhaps some sort of heuristic based off the feed itself? (For example, average time between entries with a max of 1 day.) If you combined that with a feature to manually trigger a feed recrawl, that might be good enough.
@mckinley@twtxt.net I just had a really weird mind picture of a crab-like human. Maybe something like this
?
@adi@twtxt.net I’ve seen you talk about it on here. I haven’t checked it out myself, yet. I’ve heard you talking about maybe releasing a Mac version at some point. If I check it out before that happens, is a Docker container might be the best way to try it on Mac?
@prologic@twtxt.net (#mr3lfga) its_happening.gif
@prologic@twtxt.net Oh that’s awesome! Let me know when you launch (or if you already have)! I definitely am like minded in that I think there are entrepreneurial opportunities in this space. I think you may be further than me—as I’m still looking.
@adi@twtxt.net @prologic@twtxt.net @mckinley@twtxt.net However, scuttlebutt uses a ton of crypto in its protocols and formats which means a scuttlebutt client is not a trivial thing. Its specific formats also unnecessarily complicate making a robust implementation. In addition, you sort of rehost all of your friend’s feeds on your local machine which can get expensive and be can be an onboarding barrier.
@adi@twtxt.net @prologic@twtxt.net @mckinley@twtxt.net scuttlebutt is kinda neat in that it really embraces the idea that people own their data. The source of truth for your posts is your local computer but you don’t have to be online for other people to see your posts. It’s also got PMs and some nice security features.
@prologic@twtxt.net (#bkydgsq) I agree the underlying web infra is definitely decentralized. However, I think we can agree that there is a trend towards centralization for services built on top of that decentralized infra. I’m okay with someone else hosting stuff for me as long as I control my data—meaning I “pack up and move” somewhere else without much effort. I think this type of web would make most of the Twitter/Facebook controversy moot. I think it could actually result in a more fairly moderated web as well.
@mckinley@twtxt.net @adi@twtxt.net @prologic@twtxt.net I think one thing that keeps bringing me back is the fact that this is such a simple protocol/format. I feel like it could easily capture the imaginations of many devs. I think one of the problems with other decentralized/distributed social networks is that they are built on tech that might seem impenetrable to your average dev.
@adi@twtxt.net @prologic@twtxt.net @mckinley@twtxt.net I was here before the hacker news influx. I first heard about this from a scuttlebutt user (don’t remember who). They linked to their twtxt.txt from their profile and down the rabbithole I went.
@prologic@twtxt.net I don’t find it odd. I find it sad though.
@jlj@twt.nfld.uk @mckinley@twtxt.net @darch@twtxt.net I guess I missed that it was an art project. In that case, 100 posts seems less crazy since said person probably doesn’t want to actually host the next Twitter.