mckinley

twtxt.net

A guy on the internet. https://mckinley.cc/

Recent twts from mckinley

I just typed out a message here on Yarn, undid a few things with ctrl+z, and then tried to redo something with the vi redo key combination. That means it’s time to go to bed.

⤋ Read More
In-reply-to » I noticed some unnecessary horizontal rules being rendered at the bottom of twts in the conversation view lately with yarnd's new design change. They can be seen at least when not being logged in. Not sure if it is the same experience with an active user session. I reckon there will be fork buttons then, so in that case they would be fine, indeed.

@prologic@twtxt.net Markdown horizontal rules never worked on the Web client, as far as I know.

@darch@twtxt.net, I think that specific line had a use as a visual separator between the non-interactive text and the interactive buttons, but it’s not a hill I’m willing to die on.

⤋ Read More
In-reply-to » I noticed some unnecessary horizontal rules being rendered at the bottom of twts in the conversation view lately with yarnd's new design change. They can be seen at least when not being logged in. Not sure if it is the same experience with an active user session. I reckon there will be fork buttons then, so in that case they would be fine, indeed.

@prologic@twtxt.net @darch@twtxt.net It looks like the horizontal rules are completely gone, though. I thought they looked nice…

⤋ Read More
In-reply-to » @mckinley Cool! 👌 So... Let's nuke it? 🤔 Are you thinking of writing a new spec?

@prologic@twtxt.net Yes, I’ve been thinking of writing a new, unambiguous version of the original spec with some small changes to bring it in line with how feeds are actually being constructed in the wild. The comment syntax, for example, but not the Yarn extensions.

Is there community interest for such a thing?

⤋ Read More
In-reply-to » Testing something... https://neotxt.dk/external?uri=https://example.com/~foo/twtxt.txt&nick=

Outside of this thread, this comment syntax has been used exactly twice when searching every known, currently accessible twtxt feed on the Web.

$ grep -r '@<[^ ]*>'
buckket.org.txt:2016-02-12T18:37:00+01:00	Hey @<http://vigintitres.eu/twtxt.txt>, @<teddy https://data.trafficking.agency/twtxt.txt> und @[email protected] was geht? Ich bin’s @[email protected]!
[...]
hecanjog.com.txt:2020-09-03T17:36:00-05:00	@https://tilde.town/~lucidiot/[email protected] twtxt via DNS TXT would be insane and fun.

⤋ Read More
In-reply-to » @mckinley here is the dump https://github.com/tkanos/we-are-twtxt (on the tarball all-twtxt.tar.xz)

Here are the top ten feeds by size. @prologic@twtxt.net is artificially low on the list because it’s separated into chunks, and @movq@www.uninformativ.de is listed twice. Once as www.uninformativ.de, once as uninformativ.de. I blame yarnd.

du -b * | sort -nr | head -n 10
5253921	www.lord-enki.net.txt
842733	cnbeta-com-rssding-yue.txt
755925	search.twtxt.net.txt
654717	prologic.txt
394380	jlj.txt
371632	assets.txt
246520	off_grid_living.txt
243953	mckinley.txt
225256	www.uninformativ.de.txt
225256	uninformativ.de.txt

cnbeta-com-rssding-yue.txt seems to be a syndication feed for https://cnbeta.com/ in twtxt format, assets.txt is @maya@maya.land, and the rest are fairly self-explanatory.

⤋ Read More
In-reply-to » Testing something... https://neotxt.dk/external?uri=https://example.com/~foo/twtxt.txt&nick=

For people using clients other than yarnd, does that appear as a valid mention? It’s valid according to the spec but I’ve never seen it in use anywhere.

Mentions are embedded within the text in either @<source.nick source.url> or @<source.url> format and should be expanded by the client, when rendering the tweets.

cc @movq@www.uninformativ.de @lyse@lyse.isobeef.org

⤋ Read More
In-reply-to » I guess the problem I really have, same with this whole Web3 Blockchain nonsense, is that we blur the lines between what is a centralised system, vs a decentralised system, vs a distributed system (sorry forked as we're getting off topic...)

@screem@twtxt.net I think there is value in cryptocurrencies as long as they have sufficient privacy protections. If you have someone’s Bitcoin or Ethereum address, you can see every transaction he’s ever been involved in. Not enough people know that.

The value is in being able to send a scarce resource to anyone on the planet, any time of the day, any day of the week, and have it received in 20 minutes. As long as privacy is preserved, I think it’s great.

It’s completely useless in the context of a chat service, though. The blockchain nonsense was part of the reason why I ditched Session, but it was mostly the Electron client.

⤋ Read More
In-reply-to » Interesting and tough moral dilemma: is it appropriate for Government to strip privacy away from all citizens in favour for the safety of children online to avoid the production and distribution of CSAM? I ask because there’s a podcast from the Australian Federal Police around CSAM in Australia and there was a statement that effectively the prevention of CSAM needs to take a priority over privacy.

@screem@twtxt.net It is not a tough dilemma for me. A government has no right to perform mass surveillance on its citizens, treating everyone as if they were criminals. It starts with something we can all agree is reprehensible, and they say it stops there, but history tells us it never just stops there.

In addition, computers are really bad at their jobs. How many innocent people will be punished with a false positive? How many mothers will be punished for sending a photo of their newborn to the doctor?

I’m talking about punishment not only in the legal sense, but with the time, money, and worry associated with fighting legal punishment. Do you even trust your legal system enough that it will protect innocent people in these circumstances from having their lives ruined?

There are questions to be raised about the effectiveness of such a policy for its intended purpose but I’m running out of characters.

https://puri.sm/posts/internet-of-snitches/

⤋ Read More
In-reply-to » I guess the problem I really have, same with this whole Web3 Blockchain nonsense, is that we blur the lines between what is a centralised system, vs a decentralised system, vs a distributed system (sorry forked as we're getting off topic...)

@prologic@twtxt.net It’s proof of stake, so you need to stake 15,000 units of their cryptocurrency $OXEN, worth $3118 US, to run a “full service node” and 3750 $OXEN ($779 US) to run a “shared node”. If I understand correctly, only “full service nodes” can route Session messages.

If you don’t have enough $OXEN, you can pool what you do have with other people and run a node that way.

TL;DR: Not very easy. To help route Session messages at all, you have to buy in to their cryptocurrency.

Sources:

⤋ Read More
In-reply-to » For a real feed format I would like to have a clear separation between titles and content. And more options for the content. Plaintext and HTML at least.

@lyse@lyse.isobeef.org The specification just says:

Also note that a status may not contain any control characters.

Which is extremely vague, but U+0009 Horizontal Tabulation is in the C0 control code block

I’m sure 99% of twtxt parsers don’t treat additional tabs any differently. Even Buckket’s reference implementation includes additional tabs in the message. Although, in fairness, it doesn’t check for any for control codes.

Maybe we need a less ambiguous specification documenting how twtxt feeds are being written in the wild. Did you know that the comment convention is not a part of the original spec? I feel like it’s used everywhere, even among feeds that don’t use any Yarn extensions.

⤋ Read More
In-reply-to » For a real feed format I would like to have a clear separation between titles and content. And more options for the content. Plaintext and HTML at least.

I’m sorry, I didn’t explain this properly and that has led to a misunderstanding of my actual proposal. I was not intending for the title to be a special field unless the client explicitly understood my syndication format.

The original twtxt format specification gives no special meaning to the tab character, excluding the one that separates the timestamp from the text. I was under the impression that the tab character could appear in a twt so it would be interpreted as follows, replacing ␉ with a tab character.

2022-09-22T14:53:26-07:00␉Bringing Back a Useful Browser Feature With a Bookmarklet␉https://mckinley.cc/blog/20220922.html
#^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
#|                        |
# ```- Timestamp             `- Message

Although, I just remembered that the tab character is technically a control code, so it shouldn’t be allowed.

⤋ Read More
In-reply-to » @prologic I don't know if a metadata field is strictly necessary. I think there ought to be a defined set of syntax that all clients with Markdown support can be expected to handle in the same way. CommonMark maybe? It looks like Yarn supports most of CommonMark already, though I've never seen a horizontal rule. Let's try it:

@prologic@twtxt.net Referenced links also work on the Web client, but I tried both CommonMark syntax options for the horizontal rule and only one worked on Goryon.

This is exactly what I’m talking about. We should have a concrete specification so Markdown can be rendered consistently between client implementations.

⤋ Read More
In-reply-to » who would like to play a game? https://wiki.kyoko-project.wer.ee/wiki.cgi/among-sus

I think I broke it, though, because I started a game alone and then accidentally pressed ctrl+c. When I try to log back in, it says there’s a game in progress and I can’t start a new one. Sorry about that…

⤋ Read More
In-reply-to » @prologic I don't know if a metadata field is strictly necessary. I think there ought to be a defined set of syntax that all clients with Markdown support can be expected to handle in the same way. CommonMark maybe? It looks like Yarn supports most of CommonMark already, though I've never seen a horizontal rule. Let's try it:

Nope, no horizontal rule. What about referenced links?

My website

Image

⤋ Read More
In-reply-to » Not sure which conversation you mean, @eaplmx (it's already quite late here), but here's my take: I think twtxt it's not heavy enough. For a real feed format I would like to have a clear separation between titles and content. And more options for the content. Plaintext and HTML at least. Twtxt is plaintext, but lots of folks (me included) actually use markdown in their yarns. However, the actual format being used is not advertised anywhere. To make things worse, I actually prefer reStructuredText over markdown. For podcasts some enclosure-like thing would be nice as well. Twtxt being line based also really limits structuring of longer content by hand. Just can't produce a nice source file.

@prologic@twtxt.net I don’t know if a metadata field is strictly necessary. I think there ought to be a defined set of syntax that all clients with Markdown support can be expected to handle in the same way. CommonMark maybe? It looks like Yarn supports most of CommonMark already, though I’ve never seen a horizontal rule. Let’s try it:

Some text here

⤋ Read More
In-reply-to » Recent updates to LibWeb have fixed the icons on the Yarn web client in Ladybird and the SerenityOS browser! Unfortunately, it still isn't possible to post using LibWeb browsers but we'll get there. :^)

Of course I pick the day before simple.css gets merged into main to get a new Ladybird screenshot. Here’s one post-merge.

Image

⤋ Read More
In-reply-to » Related with my current conversation, what do you think of using twtxt.txt as a format for feeds?

@akoizumi@social.kyoko-project.wer.ee It looks like that feed uses TIMESTAMP\tTITLE: PERMALINK which would be harder to parse programmatically.

This discussion has me thinking of a serious syndication format built on top of twtxt that could be implemented in normal feed readers. It would be limited, but extremely easy for a Webmaster to implement. Users could also receive updates with a normal twtxt client. I think there could be some utility in it.

⤋ Read More
In-reply-to » Related with my current conversation, what do you think of using twtxt.txt as a format for feeds?

For a real feed format I would like to have a clear separation between titles and content. And more options for the content. Plaintext and HTML at least.

I don’t think it’s a very good idea to include content when using twtxt as a syndication format. Anything based on twtxt, in my opinion, should retain the spirit of the original specification, especially readability by humans and machines. 10K of HTML in one line absolutely breaks human readability.

What about TIMESTAMP\tTITLE\tPERMALINK, like the following?

2022-09-22T14:53:26-07:00	Bringing Back a Useful Browser Feature With a Bookmarklet	https://mckinley.cc/blog/20220922.html

⤋ Read More
In-reply-to » Not sure which conversation you mean, @eaplmx (it's already quite late here), but here's my take: I think twtxt it's not heavy enough. For a real feed format I would like to have a clear separation between titles and content. And more options for the content. Plaintext and HTML at least. Twtxt is plaintext, but lots of folks (me included) actually use markdown in their yarns. However, the actual format being used is not advertised anywhere. To make things worse, I actually prefer reStructuredText over markdown. For podcasts some enclosure-like thing would be nice as well. Twtxt being line based also really limits structuring of longer content by hand. Just can't produce a nice source file.

@lyse@lyse.isobeef.org

Twtxt is plaintext, but lots of folks (me included) actually use markdown in their yarns. However, the actual format being used is not advertised anywhere.

That’s a really good point. We should formalize a Markdown flavor as a format extension on https://dev.twtxt.net/.

yarnd appears to use gomarkdown with a few extensions. I’m not sure what else gomarkdown translates by default.

There should definitely be a concrete specification on what syntax should be supported by twtxt clients with “Markdown” support. cc @prologic@twtxt.net and @darch

⤋ Read More
In-reply-to » Guys, I have a bad news, I went through the twtxt-osphere : - I found 1289 twtxt account - among those 721 are accessible ( 712 http / 9 gemini / 0 gopher) - but only 111 account are still active in 2022 :S (107 http / 4 gemini / 0 gopher).

@tkanos@twtxt.net This is really interesting stuff. Do you plan on sharing the code, or at least a more detailed write-up of your process?

⤋ Read More
In-reply-to » Recent updates to LibWeb have fixed the icons on the Yarn web client in Ladybird and the SerenityOS browser! Unfortunately, it still isn't possible to post using LibWeb browsers but we'll get there. :^)

LibWeb font handling has greatly improved. We’re no longer stuck with that terrible bitmap font. It looks like there are still a few font-related issues to iron out, but here is what the Yarn discover page looks like on Ladybird. A huge improvement in only a week, though it really can’t decide where it should put that search box.

Image

⤋ Read More

A nice chat tonight with @prologic@twtxt.net, @ocdtrekkie@twtxt.net, and @darch. Some things we talked about:

  • Markdown isn’t rendered to HTML in Atom feeds (#989)
  • A look at @darch’s branch to replace Pico.css with Simple.css
  • 100 monthly active users on yarn.social
  • Sandstorm proxy authentication for yarnd (Allows one set of credentials across programs hosted on Sandstorm)
  • Sandstorm support for Tube (#26)
  • Future improvements to Sandstorm

⤋ Read More
In-reply-to » Guys, I have a bad news, I went through the twtxt-osphere : - I found 1289 twtxt account - among those 721 are accessible ( 712 http / 9 gemini / 0 gopher) - but only 111 account are still active in 2022 :S (107 http / 4 gemini / 0 gopher).

@tkanos@twtxt.net I’m 6th, 57th, and 1280th. What metric did you use, total number of mentions? Also, is this supposed to be a continuation of mdom’s project of the same name?

⤋ Read More
In-reply-to » I have a new Atom feed at https://mckinley.cc/blog/atom.xml. Open it in a Web browser for a surprise. :)

You know, I thought you had to use a Duopoly browser if you wanted client-side XSLT but I just learned WebKit supports it as well! That means it works on Otter Browser and WebPositive. Safari too, probably, but we don’t talk about that one.

⤋ Read More
In-reply-to » I have a new Atom feed at https://mckinley.cc/blog/atom.xml. Open it in a Web browser for a surprise. :)

@lyse@lyse.isobeef.org Thank you, that’s a great compliment. I’m quite proud of that stylesheet, especially the bit that converts RFC 3339 timestamps to the friendly date format I use.

⤋ Read More
In-reply-to » @mckinley Kind of wondering whether you'd contribute your XLST to yarnd so we can use/apply it to all Atom feeds of the Pod's timeline and User feeds (presented as Atom)? 🤔 Could be interesting? 🤔

These aren’t things I could fix myself, and in my opinion some of them wouldn’t be worth it if I could.

⤋ Read More
In-reply-to » @mckinley Kind of wondering whether you'd contribute your XLST to yarnd so we can use/apply it to all Atom feeds of the Pod's timeline and User feeds (presented as Atom)? 🤔 Could be interesting? 🤔

@prologic@twtxt.net It would definitely be interesting. I don’t think there could be any substantial code reuse but I’m open to the idea. I looked into doing it, but there are a few hangups that would probably need to be addressed first:

  • Something in the pipeline is adding U+FFFD Replacement Characters to the ends of certain twts
  • Markdown isn’t rendered to HTML in the Atom feed, but mentions are.
  • <summary> elements (which should probably be <contents> instead) are type="html", escaped HTML. It wouldn’t be possible to embed twt contents in the generated page on Firefox-based browsers due to my favorite bug
    • The other options are either plain text or inline XHTML which means you have to make sure every twt is valid XHTML.
  • I don’t know how to handle different themes short of JavaScript which can be buggy when used with client-side XSLT or dynamically generated XSLT which is completely and utterly ridiculous.

⤋ Read More
In-reply-to » I have been hacking away on replacing the PicoCSS with Simple.css, since it offer a much manageable set of color variables that can ease the work of making your own team a whole lot.

@darch That’s exactly what I was thinking, it looks really nice. I can’t decide if I like the dark/blue or the dark/yellow theme better.

⤋ Read More