IRC channel logs

2024-09-21.log

back to list of logs

<dthompson>wizard: lol kinda surprising since I still don't ship a real default theme (though maybe that's a good thing)
<dthompson>glad you're having fun, though
<wizard>hey at least you ship a real default way to lay out the blog posts and rss feed with 0 extra configuration
<robin>dthompson, yeah, i'm having similar issues with markdown. i was very focused on hypertext and CSCW and wiki markup and parsing theory for a while... still have some ideas about parsers/unparsers that'd make it easier to experiment with modifications, haven't had time implement them yet, though
<robin>djot.net is interesting (designed by the author of pandoc)
<robin>in terms of more complex markup, reid's scribe syntax is quite nice. erik naggum also had a ton of ideas on post-sgml markup languages if one wants to dig around old usenet posts
<robin>(i think he was involved in sgml standardization, or at least applications)
<robin>which reminds me, i need to report a significant bug in (ice-9 peg) now that other people are reusing the underlying code (a PEG parsing system for s48 was one of my first proper scheme projects, almost immediately after they were "rediscovered")
<dpk>SGML, the Common Lisp of markup languages
<robin>but my only task today is submitting an emacsconf proposal for a guile-emacs talk :P
<dpk>alas, strangled in infancy by its own baby (if infants could have babies of their own)
<dpk>robin: what’s the future of guile-emacs? please spoil as much of the contents of your talk as possible :D
<robin>dpk, it's funny you should mention that! i don't recall the details, but erik had a proposal for something like s-expression-equivalent sgml that could also, ofc, serve as a lisp syntax, for a markup language called "salt" and a lisp interface called "NaCL", or something along those lines. i'll have to dig up the comp.lang.lisp posts sometime :p
<robin>dpk, will do! i also have plans for a new kind of publicly-funded workers coop (related to snowdrift.coop ideas, and my years of experience in "traditional" democratic spanish coops) that might interest people in its own right...if it works ;)
<dpk>on my long-term projects list, besides an accompanying essay entitled ‘SXML considered harmful’, is an angle-bracket-based-markup-language processing library for Scheme based directly on the provisions of the query and transformation languages of DSSSL
<robin>ooh. you'll definitely be interested in erik's ideas then
<robin>dpk, new TODO: dig up some relevant posts for you ;)
<dpk>:D thanks!
<dpk>i know his infamous anti-XML rant (and recall his proposal that an SGML successor should have Lisp-like macros as its main markup minimization feature) but not a huge amount more
<dpk>also: yay worker’s co-ops :D
<jfred>++ on worker's co-ops :)
<robin>it might be wishful thinking, it'd be entirely donation-funded at the beginning (focused on guile-emacs and its components) and would need revenue to support 2-3 US FTE-equivalents and overhead, say 330k USD/year within the next couple years, with nothing fancier than a simple LLC to start with. (feasible to allow external investment -- but not as lucrative as an average startup -- and may qualify for loans from ncb.coop, but there's really neither time nor
<robin>capital to start out with a nonprofit structure or support from public grants...) so it might devolve into a "keep-a-roof-over-the-maintainer's-head" model with additional democratic structure for some period of time, but it would be a very interesting model for other large projects if it does work
<robin>(i'm not opposed to traditional consultancy work for revenue in the future, but it'd be too much of distraction for a tiny coop at the very beginning)
<robin>we would likely be among the first beta testers for snowdrift.coop if things start well :)
<robin>(it'd be possible to start out as a nonprofit, but probably not a 501(c)(3) -- rather one of the more obscure ones like a 501(c)(12) -- and i'd expect that to make restructuring more difficult in the future)
<robin>(or rather i know that would make restructuring difficult from a decade as an hcoop.net director)
<robin>but i'd better finish this proposal in the next few hours and then spoil dpk as thoroughly as possible :p
<haugh>Nice release folks! Haven't played with it yet but I'm very happy to see persistence being prioritized as I think this is one of the key bottlenecks in distsys adoption. WRT UI, is there much awareness/interest of HTMX/Hyperview in this community? I haven't actually applied it to a project yet but its concept of hypermedia as data APPEARS to synergize very well with the actor model.
<cwebber>hi haugh !
<cwebber>there's a bunch of interest internally about the UI stuff
<cwebber>we haven't started evaluating it fully yet
<cwebber>but you may have seen from a year ago: https://spritely.institute/news/building-interactive-web-pages-with-guile-hoot.html
<cwebber>well, not quite a year ago :)
<cwebber>and more recently dthompson wrote: https://dthompson.us/posts/functional-reactive-user-interfaces-with-propagators.html
<jfred>The propagator UI stuff is super exciting to me. Probably needs a higher-level API layer on top to make it more ergonomic to use, but it feels like a really nice reactive UI foundation
<cwebber>Vivi has been experimenting with a higher level API
<cwebber>she hasn't posted about it publicly yet tho
<cwebber>encryptedwhispe- that is
<jfred>:D