#dev 2025-08-02

2025-08-02 UTC
bugliker02, grufwub, lurkie, gRegorLove_, sp1ff, doesnm, [aciccarello], balintm, marc_in_space and cobbland joined the channel
#
[KevinMarks] This is brilliant https://pridever.org/
ttybitnik and sebbu2 joined the channel; cobbland left the channel
#
trwnh idk if no-js should imply no-css but you can always serve plain html if you like and rely entirely on default user agent styling. it's just that there isn't really as strong of a reason to bother -- js is arbitrary code while css is declarative. there is a much more pressing reason to disable js than to disable css
#
trwnh re: graceful degradation / progressive enhancement, i do think we should try for writing semantic html first that uses no css or js, then write the css afterward
#
trwnh they are separate concerns in theory, even if a lot of people only care about layout and looks (hence how we got table layouts decades ago)
GuestZero joined the channel
#
[tantek] What is semver
#
Loqi Semantic Versioning (semver) is a method of building version numbers such that they convey meaning about the state of the underlying code - also known as MAJOR.MINOR.PATCH format https://indieweb.org/semver
#
[tantek] semver << Alternative: https://pridever.org/
#
Loqi ok, I added "Alternative: https://pridever.org/" to a brand new "See Also" section of /Semantic_Versioning https://indieweb.org/wiki/index.php?diff=103119&oldid=43323
duanin2, GuestZero, [Jo], perryflynn and pre1ude0 joined the channel
#
[artlung] I'm pregaming my HTML Day participation looking at XHTML 2 and remembering how ambitious, and yet how wrongheaded it is, despite being made by smart people. But I think I've decided not to write a document in it because the W3C Validator doesn't validate it.
#
[artlung] https://www.w3.org/TR/2010/NOTE-xhtml2-20101216/introduction.html#s_intro_differences is a parade of markup changes that make my jaw drop every time I read them.
#
[tantek] It's good that it happened to so we can use it as an example to teach people what not to when iterating on HTML
#
[tantek] Also a great counter example against the whole myth at the time that you could easily start with just XML and make your own new standard and get people to adopt it
gRegor, lurkie and thegreekgeek joined the channel
#
[social] I’m happy to have stumbled on this on HTML Day. Jen Simmons is asking about what should Apple add into Webkit for Safari - https://front-end.social/@jensimmons/114915019192417238
#
Loqi [preview] [Jen Simmons] What do you need most from WebKit (the rendering engine for Safari that runs your HTML, CSS, JS, and more)? If you are making websites, what could we do in WebKit to make it easier for you to create fantastic experiences for your users?
lurkie joined the channel
#
gRegor Random thing we discovered with the W3 validator, it doesn't seem to support the "check" urls with HTML5 doctypes. Joe's HTML2 page shows "successfully checked as HTML 2.0" https://validator.w3.org/check?uri=https%3A%2F%2Ftilde.club%2F~artlung%2F
#
gRegor But with my HTML5 page, it redirects tot he Nu HTML validator https://validator.w3.org/check?uri=https://gregorlove.com/lab/
#
gRegor It's minor, I was just looking for a "valid HTML5!" button similar to that HTML2 example.
srazkvt joined the channel
#
gRegor My HTML Day project was setting up this minimal layout "lab" section, and after realizing no valid HTML5 buttons from W3C, made my own: https://gregorlove.com/lab/
#
[mattl] My project was to put up a current homepage for https://roguetrader.us
#
capjamesg [mattl] Nice!
#
[mattl] Yeah, couldn’t do much as have to do the actual movie too, but I might have time to work on another prop later.
#
capjamesg [mattl] If you commission a movie about waffles and/or coffee I'd be interested.
#
capjamesg Sandwiches at a push, but I'd really shine in a waffle or coffee production I think.
#
[mattl] I’ll stick it on my todo list
#
[KevinMarks] Add the sites to this https://www.are.na/html-energy/websites-made-on-html-day-2025
[Zephyr], rrix, [benatwork], box4649, lurkie, rdg and gRegor joined the channel
#
gRegor what is arena
#
Loqi It looks like we don't have a page for "arena" yet. Would you like to create it? (Or just say "arena is ____", a sentence describing the term)
#
gRegor what is are.na
#
Loqi It looks like we don't have a page for "are.na" yet. Would you like to create it? (Or just say "are.na is ____", a sentence describing the term)
oodani and Ryan1 joined the channel
#
[tantek] It was also one of the first web browsers https://en.wikipedia.org/wiki/Arena_(web_browser)
#
rdg request for comments: https://dillo-browser.github.io/rfc/004-optimized-for/
oodani joined the channel
#
[mattl] rdg: I like the idea of saying ‘optimized for basic-CSS’ or ‘optimized for no-javascript’ vs a particular browser/browsers.
#
rdg [mattl]: that seems orthogonal, no?
#
[tantek] Basicss would be a great name for a simple CSS style sheet for popular types of content (index pages, streams, post permalinks)
#
[mattl] rdg: not clear what orthogonal means tbh.
#
rdg by orthogonal I mean that you are addressing a different problem that the one that RFC is targeting
#
[mattl] I guess it’s unclear to me how a page might be optimized for dillo and w3m and elinks.
#
rdg we don't support all CSS or HTML features, so you can exclude them from your page and it would look good
#
rdg but that causes problems on our side as we want to assign more priority to features commonly used
#
rdg so it is good to be able to remove the pages that were explicitly optimized for dillo when doing such analysis
#
rdg many pages state such information informally, but some dont
#
[tantek] oof, reinforcing any kind of per browser (version?!?) "optimization" feels like encouraging a bad development practice that I thought we had left behind in the late 1990s 😬
#
[mattl] Maybe its just a bad example, but it doesn’t seem like elinks + w3m + dillo have much beyond HTML support. elinks and dillo have some CSS support, not sure about any JavaScript, but w3m is text only, no CSS, no JavaScript.
#
[mattl] tantek++
#
Loqi tantek has 30 karma in this channel over the last year (147 in all channels)
#
[mattl] rdg: are Dillo+, Dillo-Plus and D+ browser the same thing? Wikipedia article for Dillo could use some tidying up for sure.
#
rdg it would be great to have a simpler HTML spec, but we don't have it yet
#
rdg so in practice, many people test their sites in different browsers, which is a good habit I believe
#
rdg [mattl]: they are all forks, some are very old
#
rdg there is an edit war at the dillo page, and I'm not interested in pursuing it further, see the talk page if interested
#
rdg the authoritative information is at our website, which has proper references
#
[mattl] Yeah, I heard about the domain name loss/hijack.
#
rdg I may talk with the EFF to go to court if needed, but for now the github pages seem ok
#
rdg we don't have enough funding to keep the domain and server for too long either
#
rdg here is more info if you are interested https://dillo-browser.github.io/dillo.org.html
#
[mattl] does GitHub itself work reasonably well with Dillo?
#
rdg no, is a nightmare
#
rdg they have been incresignly reducing the features available with HTML only
#
rdg I've been exploring other forges, and so far only cgit and sourcehut seem promising
#
[tantek] does seem weird why they need to use JS to show, checks notes, what are "simple" text files
#
[mattl] AFAIK, and it’s been a few years since I looked at this stuff but I think Savannah is the only forge which 100% works without JavaScript, but I know there was some effort for self-hosted GitLab CE to work like that too.
#
rdg they seem to use cgit: https://cgit.git.savannah.gnu.org/cgit/
#
rdg cgit is fine on dillo after some patches: https://fosstodon.org/@dillo/114927456382947046
#
rdg same history for mastodon, they refuse to even let posts be shown without JS
#
[mattl] Savannah supports a lot of stuff… I don’t know how you accept bugs but their bug tracker should work too.
#
rdg I was looking for some bug tracker as well, bugzilla seems to work fine with dillo too
#
[mattl] yeah, i don’t think they provide a bugzilla install just the barebones thing.
#
[mattl] [tantek] it’s very strange when you consider that GitHub at least has https://cli.github.com
#
[tantek] that seems, uh, ironic since git *is* command line?
#
rdg I used the cli to manage the CI, git doesn't do that
#
rdg it also allows writing and replying to issues via vim, so that is also nice
#
rdg nothing that could not be done from a plain HTML interface
#
[tantek] gRegor I bet there are valid HTML5 buttons from W3C hiding somewhere
#
gRegor I really thought I'd seen them before. Came across this page but it hasn't been updated since 2009 https://www.w3.org/QA/Tools/Icons
#
gRegor Did look in the "repository" (regular directory page, haha) very briefly, nothing jumped out
#
gRegor It was a fun thing to tinker with making, though.
#
[mattl] If not from W3C maybe WHATWG?
#
[tantek] don't think WHATWG made any validator buttons
#
[tantek] gRegor, yeah you're right, no sign of any HTML5 (or plain HTML) valid buttons here: https://www.w3.org/Icons/ (but lots more, search for "valid-")
#
[mattl] https://peerdh.com/blogs/programming-insights/implementing-custom-validators-with-html5-and-javascript
#
[tantek] an email about it from a W3C list: https://lists.w3.org/Archives/Public/www-validator/2020Jun/0015.html
#
[tantek] gRegor, these look pretty neat: https://github.com/bradleytaunt/html5-valid-badge
#
[mattl] I think I was thinking of the orange “HTML5” button/logo
#
[mattl] It looks like a shield
#
[tantek] oh yeah I remember that
#
[tantek] https://www.w3.org/html/logo/
#
[tantek] hah this is great: https://www.w3.org/html/logo/#badge-builder-header
#
[tantek] I should slap this one on my site: https://www.w3.org/html/logo/badge/html5-badge-h-css3-semantics-storage.png
#
Loqi +1
sp1ff` joined the channel