#indieweb 2021-07-20

2021-07-20 UTC
_jon joined the channel
#
[tantek] no [KevinMarks], though you did send me down the rabbithole of Hixie's blog where I found things to post in #indieweb-meta, #indieweb-dev, and #indieweb-chat
vroman joined the channel
#
[KevinMarks] https://flutter.dev/go/vector-graphics
_jon joined the channel
#
[tantek] hah I didn't bother clicking on any of the Flutter links
#
[tantek] since the topic says "format", let's move this to #indieweb-dev
Seirdy, jeremycherfas, wagle and _jon joined the channel
#
Loqi [indienews] New post: "Building a search engine for my blog" https://jamesg.blog/2021/07/20/building-a-blog-search-engine
bleb, Tomte, hendursa1 and [tantek] joined the channel
#
[tantek] do we have a common / recommended strategy for how to update posts to fix links to domains that have been abandoned / taken over?
#
[tantek] to me, linking to something other than what you intended when you published the post is worse than a dead link that 404s
#
[tantek] part of the solution may be to link to the internet archive version of the destination as of when you posted
#
[tantek] however, I'm specifically thinking of cases where domains / links were visible in the text of the post (e.g. domains / URLs that were auto-linked)
#
petermolnar I've been hesitant on doing that, but I used to ping archive.org to save all the links I'm linking to from a post
#
[tantek] yeah I've been doing that for a while too, however I have a lot of such links from before I started that
#
petermolnar I haven't been doing that for a while, but in theory, based on my initial publish date, I could find the relevant one from archive.org
#
[tantek] that's kind of what I was thinking, which is a manual task I suppose
#
petermolnar not necessarily
#
[tantek] because you have to view what's on archive.org to make sure it's the thing you meant to link to
#
[tantek] however once that is done, I'm wondering if this can be automated
#
petermolnar if you have the orig publish date of your entry and you manage to find one of the link on archive.org close enough to that date, it can probably automate most of your cases
#
petermolnar leaving only the ones without a close enough in time find
#
[tantek] I agree with that. It can at least auto-repair it in such a way that "all" you have to do is review it.
#
petermolnar there's http://web.archive.org/web/timemap/link/URL
#
petermolnar (as in replace URL)
#
[tantek] What I wonder is if that kind of "review" can be applied to a domain granularity, and then used to build up a table that the autolinker would use automatically
#
[tantek] e.g. I have a number of posts which link to federatedsocialweb(.)net
#
petermolnar we should bring this to -dev before Loqi tells us off :)
#
[tantek] yeah we were skirting that edge
#
Loqi yea!
#
[tantek] lol
[KevinMarks], rockorager, _jon, hendursaga, [grantcodes], alex11, angelo, [tantek], n8chz, [snarfed], inky, [tw2113_Slack_], [arush], n8chz1 and samwilson joined the channel