#dev 2021-08-03

2021-08-03 UTC
nekr0z, reed, calebjasik, Lohn, vikanezrimaya, Abhas[m], hala-bala[m], SamWilson[m], astralbijection[, cambridgeport90[, P[m], LaBcasse[m], lasr[m], benatkin, [tw2113_Slack_], capjamesg[d], aaronpk[d], indieweb-irc-bri, hoenir, IWSlackGateway, sebbu, Saphire, capjamesg, petermolnar and hendursa1 joined the channel
#
petermolnar tiny bit offtopic, but not that much: I have mixed feelings on using git for my www folder, because git, unlike some other solutions, doesn't have the notion of history of a single file. I was wondering if anyone decided to use something else - svn, fossil, etc - for this specific use.
#
sknebel petermolnar: does SVN have a clearer notion of "history of a single file"?
#
sknebel (I guess it might handle renames more natively? thats the main "meh" part of git for me in that case)
#
petermolnar this is what I'm trying to figure out
#
petermolnar truth to be told, I never enjoyed svn, and that's worded lightly
#
sknebel the main issue with git IMHO would be renames, so unless that specifically is something you expect to come up a lot I'm not sure something else is worth it
#
petermolnar (though still better, than ibm clearcase or that monster called MKS Integrity)
#
sknebel for me, the overhead of (mentally and setup wise) dealing with something else likely would be too annoying
#
sknebel although I guess programmatic integration (if thats something your after) is painful enough in git that maybe that warrants it
#
petermolnar yeah :/
#
sknebel (I've considered integrating git to have access to edit history of posts)
jamietanna joined the channel
#
jamietanna petermolnar what do you mean not having history for one file? `git log --follow path/to/file` should show the history for that file, including renames, for all time? :D
geman and neceve joined the channel; geman left the channel
#
capjamesg Are there any good reads on improving relevance in search results?
#
capjamesg My blog search engine isn't as good at very broad queries (i.e. coffee) and I'd like to research what the best practices are.
hendursaga joined the channel
#
capjamesg Should internal links to a page count toward ranking? Should word count be considered?
#
aaronpk I think it depends on the goal of the search engine
#
capjamesg In this case, it's a blog search engine. The goal is for queries to as closely identify with the visitor's intent as possible.
#
petermolnar what was technorati?
#
Loqi Technorati was a real-time blog search engine that provided date-ordered results for text phrases or links, typically within seconds of when people published on their blogs https://indieweb.org/Technorati
#
petermolnar oh, misread your sentence
#
petermolnar don't mind me :)
#
capjamesg petermolnar Interesting nonetheless!
Ramon[d] and [KevinMarks] joined the channel
#
[KevinMarks] Technorati was mainly date ordered, but we also ordered by blog ranking (which was a count of links from distinct blogs in the last 6 months)
#
capjamesg Interesting.
#
capjamesg I have thought about making date a ranking factor for my search engine.
#
capjamesg But not a significant one.
#
petermolnar what were those "slow" and "fast" type of entries from a few days ago? If you were to apply that thinking capjamesg: there are kinda evergreen entries that last longer (eg. a review of a music album) vs things that expire easier (how to fix problem X on ubuntu 9.10)
#
petermolnar maybe that could be a factor
#
capjamesg I think the issue would be classifying fast vs. slow.
#
capjamesg I suppose I could add some kind of numerical classifier that my crawler can interpret.
#
capjamesg Also, I would expect that word count is, to an extent, correlated with fast and slow in that "fast" posts are not likely to be as detailed, etc.
Lohn, Seirdy and [jeremycherfas] joined the channel
#
[tantek] some interesting observations on "build for yourself", it's history, and how context for it can matter a lot https://chelseatroy.com/2021/07/30/the-oxymoron-of-data-driven-innovation/
KartikPrabhu joined the channel
#
@courtneyengle ↩️ All that, and do you want to aggregate the comments. I really like ingesting all of that generally using https://wordpress.org/plugins/webmention/. I've only published a few things since adding IndieWeb back to my site, but I like having the Twitter comments and such on my site. h/t @tnolte (twitter.com/_/status/1422646199996321793)
[jacky], KartikPrabhu and [aciccarello] joined the channel
#
[jacky] there's a small JS library (or snippet) that someone has that'd prefix `http://` in a text field
#
[jacky] I forget the name or link
#
[jacky] I _think_ it might be a p3k production
#
[schmarty] For sure it is aaronpk
#
aaronpk i think i have a blog post about that
#
[snarfed] https://aaronparecki.com/2018/06/03/4/url-form-field
#
Loqi [Aaron Parecki] Improving the HTML type="url" Field
#
[snarfed] more importantly it adds http:// _after_ they enter a bare domain and hit enter (etc)
#
aaronpk [snarfed]++
#
Loqi [snarfed] has 22 karma in this channel over the last year (47 in all channels)
#
aaronpk oh and if you want to default to https: https://aaronparecki.com/2019/05/13/2/https
#
[jacky] _perfect_, thank you!
#
Loqi [Aaron Parecki] Better Default Security for IndieAuth Login Forms