#ben_thatmustbemecsarven: you are talking about permashortcitations?
#ben_thatmustbeme"ttk.me/ t4bV1" wastes an extra letter, and twitter i think will interpret the first part as a URL and create the link, which negates the original point of them
#ben_thatmustbeme"ttkme:t4bV1" is not recognizable as a URL, and also is ambiguous tt.kme? ttk.me? t.tkme? works for this case only because those others are not TLDs, but makes processing messier for sure
#csarvenben_thatmustbeme Pardon me.. I made a typo. I mean to say "ttk.me/t4bV1" (without space)
#csarven!tell tantek ..Made a typo; meant to say "ttk.me/t4bV1" (without space)
#ben_thatmustbemeit sticks the cittation to the text unless its specifically parsed out
#csarvenPoint being that you can never control how something else will interpret the characters "a b". It is just a string using an arbitrary pattern which is not inline with any existing or widely used system. This is in contrast to "a/b" which at least mimics what is known and arguably meaningful for many.
#ben_thatmustbemefor the record, i also dropped PSCs in favor of PSLs
#csarvenLet me ask this: What is the fundamental purpose of indicating the citation? And, how is the solution enriching that UX?
frzn joined the channel
#ben_thatmustbemethe fundamental reason was to give a link without making a link as there was not really UI need to return to the original post.
#csarven"a b" in status updates that are not hyperlinks is equivalent to "view only" citations. Pick up any academic paper (in PDF) and see the reference section - just some string that one has to copy/paste elsewhere to retrieve the information - bad UX IMO.
#ben_thatmustbemefirst off, this isn't about citing in a post thats entirely different
#csarvenI fail to see why bother with it in the first place. It is adding more complexity (for interpretation) than any practical use.
#ben_thatmustbemethese are on accounts on silos where the is a link in the user profile
#csarvenDoes it any practical use? Is there any documentation / survey on people copy/pasting "a b" and then changing it to "a/b" to retrieve the resource?
Zegnat and j12t joined the channel
#ben_thatmustbemewhy is that needed. its intended to say you never need to click back as there is nothing more there, but in the rare case you do actually care, i want it known that this is published elsewhere
#csarvenIt sounds like "well, this is a hypothetical solution to a problem that doesn't exist"
#ben_thatmustbemeif a large number of people did copy paste it and change the ' ' to '/' than i would say just drop PSCs and use the PSL
#ben_thatmustbemethats not what we do, its tantek and others documenting what they do
#ben_thatmustbemeon syndication, "I'm bored ttk.me/ptsxt" ant the link is clickable
#ben_thatmustbemeit would be read as "I'm bored, heres a link explaining it"
#rhiaroI definitely get people complaining about links to posts that don't contain anything extra. But I think if I broke the links, I'd get an equal amount of people complaining my links weren't right (documented on wiki)
#ben_thatmustbemeand clicking through to see the exact content is frustrating and thus bad UI
#rhiaro*but* when pulling enough replies through from different sites, the original copy of the post *is* enhanced
#rhiaroie. you can click from twitter and see replies from indieweb, facebook, whatever
#ben_thatmustbeme"I'm bored (ttk.me asdfq)" and most will ignore the PSC, especially if they are followers
#ZegnatI think rhiaro makes a good point. When is the original post enhanced enough that it is worth linking to? PSC is used for posts when there is no extra content to be had at the website, but comments, likes, etc could count as extra content
eschnou joined the channel
#ben_thatmustbemeZegnat indeed. I honsetly have removed PSCs entirely as I feel that is the case, but I do see their use. the UI of a link thats not labeled as a link to the same content is frustrating UI
#ben_thatmustbemeunless you start putting in "original post: ttk.me/zzzzz"
#LoqiA permashortcitation (or permashortid), abbreviated as PSC is a short non-hyperlinked citation to a post permalink, in contrast to a permashortlink which does hyperlink to a post https://indiewebcamp.com/PSC
#Zegnatlooks like tantek himself brought up the “richer experience” argument on 2013-11-08, documented on the wiki
j12t, wolftune and KartikPrabhu joined the channel
#kylewmZegnat: I've seen lots of people who blog a lot and measure these things say they get MANY more readers when they post on Medium rather than their own site
#kylewmwhat you say about feeds is technically true, but discoverability is a problem
#ZegnatSo, I am not a big Medium user. How does Medium help with discoverability? I often end up on Medium only because people link to good articles. People link to good articles outside of Medium too, and I read those too.
#KevinMarks_Medium will show you "what to read next" if you're logged in and email you too
#ZegnatKevinMarks_: does it email you a “what to read next” type of thing? Or only things you have explicitly asked for?
parzzix_, eschnou, parzzix__, j12t and parzzix joined the channel
#KevinMarks_Medium daily digest from authors and publications you have followed
parzzix__ joined the channel
#ZegnatSo the daily digests do not actively help author’s discoverability. Because that is the big part here: why would an author be more “discoverable” on Medium than on a personal blog.
parzzix joined the channel
#rhiaroaargh why can't I browse twitter by date ranges
#rhiarois full of regret for not going full-posse sooner
Garbee joined the channel
#rhiaroon a related note, has anyone had any look with the twitter export lately? I never get an email to say it's done
KevinMarks joined the channel
#rhiaromy timeline gives up loading at 10 Sept 2014? whu?
wolftune, KevinMarks_, j12t and scoates joined the channel
#rhiaro(never mind, twitter advanced search can do date ranges)
#Kongalooshrhiaro: how do you feel about storing your information. You use a DBMS, but it seems like there's a lot of people just storing things as html.
#KevinMarks_“You can get a site up and running in PHP in a few minutes, and that’s the problem. It used to be the terrible choice you made when you needed to get something done on the Web, but increasingly JavaScript has replaced it as the default terrible choice.”
#kylewmKongaloosh: "a lot of people just storing as html" would be an exaggeration. it's an interesting approach, but i think maybe only 1 person here does it
#kylewm(but fwiw, I personally went back and forth and use postgres now)
elf-pavlik joined the channel
#aaronpkKongaloosh: I store everything in files (yaml+markdown) but have a database which indexes everything except the post content. makes it easier to query by tags and date ranges
#Kongalooshmmm, that's the road I was thinking of going down
#aaronpkif the DB is corrupted I can easily re-create it because it doesn't store the canonical version of anything
#aaronpkalso makes it easier to synchronize between my dev copy and live site, because syncing files is easier than syncing DBs
#Zegnataaronpk, have you had enough database corruptions to warrant building such redundancy? Or is that just a lucky side-effect?
#aaronpki havent had my personal site db fail yet, and I have managed to keep quite a few mysql DBs around for many years. I think the bigger reason I like this is I don't have to worry about syncing data between my laptop's DB and the server's
#aaronpkI have definitely had other databases fail on me in the past though
#ZegnatI managed a lot of forums throughout the years and I have never had mysql fail on me, so I was just wondering :)
#aaronpkas long as you're diligent about backups and exporting to SQL text when you need to upgrade mysql versions is isn't a problem
#aaronpki'm much less confident in my ability to admin a postgres database, even though there are so many things about postgres I like better than mysql
j12t joined the channel
#kylewmaaronpk: that bastianallegier project sounds really similar to what you do, flat files with some redundant data in SQLite (intead of mysql) for indexing
#aaronpkoh nice! it looks like a generic data store rather than something content-aware
KevinMarks__ joined the channel
#aaronpki had a great time chatting with bastian during indiewebcamp germany
scoates, behind_you, Erkan_Yilmaz, wilfredh, KartikPrabhu and j12t joined the channel
#aaronpkwow, the wiki header was using an image to get that gradient since it was made so long ago
#aaronpkhrm. i normally don't like templating engines because they invent their own syntax and php *is* a templating engine, but i'm seriously considering using the Blade engine from Lumen http://lumen.laravel.com/docs/templates