#capjamesg[d]How would I go about finding the location behind coordinates in a h-geo?
#capjamesg[d]I wanted to add support for parsing h-geo locations into IndieWeb Search.
#capjamesg[d]I use a Google API for converting coordinates into locations on my micropub server but I don't want to use that approach here.
#capjamesg[d]I also wonder if it's worth adding likes / comments as a SERP feature. Maybe not because the index might not crawl a page for a while. Still interesting!!
#capjamesg[d]In fact, maybe parsing h-geo isn't very useful, other than maybe to show a special "posted in" serp.
hendursa1 joined the channel
#Ruxtoncapjamesg[d]: do you just not want to lean on googles API's? or are you trying to do it without anyones APIs? cos you need a data set of geo+names to reference, i assume this is also only for cases where name isn't set already or is set to the co-ordinates values?
#nekr0zI'd expect there to be a name af a place or something in the post itself if it matters. If it doesn't and only coordinates are given, maybe simply showing those on a piece of map is a healthier approach than trying to guess which of a dozen cafes in that building the post was written in (and fail, because it was either author's office in the same building or the pub that is comfortably located in the backyard :)
#capjamesg[d]I wonder if I could use aaronpk's Atlas to render those coordinates. (aaronpk :D)
#capjamesg[d]Ruxton I think I have found a reverse geocoding package in Python. I haven't looked into it but I assume it uses some kind of local dataset.
#capjamesg[d]Using an API adds an extra API call to any request where I want to trigger the featured snippet.
#capjamesg[d]I suppose Atlas itself is an API though.
#capjamesg[d]nekr0z I parse h-adr for addresses. This was just for an extra feature. Maybe I could add the map as a bonus in a featured snippet.
#capjamesg[d]h-geo doesn't sound appropriate for identifying exact locations.
#capjamesg[d]What special function would be appropriate for a check in? A map in a SERP? A map in a special box at the top, like the rel=me and social boxes.
#[jacky]Dat, oddly enough, isn't really _that_ invested in it. I have to ask Paul more about it but people don't seem to use it a lot (I think it's a marketing thing and since he actively aims to build things _on top_ of Hypercore/Dat that are relatively usable)
#[jacky]IPFS is more block-chainy in the sense that it uses multihash concepts and Hypercore (at its core) is like Bittorrent for arbitrary data
#aaronpki'm seeing references to ethereum name registration in order to get a "domain name" for an IPFS site, so is that the only blockchain overlap with IPFS?
#[jacky]I don't know too much about Etherenum TBH so I can't say
#LoqiIPFS is short for InterPlanetary File System, potentially a static site storage method using content based addressing https://indieweb.org/IPFS
#nekr0zI used to mirror my site to IPFS (it's easy with a static site) and have a script automatically update `_dnslink` record to point to the latest version. It kinda worked.
#nekr0zI don't do that any longer, because IPFS is a memory hog that brought my cheap VPS to its knees every now and then, and I got tired of trying to tame it.
#micahrl[m]Not directly related to IPFS, but that discussion makes me realize that IndieAuth might be a cool way to set up a community on Tor. Hostnames are free and HTTPS certificates aren't necessary in an all-Tor community that uses IndieAuth for validation, and you can even run a site on a computer at home without worrying about firewalls or port forwarding.
#capjamesg[d]Although I’m going to stick with Monocle for feed reading.
gRegor joined the channel
#[tantek]1no direct tie between IPFS and all the "web3" hype. all the indirect stuff (but you *could* use it for storage for some other random web3 thing) is not a connection
#[tantek]1just like JPEGs aren't an NFT thing, even though a lot of NFTs use JPEGs or at least links to JPEGs (which die pretty quickly themselves)
#[tantek]1aside, in going through some old papers, I found the original "Web 2.0" concept map papers / exercises that O'Reilly printed out and handed out back in the day, before "What is Web 2.0" got big. Ironically (or appropriately?) even the concept of Web 2.0 itself was somewhat (small) crowd-sourced.
#[tantek]1then page 3 was the same style of image except with heading "Web 2.0 Meme Map" and all the ovals were empty, and the lists were empty too (but still had the headings of "Strategic Positioning:", "User Positioning:", "Core Competencies:"
#[tantek]1then I have a loose page (which maybe they printed out and handed out after the session?) with *some* of the ovals filled in and some of the lists with items, also with heading "Web 2.0 Meme Map"