jackybut eh, if I can use a open solution over a bigco one, I'll try - easier to see if they're vacuuming data or opening it up for others to do it "for product excellence"
[tantek]I've definitely poked people to review bunches of the IndieAuth / RelMeAuth code over time, and I know at least some of the old RelMeAuth libraries (PHP) got review back in the day (like early 2010s)
[tantek]The other point you made is the fact that your custom one-off code for your site is unlikely to attract the attention of someone looking for an exploit
[tantek]of course that doesn't mean your site is "technically" secure, only secure from a time investment / obscurity perspective (which is still *a* measure of security even if not really a great one)
[tantek]like yeah, if you expect to be targeted, for any reason (high profile, known trolls, public speaker etc.), then you really should delegate any security related code/services to things that are actually hardened
[chrisaldrich] and dougbeal|mb1 joined the channel
[LewisCowles]Firefox has a tool if you inspect element on something using clipping paths, it allows you to interact with the page. Really needs two screens to work well
GaffenDoes anyone have any thoughts on running their own CI server? I've been considering it but am worried it may be asking too much of my weedy little server
[LewisCowles]I'm always left happy when interacting with proprietary tooling as it "just works". I'm aware that is a lazy point of view, and I wish there were easier ways to use OpenSource; but generally "more fiddling required"
[LewisCowles]Unfortunately it was such a waste of compute taking > 500USD to run CI/CD for a company with less than a million customers and not much technical excellence
[g33kcentric]cool, i did wonder about a similar setup, but it seems a little redundant to have a cms online if its just going to get piped through to a static site anyway
GaffenI've also started considering using the wordpress install for commenting functionality, though I'm not sure how silly an idea that is. Like if I'm bringing so much functionality from wordpress to a static site... Maybe I should just consider using wordpress lol
[LewisCowles]webhook to GitHub to publish the JSON & have it poll every n minutes / hours etc from your PC connecting to git, pulling, building, pushing
[LewisCowles]so, using a filter that makes a wordpress REST API call and sends that to GitHub via Github REST API to store JSON (I'd advise in a separate repo, with some signing)
[LewisCowles]the good news is you can turn it off and add a hook to grab the content on post save and send somewhere offline, like git, or S3 (with revisions on) or any storage of your choosing.
[LewisCowles]To each their own. I've only recently last year noticed Indieweb through a co-worker and run to mf2 as an alternative to schema and other structured formats as a thing to observe and experiment with
ZegnatWonder how far 50K takes you for a full audit. Guess it depends on project size. TrueCrypt Audit I think raised almost 50K and I do not remember if any money was left over after the two part audit.
[jgmac1106]all I wanted to do was display my orcid in my h-card...look how much metadata gets wrapped around an img and link and inline styling: <div itemscope itemtype="https://schema.org/Person"><a itemprop="sameAs" content="https://orcid.org/0000-0003-1035-389X" href="https://orcid.org/0000-0003-1035-389X" target="orcid.widget" rel="me noopener noreferrer" style="vertical-align:top;"><img
[g33kcentric]So i think my goal for tonight, is to build a system that sends newsletters i subscribe to, too my site where they will be saved as h-entries, possibly with custom styling for each one (obvs outside of MVP)
[KevinMarks]getting the API key was a pain. Also the API is a bit bureaucratic - you need to pass it one of it's location IDs so you'll need to do a find closest on the sitelist (I hard-coded mine when I found the location api was broken)
hs0ucy, [tantek] and KartikPrabhu joined the channel
[_iamdave]Getting some chores done presently but I’m going to be experimenting with rendering riddlers to static pages, and playing about with view templates some more
[manton]Hi everyone! Anything new going on with Instagram → Micropub solutions? We've had an Instagram import (from .zip) in the Micro.blog macOS app, but wondering if I need to spend some time on something for non-Mac folks now. I assume little chance of Own Your Gram coming back?
[chrisaldrich]thanks [_iamdave] I've been tinkering with TiddlyWiki locally for a while, but it's nice to have one on my own domain. I'll be digging into h-cards and templates for other pieces shortly.
[manton]I wonder if a browser extension would be a good solution. Visit instagram.com while signed in and click a "move everything I see on this page to my blog" kind of a workflow.
aaronpkso it's a matter of scraping. they do have a lot of JSON on the page, but they're aggressive about whether or not they return a web page if they think you're a bot
[snarfed]aaronpk i assume you still want to avoid scraping with a logged in cookie? i can commiserate, but i gave in and started doing that in bridgy a while back, and it's been solid and reliable ever since
jamietannaAdding to the conversation earlier about CI/CD - I use GitLab CI for my stuff and it's great, but I also use a fair bit of Jenkins for work and wanna fight its corner that it's all about how you set it up, it can be good to work with
[manton][snarfed] [aaronpk] Thanks… I might tinker with a browser extension if I have time. (But then there's the question of which browsers to support. Sigh.)
ZegnatCross-browser extensions are getting better. Definitely possible for something like a scraper too, where you are not interested in browser-specific preferences stuff
[Michael_Beckwi]are cross browser extensions really getting better or is it because they’re all basically running the same underlying engine, except Firefox?
ZegnatIt has always been cross browser between just Firefox and Chrome, basically. Especially what with Safari killing their extensions and the IE extension space never really getting off the ground.
[manton]Just a quick search for cross-browser extension frameworks makes me think starting with Firefox-only makes the most sense, then maybe adapting it for others. Happy to hear if there's a good framework, though!
ReneM[m]Would a IFTTT solution, which has API access to IG, the best solution for now and initiate a web request when a new photo is added to a IG profile?
ZegnatReneM[m]: can IFTTT send the photo and caption along in the web request? Because I think one of the problems may just be accessing the photo on IG at all.
[manton][Zegnat] That's cool, thanks. The problem for me is Safari, since that's my default browser. (Although I've written Safari extensions before and could handle that separately.)
ZegnatIt is all AppKit and stuff now. I’ve collected some links on that wiki page, [manton]. You as a macOS dev might have better luck with that than me as a webdev. I doubt there are any cross-browser frameworks that still include Safari left
ReneM[m]Since IFTTT has tons of ready-to-be-used Apllets for Instagram to e.g Twitter, Facebook pages that are used my hundred thousands (according to their stats) I doubt that this is a temporary solution😉