Kind replies

I found that writing short blog posts as a form of documentation https://www.jvt.me/posts/2017/06/25/blogumentation/ helps lots with reducing the barrier to blogging, as well as writing one post a day for Nablopomo https://www.jvt.me/posts/2019/11/01/national-blog-post-month/

Yeah I'd say unfortunately that's then removing the effectiveness of using a CDN. I've seen issues when using two CDNs as it can make cache invalidation etc a bit harder

Also doing NaBloPoMo (https://www.jvt.me/posts/2019/11/01/national-blog-post-month/ ) helped lots as I got used to publishing pretty quickly, which was kinda the point of it, for me at least

I used to refine things lots, but it meant spending hours on a post that took 5 minutes to read. Long posts I do still refine over time, but generally want them done at most a few days after it's finished being written.

If it's a Blogumentation post (https://www.jvt.me/posts/2017/06/25/blogumentation/ ) then I'll actively not refine it too much as a long post as the point is that it's for me in the future so if its not as readable I'll just edit it then

This is a great idea! I'm planning on doing a similar thing to this based on the "Week Notes" that Ton does ie https://www.zylstra.org/blog/2019/12/week-notes-1952/ as a visible way of recounting weekly achievements / what kept you busy

In the last couple of days, I've also added media support for my Micropub server, and have encountered the same issue.

Thankfully the changes to Indigenous are soon going live, but I was surprised not being able to use it elsewhere.

I believe https://quill.p3k.io supports it, as long as you're posting it with a note?

Glad to hear it! I've been toying with the idea for a few years but have never found the guts to do it

In 2018 I wrote 60 posts (92k words) but in 2019 I wrote 159 posts (107k words). Of those, 41 and 73 posts respectively were under the term "Blogumentation" https://www.jvt.me/posts/2017/06/25/blogumentation/ which I am a huge fan of.

Don't write for anyone else - just yourself!

I mean, I said that an hour or so ago, but that was before the site had deployed... Turns out https://www.jvt.me/posts/2019/11/11/gotcha-netlify-lowercase/ caught me out again and isn't fixed until https://gitlab.com/jamietanna/jvt.me/merge_requests/638 is in (although I've done a temporary deploy from my local machine to get the site up).

Very annoying!

This is really great stuff!

I've raised an issue on the https://brid.gy repo at https://github.com/snarfed/bridgy/issues/905 as I feel this could be a great fit for the project, as it'd handle that syndication for you, instead of you needing to set it up yourself.

It may not be needed by the community, but if it is, there's hopefully a good fit there.

If you want inspiration / don't want to write your own - https://indigenous.realize.be is great. I use it daily and it's a great Microsub client.

If so I've definitely been doing my Christmases wrong! Can't even think when the last time I saw it was

Out of interest where are you moving to?

Because GitHub is willing to work with ICE https://techcrunch.com/2019/11/13/github-faces-more-resignations-in-light-of-ice-contract/ but at least GitLab are being up front about their politics?

Although yes I agree they're not amazing

All my code hosting is on https://gitlab.com/jamietanna

Static site hosting is on https://netlify.com and VPS hosting is on https://hetzner.cloud

Although some things that I contribute back to are on https://github.com

Huge congrats Shaun!

Ah fair enough. Have you blogged about it before? Would be quite cool to read about it!

Nice! That sounds like a it's worth a blog post explaining how you did it?

I guess, that's a fair point. Would still be good if we can improve the usability of password managers for not-as-technical folks

I work on Open Banking APIs for a UK credit card provider.

A large reason I see that the data isn't made directly available to the customer is because if the customer were to accidentally leak / lose their own data, the provider (HSBC, Barclays etc) would be liable, not you. That means lots of hefty fines.

You'd also likely be touching some PCI data, so you'd need to be cleared / set up to handle that safely (or having some way to filter it before you received it).

Also, it requires a fair bit of extra setup and the use of certificate-based authentication (MTLS + signing request objects) means that as it currently sits you'd be need one of those, which aren't cheap as they're all EV certs.

Its a shame, because the customer should get their data. But you may be able to work with intermediaries that may provide an interface for that data, who can do the hard work for you, ie https://www.openwrks.com/

Have you by any chance heard of the #IndieWeb movement? We've got a great group of folks who are looking at what it means to #OwnYourData and #BeYourOwnSocialNetwork

I've written about it https://www.jvt.me/posts/2019/10/20/indieweb-talk/ and we've got a large wiki too https://indieweb.org/why

Thanks for sharing - I've been wondering about getting it since getting a Switch and may reconsider

I like that idea - automagically parsing the data from a URL is reasonable, and is mostly automatic but a little manual so there's the ability to correct issues (as mentioned in https://aaronparecki.com/2019/12/21/12/ )

Would we also be interested in iCal feed parsing, or just stick with MF2 for now?

This is really amazing - great work on this!

Out of interest, would there be any interest in the ability to syndicate events from our own websites to https://events.indieweb.org so there's a bit less manual process for adding them to the official list?

Aw thank you! It's my last day in Notts today tho so unfortunately won't be able to make it. Hope its good though - did you go see it with work on Thurs or is today your first one?

Also looks like this post doesn't have a valid in-reply-to url!

Could be that your reply is an h-cite maybe? So parsing it isn't happy? Odd as there's definitely a url in the body that references my post so it should be ok

Interesting - I'll look at giving Granary a go over the next few days and see if it's what I'm looking for

Thanks Chris - I'll look into it as both you and Aaron have recommended it.

As an FYI this post didn't send me a webmention and when sending it manually it said no_link_found - not sure if it's a known issue?

I've been using GitLab for almost 4 years now, largely for reasons stated in https://www.jvt.me/posts/2017/03/25/why-you-should-use-gitlab/ but also for all the extra stuff they've built on top over the years.

Originally it was just because of private repos, but evolved into a love of the platform being built in a way that folks can contribute, unlike GitHub.

Folks posting to their own websites and owning their content, but syndicating it to sites such as https://dev.to, https://lobste.rs or https://news.ycombinator.com for better discoverability

Sorry! Hope it didn't cause too much of an issue.

I wonder if these b issues you and Aaron have seen could be related and put into the spec for implementers?

As I've gone all in on the #IndieWeb technologies, I've been using @AaronPK's service https://aperture.p3k.io which supports the open https://indieweb.org/Microsub standard, which supports RSS among other formats - may be worth looking into as it's a great protocol for building better readers, even if it's not solving your need right now. Drop me or the folks at https://indieweb.org/discuss a line if you want to talk more about it!

I've disabled that for now - I actually didn't mean to do that and have since removed it, especially as those pages are paginated so you'll effectively get webmention'd forever as I add more content to that tag

I wonder what a reply on Twitter looks like?

I had a look at this last night and unfortunately it doesn't provide that much good data.

There's a year of listening history which may give a similar outcome, but it's not perfect

From me it's a no - I see why some people want it, but would rather prefer they stay out of the already limited space in the commit message title, and there's a level of arguable subjectivity of what an emoji means especially as different teams, projects and cultures have views on it.

But then again, so does written language, but I feel that is at least more known?

I've received my data from Spotify! Once I'm home and have had dinner, I'll get hacking on it to see what they've got for me and whether I can reproduce my Spotify Wrapped myself, for historic years, too!

Not really sure what the issue was, may be that php.microformats.io has been returning a bit slower than expected. But either way there's now webmention sending after a reboot of my post-deploy service 👍🏽

I had a quick look at the API but it doesn't seem to have a way to look at the user's play history. May require more digging, or using something like Last.FM

Thanks, I've raised https://github.com/swentel/indigenous-android/issues/262 for it

I guess it's more that I didn't want to write any posts and then not publish them, because if I've written it I'd rather have it out there than just sitting around, but have felt over the month I need to be careful in case I don't have things to write about?

What hardware are you running on? I've found it often doesn't work "out of the box" because the hardware manufacturers don't Open Source/upstream their drivers so it can't be released as part of the core distro offering.

It is definitely a pain for users, as it's not like ie Dell would say "don't buy this, it sucks for Linux usage!"

My Micropub endpoint has a fair bit of unit testing inside the Java project ( https://gitlab.com/jamietanna/www-api/tree/develop/www-api-web/micropub ) for common flows, but I've also found a tonne of implementation issues by integrating with real Micropub clients.

Some of it is an issue on a Micropub client, but most of it is something I've missed or assumed incorrectly.

I'm thinking to create a stubbed version ( https://gitlab.com/jamietanna/www-api/issues/26 ) that I can then use with https://micropub.rocks to ensure compliance.

Some of it is also a case of reading through the Micropub spec!

Ah nice! Thanks I'll have a play with it at some point then. Although I'm not sure I've got the time currently to write a spec compliant IndieAuth server too 😥

Interesting. I guess my main thought was whether it was something you could get Okta to only require MFA via Verify, or if you needed to do some hacking around it to make the flow work so you can just log in via Verify?

Did you ever write up how this works / could work? If so, I can't seem to see the post by searching passwordless

There may still be some dupes for now, looks like my async work has meant there are clashes with trying to write to the DB https://gitlab.com/jamietanna/www-api/issues/75

Looks like I've done it now - https://gitlab.com/jamietanna/www-api/issues/74 is the root cause, which I'll fix this evening properly.

For now, I've put a hacky hotfix in to get it working

But this unfortunately has not worked - I'll have a look at debugging this in the morning!

The link doesn't seem to be working BTW 🤔 seems to go to https://mrkapowski.com/2019/11/Array