Replying to MuellerSimhofer:

As user I’ve seen Spotlight implementations for document based apps (usually just using NSUserActivity), but the problem is that they usually duplicate results. You get one result from Files and then another one from the app’s index.

Also keeping the index up-to-date seems hard, since I can delete files outside the app, but the app’s index still contains it: I have to start the app, to get the index updated. I’m curious how you guys can solve this for MindNode.


The Smartphone Isn’t Evil, Chill

Ben Brooks writing about the topic of “smartphone addiction”:

Too much is getting blamed on smartphones — that people are addicted to the phone. Which isn’t actually true. People are addicted to Social Networks/Media — yes. But the phones, no.

I couldn’t agree more. People do all kinds of bullshit, like grayscaling their screens, turning off all notifications and using dumbphones from 1977. These are cheap and stupid lifehacky ways to not being honest with yourself and getting away from the real question: what are you really addicted to?

As I wrote before:

Well, I don’t want to pop the productivity bubble of blocking stuff, but what helps is deleting your account from Facebook. The same applies to phone addiction: you have to remove those time-wasting apps from your devices and then be mindful of how you’re using your phone. Turning on an accessibility setting then complaining about colored app icons that make us addictive to stuff is just stupid. Addiction is a way bigger problem than that. You have to acknowledge that your phone is just a tool and it’s your responsibility what you’re using it for.


I’m doing my last GTD Weekly Review this year. It’s gonna take a couple of hours, but it’s nothing different than doing any other review in mid-year. That’s why I think new year’s resolutions are bullshit. Do something or don’t, but don’t wait because the Earth is gonna be in a specific position related to the Sun.


Yesterday we finished watching Mr. Robot with my wife. It was a perfectly constructed story that was slowly built-up right from the start. I’ll wait a couple of months, but I’m definitely going to rewatch the show, since a lot of small details was hidden in previous seasons. These are going to get a completely new meaning now that we know how it ends.

Meanwhile, I’m just gonna listen to the original soundtrack by Mac Quayle, which is just as good in itself as the series was.



Actually @jack, you know what? Go and invent another protocol, but make sure it can’t be used to edit posts.

I disabled Gutenberg yesterday.

At first, it does look great, but after switching off the visual editor, it becomes a simple textarea. There isn’t even an insert media button, so I reverted everything to the old one. It should be supported for another 3 years, so I’m fine for a while.

The Decade the Internet Lost Its Joy:

By 2010, personal blogs were thriving, Tumblr was still in its prime, and meme-makers were revolutionizing with form. Snapchat was created in 2011 and Vine, the beloved six-second video app, was born in 2012. People still spent time posting to forums, reading daily entries on sites like FML, and watching Shiba Inus grow up on 24-hour puppy cams. On February 26, 2015—a day that now feels like an iconic marker of the decade — millions of people on the internet argued about whether a dress was blue or gold, and watched live video of two llamas on the lam in suburban Arizona. Sites like Gawker, the Awl, Rookie, the Hairpin, and Deadspin still existed. Until they didn’t. One by one, they were destroyed by an increasingly unsustainable media ecosystem built for the wealthy.

Completely unrelated post from 2004 about lurkers and social media (social media meaning blogs at that time):

Taking it one step further, maybe the ‘magic numbers’ we see in networks of humans relate these meshing concepts to our mental capacity to juggle social data.

  • 12 being the average capacity to track nodes in a totally meshed network
  • 50 being the average capacity to track nodes in an optimally meshed network
  • 150 being the average capacity to track nodes in a sub-optimally meshed network.
  • above 150 being the sparsely meshed social network where anonymity and getting lost becomes possible.

If we relate this to blogs and Clay Shirky’s power law, teenage diaries blogs are possibly primarily on the 12/total meshing levels, professional content blogs are probably all in the 50 to 150 ranges, with distinct stability levels (my blog went from 0 to 12 inbound blogs then stabilized, then grew to just over fifty inbound blogs and stabilized again.) Above 150 people are more sparsely connected and start looking for beacons or leaders to orient themselves socially. This is the range where the broadcasting type blogs are, the A-listers.

I cried inside a little when I read teenage diaries. Yes, we used to have that. They were weird looking blogs rumbling about random crap, but it was creative and fun.

Why don’t we have things like that on the web anymore?

Read “Lurking, Twitter, The Commons, and Private Posts“.

Within this setting, since roughly late 2016, I’ve been posting almost all of what I read online or in books, magazines, or newspapers on my own website. These read posts include some context and are often simply composed of the title of the article, the author, the outlet, a summary/synopsis/or first paragraph or two to remind me what the piece was about, and occasionally a comment or two or ten I had on the piece.


I know Apple has the new Pro Display XDR, but there is still a place for another 27-inch display which is made for developers. I have an LG UltraFine 4K connected to my iMac and I love it, but the design is so un-Apple.



Dave Winer made a video about how he blogs which is very unique. Basically he edits an OPML file that gets synced to scripting.com—it also looks like an outline. I like how quickly he can change anything on his blog.

I made a lot of adjustments to my blog to use IndieWeb technologies, so in essence: I can reply, like or repost tweets, receive mentions as comments here, and update my status, all from Decoding.

This is really cool, since I can keep everything under my control, but I can participate in social services.


I hate iPhone cases but I’m going to get a leather one from Apple. I just dropped my phone on the floor and it landed on top. I was lucky, it survived it without any damage, but I got scared for a second.

I’m sorry for spamming, but I’m trying to integrate Bridgy with WordPress. This should be the last test post.


I have comments and webmentions again on my blog. It feels nice.

While almost everybody is moving mindlessly into the cloud, I try to move out of it more and more and store a lot of my stuff locally (some of it are not even synced). I’m very conservative about my privacy lately. Although I use iCloud when I can, for 3rd-party syncing services, I prefer the ones where I can host them on my own server. I still use some services which store my data on their own servers (like YNAB), but I’m trying to get rid of them quickly.

One week ago I started running my own WebDAV server for OmniFocus and DEVONthink. There is an easy-to-follow tutorial from Bytemark that explains how to set up a couple of Docker containers with an automatic reverse proxy and SSL renewal service using Traefik.

I bought a cheap $5 server on Digital Ocean which is located in Frankfurt, that’s way closer to me (I live in Hungary) than OmniGroup’s sync service which is located in the US, I assume in Seattle.

When I switched to OmniFocus from Things, one thing I missed from Things was the instant and invisible syncing that they offer. Cultured Code really nailed that one: you can change anything in your Things database and it instantly shows up on your other devices, even in the background. OmniFocus’s sync was always slower for me, but shortening the distance between the server and the client looks like boosts the performance in a very big way. Using my own sync server located in the EU, I’m getting almost the same speed in OmniFocus as Things has. The good thing is that my data is now hosted on my own server.

DEVONthink also getting some speed increase compared to iCloud, but it’s not that big of a difference as with OmniFocus.