Patrick Tanguay is a self-described “generalist, synthesist, and curator of eclectic ideas.” His weekly newsletter, Sentiers, surfaces deep posts about highly relevant topics and provides insightful commentary and ideas. In this conversation, we discuss the tools and methods that enable his curation and sharing process.
- @inervenu on Twitter
- About Patrick
- The Alpine Review
- INFORMA(C)TION newsletter
- What is a static site generator? (Cloudflare)
- The Informed Life episode 54: Kourosh Dini on DEVONthink
- Keep It
- Grant for the Web
Some show notes may include Amazon affiliate links. I get a small commission for purchases made through these links.
Read the transcript
Jorge: Patrick. Welcome to the show.
Patrick: Thanks! Glad to be here.
Jorge: I’m very excited to have you on the show. I’ve been subscribed to your newsletter for a while and always find insightful links and information there. So I’m very excited to talk with you. For folks who might not know you, would you mind please, introducing yourself?
Patrick: Sure. Thats always a… it probably shouldn’t be, but it’s always a bit of a hard question to answer. I’ve started using “generalist” which I kind of resisted doing for awhile, but that’s … like my Twitter bio is “Generalist, Synthesist and Curator,” and that’s probably the best description. I’ve worked in a number of fields, and I realized a little while ago that the red thread connecting everything, was that I always ended up figuring stuff out and explaining it to others. Even when I was a front-end web developer, it was often the fact that I could explain to the client, and if I was working with others, explain across their disciplines. Like, of course the actual craft, if you will, of the front end was of course part of the contract, but kind of the selling point or that people would refer me to was the fact that I could explain it and kind of make sense of what we were going to build. And then that transferred into a print magazine, “The Alpine Review,” and I liked doing that so much that even though we closed it down or put it in a deep freeze, I try to recreate that experience with my newsletter.
Jorge: I love the three terms: generalist, synthesist and curator. It reminds me of a phrase that I believe was coined by Stewart Brand to describe Brian Eno. He said that Eno is a “drifting clarifier.”
Patrick: Whoa! That’s nice.
Jorge: And your trajectory here reminds me of that. Yeah, Sentiers is the newsletter I was referring to when we kicked off the conversation. Tell us a bit more about that. What’s the newsletter about?
Patrick: There’s kind of the… the official description and the real description. The real description would probably be, “anything that Patrick finds interesting.” The kind of official description is, “exploring technology and society, signals of change, and prospective futures.”
Which… like, “technology and society.” Technology permeates so much of the world in a growing number of areas that you end up being able to talk about anything if you look at technology very broadly. And “signals of change.” There’s so many things changing that that also brings you to many topics. And I try to — more and more — to make sense of it with an eye to where we’re going or where each topic might be going. Each field. But Sentiers is French for “paths,” and the path is taking more importance in the curation in the last year or so.
Jorge: I don’t know if this distinction is emphasized in the French: I see “path” as a distinction to something like a road, right? Like where a path is more emergent.
Jorge: Is that a part of this? Like when I say that you’re detecting signals for change, that to me implies that you’re not dictating the path, you’re somehow seeing it emerge. Is that fair?
Patrick: Yes, absolutely. And I use, for example, as many of the people I read and learn from, I use future in plural — “Futures” — because they’re always guesses at where things might be going or sometimes guesses that you’re wishing for that direction, sometimes because you’re dreading a certain direction. But there’s definitely always different potentials.
And one thing that I should have paid more attention before, but I’m paying more attention to now, is also the diversity of voices. So, some futures that we look at are already someone’s present. Like climate change. In the Western side, we’re starting to feel it, but some other people have been feeling it for years.
Some technologies… so there’s also that, someone’s utopia is always someone else’s dystopia. So, to always try to listen to a greater diversity of voices — and necessarily, as you do so, you realize that there’s multiple potential directions and futures and paths.
Picking the signals
Jorge: How do you pick up the signals that you write about? Like, what are you paying attention to that leads you to elucidate the path?
Patrick: It’s layers. Layers upon layers of people I’ve discovered through the years, or publications. It’s usually more individuals than specific publications. I’ve used Twitter. I think I’m user 6,000- something of Twitter. So I’ve been there for a while and using RSS for even longer. So, it’s, adding and replacing people as I go and feel, “okay, this person is… I realize now, was too naive about technology or too positive” or, on the contrary, “this person has evolved in their thinking and introduced me to this other person.”
And so I try to build this network, I guess, of people I’m listening to. And also using The Alpine Review before and Sentiers now to a lesser degree perhaps, but to introduce myself to those people and then to also pick up on their networks and be part of the discussions and get a better feeling for what’s going on. And then, being… I was going to say “too curious” — but being very curious about a number of topics, and adding them to the number of things I follow.
Jorge: That brings me to another question I had for you, which is this idea of spotting signals for change and another idea that I think is implicit in that, which is kind of spotting patterns, right? Like in order to detect change, we have to somehow be aware of the trajectory of something or the pattern of something, or having a sense for the context.
You’ve hinted at the fact that you’ve been doing this for a while; like you said, you were an early Twitter user and you’ve been following things like RSS. And I would imagine that you have a way not just of detecting signals, but also of building a corpus of ideas somehow, that allows you to keep track of those patterns. That allow you to spot the signal from the noise. And first, I was wondering if that was the case and if so, if you could share with us what that looks like.
Patrick: It’s the case and it’s been more purposeful in the last few years. It used to be, I guess, just piles of magazines when I was selling computers before starting the web. And then when I started doing web development, a series of bookmarks and bookmarks, and then quickly blogging, which then… it’s only recently that I’ve been specifically taking notes to refer to later.
Originally, the notes were more blogging publicly, and then as you write something, it sticks in your mind. And so for a while, the library was mostly in my mind and in the blog. And then as… I guess it’s starting with The Alpine Review, as we needed to collaborate and to keep track of whom we wanted to include, it needed to be more documented.
And then, yeah! Then Sentiers becomes a great… often even for some clients, I’ll just first go through the archives of the newsletter and re-find everything I’ve found before and compile it in a different way or see new patterns. And now more recently with the new website, the goal is to integrate the website with my note taking and my reading in Instapaper often and kind of having the information flow more directly so that I can take more notes more easily. And I was going to say, “trust my brain a little less,” but I guess it’s more expand my — augment — my brain more purposefully.
Personal knowledge management
Jorge: I actually wanted to find out more about that because as someone who publishes a newsletter myself, I have found myself doing what you’re talking about here, which is thinking, “oh, I remember writing about that in my newsletter. And where was that?” And I send out my newsletter through MailChimp, which creates a web version for each issue of the newsletter and that is published elsewhere, right? Like it’s in a different place than my regular website, so I can’t search for it using the same search engine and it’s almost like suddenly I have this separate set of information that I need to refer to. And I have the sense that you’ve recently made changes specifically to the relationship between content on your newsletter and content on your website. Can you tell us more about that project specifically?
Patrick: Sure. Well, one of my interests that isn’t often in the newsletter, but that is an ongoing interest is with PKM or “personal knowledge management.” And finding ways to find again. Because I think people trust search engines a lot, but it’s hard to search Google for, “this guy I remember seeing on Twitter was talking about this thing.” So, I try to make the haystack smaller, and the longest going tool I have is using Pinboard, the bookmarking service that Maciej Ceglowski started after Delicious started….
I’m going back — just a lot of people won’t recognize those tools. But one of the interesting things of Pinboard is that if you’re a paying member, it archives the pages. So, first of all, you don’t lose something you’ve bookmarked that suddenly disappears. And also you can do a full text search of only what you’ve bookmarked. So, to me, that’s a much smaller haystack to search and I’ll often find things through there quicker than trying to find it again with a search engine.
But that wasn’t linked to my note taking. So, when I write the newsletter, I write it to the text file in Markdown, and then I convert it to HTML and put it in MailChimp. So, when I say that I searched the archives of the newsletter, it was always the text files that I have on my computer. So, often to look for something, I would look at the bookmarks and I would look at the newsletter.
So, now I’ve tried to connect all of those things. The website used to be in WordPress, and now I’ve built it with Eleventy, which is a file based system. So it’s not a database anymore, it’s just, again, a bunch of text files. So without going into the details, or too much of the technical details, the interesting part is that the website now is a bunch of text files on my computer. And then when I want to publish a new version, it basically crunches that into an actual website and I just put it online.
And it’s… first of all, it’s much, much quicker for readers. It’s also much lighter. Because I’m trying be mindful of bandwidth and server usage because so many of those are using “dirty” electricity. So it’s good if you can save on that side. But the first reason was that it’s text files on my computer. now when I’m searching, everything is together.
You tell me if I’m going too much in the weeds, but the other change is that now I’m using Readwise — readwise.io. And that allows you to connect the things you’ve highlighted in various places. And recently it started offering a sync with the text editor I’m using, which is Obsidian. So now… for years and years, I’ve been reading either in Pocket or Instapaper, two apps I think a lot of your listeners probably use. Now everything I highlight in there goes through with Readwise and straight into my notes, which don’t necessarily make it on a website, but now it’s… so there’s more of a direct flow of everything I’ve read and the chunks I found interesting all end up in text files locally and can be oriented towards the website.
Details about Patrick’s setup
Jorge: I’m hearing you say this and thinking, not only do I want to get into the weeds with you on this stuff, but, uh, I I’m afraid we’re not going to have enough time to get as far into the weeds as I would like, because you’ve touched on several things that I’ve been exploring myself. I have been contemplating making this very same move that you’re describing — going from WordPress to what is often called a static site generator. And for many of the same reasons you’re pointing out here, I would love to have my site as text files — as Markdown specifically, which I use as well. And I recently posted about this on Twitter and a lot of folks came back to me recommending Eleventy, so it’s one that is very much on my radar.
I’m wondering about what you might lose by doing such a transition. And I can tell you two things that I’m aware of, that I would lose for my own instance. One is that WordPress provides a pretty good site search, which I don’t believe static sites have. And the other is, WordPress provides the ability for me to preschedule posts. So, I can write something and say… say on a Monday morning and leave it so that it’s published on a Tuesday afternoon, right? Are you dealing with those in any way? Is that an issue?
Patrick: Yeah. Those are pretty much the two issues. You’ve hit the two issues directly. The search, of course people can be unhappy and not tell me, but I haven’t had any people telling me that they miss the search engine. Although I did include one, but it’s… it basically searches DuckDuckGo, by specifying my website. And so it gives a result only on the website. It’s been working pretty good. There are a couple of solutions to do web searches on a static website. But it mostly ends up being work done on the client’s side. So, in the reader’s browser and so I haven’t implemented that yet.
The scheduling is more of an issue than I thought because like my newsletter goes out at 6:00 AM every Sunday. And I try to have it online exactly at the same time as the email goes out for people who want to read it online and share it. So that’s… it connects to the biggest issue, which is… it’s a lot more technical to run a site like that than it is to run WordPress. WordPress, you can just go on wordpress.com and create a blog and even have it on your own domain and you have nothing to do basically, other than use the interface, which is very broadly known already. A huge number of people have used it for themselves or at work or somewhere else.
And so this is… it’s harder. But I figured out the way. It’s like, I’m actually… I’m getting back from vacation and there’s one going out on Sunday, and it’s going to be the first one using the new automation to put it online at 6:00 AM. It’s basically, it’s… it’s going back to the command line. It’s having rsync and a cronjob running on the server.
That could probably be done some other ways, but I found that that’s actually… because the way I’ve built the new version is that my newsletter is usually four or five featured articles that I have a summary and comment on. So each of those has been split, so each newsletter has become at least five chunks — five notes. And I might issue 184, so it can take a while to transfer the whole thing. So automating it that way is a timesaver.
Jorge: That’s very encouraging. And I’m kind of desperately trying to make more time to experiment more with these things because I do find very appealing the idea that at the other end of this, you end up with this more consolidated, personal knowledge management base that you use… you used that phrase, PKM, right? And, I find the idea of having it as a set of text files on my file system very compelling. You touched on Obsidian, which is another tool that I’ve been recently migrating to. I am using Readwise and I was not aware that they had enabled Obsidian sync, so I’m very excited. Now I’m like thinking… it’s like the moment that we hang up here, I’m going to go experiment with that.
Patrick: I think it’s been active for like five days. So it’s a really, really new feature.
Jorge: That’s amazing. I was using it with Roam, to sync my highlights from Kindle and Instapaper and all these other things, sync them over to Roam. But, it’s very exciting to hear that they’ve enabled Obsidian sync. How are you using Obsidian? I’m curious. How does it play into this workflow?
Patrick: I’m hoping to transition completely to it. Right now, I ‘ve used Bear for a few years, which is also in Markdown, but it’s very visually polished, so it’s fun to use and it syncs between phone and iPad and laptop. And it’s Markdown that can be exported in Markdown, but when it’s stored, it’s not Markdown. It’s in a proprietary database. So that was one of the things that kind of bugged me. Although I would have kept using Bear if not for Obsidian and the fact that it’s pure text and you can actually open any folder with Markdown files. Open it in Obsidian and it becomes a bunch of notes and you can do back linking between the notes so that… because we often use links, but only in one direction.
So, when you get to the destination, the destination doesn’t display in any way where you came from, unless you’re staying on the same website, then there’s an indication. But if you’re going from site to site, you don’t know. And you don’t know who else might have linked to that same page. And so with backlinks or bidirectional links would be another term, then you know at least within the corpus of your notes, which links to which -which has been in Wikis for forever, and which we even had on blogs 15 years ago with trackbacks which is coming back now with digital gardens which is kind of a personal Wiki. And Obsidian supports that. And I found a way to have them work in Obsidian and when their live on my website in the same way.
And so, I’m still using Bear because it’s kind of my reflex to go to those files and client notes and articles in the works are all in there, but I’m trying to switch more and more of them to Obsidian which is so far a great surprise because it’s very modular. There’s a hundreds of plugins, and so far I haven’t seen it slow down. I’ve been wary of activating too many but so far it’s super fast. So, I’m very encouraged, up to this point, and the advantage is of course, is that I have nothing to do if at some point they start… or they stop developing it. The app is local, the files are local… everything keeps working.
Jorge: This idea of digital gardening is something that I am very interested in and we had earlier this year another guest on their show, Kourosh Dini, talking about the use of a tool called DEVONthink, which is designed for this type of personal knowledge management. And I mention it because DEVONthink too allows you to monitor folders on your computer and it indexes them and builds… it uses an artificial intelligence engine, and I don’t know the details of how this works, but it uses AI to spot relationships between pieces of content in your computer. And I have been using Obsidian. My Obsidian folder with Markdown files, I’m indexing it with DevonThink. So building this bridge between the stuff that I have in Markdown there with things like PDFs and bookmarks and all this other stuff, and it just feels like… for me, it feels like my little personal knowledge management system, which has been scattered for a long time, is finally starting to come together with these more open tools. It’s really exciting.
Patrick: Yeah. It’s… I was going to say the less exciting thing is the fact that we have to go back to old formats to get back that open function. Like Markdown files have been around forever and they’re text files, which has literally been forever for computers and PDF is also a very old standard. But it’s great to have that. I wasn’t aware of that function by DEVONthink so I’m going to have to try it.
I’ve actually… I’ve been doing some cleaning of stuff on my computer and I’ve been putting PDFs in Keep It and I’ve actually grabbed again, some old email archives that I’d archived to make the mail app snappier again. And I’ve put them in EagleFiler, which are both kind of… they both do the same thing you were explaining about DEVONthink, which is they do some search optimization and tagging and stuff, but the files remained in the finder and just on the Mac file system. So, but maybe I’m… after doing the cleanup, I’m just going to have to switch over to DEVONthink or add DEVONthink, because basically since it’s indexing existing folders, that’s the duty of it, you could have 10 applications doing different work on the same files.
Jorge: Yeah, that’s what I’m finding as well. I’ve stopped obsessing with the idea of trying to bring everything together into a single homogeneous system and more trying to find tools that are open about the data that they use so that you can get different perspectives on your information. And I can relate to this challenge you were talking about — the challenge of migrating stuff that you’ve had in more proprietary formats for awhile. We’re coming close to the end of our time together — unfortunately, because there are so many more weedy areas of this that I would like to explore or with you. But I’m wondering what the future holds for what you’re doing with Sentiers and how you see your system evolving.
Evolving the system
Patrick: Well, one of the main reason I was able to spend time doing that was that I used a grant by Grant for the Web, which is a project by the Interledger Foundation. We do web monetization. And a lot of the words they use sound like blockchain, but it’s not actually. It can be related to the blockchain, but it’s not. And they’re basically developing a standard that they want to be accepted by the W3C, to be able to stream money to the website where you’re spending time. And so the way I presented the project is that I’m already somewhat monetizing. I don’t like that word that much, but that’s… with memberships, paid memberships, but the archives and that’s the case for most anyone doing those kinds of like… another word I dislike but the “creator economy.” Often, their archives just fall by the wayside.
So, that was a way of keeping the archives evolving and accessible and useful for readers and having the web monetization work underneath and possibly be a new revenue stream. And the other reason is that by making it text files, they can be on GitHub. And that’s kind of… a lot of people have spoken about it with digital gardens, but not many have actually opened it. And I haven’t found a way yet to do it — a way I’d be satisfied with. But potentially having people participate in the notes and appearing on the website would be something interesting that could be done with GitHub. And so the goal is to… it’s kind of a forcing function for myself to note things beyond just highlighting in articles which then become notes that don’t necessarily make it in the magazine because they’re not necessarily interesting to read in themselves, but they can be super useful as you’re browsing through different notes and adding context to something and adding to the topic. So, growing the notes, making it potentially a revenue source.
The nice thing about this system is that if people are spending a lot of time, it means it’s useful for them. So then it’s a great way to transform it into a source of revenue because you’re not forcing anything. They’re just using it then. And then potentially bring in people on… I don’t know if it would be specific contributors? If it would be a way of, for example, you and I joining some of our notes, or something else that’s not… that’s kind of on the roadmap, but not planned yet as to how it would happen. But that’s another of the ways I hope to use it.
Jorge: That all sounds so fascinating. I would love to check in with you sometime in the future when this stuff has developed more just to see how that is going. But for now, where can folks find out more about you and follow your work?
Patrick: The simplest is the newsletter, which is santiers.media. So sentiers.media. Or @inevernu on Twitter. And on the Sentiers website you can subscribe, and you can also look at what we’ve been talking about. So, how the notes connect together and so far, it’s a lot, the existing archive. It hasn’t been digital garden-ized as much as I would’ve liked, but I’m adding to it constantly. So yeah, those are the two… and I write the articles about monthly. So there’s the newsletter, but there’s also some articles to read.
Jorge: Fantastic. I will post links to all of those things in the show notes. I want to thank you for being here and thank you for your work because like I said, I learn a lot from the work that you’re doing. So thank you for sharing it with us, Patrick.
Patrick: Thanks! Thanks for saying that and thanks alot for inviting me! It’s been fun. It’s always fun to discuss what you’ve been working on. It’s sometimes bring us a different perspective as you’re answering. So, it’s always useful.
Jorge: I hope that we can do it again sometime.