February was rough. Our team shipped Apple Health support, a feature I had eagerly built two years prior at Peloton’s first hackathon.
The unfortunate part—
There was a latent bug causing attempts to backfill a member’s history to be retried, in entirety, if any one network request to fetch workout data failed. Put concretely, if we’re attempting to sync hundreds of workouts to HealthKit and a single request fails, then, the entire sync would bottom out, eventually to be retried later. Finding the issue through elevated API error rates ended up being important. The client-side bug was squashed after a 2am, post-Valentine’s Day expedited submission to Apple. But, a server-side investigation revealed problems (infinite loops, high CPU usage, and malformed workout data) with the endpoints being used.
—we have our learning review (an alternative to “post-mortem”) tomorrow (this won’t be in your inbox until afterwards). Yet, the firefighting reminded me how incredibly supportive my teammates, family, friends, and girlfriend are.
I shared a couple of photos in the last entry and folks kindly mentioned they enjoyed them; here’s three of my favorites from the past month.
The next piece is less essay-y and more an announcement about a link blog I’ve maintained for the past 18 months: Substrate. I cover the “why” behind Substrate there and I’ll cover the “how” here.
I’m going to detail how I sift through web content (blogs, essays, long-form, tweets, and newsletters). If you’d rather, skip the setup, here’s your off-ramp.
The subtext of the fifth newsletter entry and backing essay, “To My Unread Books,” was emphasizing that our narrative around writing (a medium) frequently centers on books (a format). It’s rehearsed by the New Year’s chorus of “I’ve read [number greater than 12] books in [previous year]!” To be clear, this isn’t a shot at them or books. They’re wonderful! I just wish more folks ended each year also celebrating essays, papers, poetry, blog posts, and the like.
Goodreads lets folks log their book reading. When it comes to all other reading, there isn’t an agreed-upon tool. Our best options are Instapaper, Pocket, or Pinboard profiles and the link blog.
Substrate is powered by a combination of Instapaper and Pinboard alongside Reeder, Feedbin, Pushpin and Feed Hawk. I’ll cover the setup in two parts: inbound—subscribing to feeds and saving items—and the reading itself.
All web and newsletter content lands in Instapaper through two paths.
@newsletters.feedbin.com—I use when subscribing to newsletters1. Any message sent to the address will show up in Feedbin (backed by an anonymous web URL with the its content).
type="application/rss+xml"matching and even parses Medium RSS feeds).
All of my non-book reading happens in Instapaper on iOS or Reeder on macOS. To add a link to Substrate, I copy any notes into Pinboard’s description field (via Pushpin’s iOS Share extension or Pinboard’s bookmarklet on macOS). The bookmark then gets picked up by an IFTTT applet that posts to @_substrate.
Which brings me to the final—phew—note about the setup. Tags! Most of my tagging habits are borrowed from a friend (their notes are at the end). A tagging habit I’ve stuck with involves people. If a bookmark points to someone’s work, say Helena Fitzgerald’s, I’ll add a corresponding
helena-fitzgerald tag. Relatedly,
via-firstname-lastname for bookmarks of URLs sent by my dear friend, FirstName LastName.
The tags I’ve created are listed on this page. If you ⌘ + F around, you may find your name! Or, if you want to subscribe to a tag, Pinboard provides feeds—e.g. here’s the feed for my bookmarks tagged with
rebekah-cox (Web URL for viewing).
This entry is getting Very Long. The above is to say more folks should breadcrumb their reading with link blogs. There’s something to finding a writer2 through a friend’s feed or even noticing them reading your work in real time.
I hope to make Substrate worth your time and play link blogger when I’m not working on Distillations.
⇒ The story behind the cause (and recovery from) one of the toughest outages I’ve ever read about.
⇒ “And the real scary thing is because of shame and fear, many companies won’t publish why they failed, so we can’t learn from them. If people are laughing and yelling at you, you shut up, you don’t explain in detail what went wrong. And then we don’t learn. And that’s the real crime.”
⇒ “…[a] Pinboard is kind of the slow web from a reader’s perspective. It’s fine to follow an RSS feed of it to keep up with what somebody else bookmarks, but there’s also this whole other mode of reading a given tag at random, and doing that doesn’t really depend on time, doesn’t depend on ‘keeping up on it’ or anything like that.”
When a friend introduced me to Pinboard, they sent notes that I still find myself referencing. They’re mirrored below:
It’s such a great tool because it feels like you grow with it and it grows with you. It’s not in your face at all, it’s just there.
Subscribe to the archive feature. It costs extra, but it’s great to have in case a link ever goes away, which happens frequently.
Time has taught me it’s better to bookmark something than not, if you have any thought that you might care to re-read or reference it again.
This also means it’s handy to find some “Save to Pinboard” tool(s) that work for you. I use the standard Web bookmarklet (which I can invoke with ⌘ + 1 in Safari) on my Mac and I use Pinner and its Share extension on iOS.
Regarding tweets, when Pinboard archives a page it’s usually delayed by a few hours, so depending on how fast a saved tweet gets deleted…However, when Pinboard saves a tweet it usually includes the tweet text in bookmark title itself, so even if the tweet goes away, you’re not completely at a loss.
Find a tagging system that you enjoy and don’t worry about it. I like to use tags that describe the content of what I’m reading, like normal old keywords (e.g. this one’s about
advertising; that one’s about
I also have some somewhat “special” tags I use, some are like
filetype:pdffor when the file is a PDF (also have one for video)—tags that describes the kind of stuff you’re saving.
I will sometimes tag with something qualitative, like my
eye-opening-thingtag, which I use when something kind of opens my eyes or mind up to something I’d never realized before and usually changed me in some way after reading it.
Feel free to mess around as you use it, don’t fret about having a super rigid structure, find what works for you, and experiment. There are some decent editing tools out there, if you change your mind later and want to go back and update older bookmarks.
Oh yeah, I also try to make use of people tags whenever I can, so I can find all things related to (or written by) a certain person.
via:whoevertags, too—which is useful for “hmmm, I can’t remember exactly what it was, but I remember that
whoeversent me the link.”
I’m considering migrating newsletter subscriptions to my personal email address and adding forwarding rules; so, in case Feedbin shuts down—sigh—I can easily recover subscriptions and point them elsewhere. ↩
Keep an eye on Drew Austin’s Real Life contributions page. ↩