Alien Architectures

I thought I’d give an ASUS Chromebook Flip a shot as a low-cost, novel alternative to an iPad Pro with a keyboard. After playing around with a Dell Chromebook 11 last year, I got to know the world of dual-booting Chromebooks.

The Flip itself is pretty neat. It feels sturdy (aluminum body), has a decent enough display for what it is (not Retina, a notch better than the Dell’s, touch-sensitive), and it can convert into a tablet by flipping the screen back. Load up the Beta channel of ChromeOS, and it can run Android apps natively alongside ChromeOS apps, which is sort of cool.

My thought was, “tiny, rugged, cheap writing and coding device” vs. the iPad Pro (sort of large, expensive, and you can write on it but coding is tough).

Sadly, the Flip uses an ARM CPU, which immediately makes things a bit tougher if you’re going to try to put Linux on it. For instance, it was no problem to get Trusty installed via Crouton, but the ARM architecture means some things aren’t readily available (e.g. a Dropbox client). The only distro I could find with docs for dual-booting was Arch Linux.

On the ChromeOS side, there was some stuff to like about it:

The Amazon Cloud Reader app worked well when it was in tablet mode. The Android app support made it possible to play Hearthstone. I was curious how fast it would run and it wasn’t bad: More smoothly than on a first-gen Retina iPad mini. It was nice for browsing around news sites in tablet form.

On the negative side, the touch support for ChromeOS feels a little off. Google Inbox isn’t great in tablet mode (which feels like the natural mode for it), and if you’ve ever tried to use Gmail or Google Calendar proper on an iPad, you now how bad they are in a touch interface. I also hated the on-screen keyboard. I haven’t taken the time to figure out what the difference is between it and the iOS on-screen keyboard, but my accuracy on it was terrible, even when typing slowly.

So, I think I’m going to wipe it and send it back (or pass it along to someone who has appropriate expectations for this kind of thing). It was interesting to fiddle with, and it helped me think about what I’d really like in a portable computing device. iPads are still a hair too locked down for my tastes. The new Yoga Book sounds sort of amazing and strange and closer to what I’d like.

I also wouldn’t mind someone finally coming out with an iOS terminal app that can support mosh, which allows a level of persistence in an ssh connection that makes whether or not you’re locally running your Ruby interpreter or Emacs instance less material. The last iOS ssh client to provide that was iSSH, but it’s no longer under development.

Priorities v0.6

I got to spend a few hours on Priorities over the weekend. I put one thing back that the older app had, and added something new.

The thing that went back in was a conception of “effort” and a calculated combination of “effort” + “support” that yields something I’m calling “burden” internally, and “cost” in the UI. I could change it to one or the other, but I haven’t made up my mind which I like better, so I’m keeping it that way so the sense of misalignment will eventually spur me to a decision.

The idea of “effort” and “cost” is to capture the idea that not all goals take the same amount of work. By adding up the effort expended to work on the goal, and the level of relative support it’s getting, you can reflect a sense of what it actually costs to work on that goal.

The new thing I added is a little dashboard for goal health on each user’s home page:

Goals status

It’s just a simple tally of the number of goals in each state of health (good, medium, poor, unknown). If you’re using Priorities with other users who’ve shared lists with you, the tally shows up in your “Shared” list of lists, as well.

I guess there’s one other change: I’m assuming postgres for the app db in all environments. There are instructions in the database.yml file for how to set up sqlite as the db backend if you’d prefer that.

Source, as always, is up on GitHub. I’m also hosting a version I’m happy to share with others. Feel free to ping me if you’d like to learn how to get a login.

There’s a Human Tooth Where ESC Used to Be, or: Playing With a Chromebook

We have this promotion going on at work where you can get a free Dell Chromebook if you buy a Linux training course. I got a sample of one of the Chromebooks and did a little writeup about putting a minimal Linux distribution on the machine via Crouton.

Crouton’s kind of neat: If you stick your Chromebook in developer mode, it opens up a minimal Linux environment you can further leverage to install a full Linux distro on the machine, side-by-side with ChromeOS. You can choose Debian, Ubuntu, or Kali.

My writeup was about Crouton’s two CLI environments, which let you run a “real” Linux without the overhead of a GUI. The Chromebooks are there to help you learn Linux so you can get a certification, and the certification testing is conducted in a command line environment: Might as well get used to it.

The initial writeup was supposed to be about booting the machine into Linux from an SD card or USB stick while retaining ChromeOS. This particular model doesn’t allow that, and there was the looming possibility of screwing up the instructions and bricking the machine; or, worse, doing a fine writeup of the process that would lead someone less careful to blithely brick their own machine. So I settled on “minimal CLI Crouton.”

Anyhow, a few thoughts:

Dell Chromebook 11 (Hardware)

It’s not bad! It weighs a bit over three pounds, has a comfortable, spill-resistant keyboard and a rubber bumper. It feels very sturdy and substantial.

The one thing I’m not a fan of is the 11″ display. I thought maybe it was me being spoiled by Retina displays, but it’s not: I compared it to an 11″ MacBook Air and it’s simply not that crisp. That’s exacerbated by Chrome/ChromeOS, which doesn’t have great type options and tends toward tiny type in the UI.

The keyboard is a little different from standard laptops. Google preferred to pretend function keys don’t exist, so you don’t see any “f1” etc. on the function key row. It’s all ChromeOS functions (forward/backward, reload, fullscreen, tile windows) and standard laptop controls (volume, brightness). On the other hand, those really are function keys. When you have the machine in developer mode, you can switch between ttys with them.

In lieu of a caps lock key, there’s a search key. In ChromeOS, it toggles a popup Google search/Chrome app launcher. There’s no command or Win key, either, so you get an extra large ctrl key. That’s a bit cruel to an Emacs user: ChromeOS uses the more Win-style “CUA” bindings, so I spent a day or two using that giant ctrl key to accidentally “select all” rather than going to the beginning of a line.

When I compare the machine to the ASUS eeePC I bought several years back, it’s very nice. The keyboard is superior, the machine is more rugged, there’s more RAM and storage, and 11″ of screen feels capacious compared to the 8″ the eeePC had. It’s way easier to imagine this machine filling the use-case I had in mind when I bought the eeePC, which was “portable text-editing rig.”


I was a skeptic going in, but I gave myself a few days to try it out, and I’m beginning to get it. I can’t honestly consider adopting a Chromebook as my sole machine, but I can easily imagine taking one along for business travel in a Google-centric work environment.

I think squeezing the iPad into work situations made me more amenable to the Chromebook. You have to adopt a certain minimalist mindset to deal with their respective UIs. Pick up that skill for one, you can slip into it with the other.

So by the end of a week with ChromeOS, I was “just working.” The basic Google apps don’t change at all. There are a few things you can have on a Chromebook that you might be used to on a desktop machine in one form or another. A few things I found:

  • There’s an Evernote app that works pretty well. I think it’s an Android app running on top of ChromeOS, so it doesn’t behave like a browser window. Drawback: Tiny type you can’t adjust.
  • There’s a web-based Markdown editor called StackEdit that’s really well done. It can save to Google Drive, GitHub, WordPress, and Dropbox. It has a live preview and different themes. I’d have no qualms using it on the road.
  • Skype and Slack have web apps that serve for the basics. Combined with Google’s chat app, I had a way to talk to everything in use at work.

I came to appreciate the limitations by the time it was all over. I could do just enough and not much more. We can take it as a measure of how much I liked it that I considered spending the computer allowance work gave me on a Google Chromebook Pixel. Knowing I could put Crouton on it to have a development environment was encouraging, but two things stopped me:

  1. Having a Chromebook in developer mode is a little inelegant. The manufacturer does not want you to do that, so every reboot greets you with a warning and you have this nagging sense that you’re an automatic update away from the stuff you did to build a Linux environment being taken away.
  2. I was able to actually play with a 12″ MacBook, and it won me over. I couldn’t find a Pixel anywhere in Portland, so it never had a chance to make its case in person.

In its own way, ChromeOS is Google making a comment about computing the way Apple does with iOS: Both would like you to consider the hardware comfortable and pleasant to use, but neither want you to really care about the hardware. They want you to care about the experience the hardware delivers.

“But Mike, Apple is a hardware company!”

Yeah, yeah … right. They sell hardware. I get that. But they don’t want you to care about what’s inside the case in terms of RAM or clock speed or whatever any more than it takes to help their customers spend as much as they’re willing to optimize the experience.

Google’s approach is maybe more utilitarian:

They want you to realize you could take your Chromebook down to the river, toss it in, order up another one from Amazon Prime Now, have it delivered to your door two hours later, and pick up where you left off. Where the most sensible accessories for MacBooks are padded cases and maybe a grippy protective shell, I can imagine a future where the most common accessory for a Chromebook is a little holder for disinfectant wipes so that people can swab down the one they found left out on a picnic table at a public park or next to a puddle of vomit in an Old Town alley.

“How’s that Chromebook you found?”

“Oh, it’s cool. The last guy replaced the escape key with a human tooth and there was some hair and clotted blood in the USB port, but when’s the last time I needed a thumb drive, right?”

“Ha ha right … Hold it … a tooth for an escape key? Oh shit! That’s the one I lost after Murray’s bachelor party last year! Man. Wonder where the hell it’s been?”

“Oh yeah? No telling. Want it back? I found another one out on the curb this morning.”

“No, it’s cool … that one was just my clubbing one.”

Okay, that’s overstating the case. Dell just released a 13″ Chromebook that could properly be called a “business machine.” Nobody’s IT department will be happy if one ends up in the river, or left on a playground. But the i3 model is pretty inexpensive and very capable. The i5 model is pricier, but I bet it’s way less frustrating than a comparably priced Windows machine in terms of perceived speed. Maybe more importantly to an IT department, nothing lives on those things. So if one does end up in the river or gets left behind on the bus, it’s a little less costly to replace the hardware and the data doesn’t have to be “put back.” Just pull one off the shelf and slide it across the counter.

Linux on the Dell Chromebook 11

I was feeling a little poorly over not being able to try out the (somewhat scary) “flash a more flexible BIOS, turn it into a ‘real’ Linux machine” approach. I mean, it wasn’t something I wanted someone trying on my advice. I felt poorly about it because my curiosity was thwarted. I set about to fix that this morning and it was pretty simple!

  1. Pick the distro you’ll want to stick on the machine. I went with Ubuntu and used the UNetbootIn app to burn the install CD to an SD card.
  2. Crack open the case and void your warranty to remove a write-protect screw that’s designed to keep you from overwriting the BIOS. See the picture above for where to find it. Put it somewhere safe, I guess, against the day you want to restore the Chromebook to its pristine state. I kind of love the physical hardware aspect of the operation. There’s something deeply psychologically unsettling about knowing the machine you’re holding is literally missing a screw that you took out of it. I like to think they settled on the write protect screw after testing “make the user crush an ampule of cat urine over the keyboard” and deciding that sourcing cat urine would be cost prohibitive.
  3. Put the machine in developer mode. There’s a little fussing and it takes some time (they just don’t want you to do it), but it’s a simple process. At this point, if you reboot the machine it will give you this warning screen and provide you with an easy way to get back to “normal.” The screen is a clean white, and there are no sounds of screaming at bootup because that would be bad UX, and because it reduces the chances of you loaning the machine to someone who literally just taps the spacebar at boot to wipe out all your customizations.
  4. This step is dangerous. If opening the case and removing a screw didn’t deter you, this is probably the best point at which to pause and ask whether turning your Chromebook into a dedicated Linux machine really matters. Download and run a shell script that flashes the ROM with SeaBIOS. That will allow you to boot from a USB stick or SD card to run a Linux installer. With the Dell Chromebook 11, this is an all-or-nothing situation. Some Chromebooks let you dual-boot between ChromeOS and something else; this one does not. If you screw it up, you’ll risk bricking the machine.
  5. Reboot from the SD card or USB stick you burned in step 1 and run the installer.

Once you do all that, you have the basics: The machine boots into Linux, can talk to the network, and generally “works.” There’s no sound, the trackpad is sort of weird, and the special function keys don’t work. You can take steps to fix all that. Here’s a starting point, none of which have I attempted: I made the machine boot into Linux because I wanted to see what would happen.

What happened was pretty anticlimactic:

rbenv, Emacs, zsh and git are all just an apt-get install away. GNOME is GNOME. Firefox was just sitting there, and I used it to Google “john lewis restore chromebook stock ROM.” Then I downloaded the script I used to flash a new BIOS in the first place, told it to please restore me to the stock ROM, and let the Chromebook boot into its recovery media (which I made last week before I started screwing around with the machine). I’m pretty sure I’m not even going to bother putting it into developer mode again, and I need to put that screw back in place once it’s done reinstalling ChromeOS … It’s sitting there on my desk bothering the hell out of me.

Linux Anywhere But the Dell Chromebook 11

Why not even put it in developer mode?

Because I don’t have to. I’ve already got two Linux machines (one I own, one I rent via shared hosting). It turns out Chrome has pretty decent ssh client you can use, and I did for the week I was playing around with ChromeOS and doing that writeup. Stick the ssh client tab in full-screen mode, and it looks about like the ssh client in full-screen mode looks under GNOME, Unity, OS X, or my iPad.

The thing you get from developer mode/Crouton is offline availability of that more robust environment. If I traveled a lot, that would matter. As it is, I’m almost never disconnected. If there’s not Wi-Fi, my phone is a serviceable hotspot. The “little CLI workstation with Emacs and Ruby” is never far away. It just happens to be running somewhere else. That’s kind of the point of the thing.

A Few More Org Findings

So, I learned something this week. Rather, I did something that was useful to me, which is stop short of trying to get GNUS working again. Instead, I focused on seeing if I could get comfortable enough with MobileOrg to use it (mostly I did) and I kept on working on making some of the things I like about OmniFocus work in org-mode. I think that slightly more constructive behavior — pulling away from a fit of emacsimalism before going completely toxic on it — made it easier to keep on going with org-mode.

So, here’s some of the stuff I learned this week. It’s mostly about how to use MobileOrg a little better, and how to control how much of your org-mode data you have to see at a time.

Better MobileOrg

I’m learning to trust MobileOrg, but it takes a little effort to make it work smoothly, especially if you’re coming from something like OmniFocus or Things.

Save all your open org mode files from the agenda with C-x C-s

Things has the very best sync I’ve seen in a todo app: It seems to “just happen.” Others require some sort of action on the part of the user, so if you’re the type to spend some time at your desk squaring away your actions for the day, then head into back-to-back meetings, it’s a pain if you don’t remember to sync before heading out.

Org-mode can, depending on your setup, complicate matters even more. If you live in multiple files and org-mobile-push without saving them, you get an imperfect sync. The best answer (without rigging up some kind of auto-sync), is to use the standard Emacs save command (C-x C-s) from an agenda buffer: It saves all your open org-mode files. Then you know it’s safe to push to MobileOrg.

Cheat a little with emacsclient until you can remember to save-and-sync

Emacsclient is able to run elisp from the command line, so if you can ssh into a machine with your org files and emacs on it, you can do an org-mobile-push from the command line without opening Emacs:

~$ emacsclient –eval ‘(org-mobile-push)’~

Set up refiling to more easily move things out of your inbox

Refiling allows you to move a thing from one org file to another. With MobileOrg, where the default capture method dumps you into an inbox file, it’s helpful to set up your refiling targets. With this example:

   (("~/Dropbox/org/" :level . 1)
    ("~/Dropbox/org/" :level . 1))))

C-c C-w would present you with the level 1 headings from your and files as targets for refiling, meaning a given org headline will be moved to the last line under the heading you select.

Leveraging the Agenda and Sub-Tree Narrowing for Focus

I really like the way OmniFocus handles its Perspectives and Focus features. It’s easy to quickly narrow down your view to things you need to work on or think about right now. I’ve learned two ways to attain similar constrained views in org-mode.

Narrow focus with the agenda

In a tool like OmniFocus or Things, you might have a few different views into your task lists to better organize what you’re working on and when you’re working on it, but the mouse-driven interface of those tools generally means your experience is one of moving between areas of the UI, or doing one-click state changes.

Custom Agenda Commands to Narrow to Contexts

With org-mode, you can use custom agenda views to pull off something like perspectives or context views, you’ll just be doing it with the keyboard.

When I use OmniFocus, my tendency is to make a lot of people contexts and a few mode contexts, but not a lot of place contexts. With org-mode, I use tags as contexts, and tag items with people. Then I made a few custom agenda commands that make it easy to drill down to specific people (either at my desk, or when I’m using MobileOrg):

    (setq org-agenda-custom-commands
          '(("p" . "People")
            ("pn" "Nick F." tags-todo "nickf")
            ("pm" "Michelle F." tags-todo "michellef")
            ("pl" "Lauren" tags-todo "lauren")
            ("pL" "Larissa" tags-todo "larissa")
            ("pp" "Pete" tags-todo "pete")
            ("pj" "Jean" tags-todo "jean")
            ("pi" "Isaac" tags-todo "isaac")

            ("g" .  "Groups")
            ("gt" "Tech Pubs Team" tags-todo "team")
            ("gl" "Tech Pubs Leads" tags-todo "leads")
            ("gs" "Engineering Staff" tags-todo "staff")

            ("o" tags-todo "office")


These views make it possible to generate an agenda (C-c a) then tap the p, g or o keys to get pre-built tag searches by people on my team, teams I work on, or items I’ve tagged as “office” (which is my way of saying “things where I need to get up and walk over to someone’s desk to talk face-to-face.”) Those same custom commands appear as agenda views in MobileOrg, so I can walk into a 1:1 and easily see everything tagged with the person I’m speaking to.

Restrict the Agenda to a Single File

You can also restrict the files the agenda uses to generate itself by invoking the agenda (C-c a) then tapping the < key before tapping a to generate the agenda. That will limit the agenda to the file you invoked it from. If you keep your todos in “work” and “personal” files, that means you can effectively filter out one or the other with a single extra keystroke.

Restrict the Agenda to a Single Subtree

You can also restrict the agenda by tapping the < key a second time. That will restrict it to the current subtree. That’s helpful if you keep lists of single-action tasks, or want to focus on the todos for a single project.

Narrowing focus with org-narrow-to-subtree and widen

The agenda has a bunch of single-key commands to cycle todo status, etc. By enabling org speed commands, you can get the same commands in an org-file. s is useful for narrowing to the current headline (great for focusing on a single project or area of concern). Map w to widen to quickly expand the file again:

(setq  org-speed-commands-user (quote (("w" . widen))))

Use TODO label faces

You can customize the way todo faces look by keyword. STUCK, WAITING and DELEGATED each get a special face so that when I’m scanning a file, they stand out a little (white on red, orange and gray, respectively).

If you use the OmniFocus defer date, use a “scheduled” date in org-mode

Like OmniFocus, org-mode has both a start date and a deadline date. Keep the agenda clear by planning start dates for things with “scheduled.”

You can make custom links in org mode really easily

Here’s one for linking to JIRA tickets:

Then use a “jira:” URL scheme in a standard org-mode link, jira:doc-988

Opening that link with C-c C-o will open your default browser and execute the search (which will take you to the ticket).

Have Some mutt Macros

I ran through Steve Losh’s the Homely Mutt, which I am willing to stand behind as the best guide to setting up mutt on a single-user system, if you’re okay with running server-style stuff on a laptop or desktop machine. Personally, I’m not (for fiddly, neurotic aesthetic reasons). You can run mutt just fine via standard IMAP if you’re okay with giving up really good search via notmuch, but the package Losh puts together (mutt, offlineimap to Maildir, notmuch for search, postfix as a relay) gives you the very best mutt experience and gives you a local backup of all your mail you can access with anything else that groks Maildir. So if you’re in an erratic orbit around Emacs with stuff like GNUS, well, a little investment in offlineimap reaps years of futzing with email clients.

Anyhow, I do have a Linux box running under a desk at home along with mosh, which makes my logins to it from work feel pretty persistent. So given a thing I think of as a “shell box,” I’m fine running offlineimap on a cron job and postfix in dumb satellite mode if it gives me the single best email client on the planet to optimize for keyboard-centric use and versatile view filtering.

Keyboard-centric view versatility is theme number one for these macros. Theme number two is separation of mail personae, which I’ve handled with the creation of mail-account-specific profile files and macros that allow me to switch between work and personal accounts quickly, and without wondering if my signature and other profile-related stuff have been set correctly when I compose a message. So anyhow, here they are:

… and here’s my whole muttrc (Gmail IMAP variant) or offlineimap/Maildir variant. I think the scores and colors (which interact with the scoring I’ve set up) are the most useful things in there.

Three Activity Trackers, Quantified Self

Activity trackers really work for me, less for counting steps than all the stuff you do around them, like recording food and just generally thinking about activity levels.

Some of this will sound more complicated written out than it is in practice, but the benefits have been pretty clear because I’ve come to understand through observation and recording just how much activity, diet and sleep are key to my sense of wellbeing, how interdependent they are, and how being mindful of my general state where one or the other is concerned can help me make better decisions.

Three Pillars: Sleep, Food, Exercise

For instance, I know that I need about seven hours of sleep a night to feel well rested. Much more than that and I tend to be over-rested, and that triggers a bout of sleeplessness. So I try to keep my sleep steady: 6.5 — 7.5 hours a night is just right.

I learned how much sleep I needed by using a sleep-tracking alarm clock app that included a simple diary I could fill out at night (what did I do that day that might effect my sleep, e.g. caffeine intake, exercise, stress at work, working late, late use of computer/iPad), and in the morning (to record how rested I felt).

When I record my sleep and make it a habit to think about it, it can help me keep from making bad decisions. For instance, if I get only six hours of sleep, I can offset some of the fallout from that (feeling crabby and irritable) by getting some exercise first thing in the morning. Keeping track of sleep keeps me thinking about that dynamic, making it harder to say “oh man, I feel pretty bad … I’ll just skip the ride to work and take the Max in.”

Same thing with food: I know if I’m tired I’ll be more inclined to go after carbs and other cheap, fast energy. Thinking that through first thing in the morning really helps keep me from eating a bunch of stuff that’ll make me unhappier later in the day, when I crash from it.

That helps with mood and interpersonal stuff: Under six hours of sleep, and other people become pretty hard for me. Having that reminder that I’m probably not going to be receiving other people very well makes it easier to deal with things that might seem like provocations otherwise. And getting exercise generally lightens my mood.

In the big picture, it all comes together to help serve a longer term goal, which is losing some weight. When I’m in the habit of paying attention to all this stuff, I behave better, feel better, and I lose weight. When I stop paying attention, I behave worse, feel worse, and I gain weight.

The total amount of time I spend per day paying attention to this stuff comes out to maybe five minutes, tops: about .3 percent of my day. Seems worth it.

I’ve had three activity trackers in the past several years. Here are some notes in case you’re considering buying one for yourself.

Jawbone Up

I bought one of the earliest Jawbone Up’s a few years back. I don’t know what else was on the market at the time besides FitBit, and I think the Up was the first step tracker you could wear on your wrist (vs. clipping into a pocket) or wearing a giant, bulbous device on your arm.

Jawbone got a lot right about the Up: It looked nice, the iPhone app that came with it looked and worked really well, and the device integrated with calorie-counting services like MyFitnessPal or digital scales like the ones from Withings.

The Up could do a few things:

  • It could track steps, and if you tell it that some activity it tracked wasn’t walking but was actually something like jogging or riding a bike, it could adjust the calories it calculated you burned.
  • It could track sleep quality. Tell it you were going to sleep, and it would record periods of motion and let you know how well you were sleeping at night.
  • It could provide gradual alarms in the morning, gently vibrating you awake as you began to naturally stir at the top of your sleep cycle.

The Up was made of a textured rubber with silver highlights, and it looked pretty nice: Closer to a piece of jewelry than a step tracker. The FitBit Flex and Garmin Vivofit look pretty clunky by comparison.

The Up took input via pressing one of the ends of the wristband and it offered feedback via tiny embedded lights and vibrations.

In terms of ease of use, the first Ups were o.k. You had to snap an end off and plug it into your phone’s headphone jack to sync with the app. The caps become loose over time (especially if you’re an anxious twiddler who’s prone to fiddling with things like bracelet bits). Later models involved Bluetooth for sync but still require removal of that cap to charge the device (which lasts about 7 days on a single charge).

With all that stuff going for it, I wish I could still bear the thought of using the Up, but it has (or had, maybe it’s gotten better) one pretty big problem: It breaks a lot.

The issue appears to be endemic to the design. Unlike the Garmin Vivofit or FitBit Flex, which keep their guts in a discrete plastic lozenge with no moving parts, the Up hardware is built into the rubber bracelet. That means if you flex it too hard, you can break the innards.

“But Mike,” you’re thinking, “don’t flex it too hard!”

Sadly, since the Up is a bracelet with a pair of square ends that don’t clasp together, it’s quite easy to flex it too hard just by snagging it on something like a coat sleeve or a bag strap. I broke one tangling it up as I put on a backpack.

So, the first generation of Ups were pretty much a design catastrophe. Lots of people had problems with them breaking or just dying. Jawbone launched a trade-in program that it handled about as well as you can when everyone on the Internet is reporting that your product breaks when worn. I returned mine, got a replacement, and it broke, too. I returned the replacement and got another one, and it lasted a week before also breaking, at which point Jawbone just gave up and offered a refund to the entire world.

The first generation of Ups were so bad that Jawbone scuttled the product and went back to the drawing board, returning a while later with a new version that looked the same but promised to be more reliable. I broke two of those in the space of three weeks (I swear to god I wasn’t trying), and Jawbone — exhausted from its previous period of apologetic refunds — told me I could keep returning them during the warranty period or get nothing.

When the new Ups with Bluetooth syncing came out, I ignored them because a cursory glance around the ‘net showed people were still complaining about them breaking. It probably wouldn’t have mattered: Fool me once, shame on you. Fool me four more times after swearing to god you’ve figured it all out, shame on me.

Update: Just read the 1-star Amazon reviews to get an idea of how often they still seem to be failing.

Lesson: Cramming a bunch of fragile electronics into a container that’s comparable to a strip of thick garden hose in flexibility is pretty much a recipe for running an exchange program in perpetuity.

Pushback: Reviewers complain about the Up’s lack of a display or other feedback on where you’re at with your daily step goal. It didn’t really bother me. I would just do a sync at the end of the day and see where I came out.

FitBit Flex

After the Up, I got a FitBit Flex. Unlike the Up, the Flex puts all of its electronics in a non-flexible piece of plastic that slips into the pocket of a rubber wristband. It syncs via Bluetooth.

Where the aesthetics of the Up are sleek and somewhat fashionable, those of the FitBit are utilitarian and sort of dowdy. That’s the FitBit aesthetic in general, and if the life I got out of my FitBit is any indication, that’s as it should be. Sticking to a simpler design with less emphasis on making a pretty bangle means the things it would take to break a FitBit will generally involve taking it off your wrist and smashing it between large rocks or suffering an injury to your wrist that’s so severe you will not be wondering how many steps it took you to get into that situation.

Where the Up relies on button presses for input, the FitBit Flex takes taps. Sometimes that works o.k. For instance, it can tell you how far along you are to your daily step goal with a single tap, lighting up from one to five tiny LEDs. For sleep tracking, it’s not as good: You have to sort of tap-tap-tap-tap-tap the device, and it doesn’t offer the most meaningful feedback.

Since sleep and sleep tracking matter a lot to me, the FitBit fell a little short. Like the Up, you can tell it you’re going to sleep and it’ll track how well you rested. Unlike the Up, it doesn’t have a gradual wake alarm (though it can vibrate on your wrist at a set time).

Like the Up, the FitBit Flex is also a pretty good citizen of the wider fitness device/service ecosystem. It has an open API and it talks to smart scales, calorie counters and more. If you have a preferred app for tracking activities besides walking, the FitBit API can take that input and fold it into its reckoning of your daily activity.

The one complaint I have is that I managed to tear the wristband popping the little lozenge out of its pocket. That didn’t affect it too badly, but it did mean water had an easier time finding its way into that pocket. Digging a sweaty lozenge out of a rubber pouch doesn’t feel like the future at all.

Still, I had a positive enough FitBit experience that I really wanted to trade my FitBit Flex in for the next model up (the FitBit Force), which could display time. Sadly, people reported that the wrist band for the Force caused them to break out in terrible rashes. All that’s left of the Force today is artfully worded press releases that promise to get to the bottom of the matter.

One other Fitbit highlight worth noting: Lots of people I know have them compared to the other two trackers I’m covering in this entry. That makes checking leaderboards more fun.

Lesson: An open API goes a long way, even if the only thing you thought you wanted at first was a step and sleep tracker. So does a sense of reliability. I never really trusted my Up after the first two broke, and even though I loved the software and general UX, it really bothered me to think that my data would have gaps during periods where the replacement was in the mail.

Garmin Vívofit

Oh, man. The Vívofit. I’m wearing one right now and it’s a real love-hate thing.

Compared to the Up and Flex:

  • It takes a lozenge-in-a-rubber-pouch approach, like the Flex.
  • It displays the time.
  • It displays your actual steps and tells you how far you are from completing your daily goal.
  • It tracks sleep quality.
  • It uses button presses for input: Short press to switch between modes (time, date, steps, goal) and long press to activate sync or put it into sleep-tracking mode.
  • It syncs via Bluetooth or via a little dongle you plug into your computer.

One final, huge difference: It takes a standard watch battery that lasts for a year. No recharging every 5-7 days.

I really like the Vívofit as a step tracker and wrist watch. I kind of like it as a sleep tracker. Where it falls down is its API, which Garmin isn’t super interested in opening up. Garmin has partnered with MyFitnessPal, which acts as a sort of Switzerland to the many activity trackers out there by pulling in their calorie data to go with the food data you’re entering into it.

Other than that MyFitnessPal integration, the Vívofit doesn’t talk to the rest of the world much. There are services that try to sync it up with Runkeeper Pro and Strava, but they don’t work reliably. It seems to be willing to talk to my scale. In the end, if you want to get data from a bicycle ride or swim into Garmin’s Connect app, you have to enter it by hand.

The reasons for this are pretty clear: Garmin got into the activity tracker business to supplement its more lucrative GPS devices, which are under threat from GPS-enabled smartphones. Talking to a smartphone app like Runkeeper or Strava just means you aren’t going to be as interested in buying a Forerunner GPS tracker or Edge bicycle computer.

I completely get why Garmin’s got to do this, and if I were already heavily invested in their product line I’d probably be really happy with the Vívofit. As it is, I don’t need their other products, which makes their closed off Connect dashboard pretty irritating.

The things keeping me happy with the Vívofit are the watch function (I hate pulling out my phone during meetings to check the time … it feels rude) and the fact that the people sitting on either side of me at work are also using Vívofits, which means we can see each others’ step activity in our dashboards (though none of my biking activity — 75 miles a week at this point — can appear unless I key it in manually). It’s also nice to see my daily steps on the display.

Quantified Me

In the end, I pretty much live in MyFitnessPal to keep track of what my walking and biking activities mean in combination with my diet:

  • I use my Garmin for step tracking
  • I use Strava for ride tracking
  • I use a Withings scale to track my weight
  • I enter my food in MyFitnessPal’s diary

The Vívofit, Strava app and scale send data to MyFitnessPal, which reconciles their calorie information with its food journal. At the end of the day, I sync the Vívofit and close out my food diary, and MyFitnessPal tells me what my weight would be in five weeks if every day was like the one I just logged.

MyFitnessPal doesn’t care about sleep patterns, so if I want to see that data I need to use the Garmin app. Since I use a gradual wake alarm clock app (highly recommended), I can get the data from there, too.


Clearing out some old notes from the last job and figured this would be a good one to preserve. It shows how many times tags were used on a collection of posts generated and tagged almost exclusively by the user community for a site I used to work on.

For instance, row 1 shows that tags were used only once in 12,602 cases; row 2 shows that tags were used 2-5 times in 2,735 cases, etc. So out of a total of 16,169 tags, we can see that tags were used only once about 78 percent of the time, and tags were used no more than five times about 95 percent of the time.

Times Used Number of Tags
1 12,602
2-5 2,735
6-10 416
11-20 200
21-50 118
51-100 54
101-500 27
501-1,000 7
1,000-2,000 5
2,000-5,000 5

I wasn’t prepared to worry too much about the whole thing until the site (run by Drupal) started to crawl and the slow query log showed us that taxonomy-related queries were killing us. I even took to Ask Metafilter to see what everyone else had to say, and got an answer from the guy who coined the term “folksonomy.”

The thing that was maybe a shame was that a lot of those 12,602 tags were variations on each other:

  • social networking

  • social_networking

  • socialnetworking

  • SocialNetworking

  • Social Networking

  • Social_Networking

  • Social_networking

  • Social networking

  • Socialnetworking

In the absence of any discipline at all and no overarching style guide for tagging, no real patterns emerged to make the tags useful. Search engine indexation sucked because we had 12,000+ tag index pages with only a single post, those thousands of tag pages netted well under 0.5 percent of site traffic and crawl times were ridiculous. You really should not have almost as many tag indexes as you have actual posts.

It wasn’t deemed a wise use of time to try to automate normalization. In the end, I wrote a VBO that allowed us to delete the 12,602 tags that were used only once (provided they weren’t newer than a month old, so we didn’t arbitrarily blow up a trend before it blossomed). We also locked the users at large out of being able to tag at all, leaving it to the curators on staff. Yes, it helped performance.

Dark side of tag normalization: At the job I held before this one, they just gave an editor a spreadsheet with the thousands of non-normalized tags and invited her to correct them by hand. I do believe I would have gone mad.

So, here’s the thing about that Nexus 7

I think I’m probably going to have to own my last post because, you know, we’re not big on memory holes around here. And in my own defense, I meant every word when I wrote it because at the time I was comparing that Nexus 7 to some phantom device I hadn’t really held or tried out or, alternately, my iPad 3. It didn’t seem fair to compare it to the iPad 3, both because the iPad 3 is uncomfortably heavy for my main use cases (reading on the Max, reading in bed) and because the iPad 3s bigness and heaviness comes from the things that make it much, much better than a Nexus 7 for the cases in which I’d used it up to becoming a commuter (as a curiously smooth and low-functioning laptop).

I’d planned to sell the iPad 3 once I was convinced I’d keep the Nexus 7 around. So last week I found myself sitting in a Starbucks up in the Pearl District while a pair of nice but suspicious ladies made me connect to assorted networks and load web pages and generally demonstrate that I was not trying to sell them an iPad case stuffed full of nuts, bolts and wood shavings (which once happened to me while stationed in Korea, except not with an iPad—they didn’t exist yet—but a power amplifier for a radio set, apparently because a Korean depot-level tech decided the guts of an FM radio power amplifier would do him more good than they would my retrans team).

I don’t want to seem like I’m complaining about the ladies. They were nice enough and they were much better to deal with than the people who wrote me for two days straight asking if I wanted to trade my iPad for a hulled kayak, a mostly dry and ungnawed box of Life magazines from 1982, or $50. They were far nicer than the person who wrote me at two in the morning with the simple message “oh you fucking dirty scammer.” And they were less alarming than the elderly gentleman who attached a 1600×1200 picture of his smiling, stubbly face and a message that read:

$$$ how do I call !!! $$$

Eventually the ladies were satisfied, so I initiated a wipe of the device that took an awkwardly long amount of time I filled pointing out the nice features of the case I was throwing in. The transaction concluded and me with some time on my hands, I wandered down to Pioneer Courthouse Square, and then into Pioneer Place, and then into the Pioneer Place Apple Store, where what I really wanted to see was the new iMacs (not there yet) and the newest 11″ MacBook Airs (because I’m a curious soul).

My mood, on entering the store, was a little melancholy. How friendly would the greeters be, I wondered, if they knew what I’d just done? And how would I feel, looking at the new iMac and latest MacBook Air, knowing that the Nexus 7 had shaken my faith in Apple just enough to be thinking the old heresy, that life with a machine I crafted myself, like a dwarvish smith, would perhaps break through the gray sameness of life in my 40s and allow me to stand a bit straighter and meet others with the fell gleam in my eye of a full-on mithril plated nerd? These are the thoughts a technical writer who works in the corner of a pod full of competent, opinionated developers. They aren’t really worthy thoughts. But sometimes, after you’ve just gotten done being scolded for leaving trailing whitespace in a trivial docs commit, or you’ve forgotten that Ubuntu—the training wheels distribution—lets you assign arbitrary ports to sshd with no hassles while CentOS—the training wheels distribution for grownups—has iptables up and running at install, you imagine assorted geek-flavored Charles Atlas outcomes that will allow you to start socking the bullies.

I worked through those feelings in the time it took to mutter “I’m good, just looking” at the greeters, then I brushed past the iPad mini display, not even intending to stop and look because what was the point? I’d forsaken that path. But in that moment, one of the people holding an iPad mini put it down and the person who seemed to be looking at everything but the mini to keep from seeming like they really, really wanted to hold the mini while it was being held by someone else was looking away. So I picked it up.

My main thought was “is the display as bad as everybody claims?” so I opened up Flipboard, which doesn’t have many meaningful or constructive design opinions in its pretty little head but does have nice typography. Before I could consider the display, I noticed how quickly Flipboard opened at all. Just *pop* and there it was. That was nice. I remembered when I finally found a working Nexus 7 at a Fred Meyer and how I actually rebooted it because it was running so slow and I’d just subconsciously made the excuse that it had probably been handled all day and who knew what state it was in. Maybe Flipboard had recently been open already, so that’s not a scientific observation, but it’s not like I was even making some list … I was just curious about the display and noticed. Even when an app is up and the cache is warm, the Nexus 7 doesn’t seem quite as responsive. You get used to it, then you notice when something else isn’t like it.

So I flipped around in Flipboard and thought “this display is not that bad, and it’s very bright.”

I closed Flipboard and opened Safari (also *pop* and then *swoosh* when I loaded Google news), and then I just bounced around, opening things and switching between apps and noticing that the mini was very nimble and smooth and its extra size compared to the Nexus 7 (it’s .6″ wider) didn’t render it impossible to hold (which is something reviewers who have felt hard pressed to notice something have fixated on, saying that only people with giant hands could hope to hold a mini and then noting how tragic that is because their giant hands will then crush the wafer-thin device and leave them weeping and smashing SWAT teams as they blunder down the street wishing they’d bought a Nexus).

At that point, the person who’d been pretending to not stare at the person who had held the mini before me had dropped all pretense of not staring. He wanted to get at that fucking mini. So I put it down and looked at him and sort of raised my eyebrows with a Spockian “hmmm” and walked back out of the store.

By the time I deposited my iPad money and got back to the office, my thoughts had largely taken shape: I was pretty sure I liked the mini better than I did my Nexus 7. It felt faster, all the iPad apps I knew would be on it, and it was lighter and thinner. The display wasn’t as nice, but it wasn’t horrible, either. At the same time, there was probably no getting a mini any time soon. The Apple store website said they were two weeks out. A few weeks earlier, when I’d been researching small tablets, I’d asked an Apple Store lady how they were selling she told me there were lines outside the store each morning. Then she shared a little trick about scoring a mini I’d filed away more as a curiosity of Apple’s ingenious ability to sell a thing it has in stock yet also render than thing unattainable: Going to the Apple Store website at 10 p.m. (no earlier, certainly not much later) would reveal whether minis might not be in stock for pickup the next day. All I had to do was put one in my shopping cart then step all the way through to just before checkout. I asked her why people would queue for something they could just order online the night before and pick up at their leisure. She shrugged and smiled in a knowing way that kept things on this side of respectful, but only barely. Looking back, I have labeled her The Naughty Apple Store Lady.

That trick was on my mind a little on the way back to the office, but I was mostly telling myself that by the time I could just go buy a mini or just put a mini in my shopping cart at the Apple Store site, I’d be over the whole thing and happily entrenched in the Nexus. Heck, I told myself, maybe I’d even build that Linux box and forget all about it!

But that night, in bed with the Nexus, after the last email was read, the last Pocket item checked off, I thought about the trick and wondered if it was true. So I opened up Chrome and browsed to the Apple Store and put a mini in the cart. Then I visited my cart. In the split second before the page loaded, did I see the twinkle in the eye of the Naughty Apple Store Lady? I think I must have, but I couldn’t swear to it now, and either way I was definitely remembering her barely respectful smile as the Apple Store cart page told me I could have a mini held for me for pickup, and that was that. I was down there at lunch the next day.

What can I tell you about the thing? It’s smoother than the Nexus 7 for the things I use it for (page scrolling, especially, is comparatively amazing, but the apps in general are just better put together) and the battery life is great. It’s a much more familiar experience, and after over two years of using an iPad, that counts for something. It handles just fine one-handed on the Max and it’s far more comfortable for the ways I use it than the iPad 3 was. It cost more than the Nexus 7, but I think it was worth it.

Now to endure a few more days of

$$$ how do I call !!! $$$

and offers of slightly rusty but functional bicycles in trade for my Nexus 7.

Notes on the Nexus 7

I suppose the change in job surroundings probably had something to do with it, but once my Kindle 2 started circling the drain and I realized my iPad 3 was no good for commuting, I decided an iPad mini wasn’t automatically the very best choice for consolidating my tablet and e-reader, mostly because I’m not spoiled by the Retina display, exactly, but because the mini’s lack of one bothers me in a device I want to use mostly for reading.

It took a few days to find one on a showroom floor that was actually running, but after poking around at Best Buy, WalMart and a pair of office supply stores—all of which had units for sale but not for trying—I found a Nexus 7 at a Fred Meyer I could play with for a few minutes.

I don’t have a lot of time with Android. I was once sent an Android phone so I could review an instant messaging app and was left with a distinctly poor impression, mostly owing to the device itself, which was slow. I’ve played around with other peoples’ Android devices here and there, but never for more than a few minutes. So after ten minutes with a Nexus 7 at the Fred Meyer, I decided things were better enough to give one a shot.


If you asked me to describe only the good things that come to mind when I think about my Nexus, I think I’d start with its physical characteristics (I’m treating it more like a Kindle replacement, so how good it feels in the hand and how crisp its display is matter a lot) and I’d give a cautiously respectful nod to how flexible Android itself is.

Some pros:

  • Very nice size. Easy to read one-handed on my daily Max commute. Grippy back that feels a bit more secure than the metal back of an iPad probably does.

  • A lot of apps register themselves as share/data handlers in ways I appreciate (e.g. sending a URL to Pocket or org-mode for Android). I think iOS must have this backwards, because I was astounded the first time I shared a link from an app and found so many possible recipient apps for the data. I’m guessing a few things in iOS are native handlers, and it’s on developers to name other handlers from within their apps. On Android, it seems more as if anything can declare itself a handler. It annoys me that Apple knows all about the value of system-wide services and isn’t doing it better in iOS.

  • Customizable in smart ways: You can designate hotspots where the device shouldn’t attempt to download updates or automatically do background data pulls as a way to conserve bandwidth for cases where you’re using your phone’s personal hotspot capabilities. I think Apple “solved” this by forcing users to explicitly request updates, and by imposing some pretty strict limitations on how long an app can run in the background.

  • Nice display. Not “Retina” quality, but close enough that I can’t tell a huge difference.

  • Clever gesture typing (you drag your finger over the keys and it figures out your intent rather than requiring you to peck each key).

  • The Google Play app store lets you try an app out for a few minutes and delete it for a refund if it doesn’t work out. I’ve done this four or five times in the last several days. There’s a “con” lurking under there, though.

And, finally, most of the Web services I use have an official Android app, including:

  • HipChat

  • Trello

  • Evernote

  • Pocket

  • Amazon Kindle

  • Netflix

  • Twitter

  • Facebook

  • Google+

After five days with it, I liked it a lot. As an e-book reader, I’d say “mission accomplished,” at least as far as using it with the Amazon Kindle app goes. But there are a few cons, too:

  • The notification interface feels a little busy and I haven’t yet figured out how to get in the sweet spot of looking over recent notifications without getting notified. So there are times where I feel a little pestered, or just bemused because an app has decided to make some icon up in the top left corner blink at me for reasons I’m not clear on. One thing I like about iOS, now that I know to do this, is being able to tell apps they can’t notify me about anything at install. If I use them enough that I think it’d be nice to hear from them, I go back and enable them in Message Center later.

  • Many of the apps I’ve tried that aren’t official apps from large Web services feel sloppy. The icons look cheesy or aren’t very descriptive, color schemes have a nasty Windows ’95-era flatness, and ad-driven trial editions place the ads over bits of the UI in a way that suggests slipshod design. From app to app, things aren’t very consistent, either. Lots of different color schemes, icon arrangements and configuration conventions. I’ve found myself using that refund feature a lot.

  • I feel like I have to “reach” to the top of the screen to do a lot more stuff in Android apps, but that could be because some of the apps I’ve used haven’t been built with a tablet form factor in mind and elements that would be thumb-able on a 3.5″ or 4″ screen simply aren’t on a 7″ screen. I read some Nexus enthusiasts on reddit arguing that apps “just working” in Android regardless of the display size is a great feature. I think it’s not so much. More on that below.

  • Some of that flexibility/customizability I mentioned as a pro contributes to a sense that there’s an awful lot that probably needs to be managed. I don’t know if I’d have more or less of a sense of that with an iPad, because I’ve been using iOS for years now. I’m sure there are ten or twelve things I do with any new iOS device to get it feeling more comfortable that I don’t even think about anymore. I also know there are a number of things in iOS where Apple simply requires your deference since there’s no changing them short of jailbreaking.

The biggest con for me—not you, me—is probably the missing iOS/iCloud/OS X ecosystem. I’m not horribly locked in to iTunes or anything like that. I do, however, enjoy the way Apple has integrated the messaging app on my iPhone with iCloud, and that in turn with Messages in Mountain Lion. I like the smooth integration between the notepad app on Mountain Lion and iOS. I like Siri and its integration with a bunch of things, too. Google provides an ecosystem that does a lot of similar things, but Google also mostly lives in a browser, and I’m not completely sold on life through a browser viewport.

On Yet Another Hand

A lot of those notes were written a few days ago. Since then, the newness of the thing has worn off a little, I’m through that new device phase where I’m willing to accept that this or that peeve is me just not getting something, and some things just annoy.

While my main purpose for buying a 7″ tablet is to have something I can manage one-handed on my commute, I’d like to do other stuff besides read books. I’ve found a few Twitter clients, a reddit reader and an RSS reader, for instance, and there wasn’t much way to avoid comparing them to Tweetbot, Alien Blue and Mr. Reader. They don’t compare very favorably at all. They feel cramped despite having 7″ to walk around in, their type can be pretty tiny (in a way you wouldn’t mind on a phone but that feels exhausting when the amount of text in such a small font goes on for so long), and it’s pretty hard to hit the right link or button. Android does some nice things to help solve that problem (Chrome for Android does, anyhow) by zooming in on proximal links so you can hit a big, fat target instead of mashing two links at once, but in other apps it’s just a bunch of tiny buttons and links that are hard to hit just so and eventually make you feel mistrustful and burdened because it’s so easy to hit the wrong one.

One other annoyance—I’m not going to catalog every one I’ve experienced—comes from the automatic brightness setting, which is far too aggressive. I can be sitting and reading in a room where the light isn’t changing and the display brightness dips up and down every minute or so. Since the Nexus 7’s battery life isn’t stellar, automatic brightness is a common sense battery saving measure, but it’s a nuisance right now, which means screen brightness becomes A Thing You Have to Manage if you aren’t near an outlet most of the day and don’t want to worry about your battery. Either that, or just put it in airplane mode a lot.

If I were interested in having a 10″ tablet to use as a notebook replacement, I’d probably prefer an Android tablet because it would walk closer to the laptop/productivity side of the tablet/laptop divide, and I’d appreciate that. With a 7″ tablet I want to use primarily to read books or saved articles on the Max or on a break at work, the things that make iOS more limited aren’t such big liabilities, and the relatively staid and low-maintenance approach Apple pushes on iOS seems like more of a plus.

More Hands!

But I’m keeping the thing. I thought for a day or two that I might get rid of it and get an iPad mini instead, but it’s good enough for its central purpose right now, which is reading books and saved articles on the train, and I get the impression that the software catalog is improving. With the new cellular data Nexus 7s and the growing sense that 7″ is a good screen size for tablets, I think the device will improve over time as it becomes a desirable target for developers and the software grows up around it. As I noted earlier, most of the web services I use have Android apps, and those apps tend to be the most polished that I’ve seen. It’d be nice to have a deeper catalog of independent apps, but I think I believe that mainly because a 10″ iPad is smooth and powerful enough to do an awful lot more than I expect of an ebook reader with benefits. In other words, I don’t really need a deeper catalog of apps for this thing, and I need a few more weeks to let that sink in.

I also think that at some point it will either die, or I’ll decide to hand it off to Ben, and by then Apple will have improved the iPad mini’s display enough to justify a bit more of that $120 premium it’s charging now. Once that Retina mini comes along, it’ll be more tempting. Once that Retina mini comes along and is on the market for six months and starts popping up at a nice discount for refurbs, it’ll be even more tempting. By that time, though, there’s a good chance I’ll be feeling better about Android in general and won’t care. We’ll see. Meanwhile, I’ve got a decent Kindle replacement for my commute.

One Quick Update: Anandtech is of the opinion that we should not hold our breath for a Retina mini. It makes sense. I think I’m still going to keep the Nexus 7 for now and see what the next-gen mini looks like. I don’t think Apple meant to make a mini at first, so I’d like to see how it does iterating the design once it has a generation of feedback.

Evernote and Repeatable Checklists


Evernote checklist



Two of my favorite Evernote features are checkboxes and the “Copy Note To …” option. 

Since I’ve got eight sites I’ve got to do stuff on, checklists help me a lot. To make a checkbox in Evernote, you can either click on the little checkbox icon in the formatting bar for the note, or use cmd-shift-T (insert todo). A while back, I made a generic checklist of my ten sites and put it in an Evernote folder named “templates.”

When I need to use the checklist, I right-click on it and pick “Copy to Notebook” then click “Inbox,” then name the new copy something appropriate to the task I’m working on. 

Pro tip: You can search “todo:false” to find every note with an unchecked todo box in it. 

Pro pro tip: You can add a tag argument to your search to find unchecked todos in things of a given tag, e.g. “todo:false tag:work”

Pro pro pro tip: You can save that todo search by clicking File > “New Saved Search” and giving it a name. 


© Michael Hall, licensed under a Creative Commons Attribution-ShareAlike 3.0 United States license.