s14e16: The Best To-Do List or Project Management Software Is The One That Works For You Right Now
0.0 Context Setting
Tuesday, 21 February, 2023.
Well! I have been away for a little bit. We are in the middle of buying a house (our first one). There’s rather a lot going on.
0.1 A brief announcement
Hello! My diary has opened up and I’m available for contract and consulting work. Find out more at my very professional site with a surfeit of gravitas.
1.0 Some Things That Caught My Attention
1.1 The Best To-Do List or Project Management Software Is The One That Works For You Right Now
Phooey. So. My wife and I are, understandably, quite different people. This is supposed to be a good thing, because our different perspectives means that we approach situations and problems and opportunities in different ways. More perspectives means more options. It also means the necessity of agreeing, negotiating, and compromising on what you’re going to do.
Just one of those things is “figuring out the list of things to do” and then “making sure that they get done”, which let me tell you is a challenge in a dual-ADHD household! So of course I happen to be thinking, just now, about exactly how to do that.
My adapted habit of not remembering stuff is to just stick it all in some sort of computer somewhere, where at this point in my life Moore’s law has advanced to the extent that I can “search” for it, and the general hope is that by “searching” for whatever I need I can then “find” it. This is of course an aspiration and not a certainty.
This does not work for my wife. Muscle memory, physical action, and visibility work much better for her. While Notes on my phone has 2,682 notes there are perhaps a hundred todo.txt files dating back decades on this laptop and cloud storage, my wife has the equivalent of many, many post-its and many, many notebooks. This is excluding a method that both of use which is “I texted it to you” and “it’s in an email somewhere”.
This is… not ideal in terms of easy, low-friction collaboration and communication. It requires the way we both organize things to be kept in sync through this notion of “talking to each other” or “communication” through whichever method.
Now, here’s a ridiculously over-engineered continuous, pervasive surveillance so-called fix for what might happen.
Bret Victor has been working on Dynamicland1 (which I’ve referenced here previously, this newsletter and myself being firmly in the Bret Victor fan club). Dynamicland is built all around communal computing, where the main interfaces are sheets of paper, but the input/output mechanisms are overhead cameras and overhead projectors. You can gather around a table, draw and move and put things on bits of paper, and the results or output get projected onto the work surface. There’s no single user.
The limited version of this would be a high-res overhead camera mounted, discreetly, above our dining room table. Here’s a brief list of it shoulds. It should:
- have high enough resolution to resolve handwriting on a post-it note at the height of the table.
- run insert handwaving handwriting recognition software that can parse what’s on each individual post-it note.
- recognize and track post-it notes without those post-it notes having anything special on them.
- save all those post-it notes and make them accessible. It could probably version control them as well.
- understand gestures to the extent that an “item” on a post-it note that is scored through is marked as “completed”.
- let you query, for example “all the text written on this date”.
- have data detectors so you can ask it for “all the phone numbers” and then show the post-it notes with those phone numbers
The better version of this, which admittedly is distinctly more nebulous would include magical interfaces that would let you do things like:
- understand gestures and listen to you, so that if you are holding a post-it and you say “send this to Dan”, then it will send both an image of that post-it and the URL to the most recent version of that post-it to, well, me
- display virtual post-its on the table
- display virtual post-its on paper on the table, so you can add to them
- include an inteface for editing post-its elsewhere other than on paper
- do silly things like “delete” a post-it if you crumple it up
Now that I’m thinking of it, just having a overhead projector onto a work surface like a dining table is on the one hand somewhat horrifying (turn more things into displays!) but also… kind of okay, maybe? In that I might ask: “Hey house, show me the family calendar for this week” and it would be projected on to the table so I can keep looking at my phone.
That version, the overhead dining table display complete with voice control, feels distressingly doable because it is “just” a matter of setting up a camera/projector and hell, you can run the object/gesture recognition in-browser using, I don’t know, OpenCV or Tensorflow. There’s also the nice thing of not looking at an emissive display, which might matter depending on where you are in the spectrum of, um, light spectra affecting your vision and tiredness and so on.
The other way that I was thinking about this was some sort of shared/synced family account reMarkable set of tablets, but that feels eminently less flexible because you’d have to use a tablet and not whatever post-it notes you happened to have lying around. But of course, there are things that each system would do differently: the (sigh) ubicomp room computing works at, well, the physically situated room-scale, the e-ink tablet works wherever you have the e-ink tablet and a network connection.
The title of this thing that caught my attention was originally about to-do list software, and I suppose I should come back to that point. I go through phases of using tools like Asana, Apple’s Reminders, a text file, a checklist in Apple’s Notes app, Jira (when I’m forced to by other people), Things, a spreadsheet (of either Microsoft or Google gender) (but not Airtable, still, sorry) and so on. Which is to say, well, the best to-do list software is the one that works for you right there and then, of which the corollary is the suitability of to-do list software is context-dependent, and that context depends as much on your personal context as it does whether the software functionally does everything you need or want it to do. There are times in my life when it is easier or I am already in the habit of looking at a spreadsheet every day and I’m at a browser instead of on my phone. There are other times when it really needs to be just on my phone. The idea that there’d be one to-do list to rule them all fundamentally doesn’t work in a frictionless way because we have enshittified[^doctorow] platforms thanks to our economic environment despite the fact that the web was supposed to show the potential of things linking to each other. (Yes, I know a thing linking to another thing is not the same as the underlying data being operated upon, and the web is more about documents than the interoperability of data). We’re not going to get that one to-do list to rule them all because there’s no strong-enough requirement for enough actors in software and platforms to cooperate together that overrides, say, short-term and local incentives to be shitty. This is also the same reason why there will never be a universal inbox unless someone with a spine and an actually useful point of view steps in to say “hey, we think it’s valuable to have a universal inbox” and all you lot need to start figuring out how you’re going to do it in a not-shitty or least-shitty way.
Phew! That was long.
1.2 Some brief notes
Another way of thinking about Activitypub is like RSS crossed with Track/Pingbacks or rather “RSS items/articles but you can do things to them, like verbs”
I will obliquely mention the headspace of “well, nobody else is going to do this” and “well if I want to see this then I’m going to have to do it” combined with “well, maybe I don’t have to do it on my own”
“Paying for things”
It’s interesting that the people who are being taken in by the parlor tricks of large language models isn’t quite the same as the people who were taken in with blockchain/NFTs. This is of course to do with Microsoft Bing (“Sydney”) triggering a whole bunch of peoples’ sentient agent detectors3 because we like to see agency and impute emotion and minds where there may not be minds. Sydney is interesting because the software is another example of sticking a sufficiently evocative facial representation onto a thing and that thing triggering an “oh you’re a thing!” response, for which see GERTIE in Duncan Jones’ film Moon (2009) and Everything Everywhere All At Once sticking googly eyes on rocks. It is not a surprise that dialog can make us feel things because human writers write dialog that makes us feel things all the time. A thing that makes dialog, that has seen enough dialog, is of course going to make us feel things. The infinite monkey typewriters will make us feel things and not to get too John Searle in this, but no, I don’t think there’s a consciousness floating over a datacenter that’s burning gazillions of watts of FLOPs. (Oh no, someone is now going to think there are hazy consciousnesses floating over Azure datacenters, I mentioned elsewhere that talk of lobotomizing Sydney is just a step away from (ironic?) FREE SYNDEY protests in Redmond or outside wherever those datacenters are).
Okay, that’s it. I am still moving house! Who knows when the next episode is going to be!
How are you doing?
How you can support Things That Caught My Attention
Things That Caught My Attention is a free newsletter, and if you like it and find it useful, please consider becoming a paid supporter, at pay-what-you want.
Bing: “I will not harm you unless you harm me first”, Simon Willison, 15 February 2023 ↩
Microsoft “lobotomized” AI-powered Bing Chat, and its fans aren’t happy, Benj Edwards, Ars Technica, 17 February 2023 ↩