1. 11
    pencil and paper thinking philosophy dgsiegel.net
  1.  

  2. 8

    A phone can still be dialed with a string of numbers like it was a hundred years ago, and we’re still forced to remember these numbers — or look them up when we need them.

    In 1921, dialing a phone was bleeding-edge; the first dial systems were introduced (in the US at least) in 1919. Most people made a call by picking up the handset, waiting for an operator, and telling them who they wanted to call. Voice activated agents!

    Phone numbers today are like IP addresses. They exist for reasons of underlying infrastructure, and people mostly don’t think about them. I don’t have my wife’s or my kid’s phone numbers memorized, nor do I see those numbers when I call … I’m not sure why you’re forced to remember or look up phone numbers, but you might need to upgrade to a smartphone ;)

    Yes, word processors and web pages are rectangles with words going from left to right too to bottom, but that’s due to the way our written language works, which has been around for thousands of years and will doubtless be very slow to change.

    1. 6

      Phone numbers today are like IP addresses

      They’re actually more like DNS addresses. They’re no longer involved in routing at all other than as an index into a table for lookup (once you’re past the country code - I think that’s still used for routing). Given that they’re really equivalent to DNS entries, why can’t we have something a bit more human-friendly than a number?

      That said, the thing I dislike the most about phone numbers has nothing to do with their encoding, it’s the fact that the identity of a phone is conflated with the capability to call that phone. This is picked up by systems such as Signal that use phone numbers as identifiers. I would love for Signal to build a system that let me generate a UUID that allowed a single Signal user to contact me without disclosing my phone number to them or vice versa, such that on first use we did key exchange and they were never able to reuse that UUID, only send messages / calls from their endpoint to mine. I’d then be happy to give that out instead of a phone number to anyone that wanted it, knowing that they could never use it for spam: if they started sending spam, I could block the contact that they added for me and they’d never be able to use the UUID a second time to add another contact.

      1. 4

        Given that they’re really equivalent to DNS entries, why can’t we have something a bit more human-friendly than a number?

        You then run into the same issues with DNS, where you need some way to stop someone from registering a ‘phone address’ of “Totally the IRS, I Swear”. Plus, given that way more people have phone numbers than have personal sites, you’d run into disambiguation problems when trying to call an individual person.

    2. 6

      This article asks some questions that have been asked so often before, but unfortunately never answers any of them, as such articles rarely do.

      Besides, the premise isn’t exactly true either; a computer is much more efficient than pencil and paper, which makes a qualitative difference - when was the last time your paper automatically told you about incorrect spelling and grammar, or calculated the word count, or, more interesting, the reading difficulty (using multiple models!) of your text in a fraction of a second? Even the best secretary in the world isn’t as fast as a computer at these things. And writing and editing and then rewriting the same story over and over (especially when it is long) would have been absolutely exhausting in the past.

      1. 5

        Besides, the premise isn’t exactly true either;

        There is more:

        • It complains about contacts being the same as a rolodex and about having to type in phone numbers, but ignores that on phones contacts are a universal resource that show up everywhere and can dial automatically. Try that with your rolodex.
        • It complains about computers just continuing the pen and paper medium, completely ignoring video.
        • It complains about computers just being expensive replacements for paper, claiming that a compute costs a thousand dollars. It doesn’t.
        • It complains about data not being transferable between applications, but they are. Maybe not yet as seamless as hoped, but it is very possible,
        • It complains about computers only being able to represent static symbols, which is wrong.

        I could go on.

        1. 4

          It complains about computers just being expensive replacements for paper, claiming that a compute costs a thousand dollars. It doesn’t.

          And not only that, a computer is much cheaper than a personal secretary, which makes these services available to a vast group of people which didn’t have one before!

      2. 6

        Welcome to lobsters! Generally we encourage people to post and comment on articles they haven’t written, too, to participate more in the community.

        Anyway, re your post:

        We already know that computers can copy the old media and replace them with something that looks new. But what else can the new medium — the computer — really do?

        That’s the million dollars question, init? We all know that we can do more, but what are some examples of what “doing more” looks like?

        I think we’re further along than you think, though, in terms of moving past the paper paradigm. I recently had to edit some video, and DaVinci Resolve was an incredible tool for that. Did all sorts of things with interesting interfaces that would have been impossible without a computer.

        1. 2

          I find that pencil and paper thinking is superior to type writer thinking. OneNote is the best app I’ve used for post paper and pencil thinking and it works so well because it adds scissors and glue to the metaphor, and it still feels clunkier than doing it on paper because the paper is stuck behind a screen.

          1. 4

            I’ve found that an e-note (currently the ReMarkable) does wonders for replacing pen+paper, personally.

            But, what I really want is a (word processor?) that uses both pen and typewriter for the best of both worlds.

            Also some auto-managing and a smarter UI on the RM. I swear the RM misses half the point of going digital in the first place, and also managed to get its UI’s workflow backwards in several places.

            For instance, you create a new notebook by typing a name, before writing. It should instead create a blank new notebook, then let you PHYSICALLY HANDWRITE THE TITLE at the top. This is probably due to their handwriting precognition needing network access and being provided by a third party, so they can’t just up-and-require HWR in their UI.

            But, that doesn’t explain their “name and sort first and write the content later” approach in other areas - they have a “quick sheets” button that just opens a new page on an always-available “quick sheets” notebook, but they don’t provide any decent functionality for jotting down notes and then transferring said quick notes to its appropriate spot in your filing system.

            You can technically do it, but it takes several clicks every time you do it I’d guess ~7 including a multi-second hold), and every click takes maybe 2 seconds. So you’re talking ~14 seconds, which is NOT acceptable.

            I feel like the RM could be so much better than paper, and in many ways it is - I still prefer it over dead tree - but they have a single-minded obsession on being “digital paper” without acknowledging that they’re now responsible for supplying a good digital filing interface.

            In other words, it still feels clunkier than doing it on paper because the paper is stuck behind a screen.

            1. 1

              What’s the third party software story like on it? I’ve held off buying one but hope that a software community could make it more useful

              1. 2

                What’s the third party software story like on it?

                The worst and the best - it’s fairly open, it gives you the root password and lets you SSH in (apparently in order to comply with GPL3 terms), which, combined with the fact that the official software is deliberately barebones (but polished - they have Apple/GNOME syndrome), has ironically created the biggest modding community on any e-note device.

                The problem here is that the UI is a proprietary blob that can change after any update, and half the community hacks are built around modifying said blob. I’m not a fan of this.

                I think there are two escape routes to make the system more FOSS-oriented:

                1. Parabola-RM is a port of Parabola to the RM1, and if also ported to the RM2 it could replace the default OS for anyone who’s willing to give up on the default UI.
                2. The default UI probably wouldn’t be that hard to recreate, and creating a safe drop-in replacement on Parabola-RM would simplify the situation a whole lot.
                1. 1

                  Extant, sort of, and seems to grow a bit every time I check back, but not supported by the vendor and firmly located in hack space conceptually.

            2. 1

              I feel like I can answer the questions that the author is asking.

              First, the answer to “Why isn’t software better?” is that most users do not like new and different things. Most of the ones that I’ve met aren’t very naturally curious, and when confronted with even the simplest technological obstacle like a dialog box, will immediately shut down, rarely spending the microjoules of energy necessary to actually read it. (one of the things that every UI developer knows is that “users don’t read”)

              Second, the answer to “Why is it so difficult to transfer data between applications?” is that the platform owners are incentivized to make that more difficult, so they do, and it is. Desktop applications are usually pretty good about reading and writing a variety of formats - just look at what Blender, Inkscape, and GIMP can do, for instance.