PKM #1
Where we are, where we go from here
This will be a series of journal articles in which I think out loud and play with ideas about what personal knowledge management means, how to implement it best, and what kind of impact such a system has on our lives. It’s meant to be disorganized, haphazard, and badly written, so expect nothing fantastic, but I hope this will be a good outlet for raw ideas.
The information we take in and use has outgrown the 70 cubic inches inside our own heads. How do we manage all of that information? How can we make use of that information effectively, without simply trusting our brains to not forget — which, with all the information we have to manage, is simply impossible?
Many of use tools like Evernote, Simplenote, OneNote, and Apple Notes. Some of the more intense folks use emacs org-mode, TheBrain, or have some sort of organized flat-file system.
These tools aren’t enough.
Issues abound with each of them. Evernote is awkward and heavy; Simplenote is, well, too simple and not structured enough; and TheBrain, while highly structured, is too complicated to use for everyday life.
What I seek is a tool that solves personal knowledge management like Mailbox solved email (for me). One that strikes the optimal balance between being targeted for a specific workflow and one that is flexible enough to accommodate many.
One that can take in all of my information, in any format, and keep track of it in a structured, but not heavy-handed way. One that fits in to the behavior patterns of the user — like Mailbox and its “snooze” functionality — and really fits in, at a 90% level, to the needs and behaviors of the individual.
Anecdote
In Our Computers, Ourselves, an episode of the podcast Invisibilia, a guy named Thad Starner is interviewed. As one of the leaders of the Google Glass project, he has been wearing wearables for much longer than pretty much anyone else. His old system, The Lizzy, incorporated a display, keyboard, and other things so that he could type in, store, and pull up information as needed.
When he met with a friend, say, he could pull up their previous meeting and see that their son went off to college, and ask about him — something that he would have not otherwise remembered. He can, say, look up past interviews that media folks have had with him and see the most common questions that they ask him about The Lizzy.
He took this augmented memory system one step further: he coded the system to automatically look up information related to what he was typing at that moment. So, he’d be having a conversation with someone about some topic, and when he types this in, a distant, tangential piece of memory will surface thanks to the automatic related-content search.
While I do not aim to build such a complicated system (technically, socially), what I do want is the foundation and framework to enable such a knowledge management system to come about. One where I could cleanly store meeting notes, pull them up when needed, and have it act as the substrate of persistence and structure from which apps could be integrated into it — such as one that looks at my calendar, sees who I have meetings with that day, and pulls up the last 2 meeting notes from that person and emails them to me.
The ideal PKM
- Joy. Using the system should be a pleasure, not a chore when you want to input or find information.
- Balanced. Structured, but not heavy-handed.
- Integrated. Strong internal integration, like a universal tagging system, and strong external integration into other apps.
- Clean, distraction-free, WYSIWYG editing surface.
- Two separate areas: structured notes area (think Evernote), and an unstructured, loose notes area (think Apple Notes or Simplenote).
- Easy to use, traverse, and search.
Open questions
Primary structure/folksonomy. Should the system be primarily tag-based, like in the design above? Or should it be driven by categories? Are categories in a filesystem-like format too structured, making drill-down annoying? Should tags be the main organizational system for loose notes, and a filesystem-like categorization system for structured notes and files? Will that get confusing? If we have tags, do we have structured tags, as in, “keystone tags” like People or Health or something of the sort?
Approach. Do we try to build this out as notes-only first, or do we try to be inclusive of every sort of file? Should we try to figure out the philosophical underpinnings of such a system and the ideal structure of it, and then build it? Or should we try to rapid prototype and see what works?
First-class citizens. want this to be flexible, but should there be first-class citizens that are embedded into the application? Dates, tags, and all that are first-class metadata; should a summary be also first-class metadata? Should comments be allowed on every item? Should metadata itself be first-class, such that every item has a customizable set of metadata?
Approach
I’m reminded of this, which many people have heard before, and I hate repeating it since it’s obvious, but it is quite relevant:
The ceramics teacher announced on opening day that he was dividing the class into two groups. All those on the left side of the studio, he said, would be graded solely on the quantity of work they produced, all those on the right solely on its quality. His procedure was simple: on the final day of class he would bring in his bathroom scales and weigh the work of the “quantity” group: fifty pound of pots rated an “A”, forty pounds a “B”, and so on. Those being graded on “quality”, however, needed to produce only one pot — albeit a perfect one — to get an “A”.
Well, came grading time and a curious fact emerged: the works of highest quality were all produced by the group being graded for quantity. It seems that while the “quantity” group was busily churning out piles of work — and learning from their mistakes — the “quality” group had sat theorizing about perfection, and in the end had little more to show for their efforts than grandiose theories and a pile of dead clay.
(from Art & Fear, via Coding Horror)
Up until this point, my M.O. was to theorize about what a perfect PKM would look like, and then implement that. Yet, this doesn’t seem to be the right way to go about it. The idea of a PKM is still so nebulous, the categorization system so unknown, that wanting to come up with a perfect design from the beginning seems like trying to land on the moon given a photo of it. That photo, by the way, was taken by an early-2000s cameraphone jammed into an eye-hole of a telescope, and it’s a badly compressed JPEG anyway.
So what do we do instead? Rapid ideation and iteration. Using the Evernote API as a substrate — along with their fabulous sandbox environment — the goal is to create different ideations of PKM software. Build a basic version, and include minimal design styling — but make the goal testing it out and seeing whether different paradigms work or not.
Create new paradigms, test them. See what works. See what I actually use.
An initial design
- Two Evernote notebooks: Wiki and Notes
- Build app on top of Evernote API
- Wiki = more structured content. Content is backed by Evernote, but the structure is within the application itself.
- Attempt to replicate Artifact 1 (category) and Article 3 (2012 categorization system) for structured text notes.
- Then, add the unstructured/loose notes system, which operates similarly but is a stream of content, not structured in the knowledge base format.
- Think about ways to switch between these two domains and the sharing of content between them.
Artifacts
