How far are we from having a system like Memex?
Vannevar Bush's "As we may think", published in 1945, described a computer system that acted as an
intimate intellectual assistant.
As is well-documented, his article helped to inspire countless achievements in automation, such as the
World Wide Web, browsers, and the Internet communication systems we currently take for granted.
But even though we now have access to vast resources online, it's pretty clear that one of the issues
he addressed in memex, the idea of a personal "trail", is still not properly addressed today:
"It is exactly as though the physical items had been gathered together from widely separated sources
and bound together to form a new book."
"Thus he goes, building a trail of many items. Occasionally he inserts a comment of his own,
either linking it into the main trail or joining it by a side trail to a particular item."
This idea of your books, with your notes, has still not been integrated into the high-tech
browsers we use today.
As others have pointed out before, even basic capacities of a book are not respected in the web experience.
Ted Nelson, for one, points out that the web missed the opportunity to make links bidirectional. Books are
bi-directional, so if we're creating our hyper-book-like projects, we should be able to hyperlink through them
backwards and forwards. With no loss of trail. But the only remnant of this idea is in wikipedia, where a
page automatically maintains a list of all pages that refer to it. This certainly is a step towards
achieving a personal trail. But it's still not personal.
Since the web was launched, people have tried to address these issues: transclusion, note-taking, and book-making.
All are sadly ineffective in browsers, and that disrespect of basic human needs has propagated to
the products we use on the internet, which have no respect for our intellectual or intuitive orientation.
Let's think about the human needs that Bush was addressing.
How would we satisfy them today, with current technology?
I'm going to merge the engineer and the designer into one person, for the sake of this discussion.
This is like the integrated "design-build" movement in building construction.
A design-builder's sensitivity to human needs starts with their sensitivity to their own needs. It doesn't start
with "ideas about design" or "ideas about building". It starts with trying to understand human needs, and
developing an ability to listen to their own experiences, uncomfortable or comfortable, confused or happy.
They become sensitive to emotional states when using virtual or real gadgets. They ask themselves: does this
or that help our intuition? Does this or that raise my spirits, or crush them?
This is hard to do, because there are many forces that interfere with this core sensitivity, both inside our own
minds and outside of them. But this is the only way to create good software, beautiful software, which very
rarely happens in the capitalist world, because of its drive for profits and tendency to bureaucratize.
It's also hard to do because the problems are complex, and the tools rarely improve, in the ways they need to,
to help us.
I believe that, unfortunately, most people who begin to have some career success with a certain level of
human-machine interface design and software development, have abandoned this quest.
Quality is short-circuited, solutions are terrible, people don't notice when they pass thresholds of
incomprehensibility, of untenable complications. The loss of the primary point of emotionally and
intellectually supporting people in our Memex-like systems, is just a symptom of a larger problem
with our approach to interactions with machines.
Although the problems of insensitivity in computing are pervasive, I have my own case study, Gatemaker
which, though it's not about this kind of machine, is about the difficulties encoutered when trying
to use human sensitivity to enhance human-machine interaction.
I may embark on creating a simpler browser and finder combination that addresses our badly-formed memexes.