Paper has been around for centuries without significantly changing in form or function. Over that time, many stable conventions evolved for organizing informat

Paper has been around for centuries without significantly changing in form or function. Over that time, many stable conventions evolved for organizing information on paper and using paper as a medium communication. Its digital equivalents are still immature - we don't yet have common expectations of how we share the same types of information digitally. The underlying forms are still changing rapidly, and the decisions we make in how we communicate are therefore more coarse. We don't think about the minute details of how we put out different bits of information in order when we first need to settle whether we're going to use email attachments or files on Dropbox or Google docs, or send an IM, etc. Eventually, conventions will stabilize as certain types of information become strongly tied to particular methods of representation and communication. Once those become stable contexts, we will be able to vary the more minute details within each context to carry metadata in the same way Tufte does with paper. This situation demonstrates an interesting (current) weakness with using technology to manage information. When your organizational systems, methods, and conventions are purely conceptual, it's very easy to change and adapt them in real time, allowing for a gradual evolution of processes, and progression towards nuanced equilibria. But when we rely on software, we may often need to write new code to support a new idea - we increase the upfront investment needed for experimentation, often drastically. And people who are non-technical can't experiment and adapt on their own at all. Until most people can adapt their digital tools as easily as their physical ones, it may take much, much longer to optimize their effectiveness.