open/wiki commentary: both available.

In fact, do it for all books. See if BN buys the module.

Allow specifics: title, version, ...
Then distance from one note to another may be measured by a combination of the different dimensions, weighted differently:

• translation
• version
• published year
• chapter
• page
• chN, vM vs chN
• commentator
• author
• title (spelling distance)
• category
• popularity
• similar books
• similar author
• similar commentator
• similar era

Why is C++ complex, and Perl simple?
For one, the rigidity of the syntax forces me to remember lots of small details, hence increasing the complexity. C++ being a type language forces everything to have a type, again adding to the details and hence the complexity. When I use other people's library, I would have to remember all their classes and usage, this again adds to complexity.
Perl, on the other hand, has extremely relaxed syntax, thus making it simple to use. In fact, statements are essentially either regular expressions or functions.

Thinking of programming as specifying various modules through an implicit GUI interface, a complex language is like having many pulldown menus, where each has a long list of rather similar items. A simple language is like a good GUI, where one can quickly specifywhat they want, without much ambiguation.

Often we design programs with only the list of features in mind. Yet more importantly we need to ask: what will the human do in the mean time. This overall emphasis on user experience, of maximizing the objective when human and machine collaborate together, will be ultimately important.

One reason Furl will overrun Google is that Furl allows the human to feel important and in control, by offering some meaningful activity, such as the high level annotation.

Similarly, as technology progresses, and replaces society, we need to ask: what will the human do in the mean time. Then we can design an integrated system where both the machine and human will collaboratively achieve something beyond what either can do.

Ideas for a programmable alternative to LeapPad:

• A Flash program. But then that relies on a PC and its annoying screen.
• Use a tablet board, then printout the screen to be paste onto the tablet. That would avoid the screen.
• Still, portability is important. So how about printout of a flash program on a PDA, or a Tablet PC.
• Maybe just get their lisence and deal with the MindStation

The various slinky torus I created, both continuous and discrete models, could be remake into a line of innovative toys. Perhaps I should just be playing in a toy design company. Then again, toys that I appreciate often demand such mental agility, such as the algebraic and topological puzzles, that they often had to go on sale at stores, because of lack of demand.

Use Flash to construct Chinese learning systems as demos, then contact LeapPad about being developers for them. With the advent of ePaper, the future belongs to improvements over the current eye-straining monitors and related viewing systems. If nothing else, my own sight demands such system.

Aural navigation a la TV Raman is linear in time, and hence totally defined by a hierarchical tree structure, at most with links between nodes. But a 2D navigational system, such as for screen viewing, writing on tablets, etc, has yet to be formulated and solved. The HyperText system of the Web is the closest in approximation.

Furl is facing the scaling problem now. I was not able to Furl when lots more users are logged in. I was reading the article from SearchEngineWatch at the time: Create Your Own Online Web Page Archive

Then it occurs to me: a distributed Furl, not on a centralized machine, but across many stations. That is probably what they are working on. But even more generalized: people furl to their own blogs. then an aggregator will visit all the blogs to create the total furl index.

Rather than just social networks, the ability to search within the public archive of certain person is extremely powerful:
Furl - Public Archive - tranquileye (John Stevenson) has bookmarks dated back to 2001. So he must have some ability to transform all his bookmarks into Furl data. In Furl I can search this valuable bookmark set.

A generalized version of "Furl" is not just to bookmark URLs, but to bookmark individuals. Then all the resources associated with that individual (his preferences, recommendations, ideas, ...) will be lined to your database. Practical implementation: start with web pages, and try to interpret all the preferences from them.

desktop browser addon (toolbar) that looks at your history, and cluster the websites according to the navigational branching history, to provide logical units equivalent to "articles", and offer the result as potential publishable content for upload to Furl.

Bloggers Parliament has an interesting approach of Identifying problems