Avatamsaka-like: the html of this very page disposed as a colored physics-obeying tree. I'm pretty overwhelmed with all the possibilities of computer aided thinking. It is very easy to see how word processing just by itself makes writing so much easier, and we may call these developments from writing in caves to movable type and computers sort of a first revolution. The second great revolution, which is also pretty easy to see, is network communication: bountiful and very cheap information at our fingertips.

The third revolution is already happening. What people call "Web 2.0" is just the perfect combination of the first two revolutions with the very incipient beginning of the third, which is related to the organization of information: folksonomy (tagging), semantic web and what I would call the first teratogenic babies of "folk ontology" or "personal ontology". In computers "ontology" has a distinct meaning than in philosophy: it means the shape or structure of data relations, a data model, in other words, how we plan the hierarchy of information.

Tzal.org and its underlying software, InterDP, is my feeble attempt to construct one such "personal ontology", connecting anything I find relevant and tagging it – somewhat reinventing the wheel for most things (tagging in blogs is the norm nowadays, but not when I began), but really starting to consider the ontology issue from a personal perspective, which I think is a novel concept. One simpleton's way to talk about it would be "shaping the computer to work as your mind works and not the other way around". We're talking about organization of inputs and outputs, anything someone produces, in semantic networks, and not only hierarchical taxonomies.

When I say "personal", it means that the very way we relate to the computer and how we work with it is an ethic-aesthetic statement. In a world where information is overwhelmingly omnipresent, "flower-arrangements" of it are quite welcome.

That is, I intend Tzal.org as a tool for three things: storing my production, brainstorming, and automatically organizing data through an ontology. Although I mostly haven't been able to find the time to implement all the features I imagine possible, I'm currently working on three main ideas.

One is actually developing a simple semantic layer for the current tags, that is, making it possible to ascribe relationships between tags – both automatically, through statistics, and manually through a simple interface.

The other is making a fast Windows GUI for all InterDP features. This is half done. I tried for some time to integrate some existing API, like Atom or Blogger, and using some blogging client software, but it was just too much work, and these softwares lack some of the most interesting and already implemented features of InterDP (mainly tag-related). So, again, reinventing the wheel, I designed a custom API and GUI – including some Microsoft Word macros that make posting with metadata so much easier.¹ InterDP Windows GUI, text correctly formated pasted from a MS Word macro.
Finally, I wish to integrate Free Mind, a Java Mind Mapping Software, with the tags semantic network. Since the first stirrings of wanting to develop an ontology I thought I needed some kind of graphical interface for the relationships, and I already have a general idea of how to do it. In the far future I imagine using virtual physics in the tags relationship interface.

InterDP will be used in a series of websites, including eventually a Tibetan translation multilingual wiki-of-sorts. (If you wish to donate for this to be done faster, please contact me. I have many concurrent projects, so I tend to work more on those that pay for the lentils and hardware, and then for academic previously engaged commitments (philosophy course at UFRGS) – finally for programming and actually producing this sort of content, with all its little images and links and tags.)

For me it is really relevant to think these bits of information will live a little longer than my bones, but not only that, I face my overgrown notebook of alchemy – syntax and semantics both tied to a single aesthetic and ethic view, shaping me as I shape it – as the single mirror of the Great Work.

It may well be that the Stone of the Philosophers is this diamond of interconnected clarity, but I'm prone to delusions of grandeur, so you may as well read it as a plain, unvarnished and mediocre, weblog.

Tags must be easily attributed to any objects, even tags themselves. This means that any editable data field must be able to contain tags, and that tags must be easy to make.

Once there is a list of tags, a simple text file must be created which puts them alphabetically. Attributes to tags can be added in this file between curly brackets, such as this:

Amarelo {
     definition: An old dog that lives at [Caminho do Meio];
     type: being;
     class: dog;
     word: proper name;
     photo: http://xoxoxo.xom/xoxo.htm;
     language: portuguese;
     anecdotal: once I saw Amarelo being runned over by a truck;
     when displayed show: class, definition, anecdotal, photo; 
     }

vajra {
     definition: The indestructible and virtual nature of [phenomena];
     sandabasa: The penis;
     type: thing;
     class: tantric terminology;
     word: name;
     language: sanscrit;
     translations: diamond (english), thunderbolt (english), [dorje] (tibetan);
     when displayed show: class, definition, sandabasa, translations,; 
     }

Some of the info may never be used, and most entries will obvious lack any info, but since it is an easy file to edit, sometimes it will be fun to peruse and create some connections.

The ontology itself starts to work with the idea that all attribute titles become tags themselves, and of course attribute definitions can have their own tags also (in the example above, between normal brackets). When you save the file it is parsed and organized again alfabetically and including new tags posted on the site or created on the file itself, as well as refreshing statistical data (weighted number of appearences, for instance tracking relevance) as "self-created" attributes.

It is possible to stablish some filters to display only sections of the file when it becomes gigantic. You can display only a single entry or anything that is connected through an attribute. Maybe it is possible to construct a graphical display of relationships (but this I don't know yet how to do). For users, any instance of the tag shown in tzal.org will have a hover-on floating sheet of info, for the admin, clicking on it will open a pop-up with editing options.

So we have two arrays, that of tags and that of attribute titles. Attributes made tags receive technical attributes which tell them how to relate (simple procedural instructions) and how to be displayed. This self-reference also helps to build documentation on the system itself.

This way I believe it is necessary to unite keywords (tags) with the dictionary — that seems not to have worked and will be abandoned as it is (a pity, since it has costed me some dozen hours of work). Then all tags can have definitions, and more than one type of them.

Another way of doing this would be designing hierarquies of words, that is, you could start with the more general concepts, such as "thing", and evolve to particulars. The proplem with this approach (which I call the "first principles" one) is that you have to follow a pattern of design and spend considerable time deciding what is relevant or not. It would have the advantage of being more rigorous, though. I believe the organical interdependent approach depends much more on weighted relationships than in rules, and so much more feasible, although it will always be "plagged" by incompleteness and idiossincrasy (which are not so bad) or irregularities.

When I'm in programming mode, it's very difficult to do anything else. I program mostly in the bus or walking — sometimes, irritatingly, in the intervals while talking to somebody — and then when I come to the computer to write it, there's never enough time to implement all the things I already know how to implement. Also I get worried about taking care of the engine and forgetting the content (as I did for more than six months). Then there's the more experimental and hard work of design².

I believe if I had 4 or 6 months and controlled my schedule not to overwork (the 12 consecutive hours for 3 weeks in february had some impact on my health), while still enjoying leisure and research in new technologies, I could really prepare a top notch Web 2.0 software. Well, this won't happen too soon, and if I wasn't in school, people would try to get me paid jobs. Ah, how it would be nice some broadband Walden Pond. Then I would be able to take my mind from programming for some years.

In the present I have a billingual multi-user-level prospective-posting folksonomy-indexed blog with a system of rating for posts, automatic glossary, which can use templates³ to generate "books" from posts aggregated under the same name. One installation can support many different databases and users authorized can have one or more single or cooperative sites⁴.

For the near future I will complete posting by mail and posting by Microsoft Word (yes, write there, click a buttom, fill a small form: its on the web, organized with folksonomy — PHP+VisualBasic).

Then I will work somewhat on the "keywords" (the folksonomy itself), first making a simple tag cloud and then elaborating a language representing semantic relationships (homonymy, synonymy, antonymy, polysemy, paronyms, hypernymy, hyponymy, meronymy, metonymy, holonymy, exocentric, endocentric and others), establishing my own semantic spectrum. I intend to write a manifesto about the ontology of my choice. I will also be using the Dublin Core standard as soon as I migrate to MySQL.

I wish the content to grow wider in spectrum, uniting trivial ramblings, opinions, experimentation, counsels and proper articles. Oh zebu, maybe even the lowest form of entertainment, fiction. In particular I want to explore the connections between humor, language and spirituality. I will always work with a certain amount of meta noise, which I think I can turn to the benefit of my goals. The religious, philosophical and psychologic nature of irony, nonsense, in-joke, understatement and other language games⁵ is to be studied⁶. But probably I will fixate on posting old texts for quite some time, while I don't make the semantic adjustments that will certainly be so useful.