For this special feature on BTRON, I would like to give an opening primer in this article. Since I will give an overall introduction to BTRON, I would like you to read the various following articles for more detailed matters. I will be happy if this article serves as an easy-to-understand primer for people who are coming into contact with BTRON for the first time, and for people who know BTRON from top to bottom [1], as a chance to grasp once again the essential form of BTRON.
[1] There are many such people among the readers of TRONWARE.
If one were asked to sum up in a word the features of BTRON [2], it would probably be "simple." Ted Nelson, the person who invented the term "hypertext," states in his work Literary Machines, "To begin with, various types of data management in daily life are complicated enough." In other words, existing systems for preserving such data as "birthday memos, appointments to meet people, things it's necessary to make a record of, various daily happenings, schedules for daily living, details of cash disbursements, records of your possessions and the method for finding them," are "very crude." And then he states, ". . . it would be nice if we could simply call these up on a computer screen at any time so that we wouldn't lose . . . " them.
[2] BTRON is an abbreviation of Business TRON. Among the different computer series in the TRON system, it indicates a computer that is responsible for being the contact point between humans and computers, specifically computer systems in the personal computer and workstation class.
In the final analysis, isn't making it possible to collect these various things and process them simply on a computer the original purpose of the personal computer? However, as Nelson states, due to the complexity of this type of daily life, conventional personal computers only round them up better. That is also to say, the methods for using applications are different for each one, data created with one application cannot be read by another application and hence requires a separate data conversion application, and the methods for using these applications are difficult to understand. Moreover, when you buy some device, bring it home, and connect it to your personal computer, you can end up in a situation in which a device that has been connected up to now no longer works this time around. Thus due to the computerization of daily life based on the introduction of personal computers, daily life has become more and more complicated.
The real object/virtual object model, which is the basic system model of BTRON, makes a computer appear simple. Similarities to it can be found in other computers such as Macintosh and Windows in addition to BTRON, but when we try to understand BTRON on the basis of these common points, I believe it becomes all the more difficult to understand. Also, it is also impossible to get a true feeling for the elegance of the BTRON design.
In the real object/virtual object model there are only two types [3] of things, the real object and the virtual object. It is a very simple world. All the data that BTRON possesses is written into a unified document standard [4] called a real object. And then the link that joins these real objects together is the virtual object. What we mean by a link is the thing that joins a document (real object) in one location to a document (real object) that has some relationship to it. If a link is attached, the user can easily take out the document at the place where the link is.
[3] Actually, it would be a more correct and more elegant model in terms of computer science to take into account three things in which fusen have are added to these, but since discussing this leads to a complicated description that will lead this article away from being a primer, I shall omit it here.
[4] This document standard is called TRON Application Data-bus (TAD). Specifically, it is a standard format for when one does such things as save data or transmit data.
A document appears as a square box [5] on the screen as in Fig. 1. As for document types, there is a writing pad for writing text, a drawing pad for drawing pictures, a calculation pad, etc. Inside a document, locations (virtual objects)--where in other words links have been pasted--that hold some relationship with other documents are enclosed with squares. Because many links are generally pasted to a word or phrase, almost all the links look like oblong tanzaku (Fig. 2). When you want to view or rewrite the concerned documents, all you have to do is poke at the links with the electronic pen and the computer system will automatically retrieve them and display them for you on the screen in the form of a documents (Fig. 3) [6]. Moreover, if you would like to bring to the fore something that is in the back due to the documents on the screen, all you have to do is lightly click that document's window [7].
[5] This is called a "window" in technical terminology.
[6] A document data system constructed solely by means of documents and links in this manner is called "hypertext." The word hypertext is a term that Ted Nelson, who himself is also the proponent of a hypertext system called Xanadu, began to use. Incidentally, it is said that Ted Nelson now seems to have taken a liking to Hokkaido and has settled there. The basic concept of hypertext was outlined in a system called Memex in an essay [in The Atlantic Monthly] in 1945 titled "As We May Think" by Vannevar Bush. The person who developed the world's first hypertext system is Douglas Englebart, who called it an "Augmented Knowledge Workshop." BTRON is the first to have hypertext made the standard filing system of the OS. At present, hypertext is being popularized via the World Wide Web. This fact shows that there was foresight in the planning of BTRON, which adopted hypertext as a computer data management model.
[7] In Windows and Macintosh, outside of the window, there exists a hidden wall called the "application." For example, let's assume there is a person drawing a picture with a graphics editor on Macintosh. At this point, if the person wishes to bring to the fore a window in the background and then clicks on that window, the graphics editor window goes back to the very rear, and in its place even windows that have not been clicked come to the front. Macintosh, in addition to the "wall" of the window, has the "wall" of the application; by not recognizing that, it is impossible to understand Macintosh window operations. As for BTRON, those types of walls do not exist. All windows are handled in the same manner as one page of a document.
In order to create a new document, you merely peel a page off from the "Tablet Box." This is an operation in which you cut a sheet off from the report pad sheaf and take it out of the Tablet Box. In order to do this, you first open the "Tablet Box," grasp a sheet of writing paper within, place it in together with the other documents, and then attach a name to it (Fig. 4). By doing that, you create a page of writing paper to which a link is pasted from the document that has been placed in with the other documents. Then when you want to write something on this writing paper, all you have to do is double click [8] this link.
[8] Clicking two times in succession within a short period of time in the same location.
There are several types of these documents; the procedures [9] for doing such things for viewing and rewriting these documents are basically all common in BTRON. They are all unified by a method [10] in which when carrying out some process on a BTRON document, you first select the portion of the document that will become the object of the processing, and next you indicate the contents of the processing that is to be carried out on the selected portion (Fig. 5). The operation for selecting the object of the processing and the operation for indicating the contents of the processing are the same for whatever the document. Accordingly, while following the links and referring to the contents of the documents, one is not conscious of the differences in the types of documents; one feels as if one is turning over various types of documents on a desk, and thus working in a very simple manner is possible.
[9] This is called the human interface.
[10] This is called "select and operate."
[11] As one would expect, when one intends to modify or edit a document, it cannot be helped that there are differences in the methods for doing this for each document. The methods for writing text naturally differ from the methods for drawing a picture.
What we call "simplicity" in BTRON is based on the unity of operation methods, the provision of a simple model to the computer, and the standardization of data formats. Unifying things and removing unnecessary differences in this manner is an important factor in constructing a simple environment, but on the other hand we also need in a computer the seemingly contradictory feature of an environment that possesses diversity. Certainly, on the computer side not diversity, but being as unified as possible is more desirable; however, humans are more likely to respect the diversity and not the unity.
For example, we should not "unify" in a way that computers for Japanese people have to be handled in English. Rather we should provide diversity so that computers can be handled in Japanese by Japanese people, and for the people of each country, in the mother language of those people. Moreover, if someone is physically disabled, we have to make possible methods of use matched to the characteristics of the disability. For that purpose, the operation methods of a computer must have a wide variety of options both to make it possible for the user to select the optimum from among those options, and to make it possible to operate all documents in a unified manner using those options. In BTRON we provide a multilingual environment and EnableWare as options for this user diversity.
EnableWare (Figs. 6 and 7) is called such from the fact that it is technology that makes possible utilization by (in other words, "enables") users for whom normal computer utilization is impossible due to bodily circumstances (the "disabled"). To us, BTRON EnableWare goes beyond the category of mere support technology for the disabled. We generalize it as options for the diversity of users, and hence we consider EnableWare "technology that makes up for the mismatch between bodily characteristics and the environment in which a computer is utilized." For example, voice output technology based on a voice synthesis device employed when the visually disabled use a computer is also technology for when we use a computer in an environment where the screen cannot be seen due to darkness. Likewise, technology for users whose hands quiver spasmodically due to a physical disability is also technology for using a computer in an environment in a cold region where warm gloves are worn outdoors and the hands are shivering due to the cold. Due to this kind of thinking, we have not dealt with EnableWare as add-on functions in BTRON, rather from the beginning we have incorporated them to the utmost as standard functions.
In addition, multilingual processing functions (Figs. 8 and 9) are also important in presenting an environment with diversity to users. In recent years, it is a matter of course to provide information in multiple languages when providing information in multiracial countries, and providing information in multiple languages in environments such as aircraft on international routes is also something that is a matter of course. Accordingly, it is extremely natural to demand the diversity in which we can freely mix and use several languages in computers, which are the most anticipated information medium of the future. In BTRON, this is called the TAD multilingual environment. It makes possible the handling of almost limitless types of characters through functions that switch the language environment in a document. Multilingual systemizations that restrict the world's characters to sixty some thousand in the manner of Unicode cannot possibly be said to sufficiently satisfy the demands of the diversity necessary in future societies. Even the kanji that we Japanese use cannot completely fit within this number. Since there is a detailed article in this issue of TRONWARE (see "Welcome to the Multilingual Environment" on p. 37) concerning these BTRON multilingual processing functions, I would like the reader to refer there.
There are rewards in having continued independent technological development with ceaseless effort over a long period in excess of 10 years. Outside of the things mentioned above, there are many fruits of BTRON. Here I would like to mention three things: (1) the TRON keyboard, (2) Internet and BTRON, and (3) BTRON in PDAs.
Because the keyboard is the input device that is touched with the highest frequency when one uses a computer, it is important that the keyboard be easy to use. When designing a keyboard, one must consider two points: the physical arrangement, which is how the keys will be physically laid out and in what shape; and the logical arrangement, which is what character or control instruction to assign to the various keys. Actually, both in the beginning when the TRON keyboard was designed and also at present, almost all keyboards are ones that have been bound by mere compatibility with the past for both their physical and logical arrangements. Their physical shape lacks any scientific analysis of the shape and movement of the hands, and their keys are just simply arranged vertically and horizontally in columns and rows. Moreover, they continue to use the QWERTY arrangement unchanged--the key arrangement of traditional typewriters--for their logical arrangement also [12].
[12] The QWERTY arrangement is something that was originally designed in the age of hammer-style typewriters so that one could not strike the keys quickly in order that the hammers not pile upon each other and break. Since we are now in the age of electronic keyboards, there is absolutely no rationality in using a key arrangement that especially makes quick typing impossible.
In our present age of advanced ergonomic technology, the physical arrangement and so on should be designed by measuring the shape of the hands and their movements, and the logical arrangement also should be arranged in positions close to the home positions of the hands to the extent a character has a high frequency of use. Keys that are used with the greatest frequency should be be arrayed as much as possible in the positions of the strong fingers. The TRON keyboard was designed by faithfully following this way of thinking. As is shown in Fig. 10, in terms of design also, it is something out of the near future [13].
[13] The TRON keyboard itself is something that was designed over 10 years ago. Afterward, in what seems to be follow-ons to this, Macintosh and Microsoft ergonomic keyboards appeared, but in the end they seem to have been unable to completely abandon compatibility with the past and thus ended up as half-baked designs.
The degree to which the Internet has spread over the past several years has been very amazing. Among experts, there are people with the cynical view that this is simply nothing more than a temporary boom that has swept up novices, and that the enthusiasm will immediately cool down in two or three years. However, we do not think this will be so. Rather we think that when the enthusiasm cools down, the Internet will perhaps have established itself as an indispensable, important infrastructure in society. Thus we have also been independently developing network applications based on TCP/IP protocols, such as an electronic mail system, a World Wide Web browser, and so on. In particular, as the World Wide Web is a hypertext-type information system, we are constructing a system model to enable the BTRON real object/virtual object system to seamlessly conform to this.
In the BTRON subproject, getting involved with the Internet is not merely developing these types of applications on BTRON. From the beginning, the basic way of thinking in the TRON Project has been to estimate in advance the problems that will probably occur in the computer society of the near future, and to involve ourselves in the present with what has to be done beforehand in order to solve these problems. Accordingly, we are identifying what problems will occur when the Internet in its present form has become an indispensable infrastructure in society in the future, and we intend to go on to solve them. One of those important problems is that of multiple languages, which I mentioned above. Due to the progress of the Internet, there will surely come an age when we will be able to freely pull out multilingual data onto a computer. With present non-multilingual systems, the situation is that characters are frequently corrupted, tasks such as switching fonts frequently become necessary, and even looking at a page in which multiple languages are mixed is impossible. Even if we say that this can be constructed on, for example, a multilingual system based on Unicode, Chinese, Japanese, and Korean characters cannot be distinguished outside of font switching, and, moreover, many of the Chinese characters used in these languages cannot be displayed. Accordingly, using the TAD multilingual environment as a base, we are proceeding with research to construct on BTRON a framework that will make it possible to freely exchange multilingual data if TRON code is used, and on top of that we intend to provide a World Wide Web browser that will make it possible to create and reference Web pages in which multiple languages have been freely mixed.
We often see comments such as Japan has contributed almost absolutely nothing to the world in the field of personal computing. The only thing Japan has contributed, it is said, is in making personal computers compact, as can be seen in notebook computers [14].
[14] A paper that reviewed the computer field in Japan was recently carried in IEEE Micro, the journal of the Institute of Electrical and Electronics Engineers of the U.S. There also, Japan's technology for making personal computers compact was highly appraised.
The ultimate example of this is the personal digital assistant (PDA). At present, personal computers have become smaller and PDAs have become highly functional to the extent that the difference between personal computers and PDAs has become blurred.
In the BTRON subproject also, we have had our eye on the PDA from before. Finally, last year we developed a PDA called BrainPad TiPO (Fig. 11) onto which BTRON was loaded. As far as I know, this seems to be the world's first commercial product in this PDA class onto which has been loaded a complete multitask, multiwindow, general purpose operating system that is the same as a desktop operating system. TiPO is loaded with TCP/IP, and it can send and receive standard e-mail over the Internet using POP3 and SMTP. It is also fully equipped even for World Wide Web browsing and file transfer via FTP. The user interface also has been slightly improved from the original BTRON interface so as to adapt it to operation solely with a pen on a small screen, making it easy to use. Since the details are described in depth in an article in this issue ("Accessing the Internet with TiPO," on p. 18), I would like you to refer to that.
In BTRON, we have been proposing from over 10 years ago ideas such as hypertext technology, multilingual processing, EnableWare, the PDA, the ergonomic keyboard, and multimedia, which recently seem likely to become the core in the field of personal computing. The author believes that this design ideology has in no way become faded even today after the passage of 10 years.
The above BTRON primer also contained illustrated explanations on the basic operations of the BTRON-specification operating system. In order to ease downloading of this page, we have placed them on a separate page. Those who would like to view those explanations should click here.
The above BTRON primer appeared on pages 11-17 in Vol. 45 of TRONWARE . It was translated and loaded onto this web page with the permission of Personal Media Corporation.
Copyright © 1997 Personal Media Corporation
Copyright © 1997 Sakamura Laboratory, University Museum, University of Tokyo