(A book review by Robert Stewart)
I received a manuscript copy of this book several weeks ago and since then I have had about a dozen occasions to make reference to it. My queries have involved both the formatting of simple markup tags, or elements, and the much more complicated mechanics of server caching. For the most part, my questions were answered quite easily.
Even in those cases in which The HTML Sourcebook did not have the answers, it at least helped me formulate my question and know where to ask it. I wish we had had this book when we started The Virtual Mirror. (www.vmirror.com) If we had, we may not have needed to go back and make so many modifications to our pages.
The first chapter of the book is geared toward html novices, and consists of a series of lessons and examples ranging from the basic document and text formatting tags through linking URLs and a brief introduction to html forms. By the time they have finished this chapter, novices will have acquired the skills to be able to create Web pages of their own. At the end of this, and every subsequent chapter is a list of references to various resources including Web and ftp sites and Usenet newsgroups.
The second chapter is a reference list containing definitions and examples for each element used in HTML 2.0, as well as a number of the HTML+ elements. This chapter makes it easy to look up an element and see the proper format. The inclusion of the more arcane elements, such as <META HTTP-EQUIV="Last-Modified" CONTENT ="23-Sep-94 18:28:33 GMT">, make it useful to even the most seasoned html author. Chapter three describes the proper formatting of URLs referenced in html documents. This covers partial URLs as well as those for other protocols, such as ftp, telnet and WAIS.
CGI provides a way for users to access indexes, send mail or add a posting to a list like our Garden Exchange. This is accomplished through the use of forms that reference applications on the server machine. In this chapter the reader will find descriptions of all the methods, headers, responses and environment variables involved. The one aspect of client-server communication not fully covered here is user authentication. This involves the use of user names and passwords to restrict access to particular files or directories and is becoming increasingly popular on the Web.
Chapter five, "HTML and CGI Tools" covers a wide variety of topics, such as image formats, "server side includes" and cgi utilities. This is where you'll be able to find information on creating transparent backgrounds for images and where to locate cgi scripts, or programs, that others have developed and made available to the public.
The last chapter consists of first-person narratives from the creators of five Web sites. These real world accounts will provide newbie Web publishers a good idea of what sorts of problems they will encounter. This is followed by appendices on character codes, MIME and Archie. Also included is a brief glossary of html and http terms. The glossary is not as extensive as I would have liked, but it does seem to include all the commonly used terms.
For those most seasoned Webmasters who keep abreast of current developments online, any book will seem hopelessly out of date. By the time it has been written, edited, printed and distributed, several months will have passed, and when you're on the Web, several months is a long time. But for beginners and those eager to learn more, as well as those still trying to understand the more advanced concepts involved, this book will be a great time saver. Buying the Book, a Plug: Readme.Doc Discount Computer Books (www.readmedotdoc.com) is a great resource. The HTML Sourcebook, as well as hundreds of other computer books, can be purchased there at discounts of 20-25%.
It used to puzzle me how some people could graduate from High School without learning how to read. Then one day it occured to me that I have lived in Silicon Valley my entire life, work for a local computer company, get free subscriptions to computer magazines, and even my paycheck is generated by computers. Yet with all this wealth of knowledge around me, I had never touched a computer, and didn't read the free magazines because I didn't understand them.
So one day, I decided to fix this situation and put an end to my double life of shame and lies. I went to my local computer mega-store, the kind that advertises their prices as the "lowest in town".
Shyly, I told one of the salesman that I wanted to buy a computer. He began to list my choices. Did I want a Pentium or a 486; a Local Bus ISA, EISA, PCI, or a combination of everything; a Full Tower, Mid Tower, Bent Tower, Designer Tower...
It was probably the blank look on my face that caused him to pause. He might as well have been speaking another language. I looked at him and said that all I wanted was a computer. He smiled, I'm not sure if it was for courtesy or for sarcasm. Then he proceeded to give me a quote for a computer "that would meet all of my needs today and tomorrow." I was happy. After all, I didn't know what my needs were, but the salesman obviously did. Total cost: $8,472.54 plus tax. He smiled again. I realized he was smiling at the huge commission he was about to make... or so he thought. I took the quote with me and left, never to be seen again.
Sunday came, and my wife and kids wanted to go out. I have been promising them for years that we would go out. So, it seemed that the time had come to dig deeply into my pocket and take the entire family out for a day full of fun. We went to the flea-market. It is amazing the kind of things one can find at the flea-market. I bought two pairs of shoes for a dollar a piece. I was having so much fun until it started to rain. While we were walking towards the exit, my keen eyes spotted it... A computer!
Quickly, I moved towards the vendor and asked the price. "One hundred dollars," he said. I nodded thoughtfully and browsed through his other merchandise for a few minutes. Then, I pointed out that the computer was getting wet and he would have to offer me a better deal. He cut the price to $75. I could see water starting to drip into the monitor, so we made a deal for $50.00, and I took my computer home.
Back at home, I could see the excitement in the face of my children. "Daddy, daddy, turn it on!" little Jeannie said. Of course, I am an intelligent man. I explained that we had to wait until it was dry. The next day I gathered the whole family around to share our first steps into cyberspace. "No Dad, in the back," little Johnny told me referring at the switch in the back of the computer. I turned the computer on and we all stood back.
1kb OK.. 2kb OK... 3kb OK... This thing was amazing! We looked in awe at the complexity of the new world we were entering. I was so proud to be guiding my family through this new era in our life. 33kb OK... 34kb OK... 35kb OK... "Dad, what's kb?" Johnny asked. I had to think fast. Confidently, I explained that computers are used for complex mathematical calculations. "So the k means kilo, which is 1000," I replied.
"But Dad, what's the b?" little Jeannie asked, not letting me off the hook. My mind was racing. Complex mathematical calculations... It probably has something to do with that. "The b," I said, "stands for billions. Yeah, that's right. Billions. The computer is counting kilobillions"
I had to nearly close my wife's mouth, as she couldn't believe how fast the computer was counting. 254kb OK... 255kb OK... 256kb OK... After exactly 17 minutes, the computer stopped counting. We were astounded to see how large of a number this modern tool had counted in only 17 minutes. Then the computer started to make all sorts of noises and a light came on in the front, near some type of slot. As we looked at the screen, a fascinating thing happened. The monitor displayed the phrase "Non system disk" in plain English!
For much of the evening we stared at that message on the monitor, until we realized that nothing else was going to happen. Exhausted from our adventures, I pulled the plug and retired for the night. Before drifting off to sleep, I promised myself I would enroll in one of those quick-paced programming courses. I was going to learn to program my new computer to do more. Something simple to start with... assembler perhaps.
The next evening I couldn't wait to invite my neighbor, Sam, over so I could flaunt my new computer. When the doorbell finally rang and I opened the door, Sam pushed past me without even saying hello. Looking at the powerful machine, strategically placed in front of the door, he asked if I was running version one of the operating system. Clearly, he was dying of envy inside. I nodded, not knowing exactly what he asked me, but certain that a powerful multimedia machine such as mine would be running nothing but number one. That's me. Never number two. Now, if I only knew what multimedia was...
Pages 12 and 13 were ads and Internet setup info for WCO, the ISP.
Contributing Editors: Robert Holland and
Customer Service: Mel Enberg and Karry Walker
Graphics: Steve Kong
Hardware: Fred Townsend
Operating Systems: Randy Just
Operations: Chris Ward
Proofreaders: David Hayr and David Stafford
Public Relations: Steve Pomerantz
SJ Office Manager: Joseph Magdalena
Systems Integration: Paul Theodoropoulos
Technical Staff: Laurie Grey and Scott Shultis
City Racks, Bill Clark, Robert Escamilla, Lara Michelle, Mark Murphy, Pete Nelson, Laurie Newell, Rochelle Skwarla, Sean Stryker, Tiger Team, TLC, WHT.
Electronic PrePress:Fricke-Parks Press
Printed at: Fricke-Parks Press (510) 793-6543
With the Web expanding faster than Rush Limbaugh's waistline, it seems apparent that there is going to be an increase in the demand for software capable of producing html-encoded documents. Until recently, this software has come only from small companies and freeware/shareware authors. But now that Microsoft has entered the market, things may be heating up.
I selected several html editors for MS-Windows to compare. If you're looking for editors for other environments, check out Yahoo's index of html editors (www.yahoo.com/Computers_and_Internet/Software/Internet/World_Wide_Web/HTML_Editors).
I've skipped the packages of macro libraries for word processors and intentionally left out HTML HyerEdit. This application does not appear to have been updated in over a year, and is missing some important components. However, it does have an interesting way of handling libraries of URLs and you'd do well to check out the stomping grounds of HTML HyperEdit's authors for more information (www.curtin.edu.au and searching for html).
The html editors discussed here can be broken into two groups: those that are basically text editors with some additional functionality, and those that attempt to hide the mechanics of html from users with a wysiwyg interface. The former are almost all written in Visual Basic, and are usually lean applications that can run alongside a number of other programs simultaneously. The latter are large applications that will consume much of a normal-sized system's resources. Among those taking the text editor approach are HTML Assistant, HTMLed, and HTML Writer. Taking the wysiwyg route are HoT-MetaL and Microsoft's new MS-Word add-on, Internet Assistant.
The real functionality of HTML Assistant comes from its toolbar. It contains all the most frequently used tags and allows you to add new tags to a pull-down menu. A drawback here is the lack of ability to add compound tags. There appears to be a couple of other peculiarities in the way HTML Assistant handles changes to this pull-down menu and its configuration in general, that make the addition of new tags problematic. The author expects to have these problems cleared up in the next release. File size is limited to 32 kilobytes, but few Web pages exceed that size. The freeware version includes a help file and is available at HTML Assistant's Web site (www.brooknorth.com).
HTML Assistant Pro, currently version 1.5, has all the features of the freeware version and a number of additional capabilities, such as URL management and file filtering. It also shares some of the deficiencies. The problem with saving new tags remains and the solution to the 32K file size limit is to break the file into multiple parts.
HTML Assistant Pro's filtering functions include the ability to strip out all html tags and convert files to either DOS or Unix text. Especially useful is the ability to extract all the URLs along with the text to which they are linked. This makes it simple to create libraries of URLs that can later be incorporated into other documents. HTML Assistant Pro is available for $99.95 from Brooklyn North Software Works. (www.brooknorth.com).
The greatest attribute of HTMLed is its floating toolbars. The principal toolbar across the top of the window offers only a limited number of functions, but there are three floating toolbars as well. The first supplements the functions of the main toolbar and the second provides the tags for the extended character set. The third is user-configurable and can include both simple and compound tags. This is an important feature since it's unlikely the authors of these editors will be able to keep up with the new html tags as quickly as they are implemented.
The latest version of HTMLed available, version 1.2d, seems to be incompatible with a wide variety of other applications. The program sometimes crashed immediately when these other applications were running, often taking down my mouse driver with it. An earlier version never had this problem and I suspect this is caused by a simple bug introduced in the last build, and will probably be cleared up shortly. HTMLed has the same 32 K file size limit as HTML Assistant. Minimal help is available in a short text file. HTMLed is shareware and registration costs $39.
Although HTML Writer was also written in Visual Basic, it has somehow beaten the 32 K file limit. HTML Writer uses the menu bar as its primary method of inserting tags. There are also a number of dialog boxes to make it easier to use the more complicated tags, such as those used in forms. I found its biggest deficiencies to be the limited toolbar and the lack of an option to add new tags. There is a help file available. HTML Writer is authored by Kris Nosack and version 0.9 beta 4a is available as "donationware" (no payment is required, but a $10 donation is suggested). Visit the HTML Writer site for more information (www.public.asu.edu/~bottger).
HoTMetal is a stickler for rules. If you have a page that it feels is incorrectly coded, you won't be able to open it. I found this was often true even if I had turned rule checking off. The commercial version will attempt to "filter" out tags it doesn't recognize so you can open these files, but this didn't always work. HoTMetaL will not allow you to use the various extensions to html implemented by Netscape and others.
One of HoTMetaL's biggest sins is that it's a huge memory hog. The literature says you will need 6 megabytes of RAM, and even with that much you might have trouble running other applications at the same time.
But HoTMetaL is not without its good points. The commercial version can set
up html tables, which none of the other packages I explored can do. This will
be an important feature in the future. It also comes with both a spell checker
and a thesaurus. Of course, the word "Internet" was flagged as misspelled,
but I wasn't going to even mention that.
The freeware version is available at (ftp.NCSA.uiuc.edu/Web/html)
The commercial version is available for $195 directly from SoftQuad.
As an authoring tool, Internet Assistant offers a somewhat revised MS-Word interface. The menu and toolbar will look familiar, but now have the ability to create html tags. The idea here is to make the creation of documents as seamless as possible. You create your document in Word, adding graphics and different font sizes in a wysiwyg interface and save it as either an html or an MS-Word document. While this seems like a great idea, the implementation is just not quite there yet. Many of the most popular tags are not available, and other buttons that are available shouldn't be. For instance, the button that allows you to indent a paragraph is displayed. If you use it, the text on your page will appear indented, but no tags are saved in the html file to implement this.
The most frustrating thing about Internet Assistant is that there is no way to view the page as text and modify the tags directly. Using Internet Assistant as a Web browser is a little like taking an eighteen-wheeler around the corner for a quart of milk. It took longer to load our home page off the hard disk with Internet Assistant than it took Netscape to load it from our site a thousand miles away. And when it did load, several codes were misinterpreted.
While not there yet, Internet Assistant, and applications like it, hold the best hope for a wysiwyg html authoring environment. It almost pains me to say it, but Microsoft is on the right track here. It really won't take much to bring the authoring capabilities of Internet Assistant up to snuff. Its use as browser, however, may have to wait until we're all running machines with 64 megs of RAM. Internet Assistant is available free from Microsoft's Web site (www.microsoft.com).
Robert Stewart publishes The Virtual Mirror (www.vmirror.com)
The bottom of page 14 had an ad for California Systems
The live video shows the dance floor and the bandstand. The image switches among several cameras mounted from the ceiling. You can go out on the floor and find yourself on birds-eye video. The band takes a break, canned music plays and the video changes. Now they play a compilation of computer-animated video short subjects. You may have seen several of these on public television or at tradeshows.
Great strides in horsepower have made computer graphics tools commonplace. Today's technicians can knock out a wire-frame representation of a common object, give it movement, slap on some surface material, and add light, shadow, and reflection. Creating and animating three-dimensional graphics is no simple computer task. It has taken many years to work out the algorithms for color, movement, shading and texture. The process burns up a lot of CPU time, which until recently was too expensive for the desktop, at least for such proletarian use as entertainment.
Today's computer animation pioneers delve into imagination, creating colorful 3-D worlds chock full of bright colors and animated common objects. These are worlds where plants spit seeds across the universe, birds fall in love with fish, and dogs dance with men and birds.
With all this onscreen glitz, why are today's animations so boring? Being strictly computer-generated animations, there are no human actors in these films. The most interesting image to any earth-bound audience is the human face, which computer animators have not mastered. Humans in film bring with them character, dialog, and emotion. Even the best computer-modeled faces look mechanical and cold. Parts may be parts, but they should have some connection to the body. I suppose animating necks and ankles, is as difficult as drawing hands is for art students. Nevertheless, these missing connections distract your eye from the intended subject.
Computer animators shouldn't go it alone. Just because you have the skill and tools to animate a chrome-surfaced mannequin does not mean you should. Computer animation is a tool to tell a story; no matter which tool you use, you must have a compelling story to tell. Judging from this crop of animations, we are in the gee-whiz stage, when animations are created to establish bragging rights.
There's a lack of filmic discipline among early computer animators, especially in using the camera's point of view (POV). Filmmakers understand very well the audience must be able to have a visual foothold from which to interpret the image, and placement of the camera is critical. Overuse of zoom, pan and dolly shots tire the viewer, becoming visual nonsequiturs.
Computer animators love the flying camera. The camera flies up, down, around and through scenes continually, moving at random as if we are house-flies. Animators have their reasons: first, the computer makes the housefly POV easy to implement. Second, moving the camera makes up for lack of interesting action in the scene. Finally, animators want you to see their handiwork from many angles. The moving camera is a flaw common to neophyte filmmakers. As animators gain maturity and discipline, I hope they learn to deliver a POV that makes sense to the audience.
Experienced filmmakers have stepped in to take charge of computer animations created for profit. For example, the children's TV show Reboot, which is wholly computer animated, is produced and directed by established filmmakers. Although Reboot shares flaws common to the technology, it does not suffer from housefly POV. The camera is mounted in a logical, easy-to-understand manner. You can see similar differences between professional and amateur computer animations. If you've seen Shell's dancing gas pumps commercial, notice the placement of the camera. It stays at eye level, as if you are standing in the scene. The POV draws the viewer into the scene.
Computer-animated worlds look too perfect there's no dust! Examine a wood surface in real life. On flat surfaces you can easily see flaws (knots), dirt or damage (scratches). Look at a carved object and spot the craftsman's flaws. You won't find such imperfections in computer animations. I don't know the history of the algorithm for generating a wood-grain surface texture on 3-D solids, but it needs work. The wood grain you see in animation appears flawless, without texture. Animations of objects carved in wood appear symmetrical and plastic. Rare is the computer animation that doesn't feature chrome objects. Items mirrored on chrome showcases the animator's attention to detail, with one annoying distraction. There are no fingerprints, smudges or rust on the perfect chrome surfaces.
Numbed from watching a dozen or so short flicks, I go back to the dance floor. A young woman nearby has perfected her moves with a lot of mirror time. She's doing her Solid-Gold dance as best she can on the crowded floor. You know the routine - big pout, run hands through hair, shimmy and shout, with plenty of hip action. She's fun to watch. Surrounded by imperfect, live action, the animations play on unnoticed. I'm no longer bored - the story playing out in front of me has depth and emotion more than the sterile worlds from the animators' fertile imaginations.
If you want to see the future of culture and consciousness, look to the edges of human experience. Find the cracks where the boundaries of experience are extended. Cultural change begins like a crack in the wall of our ordered and highly structured existence. Sometimes the change is successful and survives long enough to generate a viable foundation for community. Sometimes a crack in our social existence grows into a force strong enough to drag all of reality through itself and on into an altogether different paradigm.
For over fifteen years, the Internet, and the larger world of cyberspace itself (the totality of non-spatial and non-temporal electronic culture), have existed on the edge of dominant culture. Until as recently as last year, the Internet has remained invisible and beyond both the experience and scrutiny of the majority. Today, the Internet has permanently entrenched itself within the landscape of alternative culture, and has a steadily growing presence in the larger, more prevalent world of mundane reality. The Internet is, I believe, a cultural phenomenon that is destined to be the seedbed of a new form of consciousness and a new type of self - the uncensored self.
The Internet will have a dramatic effect on the cultures and individuals that interface with it, due to the relationship between geography and consciousness. Both communities and individuals, cultures and psyches, are defined by the physical geography of their community and the physical shape of their bodies. The principal is simple: change the geography of existence and you change the nature of the self.
It is not every day that we see a massive shift in the foundation of our existence. This is simply because the majority of individuals live within a relatively stable and narrowly defined social geography. Excluding for the moment, nomadic societies, it can be said that the further you go back in time, the more physical (and social) mobility decreased. As a result, cultural paradigm shifts were rare. In pre-industrial society, and for the vast bulk of the record of human civilization, the geography of existence was defined by a day's walk from one's village. This radius was the scope of the peasant's life. All else outside this familiar landscape was myth and danger.
Five years away from the dawn of a new millennium, we are faced with nothing less than massive global immigration into cyberspace. One million new electronic citizens are initiated into its mysteries each and every month. By the year 2000, there may well be half a billion homesteaders on the virtual frontier.
Cyberspace immigrants enter into a global, multicultural social context. A virtual, but nonetheless real community where time and space are of little help in mapping presence and relationships. What, then, are the characteristics of the geography of the Internet? Can we map the social landscape of cyberspace? If so, then we will have gained a glimpse into the future state of the human animal - a state I have named the uncensored self.
It is the unique nature of Internet communication that provides us with a point of reference within the landscape of cyberspace. Internet-facilitated communication is an altogether new form of human behavior: uncensored and accessible (at least to the middle class), bi-directional, mass communication. The technology of the Internet has enabled an entirely new technique of existence - mass participation in bi-directional, uncensored, mass communication.
This is critically significant when we realize that community is fundamentally based upon communication, and in cyberspace we have an entirely new form of communication. On this new form of communication a new culture is emerging. This new culture will be the birthing grounds of a new manifestation of the self. Communication, culture, and the self all hang in the same web. Any innovation within one element will have a direct and inevitable effect on the other elements of existence.
Consider that throughout history, mass communication has always been tightly controlled. In pre-industrial society, a crowd was always perceived of as a threat by the elite. In post-industrial society, the ruling elite have maintained almost total control over all vehicles of mass communication. As a result of the rise of cyber-communication, the controlling institutions of society have, for the first time in history, lost control over mass communication.
From this point onward, every one wired to the Internet owns a printing press (and soon enough, a radio and TV station). The means of mass communication has been democratized. The state has lost control over the means of production and distribution of knowledge at the very point in time when we have entered into the digital Information Age.
If you want to see the future, look toward the edge of the Information Age, look into cyberspace. When you have arrived there, listen to the multiplicity of voices. Watch for the appearance of those who become empowered through bypassing the gatekeepers of mass communication. Recall how the Gutenberg Press empowered a few critical thinkers to change the course of nations with their writings.
How much time will pass before we stand witness to cyberspace writers who re-engage the one constant historical force - the power of uncensored communication, the authority of the compelling voice? The new technology of communication, the new geography of consciousness, the new technique of existence combine to form a linchpin on which the whole world is about to turn.
Within the last few years, desktop systems have included windows and icons. At times, it seems that there is too much visual information presented to us, and too many icons. The development and future of "software agents" is one of the most exciting and potentially powerful tools present in the realm of software. Some predict that the concept of the interactive agent may replace the familiar desktop structure.
The interactive agent allows a user to request information or carry out an activity on a host computer. The agent resides in the users computer software. On behalf of the user, the agent travels through cyberspace and contacts relevant databases, gathering up information and reporting the results or the findings back to the user.
Telescript, from General Magic, is designed so that agents cannot cause havoc on machines they interact with. Agents also can vouch for the identity of their masters, using a digital imprint. Access to machines is barred if the agent does not meet the criteria established by the host machine. Any process within the host that controls or stores data is off limits to the agent. Time restrictions are also placed on visiting agents.
In a simple application, an agent can accept the phone number of a client, step out to a directory and find the matching email address for the client, then make the connection or send the message. Users won't have to remember things like: firstname.lastname@example.org.
Traditionally, software agents have shunted information around, sometimes keying a program that already existed on the host computer. General Magic's agents will run as full-fledged programs on any computer that has Telecript installed. Instead of a passive system waiting for a user's mouse-click or macro, an agent anticipates the user's most likely actions and takes unattended action. After some experience with a user, agents learn to perform routine tasks automatically. For less predictable tasks, agents ask the user what they desire before carrying out a function. Agents learn by watching over the shoulder of the user. They detect patterns and consistency in the user's behavior. It takes time for the agent to learn how the user works, so time has to be invested in the agent's training before they are productive.
A collaboration between two or more agents working for different users is plausible and can shorten the learning time required of each participating agent. Agents can also negotiate with each other over which of the other agents are worth learning from. One of the most interesting aspects of the agent-to-agent relationship is that the more agents there are, the more probable the interaction between agents. Agents can work together on a project, and when their initial instructions differ, as they might when reaching a decision point, they part company.
In order to improve their performance and adapt to change, agents must be in a feedback loop. Two methods are used to correlate the actions with any changes in data that is gathered. The complementary methods are perceptual selectivity (PS) and cognitive selectivity (CS). PS restricts the set of sensor data the agent pays attention to at a particular point in time, while CS dictates when the internal structures are updated.
One agent in development by Patricia Maes at MIT's Media Lab is designed to deal with electronic mail. Email overload is an increasing problem for people in academia and business. The agent can learn to sort mail according to priority. For instance, a message from the users' boss might sound an alarm signaling them to read the note immediately, while a routine request for information might be filed away for later viewing.
Email agents could figure out how to answer common requests. In response to a memo calling for figures from the past week, an authorized agent could automatically retrieve the figures from the user's hard disk and forward them to the person who requested them. The agent would log its action in a file that could be checked later to see that the task was carried out.
Another Maes' device helps people schedule appointments. This agent negotiates electronically with similar agents elsewhere, setting up the time and place for a meeting and sending out invitations. The agent system works by building a database of different situations and the user's response to them. The agent looks through the database for actions that most closely match a new scenario, then carries out the task.
As agents work, they employ graphic symbols - drawings of faces whose expressions convey the activity underway, such as thinking, offering a suggestion, and looking pleased if a suggestion is followed. This provides the user with feedback on an agent's activities. This brings to mind what Microsoft's BOB might look like as it does its work for you.
Maes and other researchers are seeking to create agents that will scan newspapers, magazines, books, and electronic databases for material that fits a users' general interests or satisfies specific requests. Such systems would have to be smart enough to recognize that if someone wants to know about film, an article on cinematography is relevant, even if it doesn't mention the word cinematography.
One of Dr. Maes' goals is for agents to program themselves. This will require a massive knowledge and rule-based database. The most ambitious project in this area is now under development by the Microelectronic and Computer Technology Corporation in Austin, Texas. The earliest versions of this database may appear as part of some software products within a year or two. General Magic's approach differs from that of Maes' in that the user must be very specific about the instructions given to the agents.
During the next decade, many industry observers predict that computers using agents will also have speech synthesizers and devices that can make sense of everyday spoken or written language. Mitch Kapor imagines a "...user friendly, multimedia internet over a switched fiber-optic network: practically infinite capacity and infinite reach in a world of cheap video equipment, subtly interactive software and so on."
Telephone companies are showing interest in agent technology. AT&T has developed an agent network called Personalink. Japan's NTT has something similar planned for the Japanese market. Some web links to agent-related topics:
Afterwards, the American team became very discouraged by the loss, and morale sagged. Corporate management decided that the reason for the crushing defeat had to be found. A Corporate Steering Committee was set up to investigate the problem and to recommend appropriate corrective action. Their conclusion:
The problem was that the foreign team had eight people rowing and one person steering; whereas, the American team had one person rowing and eight people steering.
The American Corporate Steering Committee immediately hired a consulting firm to do a study on the management structure. After some time and millions of dollars, the consulting firm concluded that too many people were steering and not enough rowing.
To prevent losing to the foreign team again next year, the team's management structure was totally reorganized to four Steering Managers, three Area Steering Managers, one Staff Steer Manager, and a new performance system for the person rowing the boat to encourage him/her to row harder. The next year the foreign team won by two miles.
Humiliated, the American corporation laid off the rower for poor performance, sold all the paddles, canceled all capital investments for new equipment, halted development of a new canoe, gave an award to the consulting firm, and distributed the money saved as bonuses to the senior executives.
Pages 19-24 of WCO magazine listed thousands of BBSs and web sites.
Pages 19-24 had ads for Grafix (www.grafix.com), Byte Communications (www.bytenet.com), IBBS West,Construction Bid Source, Atlantis BBS/Internet service (www.atlantis-bbs.com), the Sacramento Pacific Exchange, the Searchlight of San Luis Obispo, AutoPC, Sysop News (www.sysop.com), and West Coast Online - the ISP.
End of Issue 27. Go back, or to Issue 28, or to Mark's home page.