Hypermedia: the Multiple Message Development Opportunities on the Macintosh

by Craig Ragland
MacTech Quarterly
Spring 1989 — Page 32

Hypermedia is a new form of software that blends information of different forms with high levels of user control. Unlike most software, the production of hypermedia is limited less by technical expertise or programming skills and more by access to data and design abilities.

This offers significant opportunities to developers of widely different levels of sophistication. The technically naive can focus on assembly of information, while advanced programmers can extend available delivery environments, create new environments, and add the concepts and power of hypermedia to other classes of applications.

Hypermedia is likely to encompass many existing media types and escape the bounds of “computers”. It seems an appropriate method of managing personal, business, and entertainment data. If adopted in near-future homes and businesses, it will have the largest market of any existing form of software.

Hypertext, Hypermedia, and Interactive Multimedia

Ted Nelson, coiner of the term, defines hypertext as “non-sequential writing”. Hypertext is any form of writing or reading which is done non-sequentially or non-linearly. This includes the common practices of writing on index cards, gathering notes on different articles, or taking a telephone message. Reading examples are more obvious; you read non-sequentially when you use a phone book or an encyclopedia or cookbook or a newspaper — you dip into a highly structured document to grab the little chunk of information you need or find of interest. Such paper–based hypertexts force readers to either use lots of steps to locate the desired information, or to follow known, arbitrary conventions (such as alphabetical listings or pagination rules).

Many writers (principally those in the popular press) have tried to narrow the idea of hypertext, restricting it to “a collection of text nodes connected by links”. Nelson rebels at these restrictions, stating that many are “over-simplifying it and over-restricting it. Hypertext is non-sequential writing of any kind with links of any kind”. The restriction of the concept has been fueled by the widespread use of HyperCard, which tends to reinforce the linked-chunk model of hypertext. There are alternative hypertext and hypermedia models which may prove more powerful and popular — more on this later (since this is a sequential document and your author controls your access patterns).

Hypertext, or non-sequential writing, additionally makes a basic requirement for reader interaction not present in sequential text. Readers or users of hyper-texts make a series of decisions to determine which material is accessed. These decisions require greater cognitive overhead than simply reading sequentially down a predetermined path of linear writing. As hypertext creators, we let readers determine what they read next, instead of assuming they will follow the flow of our linear writing. While these decisions provide enhanced control over the reading experience, the added overhead can also be annoying. Adding lots of decision points may have profound impact on the type of written materials appropriate for delivery via hyper-text.

It is unclear, for example, that very many novels with distinct plots and sequential time frames would benefit particularly from free-form, non-sequential access. On the other hand, it would add value to a book review, if the commentary was linked to specific points in the subject work. Non-sequential fictions might also constitute an exciting new art form.

Hypermedia extends the concept of non-sequential writing by adding graphics, sound, animation, video, and any other expressive medium. Apple Computer generally uses the phrase “interactive multimedia” in place of hypermedia. This helps emphasize the importance of human interaction in controlling the flow of multiple forms of data. It also avoids the all-too-easy suggestion (or accusation) that “hypermedia” is mere “hype”.

Just what is sequentiality, and is it inherently good or bad? It could be argued that our experience of life is entirely sequential. There is a constant stream of information impinging on our senses, all of it linear, at least along the time domain. However, as data is encoded in the brain it loses this linearity. Brain science suggests that separate ideas, images, and feelings are stored throughout the brain in widely disparate regions. This data is subsequently accessed simultaneously via different neural networks. Some have emphasized this apparent similarity between non-sequentially organized information and brain architecture/function. However, the particular benefit of structuring information using neural models is obscure at present.

So does hypermedia just present a number of choices at any given instant? No. For most hypermedia designers, it additionally requires higher levels of organization, integration between choices, and some depth of user control. For example, existing commercial television offers a number of choices of stations at any instant. Television, however, offers no higher level of organization (other than channel assignments by FCC or your local cable company). Current-day television offers virtually no integration between the competing programs. Finally, the depth of user control over their individual data stream is limited merely to selecting between alternative hard-wired, linear programs. While remote controls let us ZAP commercials, all we are doing is flipping from what one company wants to sell us to what a different company wants to sell us. How might this most pervasive of media forms be made into hypermedia? — more on this later…

Is sequentiality going to disappear as technology increases our abilities to work non-sequentially? Absolutely not. Many people absorb step-by-step sequential instructions more easily. Such “howto”, cook-book approaches work and they will not disappear. In addition, some forms of writing (such as argumentation or plot-dependent stories) probably benefit from linear progressions. At present, very little evidence exists on how linearity affects the user’s perception, ability to learn or to access information. Furthermore, little is known about the impact of enhanced interaction with information. Do higher levels of interaction improve retention? Hypermedia has leaped into the commercial marketplace too quickly for academic research to respond. However, these issues are now being widely studied.

Future Hypermedia Development

Hypermedia developers come in a multitude of flavors, but there are two basic classes: (1) those who work on creating development systems — System Developer — and (2) those who work within the development systems of others — Hypermedia Developers. As you will see, these categories overlap for extensible software development systems like HyperCard.

At the present time Apple’s HyperCard is the only hypermedia development system with a significant market penetration (1 million+). There are alternative development environments, on the Mac and on other platforms, but none of these offer large installed bases of informed consumers. This means that companies must sell a particular hypermedia delivery system in Pen & ink on scraper board illustration by John Laney addition to selling a particular product or project. At the present time, none of the competing programs offer development environments as rich as HyperCard. That HyperCard includes the embedded language HyperTalk, which gives it a distinct advantage in flexibility. This does not mean that hypermedia development opportunities on the Mac are limited to HyperCard stacks. Indeed, HyperCard is not the final word in Hypermedia and there are many opportunities for application development.

At least three areas of technological and market development are of great importance to hypermedia developers:

• The imminent release of alternative development systems for the Macintosh and other microcomputer systems.
• Multi-user hypermedia systems that allow multiple access and development of a common data set.
• Invasion of the home market with optical media-based interactive technologies.

Alternative Development Systems

Several companies intend to deliver alternative development environments on different microcomputers. On the Macintosh, existing alternatives that might be labelled hypermedia include Macromind’s VideoWorks InteractiveTM, AuthorWare’s Course of ActionTM, OWL International’s GuideTM, and EastGate Systems’ HyperGateTM. In addition, there are persistent rumors of to-be-announced HyperCard-compatible or HyperCard-like PS/2 hypermedia systems. Rumor also has Apple soon to announce an Apple IIGS version of HyperCard. Developers are known to be working on HyperCard-like systems for the Amiga. Some rumors involve extremely large software publishers and hardware manufacturers. Representatives from one company, for example, were anonymously recruiting HyperCard developers for PS-2 development at the recent HyperExpo in Boston.

If you are a system developer, these competing systems are of critical importance. You need to inform your own work, in part by carefully analyzing the work of others. If you are a hypermedia developer, you should carefully attend to alternative systems which might allow an inexpensive expansion of your market. While traditional software developers (what an odd notion for such a young industry!) must focus primarily on the technical aspects of implementing a complex system, hypermedia developers attend more to integrating various forms of data and structural design. Apple would certainly prefer the Macintosh to remain the only viable system for delivering hypermedia data. Hypermedia developers, however, must carefully consider all of their options — while Apple has a great image, it is known to destroy small developers through minor product changes or releases. As data interchange formats develop, the porting of hypermedia data and structures will be trivially easy, and any company not selling hypermedia software for multiple delivery environments will be at a competitive disadvantage.

Multi-User Hypermedia Systems

Hypermedia development systems that support multiple simultaneous users will be very significant. Quite simply, people tend to work, learn, and play in groups instead of as solitary individuals. This is true in virtually all existing markets. In the corporate community, the winning technology improvements will increase the productivity of work groups and departments as a whole, as opposed to just the individual worker. In the education market, it addresses the basic fact that groups of students interact with teachers and each other. Educational experiences which involve other thinking people are richer than solitary study. Individual knowledge workers will also benefit by increased access and ability to use others’ work more directly. This article, for example, would be dramatically improved if all the other researchers and hypermedia developers in our widely distributed community could contribute their comments.

System developers have two alternatives regarding multi-user development systems:

• Extending existing systems.
• Creating new multi-user systems.

The release of HyperCard version 1.2 added read-only, multi-user access to a stack which is mounted in an AppleShare folder. At least two significant extensions to HyperCard add read-write, multiuser capabilities. These are HyperComTM by GAVA, Inc, which does a very nice job of adding AppleTalkTM communications abilities to HyperCard, and Oracle Corporation’s addition of HyperCard XCMDs, which let a developer create hooks into an Oracle database.GAVA has created an AppleTalk Driver which takes care of most of the trivial details of AppleTalk, allowing the developer of a HyperCard-based product to convert it fairly easily into a multi-user system. This is significant for many educational products as well as business-oriented systems. GAVA includes a multi-user address book with HyperCom which illustrates an immediate practical application, allowing more than one user to share the common address data.Oracle is the most commonly used mainframe and minicomputer data base system in the world. Using Oracle Corporation’s new HyperCard XCMDs, companies will be able to add HyperCard-based front ends to their existing data bases. This lets effective interactive designers create highly customized and easy-to-use front ends for different roles within a company. A shipping clerk who has finished processing a particular package might pick up an icon representing that package and move it to an Out Basket. The clerk’s direct manipulation of that graphic representation would report that step to a centralized data base running on a Mac or other machine. (By the way, the same thing could be accomplished for smaller organizations using HyperCom along with HyperCard.)

System developers should also consider designing new hypermedia systems for multiple users. HyperCom-enhanced groupware products using HyperCard face all the traditional design problems of multi-user database systems, including record vs. field locking, network traffic issues, and other very basic multi-user design constraints which are built in to most multi-user products.

Hypermedia developers interested in creating multi-user hypermedia data spaces face numerous interesting problems. Should all users have equal access to all data? Should private data be allowed in the system? Can anyone come along and change “your” data, or are they restricted to linking to “your” data? The degree to which these questions are addressed by the system designers will inversely determine the flexibility and difficulty for hypermedia developers.

GAVA’s HyperCom and the Oracle hooks also exemplify another class of hypermedia-related developer. Are they system developers or hypermedia developers? Both are adding new, system level capabilities to HyperCard. There are several other XCMD-oriented products which fall into this class, by adding new, basic abilities to HyperCard. These include “101 Scripts & Buttons for HyperCardTM” published by Individual Software (which adds several basic user interface extensions, and was created by the author of this article), HyperPress’s “Icon Factory” (which lets developers more easily create ICON resources), Farallon’s MacRecorder (which lets developers add digitized sounds to their stacks), etc.

The Home Market

The home market is often portrayed as a gigantic opportunity for existing software developers. At present, its exploitation has been essentially limited to games, which have been ported from microcomputers to Nintendo and competing game systems. The missing element for hypermedia has been appropriate delivery systems. While Nintendo systems offer some possibilities, they are limited to ROM and RAM-based data. Another home system which should be released in 1989 is called CD-I for Compact Disc Interactive. The CD-I format is supported by Philips, Sony, and others. It defines a standard format for images, sound, and data. Apple has not joined in supporting CD-I, which (along with their future-oriented video tapes) suggests they see a different path for addressing the home market, presumably one with proprietary hardware.

Attending to the home market is important if you are interested in the “big payoff”. The potential market for a really spectacular, microcomputer-based product is miniscule in contrast to one which could be used on any television. The amazing abundance of VHS-formatted video tapes and video stores provides some insight to this market. Home-oriented hypermedia products do not yet exist, but they will. Will the CD-I player be the vehicle which brings interactive information into the home, or will the invasion await a richer, more powerful standard? Nobody knows, but lots of really BIG companies are betting many millions on CD-I and competing formats. Presently, there are CD-I authoring systems available from Sun-Phillips and Matsushita. The work required to create a CD-I product and that to create a large microcomputer-based hypermedia product are quite similar. It would seem prudent for smaller developers to adopt a wait-and-see attitude toward CD-I. It seems clear that eventually a market will be developed for home-based large-scale interactive entertainment/information products. However, small developers cannot create this market, and are better occupied developing viable products for the present market.

The home market should really explode when we’re able to distribute hypermedia projects using broadcast media or cable systems. There is a fundamental problem associated with the bandwidth required to allow thousands or millions of local users to control the flow of data. We can assume (perhaps naively) that these difficulties will be surmounted. Perhaps local storage and control via advanced, in-home entertainment systems will hold the solution — as a user moves into a given category of data, a block of data will be transmitted from a central storage library to their local storage unit.

Macintosh Based Opportunities

For now, both system and hypermedia developers are pretty much limited to microcomputer-based delivery systems, and if we are to build businesses producing hypermedia, we had better make the best of the current situation. The remainder of this article focuses on short-term opportunities for developers on the Macintosh computer.

There are several different ways in which developers can exploit the concepts of hypermedia. These include:

• Adding hypermedia to existing products of other classes.
• Creating new hypermedia development products.
• Extending or creating tools for existing hypermedia devel-opment products.
• Creating information management applications usinghypermedia products.
• Creating custom hypermedia projects for corporate clients.
• Creating information products using hypermediaproducts.

Hypermedia and Existing Applications

As the Macintosh has become more sophisticated, so has the average application. Increasingly, help systems are being perceived as crucial components. This trend will continue, and hypermedia offers an excellent way to add value to existing software which goes beyond simpler models of help. Others have noted the similarities between context-sensitive help and hypertext. Embedding hypermedia-based help into your applications will let novice users access specific, graphical data on steps required for particular operations.

An example of hypertext-like help is found in the popular tax preparation product “MacInTax”, published by SoftView. This package provides pop-up instructions for each of the different lines on the IRS tax forms. This makes it relatively painless to correctly complete many of the various tax forms. While this is a step in the right direction, simply placing instructions into a product is a small step. Watch for applications which provide more — much more. MacInTax could, for example, include the texts of popular tax preparation guides, which provide more understandable instructions and advice which helps users make decisions (though there may well be legal constraints on providing tax advice). While Softview created its own method of popping up help, HyperEngine by Symmetry Software lets developers add limited HyperCard-stack playing abilities to applications.

HyperCard and VideoWorks Interactive are currently in use by several software publishers for guided tours and product demonstrations. This relates to their excellent abilities to simulate other applications, providing users with a simulated experience in the product. For these tasks, Videoworks is probably better when animation is required and HyperCard is probably better when the project is more information intensive. It’s also much cheaper to produce a high quality HyperCard stack than a VideoWorks movie.

The ability to simulate a software package using hypermedia products is also being widely used in the prototypingphase. You can create a functional simulation of almost any product using HyperCard at a fraction of the cost of actually coding it out using C or Pascal. This use of HyperCard has already proven popular in-house at several software publishers. A speaker from Claris Corporation spoke at the June, 1987 HyperExpo about their use of HyperCard for prototyping.

As hypermedia technology improves, we will be able to embed complete, comprehensive training packages with simulated trainers right into the applications. Such systems will include voice-over, how-to examples which take the users by the hand and show them how to solve their problems. Hopefully they will begin to move away from the current command-oriented model of Help, and toward user problem orientations.

Another opportunity appears to exist in binding relevant data sets into applications. At present the basic software model is to provide just the tool and let end users generate or acquire their data separately. As the market for software matures, customers will come to appreciate one-stop shop-ping in their software products. Products that embed useful data, which makes it easier to effectively use the product, will enjoy a competitive advantage. A highly successful product that illustrates this approach using clip art is Broderbund’s Print Shop. Print Shop doesn’t use a hypertext or hypermedia model, but a very attractive graphic package could be created by extending this concept with hypermedia.

Opportunities for embedded data sets include management tools with advice on human interactions, bibliographic tools with selected data sets, construction estimating tools with product information, business plan creators with alternative business strategies, travel planners with travel information, CAD packages with embedded building codes, sales support tools with methods of addressing sales objections, etc. Almost any task requires basic information as well as the software tools for managing the information. When the information is common to all users of these tools, an opportunity is created for applications which include hypermedia-based data sets for the end user.

Creating New Hypermedia Development Products ∆ HyperCard is bundled free with every new Macintosh. This is great for purchasers of new Macs and HyperCard developers, but its effect on existing and prospective hypermedia system developers is chilling. How can you compete with free software from Apple? It is possible, but difficult. The bulk of the existing companies with alternative Macintosh hypermedia development products already had their products on the market before HyperCard’s release or were so far along in the development cycle that to terminate the product was not reasonable. These include OWL International’s Guide and Eastgate System’s HyperGate.

To create a viable alternative, your product will require sufficient distinguishing characteristics that it will attract prospective hypermedia developers. Not only must it be sufficiently different from HyperCard as it is currently implemented, but it must not be made obsolete by the upcoming HyperCard version 2.0 and all the increasingly successful HyperCard add-on products. This would seem a rather scary race for any company to enter.

Another issue for those who would create and market an alternative Macintosh-based hypermedia application is that your product users will be standing alone. HyperCard developers currently enjoy an unprecedented symbiosis and acceleration of the state-of-the-art in HyperCard development. Through learning-by-example, hundreds of different approaches toward any given topic have probably already been explored. Not only are there many great HyperCard stacks in public domain, but there are dozens of outstanding commercial stacks. This level of user-support for a development environment would be hard to replicate on the Macintosh.

All this, unfortunately, is likely to drive would-be hypermedia system developers on the Macintosh to other development platforms. However, an alternative is to consider creating a HyperCard add-on product. This market actually offers many similar development opportunities (see next section).

Extending or Creating Tools for Existing Hypermedia Products

One of the most exciting product categories for HyperCard has been developer tools and extensions to HyperCard. Currently available products let HyperCard developers:

• Run relational data bases.
• Control higher quality animation.
• Add digitized sound.
• Add new visual effects.
• Add new interface capabilities.
• Generate ICON resources.
• Generate CURSor resources.
• Generate MENU resources.
• Use color on properly equipped Mac IIs.
• Create fractal graphics.
• Print more effectively.
• Control PICT resources.
• Play MIDI instruments.
• Dump PostScript.
• Open full-sized Paint Documents.
• Move resources among stacks.
• Communicate across local networks.
• Control serial devices.
• Collect data from laboratory instruments.
• Scan documents on flatbed scanners.
• Grab data from video sources.

The majority of these products are aimed squarely at HyperCard developers. HyperCard developers create HyperCard stacks for personal use, use within their businesses, and for other end users. To the extent that a tool increases the productivity of a user, it will be appreciated. There are several opportunities for basic new capabilities which have not been commercially exploited. Given the focus of this article it is important to note that only a portion of HyperCard developers create hypermedia. The vast majority of public domain stacks follow data base models and are based on structured records, rather than linked non-sequential data.

A critical issue, which a prospective developer of HyperCard extensions and tools must consider, revolves around copyright and licensing. If you provide users with a new capability, will they be able to redistribute your work? If so, on what basis? There are literally hundreds of XCMDs which are freely distributed in the so-called “public domain” or “shareware” markets. The vast majority of these carry copyright notices by their authors. This means that they are NOT public domain and that the author retains the right to control the distribution of the product. If you wish to incorporate their work into your own projects or products, you must get (preferably written) permission from the copyright holder.

It is important that you clearly address this issue if you commercially provide HyperCard tools or extensions that others may want to distribute as part of their own work. It has become fairly common to distinguish between site licenses and commercial distribution licenses. Some products have established flat fee rates for site licenses, and royalties for commercially distributed extensions. In the commercial product “101 Scripts & Buttons for HyperCard” (which was created by the author of this article), the publisher has bundled redistribution rights into the purchase price of $69.95. This is an extremely aggressive posture which is likely to result in widespread use of the included extensions. Since the product includes such basic user-interface extensions as the ability to add menus, pick up and move buttons and fields, selecting screen regions, and resizing standard HyperCard buttons and fields, without leaving browse mode, it enables hypermedia developers to add many Mac-like capabilities to their HyperCard stacks.

There are many other fairly basic capabilities which are missing from HyperCard. When considering possible HyperCard tools or extensions you are forced into playing a guessing game about what Bill Atkinson and the Apple HyperCard development team will add to future versions. After all, there is little reason to expend time, effort, and money developing capabilities which Apple will give away free in a future release. Becoming an Apple-seeded developer is one way to get some advance notice. The best way to accomplish this is by releasing a commercial product which demonstrates you have a clear need-to-know.

As various articles have pointed out, HyperCard does have limitations, and there are many capabilities which would improve it. Some which would truly benefit many developers and end users include:

• Navigational support in the form of graphical browsers.
• Enhanced textual or even graphical searches and sorting.
• Independently manipulable, field-based graphics.
• Real hypertext within text fields.
• A dialog box editor.
• Floating pallets.
• Any number of higher level development tools.

Apple is likely to add some of these features to future HyperCard versions. If you’re able to second-guess them and also successfully market any of these extensions, you’ll make some money.

Creating Information Management Applications

HyperCard is an effective tool for creating some types of information management applications. The principal constraint is the relatively poor performance of HyperTalk for some operations and the lack of some common database features (most significantly the lack of common data base search or sort criteria and the lack of data validation). HyperTalk is particularly slow at multi-record processing. If a particular application requires that frequent totals be computed, then HyperCard would probably be a poor choice. On the other hand, it might well be acceptable to periodically dump data out of HyperCard and load it into a predefined spreadsheet template for multi-record data processing.

The lack of any runtime fees for HyperCard-based applications offers a significant financial advantage. While other data base environments often offer better performance, the run-time fees involved may price the resulting applications out of range for many prospective users. HyperCard is also considerably more flexible than most data base development programs in letting the hypermedia developer control the types and forms of user interaction. While HyperCard developers can create new styles of interactions (such as knobs that turn or toggles that flip or sliding controls that slide), this level of extreme flexibility is rare in other development environments. In addition, the widespread availability of predefined capabilities that can be directly plugged into your application (see section on HyperCard extensions and tools) can make the construction of applications quite inexpensive.
Danny Goodman’s Focal Point is the best known example of a HyperCard-based information management application (actually several different, slightly integrated, applications). Recently, several competing products have been released including: Organizer+ published by Dazzl; Client published by Softworks, and Hyper-Action published by Multi Solutions. These products all extend some or all portions of Focal Point; a new version of Focal Point is due out shortly as well.

HyperCard-based information management applications are somewhat like toys built out of Legos. Just as it is possible to take a Lego toy and modify its basic structure and function, HyperCard-based applications can be extended and modified to meet varying needs. However, to date we have yet to see HyperCard stacks designed explicitly to be modified. Existing applications have followed traditional software models and attempt to deliver finished applications and not user-modifiable capabilities. Just jump into Focal Point or its clones and delete various fields and buttons — it destroys function. If you delete critical components, the loss will ripple throughout the different stacks. This is much less friendly than a Lego Spaceship, which can easily transform into a Lego Submarine.

Some end users would love information applications which are really easy to modify — and this means with-
out any scripting. This would allow a general information managment program to be customized for a particular vertical market. Interface extensions could be used to let users view the function of a particular button or field using a higher level perspective, and then add, delete, or change that object’s functionality. This could be accomplished using an icon-based representation scheme. A couple of interesting products which use such icon-based programming include Odesta’s Helix and Mainstay’s VIP.

There are ample opportunities for vertical market information management products. Knowledgeable developers can tailor HyperCard stacks to meet specific information needs. This makes it possible to produce much less expensive vertical market solutions than previously feasible.

Creating a vertical market “product” can also result from creating a custom “project” for a business client. It is sometimes feasible to arrange a joint venture with an existing business in the vertical market of interest. This can be quite beneficial to both parties, as it should raise the value of your work. In your client’s eyes you are raised above a simple work-for-hire contractor, as your work becomes a part of their existing and future business. Clients typically benefit by receiving more software development for their financial investment, and the potential of a highly valued product. As a developer, you will benefit from having extensive access to an expert and real-world test ground in the vertical market, as well as support for a product’s development.

Creating Custom Hypermedia Projects for Business Clients

Presently, it seems likely that many more developers are employed creating hypermedia projects for particular businesses than are producing general products for a mass market. Project types are as varied as all the other categories in this section. The key advantages to creating a custom project for a particular company are (1) you are assured your project will have a market, (2) a series of successful projects steadily reduces the amount of marketing required to bring in further work, and (3) you can often rely on the income of long-term or large projects.

Creating custom projects as an outside consultant, instead of as an employee, can also be quite lucrative. The basic financial alternatives are to bill at an hourly rate or to bid a price for a total project. Hourly rates for HyperCard developers in the Seattle area ranges from about $10/hour for part-timing moonlighters to about $75/hour for established consultants. Bids are based on the project scope, but range from a few hundred to tens of thousands of dollars. The downside of developing software independently is that you have very little security. The upside is that the income can be significant and you own all of your blue sky.

The best way to maximize your leverage in creating custom work is to develop expertise in a particular vertical market, and then deliver custom hypermedia projects tailored to that market. This lets you more effectively understand and fulfill your client’s particular needs. It also makes it easier to reuse work created for one client in future projects. Focusing on a particular market segment, however, may be a luxury which only comes after you routinely meet the basic costs of doing business.

Creating Information Products Using Hypermedia

Information products are those which focus on the informational content of the product rather than the product’s data management utility. There are relatively few commercial products in this domain, through this is definitely an area of tremendous potential growth. Some existing or soon-to-be released information products include: DTP Advisor by Broderbund, Business Class by Mediagenics, The Manhole by Prolog Software, and Beginners Guide to HyperText by Intellibooks. Other information “products” which have been distributed by Apple Computer include the HyperCard version of The Information Exchange (which is freely supplied to Apple Certified Developers) and the MacWorld Public Information Kiosk stacks.

Information products can also be separated into a different classes: Public Access, Instructional/Educational, Reference, and Entertainment. Each individual project or product has different design requirements and constraints. However, the similarities within each of these classes call for some individual discussion.

Public Access Hypermedia

For public access systems, the chief design constraint is the lack of user experience with a Macintosh or any other computer. Fortunately, HyperCard makes it fairly simple to create easy-to-use systems. In general, public access systems must abandon almost everything Apple has attempted to mandate in their excellent text Human Interface Guidelines: The Apple Desktop Interface. While these guidelines are critical for creating consistent user interfaces for software intended for Mac users, the general public is not Mac-literate. This means that systems which should be immediately usable by naive users, you must abandon any interface item which is not intuitive and immediately recognizable. Some examples of interface components which are common in HyperCard (and other applications), but are difficult for the Mac-illiterate include: modal dialog boxes (users don’t know that they have to deal with the dialog box before continuing), scrolling text fields (users don’t know what they are and how they work), standard check boxes (users don’t know it’s really a toggle switch between two conditions), and radio buttons (users don’t know they are mutually exclusive switches). In HyperCard, it is often difficult for users to distinguish between what is hot and what is not.

All of these issues are best addressed by replacing the Macintosh standard items with highly graphic alternatives. For example, to let the user toggle between two alternative states, use a graphic toggle switch which looks and behaves like a light switch. This can be fairly easily created using alternative ICONs, with appropriate scripting. Examples of these and numerous other functional and visually effective buttons are found in Stack Starter by Robertson Smith and 101 Scripts & Buttons for HyperCard published by Individual Software (and created by the author of this article).

Another constraint for public access systems is the potential for vandalism and theft. Macintoshes are valuable and must be either used in safe environments, or sufficiently protected for use in a public setting. Related to this is the vulnerability of the mouse. Mice are also more difficult for first-time users than alternative pointing devices. Designers of public access systems should consider both touch screens and industrial strength track balls. Touch screens present a different set of constraints for designers — principally that objects must be larger to hit properly, and that fingers obscure an awful lot of the screen.

The opportunities for public access information systems seem quite extensive. There are many information-intensive settings which would benefit significantly from effective hypermedia products. These include many different medical and dental settings, public information desks in hotels and stores, trade show directories, public transportation, banking information, telephone support systems, and in-store directories. This is another area where custom development and product development overlap.

Instructional/Educational Hypermedia

One of the heaviest areas of HyperCard development is in education and training. HyperCard offers an extremely rich environment for development of training materials for almost any subject matter. The third annual CD-ROM conference book. “Interactive Multimedia: Visions of Multimedia for Developers, Educators, & Information Providers” describes several different education projects, using HyperCard as well as other multimedia software products. Computer-based training has come a long way from the simple programmed text models of Skinner and others, though some have noted the similarities of highly interbranched, computer-based courses and the linked node hypertext model.

There is need for products which support developers of HyperCard-based training projects and products. While creating simple, branching card trees is easy using the vanilla version of HyperCard, the creation of more complex systems requires some sophisticated programming. Hypermedia-based training materials should treat users with more respect than forcing them to wade through long paths of inappropriate materials. The specific materials presented on a particular subject can be dependent on one or more previous decisions, instead of being hardwired, as simple CBT programs tend to be. In addition, hypermedia challenges us to develop different approaches toward tracking and monitoring a learner’s progress through material. While traditional approaches would test and quiz, hypermedia approaches might let users grab data along their trips for later synthesis and reporting.

HyperCard-based training software yells out for some form of higher level branch management aids and development tools. This approach is used heavily by Course of Action by AuthorWare, which also offers a significant alternative development environment for this class of applications.

Training products, which provide training materials for particular processes or tasks, are another potential development opportunity. To date, the bulk of training software seems to be oriented around training end users on the use of other commercially successful software packages. One of the leading publishers in this field is Individual Software, which publishes Individual Training for PageMaker, as well as numerous PC-based products. Another interesting product, HyperTutor by Channel Mark, is an interactive training product for HyperTalk. HyperTutor is, however, highly text-oriented and provides little for anyone who has gone beyond the basics of HyperTalk.

There may well be significant potential for HyperCard-based training products. Producing training stacks is an extremely active area within the business community, where the costs of training are increasingly well understood. Just as using Macintoshes instead of PCs saves significant training costs; creating HyperCard-based training materials is more efficient than paper-based or classroom training. The utter effectiveness of using skilled teachers to produce outstanding materials which benefit students over and over again will eventually result in extensive development of hypermedia-based training and educational materials.

Reference

Many users have placed their personal reference materials into HyperCard stacks. Its high-speed Find command and ability to branch according to the idiosycratic wishes of an author make it an excellent environment for storing cross-referenced information. Several large companies have also produced extensive product information stacks for use by in-house sales personnel. In addition, several HyperCard-based CD-ROM reference projects are currently under development by major publishers. One of the more interesting projects is a HyperCard version of the Whole Earth Catalog, to be published by Broderbund. A sample version of this was distributed on the Apple Learning Disc (a demo disc produced by Apple and given away earlier this year at the unveiling of the Apple CD ROM player), and it is marvelous. It makes highly effective use of different graphic styles for the various sections.

One of the more extensive HyperCard-based reference projects in widespread use today is Cameo II, produced by NOAA. This innovative system contains a data base of several thousand hazardous materials, and allows users to develop local geographic information systems. It also helps manage the hazardous material containment process by projecting wind plume area for spills. This aids firefighters in decisions regarding protective clothing, breathing apparatus, evacuation, and neutralization procedures. Several fire departments have equipped emergency response vehicles with on-board Macintosh systems and are able to identify and cope with toxic chemicals right at the site.

There are many kinds of reference materials which could benefit from release in hypermedia formats. However, the tasks of moving the data into digital form and structuring it for effective use can be immense. Such projects often require very large budgets, particularly if the legal rights to the data must be purchased, or if significant reworking of the data is required.

On the other hand, some extremely valuable reference materials may be readily available in digital forms. For instance, extremely time-sensitive reports and papers are sold for several hundred to many thousands of dollars. These include reports on emerging markets, financial information, investment research, feasibility studies, etc. When reference material is that valuable, it can be quite cost-effective to both produce it using an information processing tool and deliver it using an information environment which adds value to the data. Today the vast majority of such reports are produced on computers (using word processors and page layout software) and delivered in paper form. Tomorrow they will all be delivered digitally and probably utilizing hypermedia software. Some innovative companies will begin publishing such materials in HyperCard today.

Entertainment

The single largest class of information consumed by the average member of our culture is within the realm of entertainment. We are inundated by a barrage of media in the form of television, radio, films, and print. While some media is oriented around informing the user, many others are pure escapist entertainment. Hypermedia, with its capability of letting users actively participate, instead of just passively viewing other’s interactions, should prove a very popular alternative for a great many people.

There are numerous commercial Macintosh games and entertainment products which could be easily created using HyperCard. One of the more attractive existing entertainment products is The Manhole, by ProLog Software. This marvelous product is aimed at kids from 3 to 8 years of age and links hundreds of beautifully crafted graphic cards with high levels of care and craft. Kids and adults alike respond to this attractive package with a true sense of wonder and excitement.

A major constraint on delivering HyperCard-based entertainment products is the size of highly graphic or audio-oriented stacks. The Manhole fills five 800K disks, which makes it a product with a pretty high overhead cost. Entertainment products demand extensive use of graphics and sound. Using existing technologies, this requires large data spaces. The high price Apple attached to its CD ROM player ($1,200) has made it a still exotic peripheral, so producing CD ROMs for this market segment remains difficult to justify, though many software companies are exploring this market. The problem with selling CD ROM players is that there are so few interesting CD ROM discs. If you need a particular CD ROM disc, then you will buy a CD ROM player. Very few will buy expensive peripherals for entertainment reasons.

While data space limitations are problematic today, we can anticipate these will be solved by hardware innovations in the fairly short term. The 256 Meg Read/Write optical disc bundled with the NeXT Computer is a nice example of where we are headed. While a $50/disc overhead on the Next distribution media presents difficulties, we can expect hybrid optical disc players which read mass produced discs (like existing CD Audio discs) and also let end users write to user discs of a different format.

While it is nice to hope that hypermedia entertainment products will be of a higher caliber than existing media, this seems rather unrealistic. The beauty of well-designed hypermedia is that users can control their access. However, if the quality of the content is low, a high level of control over access adds little — though it does allow one to more easily avoid vast amounts of uninteresting data.

The Software Designer of a New Type

Hypermedia development is in its infancy. It lets information workers, as opposed to just programmers, assemble interesting software, much of which has significant market potential. When Bill Atkinson spoke at a recent meeting of the Seattle-based Downtown Business Users Group, he asked how many people there were HyperCard developers who were thinking of commercially distributing their stackware. About 200 out of the 800 attendees indicated they had such plans. HyperCard has created a new class of software designers and many will succeed in several different software categories.

Craig Ragland is a principal with INTERACTIVE DESIGN, a Seattle-based Hypermedia development company. In addition to extensive experience developing custom HyperCard stacks for the corporate community, INTERACTIVE DESIGN created the popular HyperCard add-on “101 Scripts and Buttons for HyperCard” published by Individual Software: (800) 331-3313. INTERACTIVE DESIGN seeks creative hypermedia development opportunities and can be reached at (206) 542-9000.