Summary

This article consists of three main parts, all related to the SuperBook project. The first part explains some cognitive issues that are relevant for the SuperBook project, like information retrieval, illustration presentation (pop-up or inline?). These were used as design guides.

The second part is about the SuperBook project and its evaluation. They talk about flaws, like too low program performance, and their solutions, e.g. the reimplementation in C.

The third and last part consists of a study of the CORE (Chemistry On Line Retrieval Experiment) project, where print, a article retrieval system (Pixlook) and SuperBook were compared.

In short, the product was quite effective. The problems in the early versions showed that producing an effective reader is a difficult task, and that constant evaluation is necessary in order to succeed.

Best Things

Not only the product was described, but also some aspects why they decided for certain features.

Not only the evaluation of the final version was shown, but also how it progressed. This clearly showed the advantage of evaluation.

Improvements

Although they talked about certain aspects of the design, the description is in no way comprehensive. For example, the choice of the programming language proofed to be a mistake. But why they dicided for the language in the first place is not mentioned.

No project efforts and costs were mentioned, but these would be very useful for getting a feeling for these types of projects. They were probably left out because the benefits of the products were not quantifiable.

Question to Authors

What's the status of the SuperBook project? Why have I never heard about it?

Article Questions

1. The SuperBook project was carefully designed. Was it effective from the very beginning? What do you conclude from your answer?

2. Name a program you know that closely resembles the SuperBook. What do they have in common, where do they differ?

My Opinion

I am not very happy to work on articles of this length/time ratio for class.