• Both expository and research papers with lots of hypertext or graphics. Graphics on computers, as opposed to that inserted into conventional journals, is inexpensive, and we'd like to see more imaginative use made of it. We'd like to call attention particularly to Java, both applets and applications, a Language in which it is simple to create interactive graphics. Animation also is not difficult in Java, although performance limitations have to be kept in mind. On the main part of our site we intend to publish a series of tutorials on practical computer graphics for mathematicians.

• Algorithms in almost any field, but particularly number theory and algebra. We'd like these to be demonstrably practical, preferably accompanied by programs which can be compiled and run in a wide variety of environments. We'd also like to see algorithms explained interactively or with animation. Since few mathematicians are familiar with many of the sophisticated algorithms well known among computer scientists, we'd like to see good explanations of these with mathematical applications in mind.

• Mathematical tables and collections of data. These should be accompanied by a discussion of how they were generated. If they are complicated or extensive, we'd like to see some discussion of how they can be interpreted by a computer and here, too, accompanied by practical interpretation programs.

• Discussion of issues concerning both mathematics and computers. The biggest issue we face is how to maintain the journal itself - i.e. to maintain readability over a long period of time! All advice is welcome.

• Enhancement or annotation of classic mathematical works as well as recent conventional papers. This could be done either by the original author or someone else. We feel that a lot of mathematics can be remarkably improved by the addition of graphics. Older mathematics often has no copyright attached to it, so that old text can be copied and annotated electronically.

Examples already on the Internet of the sort of thing we have in mind:

• Jim Morey's Java Cup winning version of Euclid's proof of Pythagoras' Theorem
• Several more Java versions of proofs of Pythagoras' Theorem
• David Cervone's exposition of tight polyhedral immersions
• Xah Lee's pages on conic sections

Submissions can in principle be of any length, but for the time being we strongly suggest short articles with a single theme. As time goes on, our authors and our readers will all become more experienced in the sort of exposition we have in mind, but we expect that like all software shops we will face a myriad of passed deadlines and incomplete projects. Short submissions will reduce complexity.

Of course the possibilities cover a huge range. In the long run we will probably narrow our areas of interest, but for the moment we would like to see almost anything fitting into the scheme outlined above. Our lists are not meant to be restrictive, and we shall happily entertain further suggestions.

The examples given above are mostly of an expository nature rather than an account of new work. Expository material is certainly acceptable, and indeed we expect one of the main uses of articles to be for use directly in class rooms. But we hope also that people will submit projects explaining new mathematics, since at the moment it is precisely this category of mathematical publication - concerned with current research, but using electronic formats in an essential way to explain it - which has little opportunity for expression.

All submissions should be as portable as possible, in the sense that they should behave well on a large variety of platforms. Documentation of all kinds must be readable from one of the major browsers. We strongly recommend .pdf instead of .dvi or PostScript for the normal mathematical component, for example, since Adobe's Acrobat reader is freely available and of very good quality. We will accept mathematics symbols in .GIF form, of course, as produced by a program like Latex2HTML.

The amount of material already on the Internet is also huge, and a lot of it of very high quality. Why are we starting something like this? Our aim is to get people to work a bit harder - to think a bit more about what they are doing - than they would if they just post their own stuff. We hope to encourage people to carry their work to a stage of completion more useful to others. The biggest mark of our success would be to spawn the next generation of similar electric journals.

As far as software is concerned, we are not at the moment too concerned about whether it doesn't run on this or that machine, but it must be as clear as possible what the platform limitations are. Precise documentation and specs for software are extremely important. Example should be shown. If your choice of platform is narrow, then the documentation ought to be good enough for someone to implement it on other platforms.

Submissions will be refereed in the usual way as far as mathematical content is concerned. Because of the unusual nature of what we have in mind, it might very well happen that submissions are rejected for technical reasons not related to mathematical content. We can only ask for patience. We ourselves have some limited experience in dealing with technological questions concerning Internet publication, and we will try to offer advice as problems of presentation and style arise. We suggest that you write to us if you have any ideas that you think might be of interest, even if you are not sure of technical feasibility.

Bill Casselman
Djun Kim