Speaker: Ralph Youngen (American Mathematical Society)

Panel Discussion: Current Thinking Panel

Well, this first slide is fairly unimportant except for the fact, please note, that I am from the American Mathematical Society. As such, I am part of the Mathematics community, not the physics community, and I personally will have very little influence on what happens in electronic publishing in the physics community. So keep those rotten eggs and tomatoes for the later speakers today!

What I do want to do is just tell you a brief history of what the AMS has done with electronic publishing, tell you what our current disposition is with regard to electronic mathematics preprints, and tell you what we see around the corner for electronic publishing.

One of the first things that the AMS did in the late eighties was to put up an Internet node called e-MATH, which has become a repository for mathematics information for the mathematical community. It was supported by an NSF grant, and has been available now since late 1990. Initial access was by Telnet only. 1990 predated Gopher and certainly the Web, so we had to write our own menuing system to give you access to things like membership information, publications that we put up on e-MATH, etc. We then migrated toward Gopher, and are now in the process of migrating toward the Web and will probably abandon Gopher at some point in the future. What this really has done for the AMS is it has given us an infrastructure on which we can base future electronic publishing initiatives.

Probably our first endeavor into electronic publishing was to make the Mathematical Reviews database available electronically on CD-ROM and through on-line vendors such as Dialog and Compuserve. What we also did on e-MATH was to allow you to look up author names in Mathematical Reviews. Of course, we didn't give you the full review text or any bibliographic information. We just provided a reference (an MR number) that pointed you to the print publication that you went and pulled off the shelf, and looked up the review.

Our next step in electronic publishing was to put up the Bulletin of the AMS on e-MATH. This was a relatively easy thing for us to do. The Bulletin of the AMS, like all of our journals, is produced using TeX and we simply posted the TeX files onto e-MATH and allowed you to download them with an associated macro package and process them locally. Another reason this was so easy for us to do was that the Bulletin is a privilege of membership. As such, there is no revenue stream attached to it.

Another major project under development is a prototype of an electronic preprint server for electronic mathematics preprints. This has been available since April 1994. Within the mathematical leadership, there were really two factions. One side felt that the AMS should really not be involved in mathematical preprints whatsoever, for to juxtapose the AMS name with these preprints would perhaps lend a little more credibility, if you will, to mathematics preprints than what some people felt was deserved. That faction was outnumbered, or at least overpowered by the faction that felt that mathematics preprints are here to stay. They are a valuable research tool and the AMS should take the lead in coordinating this effort in the mathematical community. So that's what we are currently charged with doing.

My group is currently charged with putting up a mathematics preprint server on e-MATH that will pull together all of the different preprint activity that's going on in the mathematics community already. These little pockets that are springing up very much like what I hear from the physics community. The difference is that the AMS is taking the lead in coordinating that effort. So what we are doing is we're creating kind of an umbrella for all the mathematics preprints that are out there on the net. The first easy thing to do is that we simply will set pointers to other mathematical preprint servers which have already been established.

There are preprint servers using Paul Ginsparg's software already, and we on e-MATH will simply point to those so that people can find preprints in certain specialty areas in that way. More significantly, however, we hope to post abstracts of every mathematical preprint on the net. We hope to have an abstract and the bibliographic data for all preprints on e-MATH with a URL pointing to the location where preprint actually exists. That's one method. If the preprint doesn't really have a home out there on any of the other specialty preprint servers, we will also allow that preprint to be stored locally on e-MATH. But the idea is that the AMS will have all of these abstracts in a single collection. People can view these abstracts, they can search these abstracts, and then can choose whether or not they want to go on and find the actual text of the preprint.

We expect that you will be able to view these abstracts by date posted and also by Mathematics Subject Classification. In mathematics we have the mathematics subject classification scheme which is around 60 different primary classifications. We will allow viewing by these primary classifications so that you can easily find the preprints in which your interest lies (i.e. in your specialty area). We also will do some kind of a WAIS search across all of the abstracts posted, and we expect to have just an e-mail interface in addition to the Web interface. With the e-mail access you'll be able to subscribe to different areas by math subject classification. So when a preprint gets posted we will send out the e-mail notification that lists the headlines, like the author, title, and things like that of the preprints. We'll also have an ID number so that if you want to use e-mail you could simply say "get" and that ID number and our server will mail back the abstract for you or the full text of the preprint, which ever you prefer.

Some other issues with our preprint server: copyright will remain with the author. We will be explicit about that up front. When preprints must be withdrawn because a publisher requires you to do so, we expect that the author and not the AMS will be withdrawing these preprints from our preprint server and to facilitate that, what was just discussed, we do anticipate sending out some kind of a private security code to the author posting the preprint. Any further actions taken on those preprints, like any revisions or any withdrawals, must reference that security code in order for those actions to take place. If you do withdraw a preprint we will allow you to put a pointer to the final published work in the abstract.

The system will be heavily automated, but it will not be automated to the point that if you submit something it will go up on our publicly available preprint server without human eyes first looking at it. We are going to have a very brief editorial read of every preprint coming to us. We hope that we're going to get many more just abstract-only submissions with URL pointers to preprints out there on the net someplace. And in that case, we will simply verify that the URL does indeed work before we post that abstract to the server. But there will be a cursory editorial read before anything goes up there.

I mentioned before that the Bulletin of the AMS has been available electronically, but we plan to significantly enhance this offering by first making it accessible over the Web, and then to also include DVI, PostScipt, and perhaps PDF files. We found that a lot of the mathematicians are somewhat annoyed by the fact that they have to actually download the TeX file and run it through the macros locally. What they really want to do is download it and print it, and so we will facilitate that.

Plans also call for distributing the Notices of the AMS electronically. The Notices is the AMS's journal of record. It is also a privilege of membership, there is no revenue stream attached to it, and we anticipate putting it on e-MATH in 1995.

And then we have to bite the bullet and put up some things which are going to be subscription-based. Starting with the 1996 subscription year, we will be making electronic versions of our current primary journals available. At the onset we anticipate this to be mostly a mirror image of the paper copy. We will not be adding a lot of value added in the electronic realm to those journals. We certainly will in the future, but at the onset they will just be available as they appear on the printed page. We also will be putting up some electronic-only specialty journals as they are acquired, and those will have a lot of added value electronic sense. And Mathematical Reviews will be available on the Internet starting sometime in 1996, most likely. That also, obviously, has got to be subscription based. And finally, we anticipate putting the catalog of publications on e-MATH, making it searchable through a WAIS search and allowing you to order our publications electronically as well.

That's pretty much all I wanted to say. Probably one of the larger issues facing us is the close proximity between electronic preprints on e-MATH and the electronic version of our primary journals on e-MATH (in 1996). The fact that you can view a paper in the preprint stage freely, but as soon as that same paper moves over into the journal side of thing - sorry, you're going to have to come from an institution that has already paid subscription to the journal in order to be able to view it.

[Bob Kelly] I see a bunch of hands raised. I'll entertain three questions for Ralph and then I'm going to bring up the next speaker. Annette, you were the first hand I saw so we'll go for you, then Maria and then Ann Okerson.

[Question from the audience] A short question. Are you going to demand extra charges for the electronic version of the journals and will it be possible to subscribe only to the electronic version?

[Ralph Youngen] Yes, we do anticipate being able to subscribe both to the electronic and the paper copy. I don't really know if I can say whether there will be an additional charge or not. We haven't set the pricing policies yet.

[Statement from the audience] Ralph, if I understood you correctly you said that you were looking forward to putting some of your current journals available electronically in the future, as well as having some specialty journals. But if I also understood you correctly, you made the distinction that there would be very little valued added to the current.

[Ralph Youngen] Initially.

[Question from the audience] So your initial is just, are you going to bring it as a bit map image of your page?

[Ralph Youngen] No, it won't be bit map. No, there will be TeX, DVI, PostScript, and maybe PDF which is, you know, Acrobat's file.

[Question from the audience] Okay, and then you'll add functionalities, etc. in the future?

[Ralph Youngen] Yes, functionality we might add later could include the references to be pointers directly into the Mathematical Reviews database or navigation within a document (say from an equation number in text to the actual equation). But that will be added later.

[Bob Kelly] Okay, two final questions. First, Ann Okerson and then Jeffrey West. As our host I have to give to Jeffrey.

[Question from the audience] I just have a question [INAUDIBLE]

[Ralph Youngen] I guess it is the case that we have a few subscribers to the Notices, maybe to the Bulletin also.

[Question from the audience] My follow-up question really was [INAUDIBLE]

[Ralph Youngen] Not that we're at all aware of. As I said, one of the reasons we experimented with the Bulletin and with the Notices is that, perhaps it was erroneous for me to say no revenue, but very little revenue is generated from them directly.

[Question from the audience] First, how much money do you get from the NSF?

[Ralph Youngen] Oh, I don't recall. Sorry.

[Question from the audience] Maybe we could ask how much money they're getting from the NSF.

[Bob Kelly] Who's they?

[Question from the audience] We are getting roughly $225,000 a year for three years.

[Ralph Youngen] Yeah, ours was on the order of $200,000 also.

[Question from the audience] The second question is, or comment maybe, it would seem to me if you have an e-print system that is accessible for free and then you have a journal system running concurrently, free market will clearly determine that no one will every read the journal.

[Ralph Youngen] You have to consider what is the value added in the refereed journal.

[Statement from the audience] No, but then you have a serious question of revenue.

[Ralph Youngen] That's right.

[Question from the audience] So, I'm very interest in what the American Mathematical Society is thinking of in terms of how they're going to cope with what I think the APS is going to have to deal with, in terms of this revenue question.

[Ralph Youngen] Right, we do expect journal revenues to fall off. Currently journal revenues are 70% to 80% of our publishing revenues. We hope to be able to counter the fall off in journal revenues with a substantial increase in our book program. And we have hired two people to do much more proactive acquisitions of monographs than we have ever done in the past. That's one way we hope to offset the loss of revenue. The other way is through our sale of service effort. I don't know if you're aware that the AMS offers sale of service in TeX-related publishing services and we're currently conducting a market survey to see if there's a market out there for us to aggressively pursue that effort as well.