Moderator: Peggy Judd (AIP)

Panel Discussion: Tracking of Papers

Our panel is composed of four speakers and we arranged this panel in order to elicit some response from the audience here and get some answers about how to build a model to track papers. We are going to ask many questions. This is how I would like to approach this, questions are what I think we all have. The title of the session is, "Tracking Papers from E-print through Archive" and our session is predicated upon the premise that the archival version is the definitive version, that it has been refereed and reviewed, that we are trying to retain that process and that all scientists really feel that review is still very important.

I think from this morning's session on current technology and current thinking, technology is not the problem at all. Our panel is not up here to offer technical solutions or even to discuss that technology has got the answers. What we are trying to find out is, what model are we supposed to be putting into place? I think there are a great many key players. Some of them are playing certain roles right now. Those roles will change, but what the technology people need to know is, "what model needs to go into place?"

We pose the following questions: How will electronic dissemination, namely E Print Phenomenon that is taking place, going to affect the traditional archiving process? How are people going to cite articles? How are we actually going to know what the definitive version is? We are going to have to have some model in place, all key players are going to have to be part of that model.

Second question, how will the researcher, the reader, Scientists and Publishers want the readers to find and read the material. How are they actually going to locate it and identify what version they are reading. Right now, with a certain number of data bases, you pretty much know what you are getting when you go there, and you go there for valid reasons.

But if those preprint data bases proliferate and grow in number, if you can put certain articles on many of those data bases, not just one, then pretty soon the reader is going to have a hard time finding the definitive version The reader will not know what version they are reading when they get there.

The third big question is, Who will be (or should be) tasked with tracking or monitoring the various versions and identifying the archival version? And who is exactly going to be responsible for putting this model into place? I would like to say that I think it is going to be a collaborative effort. I think that has been said before today, but somebody is going to have to be responsible for tagging, identifying, and monitoring versions of articles. We have seen and heard discussions here today that show that effort is starting to take place, but we are going to have to answer many more questions and start designing the model.

We decided that we would look at these questions from a variety of viewpoints and from four different key players.

1. Author and/or Scientist
2. Journal Editor and Administrator of a Co-Print Database
3. Database Provider/Publisher
4. Librarian or Information Specialist

Each speaker on our panel is going to speak from their particular point of view. Of course, this panel is qualified to address all of the issues, but I have asked them to try to focus on their assigned topics. At the end of the session general comments from each panel member may be appropriate.

Someone from the Abstract and Indexing community may see things considerably different from the viewpoints presented today. Their comments would be welcome at the end of the discussion if there is a representative from that community in the audience.

In our panel discussions prior to today we found out we have more questions than answers. We also discussed many issues related to the above questions. To help structure our presentation we all agreed that we could view the publication process as a simple triple-layered structure, with each layer representing a value added:

1. The preprint or e-print layer
2. The peer-reviewed layer
3. The archival layer

Paul last evening asked when the Society (and/or publishers) were going to develop this new model. He thought that everyone was at Los Alamos to discuss how to get to a new model and what to do with eprints. He asked that we please don't duplicate what is being done right now in print and I think that is an important point. I think we should heed his message. I think that what electronic publishing is, if you want to use that global term, is not another layer or many layers, it is frosting on each layer. I think that it is already impacting every one of those layers. We should consider electronic publishing as an enabling technology. We should expect to take advantage of every piece of technology that we can and utilize it to our very best advantage.

I think it is important that we not lose the sense of history, "the document history", the evolution of that final research. Electronically, we probably can take care of that. We could have living documents, we could annotate documents. Someone earlier today said something about that. Those notations would exist as part of the document and could be there, both prior to and after referring.

UNDERLYING PROCESSES ARE COMPLEX

However, each simple layer presents numerous questions and concerns because the underlying processes and connections to our traditional publishing process are complex. Who should administer each layer? Who should label each version within each layer? Who will replace outdated versions?

Simple answers: The author for layer 1? The Editor for layer 2? The library or publisher for layer 3? These suggestions are reminiscent of the traditional publishing process and most likely will be the way many see, and want to see, this whole situation evolving. However, I believe that this model will fall way short of what will be needed. I am also convinced that technology will eventually enable us to dramatically change the way we all communicate. And the overload of available data will demand some navigational help, sorting, authentication, validating, and labels or pointers.

The methodology whereby these services can be economically provided is material for an entire new workshop. Hopefully our speakers, along with others today, will clarify the important concerns so that a truly useful methodology can evolve.

>From a technology point of view we probably tend to be a little bit more radical because we can. We are not really worried about the revenue streams, however, the publishers, the Societies, even the authors may be very concerned about the cost.

Each one of these layers is very, very complex and, if someone said right now to the publisher or to APS, "Okay, go head - I mean APS has even said they are going to look at e-prints as part of their process and try to help that process," I think we would be looking at a huge investment of resources to establish working procedures along the lines we have been discussing. If you look at these layers and start to pull them apart you quickly recognize the complexities. Each layer is connected into this traditional publishing paradigm that we are living with and that we seem to believe has supported us for many years. The paradigm is shifting, as we all know, that is why you are here. There could be simple answers for this, but I'm not sure there are. I do think we are going to need some navigational help and some labels. I think Paul has included labeling as he has started to build this. The home pages have started to point and navigate for the reader. It is completely different from last year and technology will continue to enable changes.

What I am hoping to derive from today's discussions are to identify the real concerns about the "new model", not haggle about the small details where we tend to disagree. The real concerns should be reiterated by everyone in the audience, and this panel, and speakers on the other panels. After listening all day we should walk away knowing the important concerns and important parts of the model and that is what we should look at and try to put into place.