perlman@giza.cis.ohio-state.edu (Gary Perlman) (11/24/90)
As you may recall, I sent out news articles (Dec. 3, 1989 and Aug. 16, 1990) about compiling a public domain bibliography. This is a status report and a request for people (like me) willing to do some grunt work. The next posting will describe how you can access the online databases using anonymous ftp. First, a truly public domain bibliography is not possible because of the rights of publishers. The next best situation is that researchers can make free use of the bibliography, provided that the material is not sold or republished. Some organizations, like the ACM and IEEE, grant such permissions as part of their copyright notices, and I have been successful in gaining similar permissions from publishers (e.g., Lawrence Erlbaum Associates has granted such a permission for the journal Human-Computer Interaction). So, for people who want such a bibliography for use in research, publisher copyrights should not be a serious restriction. My request to the net for abstracts from the journal HCI was unsuccessful. I got no responses, except for a few people sending me their extended entries with abstracts. I mailed the same information the the SIGCHI mailing list, with similar results, except that a few people wrote long replies that they no longer had the abstracts online, and why should they provide them in any case? Their replies almost as long as their abstracts! So much for altruism and/or self-promotion. One extremely fortunate result was that I got connected to the administrative editor of HCI and got all the abstracts at once, and these have been incorporated in the database. One problem with that source is that the online abstracts had spelling errors and did not exactly match the printed versions because of some word-smithing, I assume. Still, I have found it useful to have them, and I hope you will, too. The data validation issue is addressed later. I was fortunate to get two work-study students to do a lot of typing, and as we sit here, they are entering the complete 82-83, and 85-90 ACM CHI conference entries: regular bibliographic information and abstracts at first. These will also be made available. All the information is being kept in refer format, mainly because most people responding to my earlier postings suggested it. The format looks like: %A author name %D date %T title %P pages %X abstract The data translation people at OSU are going to help provide data translators for other formats such as Scribe, troff, LaTeX, and perhaps RTF. We will also have the ability to translate from other formats to our basic format. (For those who care, we plan to move to an SGML base later on and provide translators and translations to other formats.) We currently have the following online in refer format: 58 51K CHI83.bib Proceedings of CHI'83, with abstracts -- there was no CHI conference in 1984 38 34K CHI85.bib Proceedings of CHI'85 with abstracts 34 25K CHI86.bib Proceedings of CHI'86 with abstracts (incomplete) 51 42K CHI88.bib Proceedings of CHI'88 with abstracts 53 69K HCI.bib Contents of the journal: Human-Computer Interaction 116 46K books.bib Books in HCI, with some tables of contents There are some other refer format files, donated by a few kind souls, and there are some files that will eventually get translated to refer, such as the contents on Helander's Handbook of Human-Computer Interaction and Salvendy's Handbook of Human Factors. If you have some information like the above, in any online format, please get in touch with me. Now, I would like some input and some help. How should this information be made available to others? The readers of this article are obviously biased to say anonymous ftp access or mail server, and these are almost certain, but I would like other ideas for other types of users. We could, for example, put the records into HyperCard stacks. What sort of collections would you like to be able to get? One possibility is to provide a query server, but I am concerned about the load this might create. We could restrict the hours of access to early mornings. Any suggestions would be welcome, especially from people with experience doing this sort of thing. Given the problems with the accuracy of data, and our goal to provide a high-quality information base, I think we will need some help with data validation. I am looking for volunteers who will take a chunk of entries, such as one journal volume or one conference proceedings, and check all aspects, sending back corrections. The benefits to the volunteer for doing this will be negligible, amounting to being on a list of validators. Still, it is a good cause, and the online bibliography, once done, will be available for the world, forever. Please volunteer to do the validation task for something on the order of 50 articles and abstracts. You'll probably regret it, but by that time, you'll be done. And again I ask for people to volunteer their databases, especially once that completely cover an area such as a journal or the proceedings from a conference. Abstracts would be great, but work-study students can do a lot of the work. I am especially interested in getting the INTERACT'84'87'90 proceedings. Except for the data validators, who will get their bibliographies immediately, I estimate that the first bibliographies will be available via ftp in January. This should include the complete conference proceedings from SIGCHI from 1982-1990, and some rudimetary search and formatting software to complement/replace refer tools. -- Name: Gary Perlman | Computer and Information Science Department Email: perlman@cis.ohio-state.edu | Ohio State University, 228 Bolz Hall Phone: 614-292-2566 | 2036 Neil Avenue Mall Fax: 614-785-9837 or 292-9021 | Columbus, OH 43210-1277 USA