The OLC coordinators realized that one area of cooperation that had not been investigated was government documents. On June 20, 1991, an open meeting was held to discuss areas of mutual interest. A number of projects were discussed--at first not limited to Federal documents. Proposals included an online catalog for government documents, a cooperative serials list, sharing expensive reference items, or a CD-ROM network.

It was decided to investigate the possibilities of a CD-ROM network partly because Pitt had received a grant from the College of the Future to add their government documents to their online catalog. Also, both Pitt and CLP had made arrangements for CD-ROM versions of the Monthly Catalog.

A task force was established with one member from each library. All of the Government Information Services Task Force members have a reference/public service background so my apologies to all of the techies, semi-techies, and mini-techies in the audience if my descriptions of the project do not seem to be particularly technical in nature.

The original plan was for the Task Force to hold additional discussions, decide on CD-ROMS to be included in the network, and put together a proposal to send to a foundation by October, 1991.

The task force met and presented five possible configurations to the OLC coordinators. This was happening just as the first Census CD-ROMs were being released, so the original plan included various numbers of Census CD-ROMs, National Trade DataBank, and the County and City DataBook. The coordinators approved one of the plans and suggested that we have an open meeting with the techies from the various libraries to see how the project would work.

When we had the meeting, the first questions we were asked were whether the system could support 100 simultaneous users? How could CLP be connected since the library was not on the Internet?

We also had meetings with the public services staff--since more CD-ROMs were being released the staff wanted to know whether the configuration we had was "the best."

The task force had had some preliminary estimates of costs for a system with approximately 8 work stations split among the three libraries and 12-15 CD-ROMs plus some slots for especially-requested items. This proposal had an estimate of over $40,000 for the file server, CD-ROM server, PCs for work stations and software.

In the fall of 1991, we wrote a proposal to a local foundation for a planning grant so that we could bring in an outside consultant. The techies at the various libraries felt that a project with approximately 100 simultaneous users was possible, but none of the staff could put in the time to evaluate all of the needed components. Also, in meetings with the public service staff, they had requested some of the just-released census CD-ROMs be substituted for the PL 94-171 CD-ROMs originally included in the project. Adding more CD-ROMs could mean more towers and additional linkage to make the project work.

During this whole process the task force looked for similar activities so that we could get more ideas for our CD-ROM projects. We asked for help on several electronic bulletin boards as well as talking to anyone we thought might be of some help. Some of you may remember my asking about networks last year at this conference. We heard about several networks, but more on the scale of our original proposal than anything with a large number of multiple users.

One project in the midwest used three workstations plus phone-in capabilities. However, when an outside user was connected to the system, one of the in-house machines could not be used.

The second effort at a local university had connected a CD-ROM network in the library but had been unable to get the remote dial-in procedure to work.

During the fall we waited for the decision of the foundation. When we received approval of the planning grant, we began arrangements for hiring a consultant. One of the reasons that the consultant gave for taking the job was that although he had done extensive work with local-area networks (LANs), he had not worked with the type of wide-area network (WAN) that we were describing. He said that the project was "interesting."

By the time we had hired the consultant the earliest that he could be in Pittsburgh for meetings was July, 1992. As we got ready for his visit, we sent him information on the various computer systems at each of the libraries including the types of workstations or dumb terminals involved. We also kept adding questions that we needed answered. Should the CD-ROM information be loaded on magnetic tapes? Would we have problems paying for software licensing needed to run the system with a large number of users?

In July, the consultant met with us. We held an open meeting for all staff where he discussed both LANs and WANs. A final report was issued in the fall.

The final proposal came with a cost estimate of $250,000. This figure did not include any workstations, and although we now knew that CLP would be joining Internet through a Prepnet connection it would not be the same speed as the Internet connections at Pitt and CMU. This could mean that we would still have problems maintaining the connection.

When the techies at the libraries saw the proposal, they still saw problems with implementation. In fact, the consultant had listed several products still to be beta-tested among his possible solutions. He had also used IP tunneling in his possible solution and the staff at Pitt was opposed to its use.

Today, while I am still in shock from the jump from approximately $40,000 and 8-10 simultaneous users to $250,000 and 100 simultaneous users, the fate of our project is back in the hands of the OLC consultants.

Now, on to Mary Beth Miller and some information on a CD-ROM network that is actually up and running--even if it's a LAN and not a WAN. [See Administrative Notes, v. 14, #14, 6/30/93, for Miller's remarks.]