SUBSCRIBE NOW!

Vol. 12 No. 4 — April 2004

In the January 2004 Searcher, Richard (Dick) Giering spoke about his early military and Data Corporation experiences in developing full-text retrieval technology. We pick up Mr. Giering's story in 1968, when Data Corporation was acquired by Mead Corporation.

An Interview with Richard Giering Continues

The Mead Merger

Giering: A major event occurred in late 1968. The story goes that Lysle Cahill, who was president and partner with Bill Gorog of Data Corporation, was also a friend and neighbor of Jim McSweeney, the CEO of Mead Corporation. They got to talking one day about acquisitions and new business. Those were the days when everybody wanted to be a conglomerate. Among a lot of other entities that Mead thought about acquiring was Data Central. They realized, "Oh, there's a think tank out there. They've got equipment and cameras, and they're doing work on printers (ink jet printers), and some computer work, and some highfalutin' target stuff. Maybe we ought to acquire them." So Mead bought out Data Corporation.

I remember getting called in with all the rest of the employees. We were told that Data Corporation was now a subsidiary of Mead.

Bjørner: You were told after the fact. It wasn't that Data Corporation went looking for more funding?

Giering: No. By this time, Data Corporation was already hot. The Air Force funding for Data Central had been cut off, and Bill Gorog had gone out and put Data Corp in debt and obtained money. We had demonstrated at ASIS and built a center in Washington and gotten some business. We had done all this, but only in our small niche, and there were a lot of other divisions in Data Corporation.

Shortly thereafter, we were invaded — I use that term deliberately — by a group of people from the Mead Corporation. Basically, they said, "We got you. Now what did we get?" They started looking at the various parts of Data Corporation.

They said, "We know nothing about ink jet printers; we know nothing about computers." Mead Corporation was a forest-products, paper-processing, pulp-making company. What did they know about what they were looking at?

They didn't know what they had; they didn't know how to evaluate it. One of the things they did was to zero in on the Information Systems Division (our division). They decided, "You've got all these government contracts. You've got this Union Carbide installation. You've got OBAR spending money to convert data. You've got these computers running. You've got terminals scattered around. Maybe you've got a business. We don't know, but how do we find out? Let's get Arthur D. Little involved."

Focusing on Legal

Giering: They contacted Arthur D. Little, and ADL brought in a study team. The study team was headed up by H. Donald Wilson; second in command was Jerry Rubin.

Before I go further, I'd like to tell an anecdote about Arthur D. Little and its studies. Don Wilson told me the story — after the fact. Years earlier, there was an Arthur D. Little study conducted for IBM on something called xerography. Battelle had this new process and offered it to IBM. IBM called in Arthur D. Little to do market research. ADL went back to IBM and said, "Don't touch it. We feel that the market for xerographic copy machines is something like 200 units over the next 5 years." Basically, they looked at the wet copiers and the rest of the copiers available at the time and they said it would take 190 machines to reach the break-even point. They figured 200 machines were the most they could sell. So they said it was not a good business to get into. IBM killed off the product, as Arthur D. Little advised them to do. We all know what happened to xerography.

The problem was that ADL did not take into account the changes in office processes that would take place as a result of this new technology. They did not think outside the box. Remember this story.

Arthur D. Little was called in to look at this whole thing called Data Central. Both the men were lawyers. "Government contracts? You're not making any money on government contracts. Union Carbide? Piddling. OBAR? Well, that probably could make some money." They came back and said, "We think that if we study the legal research market, we can probably give you a better business." All of a sudden, the scope of the study changed. Now the study takes off, but it was going to be limited to the legal marketplace.

By this time, Mead had decided they were going to have a business. In February 1970, they took the Information Systems Division (my division, the one of which I was a co-director) and spun it off as Mead Data Central. The rest of Data Corporation continued as a wholly owned subsidiary of the Mead Corporation. Don Wilson of Arthur D. Little became president of Mead Data Central; Jerry Rubin became one of the vice presidents. Peter Vann and I, the two co-directors of the Information Systems Division, became two vice presidents, and a man named Ralph Welch, who was the contract administrator for Data Corporation, became another vice president. That was the nucleus of what was Mead Data Central. The initial charter for Mead Data Central, Inc. was to take the total Data Central business — all of it — and to start running. Use the Mead financial wherewithal and make a big business out of it...although the study, not yet 100 percent complete, was aiming toward the legal research marketplace.

System Enhancements

Giering: From my standpoint, we were still in the throes of enhancements. One of the major enhancements that we did was Keyword in Context (KWIC). We had enhanced one of the terminal capabilities to color terminals. The Keyword in Context became Keyword in Color. Keyword in Color was the following (you could use any color, but we chose the following): the keyword, meaning the word used by the searcher at the terminal in the search process, was in red; the surrounding context was in yellow; the normal text was in blue. The way it worked was if the keyword was going to occur on the top of a screen, the bottom of the preceding screen would turn yellow to warn the reader that he was in the context of the word used in his search. And vice versa. If the keyword was on the bottom of a screen, the top of the succeeding screen would be yellow to remind him that he was still in the context of the word he had used. The number of words was specified by the database administrator. Normally, we used something like 15.

Bjørner: This was brand new. Nobody had done this before.

Giering: Not only that, when we demonstrated it at the ACM (Association of Computing Machinery) convention in the fall of 1970, in New York, we were a laughing stock. Nobody would give us the time of day. Everyone said that no one would ever use color terminals. Color was too expensive, was no good — nobody would ever use a color PC. We had a big booth with a color terminal and operator at each of four corners, and we brought people off the floor to operate them. Key Word in Color. People, literally, looked at it, laughed, and said, "That's nice, but nobody will ever buy it." It was too early. Today, do you ever see a black and white PC?

Highlighting. In the noncolor world, we were highlighting by underlining, by the use of italics, etc. — we used all kinds of techniques to highlight. But we found that color was the most eye-appealing.

Bjørner: I wonder whether that came about because of the history the rest of the company had with photography.

Giering: No. We were completely independent of photography.

Bjørner: Why did you pick color?

Giering: I don't really know. It was just a natural.... We used terminals from Computer Communication, Inc. of America — CCI. They contracted mainly with the stock market. Their progression of terminals went from black and white to color, so I got a color terminal. Remember the 5-inch Sony Trinitron TVs? I built a desk set around one and had it on my desk. It was a color terminal, and I had a keyboard. It was only good for a short distance. If you were further away from the terminal than my arm, you couldn't read it.

Bjørner: This is a very different development from the bibliographic systems, and I can see that it was driven by the fact that you were dealing with full text.

Giering: Not only full text. We were driven toward the end user. Not toward a librarian. We were interested in the lawyer, the doctor, or whoever the end user was going to be. We did not want anyone interposed between the person needing the information and the person having the information. That was our goal.

Bjørner: There's where you are forward-thinking precursors to the Internet, because the whole thing with the Internet is that anyone who wants information can find it himself.

Early Telecommunications Challenges

Giering: Let me comment on something about communication before the Internet. In 1969 or '70, I remember that I would put a terminal in the trunk of a car with an acoustic coupler, drive around central and western Ohio and northern Virginia, and go to a hotel lobby or a gas station. I'd put a quarter in a pay phone, get an operator, use a credit card to call the computer, and go back to the trunk of the car to communicate through my acoustic coupler to find out about the dial central. At that time, there were three types of telephone systems — dial centrals — available. I don't remember all three. One was electronic, one was step by step. We knew that if the dial central in a particular town was electronic, we had no problem; we would be able to support any user in that town. If it was the third one, forget it. There was no way we could ever support a user in that town. If it was a step-by-step network, we could get the Bell system in that town to put in a special line that we called "conditioned" for that user's modem and support it.

The only way we could find that out was to go around and test a particular dial central — that is, the three-digit telephone number prefix — to actually call and see whether it would work. If it worked, we'd write down that three-digit prefix and go on to find another three-digit prefix and try that one. That's the only way we could say when we went to OBAR, "This law office in this town, forget it; we can't support it. Not because we don't want to, but because, technically, communication-wise, we can't do it. On the other hand, that law office in the other town, no problem."

We were covering the state of Ohio for OBAR and northern Virginia for our Washington service center area.

Bjørner: Were they adding statutes from other states at that point?

Giering: No. We were still using Ohio Bar money; it wasn't until after the 1971 split that things changed.

The Split

Giering: In 1971, we had three businesses running. But there was a very distinct difference of opinion — and that's putting it mildly — between Jerry Rubin on one hand and me on the other. Jerry Rubin felt very strongly that the only way this business — whatever that meant — was going to succeed was to concentrate completely, 100 percent, on legal research. I believed very strongly that the only way the business was going to succeed was to move adroitly into all three business/market areas, including the law. The three areas included: 1) the selling of data, all data, wherever we could find it, including the law; 2) the servicing of data in a service bureau; and 3) installations.

Bjørner: Tell me more about the servicing of data.

Giering: The people who have data but don't have a computer. ... bring their data to us and we put it up on our computer. We give or rent them a terminal, and they use their private data to which only they have access. In today's Internet environment, the service bureaus of that time would have been called Internet Service Providers (ISPs).

In today's world, installations would be not unlike Computer Associates, where they install their software on a computer system, furnish documentation, train users, and leave. Billing includes maintenance and upgrades.

But Jerry Rubin felt that we needed to concentrate. Throw away those other businesses and concentrate only on the selling of legal research data. Period.

Bjørner: It's a different way of thinking, a different philosophy. I am reminded of a story from Roger Summit. Dialog was starting to sign up databases in different subject areas. Bill Knox, the head of NTIS, said to Roger, "If you don't know what business you're in, Roger, you're not going to be in business." Of course, Dialog continued to add databases in different subjects and they stayed in business.... So, you had this difference of opinion.

Giering: That's right. It manifested itself in a very significant area, the man-machine interface, the communication. With a diverse customer community, it was clear to me that we had to have a generalized interface, but Jerry Rubin felt strongly that we had to have a very stylized interface for the legal community. That was the very significant factor. That was the essence of our disagreement.

Bjørner: I remember Jerry Rubin was quoted as saying that lawyers don't know how to type or spell, so he wanted to have a terminal that essentially typed for them. This, of course, became the stylized UBIQ red terminal.

Giering: We started a series of meetings at Mead corporate headquarters on the subject of "Where do we go?" And to make a long story short, Rubin won and I lost.

At that instant, the Data Central software was copied. One copy went back to Data Corporation with me and the Information Systems division was reconstituted in Data Corporation. We took what was euphemistically called the "non-legal business" — some referred to it as the "illegal business" — we took all that non-legal business back to Data Corporation. Jerry Rubin became head of Mead Data Central and revised everything into what became Lexis.

I want to make sure that something goes on record here. Jerry Rubin has been adamant in saying that I had nothing to do with what the Lexis user saw and had access to at that time and since. He's absolutely 100 percent correct, in that the communication interface, the man-machine messages, the search engine, etc., were all redesigned to be oriented to the legal research marketplace. That's what they did. That's what became Lexis. However, and this is where I differ, a lot of the back office work, all the internal file structures, remained constant. So the technology that backs up how they did what they did remained the same. But what Jerry Rubin says about what Lexis was and what Lexis is — that Dick Giering did not have anything to do with Lexis — he is correct. The part that he's talking about, he's absolutely correct, and I don't want to take any credit for what he's talking about.

The man-machine interface and the specific search engine that used the technology were his [Rubin's], but the underlying basic technology was mine.

Bjørner: So we have the split. One copy stayed with Mead Data Central...

Giering: ...and the front end communication and search engine were revised into Lexis...

Bjørner: ...and you took the other copy and went back to Data Corp. This was in 1971?

After the Split

Giering: Yes, and now we move into a new stage. During this split, some words were spoken that should not have been spoken, by some of the people who worked for me, and by me, because I support my people. As a result, Bill Gorog, again my mentor, had to "hide me" for awhile. I spent a few months writing documentation. Here we had a system — by this time, it was a real system; but there was a lot of stuff in the back end, in the back office, that was still pretty flaky. We had done a lot of work, especially on the database administration functionality, to make it easy to add new databases, to add new capabilities. We did it all in a hurry. We had customers coming in the door, so we had to get it done fast. There was very little documentation.

So, while I was "hiding," I used the time to write documentation. I don't remember whether it was 6 months, 4 months, 8 months. The Washington service center had been closed and Washington contracts moved to the Dayton service center. By use of modems, it really didn't matter where the service center was — a telephone call is a telephone call, especially with toll-free 800 numbers. At one point, a couple of the customers in Washington were having problems and Bill Gorog called to ask if I could go to Washington to calm some troubled waters. I did, and as a result, I was "surfaced" and given a new charter.

The charter was to find some new business. I didn't have to make any money, but I couldn't lose any. I could take any profit that I made and pour it back into the business. I didn't have to show any profit in Data Corporation, but I couldn't show a loss. Part of the deal was that if I found a new business, I could have it this time. This was to be my business. If it was spun off, I could have it. This was the middle of '72. Unfortunately, none of this was put in writing.

We went out and looked for additional installations. We had a lot of work to do. We got some contracts. Some made it, some didn't make it. We had the Recon Central contract, by this time called Avionics Central, because the Recon Lab changed its name to the Avionics Laboratory. We had a little bit of profit. Between that and some of the Washington contracts and service bureau, we were able to eke out some profits and do some advertising and get some additional business. We didn't show a loss. We didn't make a profit — we poured it all back in.

The Boston Globe

Giering: Then in late 1975, I heard about a man at The Boston Globe who wanted to do something with his library. George Collins. I went up to see him. I carried a terminal with me — a color terminal. (Parenthetically, by this time, Lexis had done away with color terminals; they'd done away with sort; they had done away with arithmetic capability; they had killed a lot of the capabilities that Data Central had, because in the legal world, they weren't necessary.) Anyway, we still had those capabilities. I took a color terminal to The Boston Globe and demonstrated what we had.

George Collins said, "I think this is what we've been looking for." They had just — a couple months earlier — converted their daily process to computerized typesetting. This was the first of many visits. About the third or fourth time I was there, I convinced him to give me a purchase order for a test. I said I could take a daily tape in which his people would select the stories and send me the tape from their typesetting system, with only the stories he wanted on the Data Central system — our service bureau.

By this time, I had a specification of their capability. I said, "I can program our front end. Give me a purchase order, $5,000 to program our front end to read a magnetic tape off your daily take. We will read one day's tape and pour your data in and show you your data online."

He said "I don't believe you." I said, "For $5,000, what do you have to lose?"

I walked out with a purchase order — one piece of paper (8-1/2 by 11 inches). The purchase order read something like "Test of conversion of Boston Globe typesetting material, per discussion, George Collins, Dick Giering. $5,000." Period. Signed. In an envelope that I had in my briefcase. I walked in the next morning to our contracts department, handed it to the contracts administrator, and he said, "What's that?" I said, "A purchase order."

This contract administrator was used to government contracts, packets that were inches thick. He said, "That's all?" I answered, "That's it." He said, "That can't be it." I said, "That's it. $5,000 right there." He said, "I don't believe it." I said, "That's it. I want a project number right now." He said, "I've never heard of such a thing."

He went down the hall to the finance people. The head bookkeeper (Mrs. Honeycutt) and I had been working together to make sure about my accounting numbers. Because every year, I would be saving up my profits, so that at the end of the year, I would be able to put them into advertising. So, she and I worked very closely. He took this piece of paper down — I went with him — and showed it to her. He said, "What do I do with this?" She said, "Give him a project number!"

A week later, we had programmed the front end. I called George and said, "Make arrangements to send me a tape." A couple days later, he called back, and said, "When do you want a tape?" I said, "When do you think you can FedEx it?" He said, "How about tomorrow night?" The following morning, we got a tape. We sent it down to the computer center. I don't remember whether it was that night or the following night, it came up online. We saw it; we checked it out; it looked pretty good. I made arrangements to take a terminal; I went back to Boston. This was 2 days after they had sent us the tape. I walked in, and said, "George, ready?" He said, "Already?" And I said, "Yes."

So, we went to a conference room, and I set up a terminal. I said, "George, sit down," and gave him a keyboard. I told him what to type, and out came one of his documents — a Boston Globe story. He asked, "Can I send some more data?" I said, "Yes, if you send us some more money."

They started sending data and gave us more purchase orders — I guess this was 1976. About June or July of '76, they started running dual, where they were still clipping the newspaper for the morgue and also sending tapes. In July of '77, they stopped clipping and ran completely electronic. They were the first electronic library. This was our newspaper service bureau business.

More Newspaper Developments

Giering: At the same time, we found another newspaper, the Philadelphia Inquirer, that wanted to do something similar, but they wanted it on their computer. We started negotiating to install — the second type of business, an installation — on their computer, to do exactly the same kind of thing with their data. We set up the system to read from their typesetting system; they would take their tape onto the Data Central system on their computer into their system.

By the way, George Collins at the Globe would bring in his editors. We set up four terminals in his library, on swivels. The editors would come in, and sometimes they would ask the librarians to do the searches, and sometimes they'd swivel the terminals around and they'd do their own searches.

The Inquirer set it up so that they would do exactly the same thing, except they wanted the terminals in the editorial department to be split screen. The editorial department was going to be able to edit and, with a split screen, simultaneously search the library. This was in late 1977. Then we got into the next donnybrook.

Bjørner: I have to ask you one question first. You had two different newspaper libraries. Did you create a database structure that matched for each of those systems?

Giering: Similar but not the same.

Bjørner: Were the customers the people that determined the structure?

Giering: Yes. Something we did back in the pre-Lexis days was we had the database definition. After the split, I wrote a lot of documentation. One of the reasons we know that a lot of the internal structure and a lot of the back office operation were still being run at Lexis was because we had the documentation, and we would get telephone calls from the people at Lexis asking for information. They had no documentation. We had the documentation. They would ask us questions, especially in the area of the database definitioning and data update.

When it comes to your question about the newspaper, with the database definitioning capability, we were able to set up separate yet similar-looking Electronic Newspaper Libraries. Let me give you an example — noise words (i.e., stop words). At the time we set up The Boston Globe, we sat down and established a set of noise words — if, a, but, for, an, the, what, of, nor, who, etc. — for the actual text of the story.

George Collins called me on the phone one day and said, "I've got a problem. My editors came to me and can't find something." I said, "What is it they can't find?" "We have a rock group that played here, 'The Who.'"

We already had the capability for setting up different sets of stop words by field, by group. We had index terms, where the librarians could say that this story, although it didn't use these terms, had to do with this subject. In this instance, although the main text said "The Who" (both words of which were noise words), we specified "No stop words — search every word" in the subject area. The solution was for the librarian to put "The Who" in the index terms group. So now, a search for "The Who" found that document.

Subtle things like that came as a result of the database definition subsystem that we had set up a couple years earlier. This is what I meant by saying "similar but not the same."

Another anecdotal sidelight. One day I was playing around with some Boston Globe material on my desk. I don't remember how, but I came across a misspelling. S-U-R-B-U-R-B. I searched on it and found quite a number of stories using the word "surburb" in the Boston Globe. Some would say this is not unusual for New England! But I thought to myself, "Why is that there?" I picked up the phone and called George and told him to get online. I said, "Run a search on 'surburb.'" He found something like 300 hits. "What in the hell is this?" he asked. I said, "I don't know — that's what I'm asking you."

One of the things they had in their database definition was a Source field, a place for backgrounder information (not normally seen by the average viewers), where they kept data about where the material came from — the editor, the typesetter, etc. George said, "I'm going to check this out. I'll let you know." He started checking, and it turned out that when one of the editors sent down the word correctly spelled, the typesetter "knew" that it was misspelled, and the typesetter would change it. The editor ran a couple of tests, followed it, and found the typesetter doing this. Only through the online system did they find that error. "Surburb" has now been corrected.

Bjørner: You talk about the next donnybrook.

The NewsLib Business

Giering: I now had a new business and I was trying to find other business opportunities, including additional newspapers. I thought we could probably find medical, education, other markets. But I needed money. I needed to bring in some profits to be able to do more. I talked to New York Newsday (Long Island), Time magazine (Peg Fischer) — I had lots of conversations with Peg Fischer at Time and also with people at Newsday in Garden City. Went out to The LA Times and talked to people in Milwaukee. I spoke with a lot of people. Some of them were indecisive about this new technology, and some of them looked as though they might do something. Things were starting to jell. Realize that the newspaper activities were a subset of two of the basic businesses that I had postulated just before the split from Mead Data Central in 1971 — service bureau and installations.

At some point before, the Data Corporation subsidiary had its name changed to Mead Technology Laboratories. They had new management. Lexis was running beautifully; it was a very significant success, sitting off on the side. The management of Mead Tech Labs was looking at us, and I wasn't showing a profit, though I wasn't showing a loss. There was significant income that I was pouring back into the business. The rest of their divisions were showing losses. They didn't like that. They wanted to take the money that I was generating and do things with it. They got an idea: Why don't we take this newspaper business and maybe we can make another Lexis called NewsLib and sell newspaper data? Like Lexis is selling legal, maybe we can go to The Boston Globe and convince them to let us sell their data, and go to the Philadelphia Inquirer and sell their data. Maybe we ought to build a business plan and present to Mead something called NewsLib.

They hit me with this, and I said, "Wait a minute. You're killing the golden goose. The minute you do that, Mead is going to ask, 'Why should we have two businesses selling data in the same corporation using the same technology?' They're going to steal this thing right out from under you. Don't do that!" Their response was, "Oh, you're full of it. They're not going to do that. This is yours, this is ours."

They went ahead anyway. And that's exactly what happened. They built a business plan called NewsLib. I had nothing to do with it. It was against my better judgment. They presented the NewsLib business plan to Robert O'Hara, the group vice president, who was also group vice-president of Lexis. He said exactly what I told them he was going to say. He said, "We're going to take this newspaper business and turn it over to Lexis."

I said, "Wait a minute. Bill Gorog promised me that if I find a business, it's mine." O'Hara said, "Show it to me in writing." I didn't have it in writing.

The upshot of the whole thing is, I resigned. I said, "Have me once, shame on you; have me twice, shame on me." I knew I was persona non grata in that organization (Mead Data Central). Jerry Rubin was never going to put up with me.

A man named Paul Beck had come back from Mead Data Central with me when we split. He was with me the whole time while we looked for all these other businesses, helped me with the development, then operated The Boston Globe, the Philadelphia Inquirer. When they took the newspaper business away, he went with the business to Lexis but didn't stay very long at Mead Data Central. He might be able to tell more about that transition.

When the newspaper business was transferred from Mead Technical Laboratories to Lexis, it became the basis of what was introduced by Mead Data Central in 1980 as Nexis.

After NewsLib

Bjørner: You left Mead Tech Labs, but you continued in newspapers.

Giering: Luckily, in the interim, I had been approached by another organization to form a partnership, with a company named Atex. They built newspaper typesetting systems. We formed a partnership to build library systems for newspapers. We built on the PDP-11, which was too small a system, and installed the first one at The Boston Globe, believe it or not! We were up and running. Then we got the Chicago Sun-Times. InfoTex was a partnership and became an incorporated entity later. We realized the PDP-11 was too weak a machine, so we started reprogramming it for the VAX.

When I left Mead, I had to sign a non-compete agreement, and I could leave only with what was in my head. Basically, we designed for the PDP-11 everything we had already done, with one exception. I added the arithmetic extraction from text with unit of measure capability. We were able to extract arithmetic information together with unit of measure from the text.

Let me give you just two examples: A searcher could ask the question, "Give me all information that contains the word fire or conflagration within five words of five persons dying...five persons killed." Another example: "Give me all articles/material about airplanes or aircraft or spacecraft within 10 words of speeds greater than 500 knots." And regardless of whether that speed is in miles per hour, kilometers per hour, knots or whatever the unit of measure it is, the system would find it.

Bjørner: You added that in the late 1970s? That's quite a capability.

Giering: Yes, we added that to the PDP-11 system. But now, it's all dead and gone.

Bjørner: But you did a lot. Can you tell Searcher's readers what you think your particular contribution was?

Giering: In my view, my contribution to this industry was moving the technology from bibliographic access to full-text access. At the time, nobody believed it was financially or technically feasible. What I did was prove that full text could be commercialized. That is what I believe I did. u

[Mr. Giering was interviewed on September 10, 2003, in Libertyville, Illinois. Lysle Cahill and Paul Beck contributed comments by e-mail and telephone in January 2004.]

The next story in this series will focus on Bibliographic Retrieval Services (BRS), with an in-depth interview with Jan Egeland, one of the company's founders.

 

Lysle Cahill on Working at Data Corporation Data Corp. was issued a contract by the Foreign Technology Division of WPAFB [Wright-Patterson Air Force Base]. The statement of work was to write a series of classified programs that would encompass all research being done by all military forces of the U.S. The contract was broad in that a number of in-house programs already underway were included in the statement of work. Our entire facility carried a Top Secret rating. There were additional coded ratings that included special assignments, particularly with SAMSO [Space and Missile Systems Organization] on the West Coast. We had an ingenious systems programmer, Dave Behane, who wrote all of the guidance and tracking programs for the satellite tracking systems. His programs were unique — and always undocumented! We accordingly assigned a group of programmers whose only work was to document Dave's programs. I was given an approved storage file holder that required a crane when it was moved into my office. It was a four drawer file cabinet with a double combination lock. All of our classified contracts were kept in this cabinet, which was actually a safe. Strict orders were issued that the safe was to be unlocked any time I was in my office. A drop-in Open sign was always placed in the file drawer handle when open. One of the work statements was that we were to write a classified database that would permit all of the government military offices to have easy access to the classified work being done by various military agencies; thus was born the classified storage and retrieval system. I had orders to destroy all of the classified documents after 10 years. As a consequence, when I moved, the safe moved with me. Al Fluery was our designated contact man for the Air Force, and he had an access terminal that only he could touch. This was the reason Dick mentioned [in Part 5, January 2004] that "Pappy" Al Fluery would disappear into the basement when certain information was requested by our military customers. I did dutifully destroy all of the classified material after 10 years and breathed a sigh of relief to be shed of the burden. I can indeed verify that Mac [Jim McSweeney] was our next-door neighbor at the time the Mead purchase of Data Corp. was initiated. I invited him to see our facility and we started negotiations about a week later. v Paul Beck on Data Corporation and the Newspaper Business I agree with Dick Giering's comments about Bill Gorog — an insightful businessman. I think of Dick as a technical genius and even a hard working sales person. He was a "driven" man — and very difficult to manage! In comparison, I am a detail-oriented person rather than the visionaries that Dick and Bill Gorog were. For example, I could still tell you the make, model, and speed of the acoustic coupler and terminals we used. I started at Data Corporation about September 1966, but worked on other reconnaissance-oriented projects until maybe September 1968, when they needed a Recon Central program manager. I knew almost nothing about computers, but I did know the Air Force Recon Central funding source, Mr. Bob Roalef, who had been my boss when I started in 1956 as an Air Force second lieutenant. Bob did not know computers well, but he wanted to do recon research better, and if the Recon Central system would help him do that, he was willing to spend some USAF dollars. I acted as a "translator" between Bob Roalef and Dick Giering and the programming staff, trying to tell them what Bob wanted. I consider myself very "visual," so I could comment on what the programmers were producing vs. what I thought the USAF and Bob Roalef wanted. Later, I was on the team that developed the custom terminal for Lexis. Function keys, as we know them today (F1, F2, etc.), mean nothing to an ordinary human being, so developing each key engraved with its specific function in the Lexis world (i.e., Page Forward, Page Backward, Case Forward, Modify Search, etc.) was so much fun. I have detested Function keys for about 30 years now; they are a "nonhuman interface." When Mead decided to pull the NewsLib business back into its court, we at Mead Tech Labs were three-quarters of the way toward installing a system at the Philadelphia Inquirer. I believe that Mead paid the Inquirer to get out of the contract, because Mead Data Central didn't want to be in the installation business. I stayed for awhile servicing the closeout of The Boston Globe service bureau contract, but I was tainted by my earlier association with Dick Giering, and I was soon invited to find a new job for myself. Who's Who: Key People Mentioned in This Installment Collins, George M., Jr. — 1939-1988: member of The Boston Globe staff. 1973-1980: newspaper's librarian. Died in 1998. Gorog, William F. — Executive Vice President, Data Corporation, 1956-1963. Chairman, Chief Executive Officer, Data Corporation, 1963-1975. Vice President, Mead Corporation, 1972-1975. Advisor to Gerald Ford, 1975-1976. Died in 2002. Rubin, Jerome S. (Jerry) — 1970-1971: Executive Vice President, Mead Data Central, Inc. 1973-1981: President, Mead Data Central, Inc. 1983-1990: Group Vice President, Times Mirror. 1990-1992: Group Chairman, Times Mirror Co. Introduced LEXIS and the National Automated Accounting Research Service (NAARS), a tax database from the American Institute of Certified Public Accountants. 1985: Information Industry Association's "Hall of Fame" Award for creating LEXIS and recognizing his "pioneering efforts, inspiration, and determination to create, develop, and bring to fruition a major trend setting business servicing professional information needs." Wilson, H. Donald — February 1970: First president, Mead Data Central, Inc., which was created as a subsidiary of Mead Corp. Preston, James — Ohio State Bar Association President, 1965. With William Harrington, developed OBAR. Further Reading "ACM Fact Sheet," http://www.acm.org/announcements/fact_sheet.html. "Ballistic Missile Organization," http://www.globalsecurity.org/wmd/agency/bmo.htm. Bourne, Charles P. and Trudi Bellardo Hahn, "Computer Searching for the Legal Profession: Data Corporation, OBAR, Mead Data Central, 1964-1972," in A History of Online Information Services, 1963-1976, Cambridge, MA: The MIT Press, 2003, pages 229-257, 300-304. "Historical Sketch of Los Angeles AFB and Fort MacArthur," http://www.losangeles.af.mil/ABG/history.htm. "LexisNexis: Celebrating Innovation, 1973 ­ 2003," http://www.lexisnexis.com/anniversary/. "The LexisNexis Timeline," http://www.lexisnexis.com/anniversary/30th_timeline_fulltxt.pdf. Marquis Who's Who (Dialog File 234): Giering, Richard Herbert; Rubin, Jerome Sanford. Long, Tom. "Obituary: George Collins, Jr., Newsman Who Served Globe in Many Roles; At 75," The Boston Globe, March 2, 1998, p. A7. "PDP-11: Dedicated to Preserving the History and Legacy of the PDP-11 Series of 16-Bit Minicomputers Produced by Digital Equipment Corporation from 1970 to 1990, and by Mentec from 1994 onwards," http://www.pdp11.org/. Provenzano, Dominic. "Where Are They Now?," ONLINE (Special 10th Anniversary Issue), vol. 11, no. 1, January 1987, pp. 41-42. "The Story of Xerography," http://www.xerox.com/innovation/Storyofxerography.pdf. "WhatIs.com," http://whatis.techtarget.com/ (dictionary of information technology words).

Stephanie C. Ardito is the principal of Ardito Information & Research, Inc., a full-service information firm based in Wilmington, Delaware. Her e-mail address is sardito@ardito.com.