Dr. Alan Porter oral history interview, 2018-11-15

HAGENMAIER: The following interview is conducted as part of the Georgia Institute of Technology's retroTECH Software Preservation Oral History project. Today is November 15th, 2018. The interview is taking place in the retroTECH Lab in the Price Gilbert Library. The interviewer is Wendy Hagenmaier. The interviewee is Dr. Alan Porter, Professor Emeritus of Industrial and Systems Engineering and Public Policy at Georgia Tech. Thank you for participating in the project.

PORTER: My pleasure.

HAGENMAIER: We'll start with just a little bit of background. Um, can you tell me where you were born and where you grew up?

PORTER: Born in New Jersey, moved to Atlanta at about three years old, back to Philadelphia at about eight years old, and lived there until the middle of high school, when I went out to Los Angeles.

HAGENMAIER: And then can you tell me about your, your education pathway and sort 00:01:00of how that led you, uh, here at Georgia Tech?

PORTER: Okay. Um, my undergraduate degree is in Chemical Engineering from Caltech. Senior year, Uh, I got intrigued in a psychology class, uh, split brain research and stuff. Went and got a Ph.D. at UCLA in Psych, uh, meandered all over that field, ended up in human factors as a degree, research which today would be neuroscience. And then went off and did technology assessment as a post doc at the University of Washington. And from that, the Sloan Foundation had given about a dozen colleges funding to try to get some public policy, social science connections into engineering. And I was one of the folks who was, I was brought here to do some of that.

HAGENMAIER: Okay. Here to Georgia Tech.

PORTER: Yeah.

HAGENMAIER: Um, apart from your educational pathway, what would you say are some 00:02:00of your interests or hobbies?

PORTER: Well, non-academic stuff. Uh, playing tennis and playing poker are probably at the top.

HAGENMAIER: Um, can you just tell me about, um, one significant person or event that has influenced, uh, who you are today? Just anything that comes to mind doesn't have to be the answer.

PORTER: Um, probably important father figure was my father. Um, um, very kind and um, dedicated person.

HAGENMAIER: What did he study? Did He, uh, have, what was his career or what did he study?

PORTER: Eelectrical engineer, worked for bell labs, um, became a field rep and that's how we bounced around Atlanta and back to Philadelphia.

HAGENMAIER: Great. Um, so you talked a little bit about how you ended up here at Georgia Tech. Um, what specific programs or units or initiatives, um, have you 00:03:00been involved in during your time here? I know it's probably quite a few.

PORTER: Well, the interest that it sparked for me as a Grad student, just coming upon an article, kind of bridging some of the, the, the training and backgrounds, but loosely was technology assessment. So that's trying to understand the potential impacts of new technology as it's developing. So good and bad impacts to watch out and do something in advance about that. So that's been kind of a, a bit of a unifying string of, of my interest in research here and I started in Industrial Engineering in, uh, January 1975. The, my close, closest colleagues were probably social science, which then became a part of it became Public Policy in the early nineties.

HAGENMAIER: Yeah. And in terms of, um, you've been involved in different, uh, sort of interdisciplinary research projects here at tech as well, right?

PORTER: Yeah.

HAGENMAIER: Yeah. What are some of those projects called?

PORTER: Well we've had an enduring interest for about 15, well, 15 years now, but also dating back to the late seventies, early eighties, in studying the processes of interdisciplinary research. So just I was filling in today in a class on bibliometrics in Public Policy, talking about how our measures of interdisciplinarity in terms of research based on the diversity of references cited, calculating from that and then doing some mapping to show how that work splatters across, uh, a map of, uh, different subfields across all the of the 00:05:00sciences. There's a bit.

HAGENMAIER: So let's talk about the, uh, the Knowbot. So the Technology Opportunities Analysis Knowbot. Um, what motivated you to, um, create this software or where were you when the idea for this or the need for this arose? What were you working on and how did this come about?

PORTER: Okay. I think the date is about 1992, and we had begun with a filing cabinet, collecting uh, bits on different emerging technologies, little pieces of paper. And then the light bulb went on saying, that's a tough way to go, much more efficient to tap into electronic information resources, things like, uh, databases such as Web of Science, which Georgia Tech licenses. And I was using 00:06:00it as an example in class today. So software that would help one do a search in these global databases and pull down research on a topic. Um, um, one example we've used, it comes to mind because we talked about it today, is nanotechnology-enabled drug delivery. So do a search on terms relating to that, download, uh, from the Web of Science to local, uh, desktop computer, Uh, I think our number there is on the order of 50,000 abstract records, and then software to help analyze that. So basically stack up your 50,000 records and look across them. So who has authored the most, uh, what years, what's the pattern over time? Is it accelerating? Where is it taking place? U.S. or places like China and so forth. So answering the basic reporters' questions, who, what, 00:07:00where, when, uh, with an eye toward how and why. But those are kind of tougher. So the software was to help analyze these field-structured abstract records.

HAGENMAIER: And what was the process of creating this software like? Who did you work with? How was it actually built?

PORTER: Um, the start was summertime, I believe, 1992, when my son, a computer science undergrad at Virginia Tech was home for the summer. So we talked about this a bit, and he wrote the original coding to help, you know, go through a stack of these records and make essentially lists of the leading authors, the terms that are appearing and so forth. That got blended in with some other ideas, with colleagues at a small company, Search Technology Incorporated, which 00:08:00had been founded by two folks who were faculty at Tech [Bill Rouse and Russ Hunt]. So put together a proposal and got funded by DARPA, Defense Advanced Research Project Agency, for an STTR [Small Business Technology Transfer] award. I forget what the acronym stands for, but it's small business program, but with a university partner and a small business, they gave us a little bit of money, six months of support. We delivered our results, and it's a grant, so it's not a deliverable to the such and they evidenced zero interest. Uh, we managed to make a presentation at a Tech Transfer Society meeting and a fellow in the Army said, I think some of that looks real interesting. So he went back to DARPA, uh, and with somebody who's interested, it was, oh, that sounds pretty good. Where'd it 00:09:00come from? It was you guys. Oh. And then they co-funded the more substantial stages of STTR support through phase one, two, and three. And that went probably about '92 to '99, thereabouts.

HAGENMAIER: And so the focus at that time, what types of data would be mined? Um, for the military?

PORTER: It wasn't so much for the military. Um, it was from the get go, intended to be open, uh, with the small STTR to commercialize. But in fact, early users, uh, were the military. Uh, A particularly notable one was Edison House, which is a, just a physical location in London, England with Army Air Force and Navy representatives. Essentially their role was to be matchmakers, some research 00:10:00going on in the U.S., probably with some military connections, some issues, look for folks in Europe working on it, that you might get them together at some point. So this was a tool to help screen. So this topic, who in European countries is working on it and get some candidates to go knock on the door and check out their interest.

HAGENMAIER: Um, if you just can think back when you were first using the software, do you remember anything about what kind of hardware you would have been using it on or what did it look like at the time?

PORTER: My memory is sloppy on that. Uh, the software actually has been re-coded over time. The lead programmer through the whole thing has been my son Doug Porter, who continues to work full time on it for Search Technology now. It's 00:11:00migrated, it's now written in C++. I remember an era when it was a Unix side, we had to split our desktop from the Windows world to the UNIX world, Linux operating system. But I don't remember much about the coding and cause I was never, I was the user, not the the doer.

HAGENMAIER: Right, right. You're defining the requirements for the software.

PORTER: Yeah.

HAGENMAIER: Um, so, um, so can you tell me a little bit about, you know, you've talked about some of your collaborators, um, and the sort of public-private partnership for the grant funding. Um, so the software eventually became just a private, a private product or private enterprise or was did this sort of public research side continue with the software as well? Um, on the Georgia tech campus 00:12:00or elsewhere?

PORTER: Kind of yes. To all of that. Yeah. The software, uh, was licensed by Georgia Tech to Search Technology. So they market it, they continue to develop it. Uh, there's two commercial versions. There was a government version, Tech Oasis, um, that was maintained for about 10 years. And then, uh, no more. So there's the two commercial versions. One is VantagePoint that Search Technology, uh, develops and works on, Georgia Tech uses freely. It's on the, the OIT shelf to be downloaded for academic and scholarly use. Um, and we continue to do. And the other commercial version is Derwent Data Analyzer, which is, uh, marketed 00:13:00by, uh, Clarivate formerly was Thomson Reuters. It spun out. So royalties come back from Clarivate to Search to Georgia Tech. So there's actually a little pass back to Tech, and there's ongoing research. We have a project right now to Search with Georgia Tech ss the major sub on it. So about an almost a 50, 50 type deal to develop indicators of technological emergence with these sorts of research publication or patent data and using the software in the process, develop new scripts and capabilities to enhance the software. It's a nice byproduct.

HAGENMAIER: So who have you seen, um, use this at Georgia Tech? What kind of research questions would they bring to the product? Would it be, is it largely 00:14:00R&D type questions? Um, like you've discussed or, um, yeah, what kind of research questions?

PORTER: Yeah, so like R&D or R&D management or policy. So it's kind of a spectrum there, but it's take a look at some technical area. Um, particularly on the academic side, what we would do, so one that we've been following quite a bit is nanotechnology. So we update our search. How do we get related records now up to from Web of Science, for example, about 2.2 million abstract records. And then we, uh, identify hot areas of research. So a researcher might take a look and say, oh, let me see if I want to orient my research that way. Um, a funding agency, like NSF might take a look to see what seems to be emerging topics to maybe emphasize funding and those directions. It just triggers the 00:15:00thought that, um, we've had collaboration with some folks in China. Uh, one fellow visited Georgia Tech in the 1990s, uh, Professor [name?], a lot of contributions into the software, comes to mind because just the last week we've been trying to track down exactly what algorithm did we put in to do one of these statistical routines called PCA, Principal Components Analysis and tapping [name?'s] memory as best he can. And then looking for a CD that we read yesterday and then found some paper and we're hoping to get some answers to that in the paper. But it's been, uh, an ongoing routine. On the commercial side, and most of the clients for the software are large companies doing competitive technical intelligence. So there, the simplified model would be Ford Motor 00:16:00Company takes, keeps track of what Toyota's doing with say fuel cells, can look at, who's doing inventions, if they publish, if they're publishing articles, if there's teaming and collaboration to look for potential competition or collaboration.

HAGENMAIER: This software has an archive of its own sort of like you were talking about, you know, the memory that everyone has as well as the documentation. Yeah. And the name changes through the years. Yeah.

PORTER: More like an accidental partial, oh, we have something archive.

HAGENMAIER: Yes. Yeah. Um, do you remember at the time that this idea was coming into being, was there anything like it out there? Um, or was there any competitive environment for a product like this?

PORTER: Um, I think relatively novel, but depending on how you define, like it, 00:17:00uh, in the broadest sense of trying to extract usable intelligence from these collections of metadata information, abstracts on records and so on, uh, sure. There was some software coming out of France, uh, kind of doing that, um, general tools that people could use. But then we commercially, the primary marketplace has been people doing patent analysis and so we'll be a family of kind of competitors who see each other, at conferences and so forth. So depending what scope you look at, um, on the academic side, a different collection of tools looking toward open source, which is tough competition when you're commercial. Um, and then on the commercial side toward the patent analysis.

HAGENMAIER: Um, so you've talked a little bit about the, the prime customers, um, on the sort of corporate side, being large corporations, like say a car company, um, on the academic side really quite broad. Um, but especially researchers say in public policy or, um, technology trends or R&D. Are there any other customers that, or users, end users that, um, you've seen through the years?

PORTER: Um, good question. Um, it's very uneven. So, at Georgia Tech, clearly the main use has been science and technology, policy analysis type folks. Um, with some others, little outposts kind of pop up, kind of like a mushroom 00:19:00growing somewhere. Um, globally, there's two countries where academic uses is way highest, not the U.S. Brazil would be way up there. Some consortia that have spread the word with innovation interests to technology transfer, technology, um, management issues. China, the lead organization has been Chinese Academy of Sciences, which is a strong user and it's got like over a hundred institutes, the library in particular, but doing things like just use it to help justify a new proposal by showing the landscape of the research profile of who's doing what, where and why we think it's an opportunity. So those come to mind.

HAGENMAIER: So I guess, how has the software been received by users? Uh, you 00:20:00know, originally how was it received? Have you, what has the process been like of I guess getting user feedback and sort of iterative, iteratively changing the software to meet the needs?

PORTER: Um, from the start deciding how to price and so forth has been an issue. We ended up latching onto a price, and it's been the same price for 20 years, going to change next year, the model, but a lifetime license for a set fee and then a yearly renewal. Uh, um, so you could keep using it without support perpetually. But if you wanted support, there was an annual fee that was about 40% of the purchase price and that was to get really amazing tech support. So 00:21:00the idea of you've got some issue, you're trying to use the tool to do x and y and trying to give you serious help within a day. So that's been probably our, our key strength and we would get good user feedback favorable on that.

HAGENMAIER: Yeah, that's quick, quick support for sure. Um, so, um, obviously the software still exists today for sure and going strong. Would you say that this, um, software has inspired other software, either that you've seen grow up here at Georgia Tech or elsewhere in the world? Certainly, you hear so much about mining big data and being able to analyze data sets that are too large for us to read. It seems like this idea has, um, inspired or been part of a movement of other ideas as well.

PORTER: Yeah, and we've been doing this for quite a while before the big data boom of whatever, 2008 or so onward. Um, and we're probably more in that kind of large data domain, so not huge volumes of real time data, but tens of thousands of records, too much to read, uh, sensibly. So, so that's been good. The inspiration unfortunately is toward competitors. Uh, I can think of one out of India that we can recognize a lot of features and so forth. So that's challenging. And as a small company, that whole world is, remains challenging. We've also seen a lot of academic spin-outs into this, let's do stuff, uh, get some value-added from information resources and we've outlived a lot of them. 00:23:00Their life has tended to be, in the ones we've come across, pretty short.

HAGENMAIER: So this is, makes this project stand out as one that has been going for so long from that, that origin. Yeah. Yeah. Um, what do you see as the road ahead for VantagePoint? Um, and the, the Knowbot as it exists today?

PORTER: Well, back in the Knowbot days, there were two versions, a shorthand with TOAK (Technology Opportunities Analysis Knowbot) and TOAS (TOA System). One was Search side [TOAS], one was the Georgia Tech side [TOAK]. Um, with the Georgia Tech side, trying to retain, you know, let's say we're trying to cluster 00:24:00different terms to make topics to include several clustering routines and so forth. The TOAS side, tending to be leaner and trying to make for a simplified interface so it's easier to use. That's the one that flipped into vantage point and Derwent Data Analyzer. TOAK side, unfortunately was killed, um, basically by, uh, Georgia Tech concerns, over conflicts of interest, unfortunately. Yeah, Georgia Tech continues to use VantagePoint,

HAGENMAIER: Which is the TOAS side?

PORTER: Yes.

HAGENMAIER: Yeah. Um, do you know if the original code still exists anywhere, the documentation for it, or I don't know if it would have been on stored on 00:25:00floppy disk at any point or

PORTER: I don't know.

HAGENMAIER: Yeah. Um, how would you say, just sort of looking back, um, how the creation of the software and its, um, evolution has affected your growth as an academic or, um, your career?

PORTER: Um, it certainly the prime part of my, incorporated into most all of my research, and kind of a strange situation. I retired at the start of 2002 or semi-retired, so not teaching and just part time, hourly, if we have support on a project like from NSF typically, uh, and most of that is VantagePoint-based. In terms of productivity, I haven't really counted. But in terms of articles published and so on, definitely the rate has gone up since retirement. Um, and 00:26:00it's using the tool and with good colleagues.

HAGENMAIER: Um, how would you describe what the experience felt like to be at the, at the point of creation of this software? What were sort of, um, how would you characterize your, your experience of creating something new? What did it feel like?

PORTER: Um, that's a good question because it wasn't neat and clean. It was this very little small scale effort in summer, part time project by, um, an undergrad. Um, and then we got the support from DARPA and there it was a component in a program managers associate, was what the research was called. So 00:27:00it was never a crisp, clean, ah, look new tool, maybe until it was commercially introduced, which I think would be 1999 or 2000.

HAGENMAIER: Um, since, um, or in addition to working on vantage point, um, have you worked closely with or been involved in the creation of any other software programs during your time at Tech?

PORTER: Ah, nothing approaching that scale. So, no, I don't think so.

HAGENMAIER: Um, is there anything else that you'd want to share about your experience, um, here at Tech or with the software or, um, like any, um, specific, uh, times that you've used the software, um, to identify, um, or to illuminate something that you didn't expect to find or anything that comes to mind?

PORTER: I think the notions of the software, if not the software per se, or if it was, and my timeframes aren't quite sharp here enough to tell, it would have been very early stages, was to take a look at two emerging technical domains and provide a little input to the Vice President for Strategic Research at Tech and then into president Crecine at the time, so that's the era, like early nineties, so from our perspective saying, hey, look, nanotechnology really looks like it's got some great potential. And the other was looking at biomedical engineering as some real opportunities there. So we had, you know, tiny voice you're not sure how much you were heard or a factor, but two areas that Georgia Tech has really 00:29:00picked up and become prominent in. Yeah. So you, maybe our tech intelligence might've helped it a little bit.