Robert Morris

[completed 2018-05-29]

Bob Morris was involved in the birth of the TeX Users Group in 1980, and he led state-of-the-art digital typography research at University of Massachusetts Boston.

 

 

Dave Walden, interviewer:     Please tell me a bit about your background.

Robert Morris, interviewee:     During and shortly after WW2 my father worked as a radio electronics engineer at the U.S Army Signal Laboratory at Fort Monmouth, N.J. Around 1952, Senator Joseph McCarthy began his anti-Communist witch-hunt, moving it from targeting the Hollywood movie industry to targeting the civilian employees at Fort Monmouth and nearby labs. My father lost his security clearance and got little work as a radio engineer until McCarthy was censured by the Senate and several years later my father was re-employed at Fort Monmouth, but only after several years of court action almost all the way to the U.S. Supreme Court.

DW:     Please say a little bit more about your high school years.

RM:     In suburban high schools it was common to have three study tracks: one for college bound, one for “industrial arts” and one for “home economics”. It's historically unsurprising that in that era, the industrial arts students were boys largely from low-income families, the home economics students mainly girls, and the college bound students were boys and high achieving girls. Red Bank is in Monmouth County which in the early 1960s was seeing an explosion of middle class housing development. Behind my house was a peach orchard, which was razed and turned into such developments. The college bound science track was a year each of General Science, Biology, Chemistry, and Physics. The General Science curriculum covered little more than I taught myself from Scientific American so my parents argued — at length — that I should instead be allowed to take the first industrial arts course, where I learned woodworking with electric tools, notably saws and lathes. Rough counting of pictures in the high school yearbook reveals that of the entire faculty, only two were black but about 75 students were black.

Some of the teachers were innovative. Marion Olson, in her first year as a Chemistry teacher, put half a dozen of us in a separate room and let us go through the text at our own pace, returning to the chemistry lab on days when lab work was to be done. At least one of those, Hugh Wilson, I am still in touch with. In fact, Hugh is one of the world's leading vision research scientists. In 1989 Hugh hosted me for the fall semester in his visual psychophysics lab then at the University of Chicago. There I learned how to design and implement experiments that I could apply to reading under varying visual conditions.

The following is from the 1961 Red Bank High School yearbook Mathematics staff page.

The Atomic Age has clearly indicated the importance of science in our modern world, but science without mathematics doesn't exist. Therefore, Red Bank High sees to it that each of its students acquires a math background.

In addition to Algebra, Geometry, Trigonometry and General Math, a new extra-curricular class in Modern Mathematics was instituted and instructed by Miss Mary Larsen. This class dealt primarily with the mathematics used when working with computers.

Throughout the year, capable students participated in mathematics contests and events sponsored by New York University, Rutgers University and various New Jersey high schools.

I don't recall any extra-curricular class, but I would have been fully occupied with my after school activities with the school newspaper. In Red Bank High School, I rose through the editorial ranks of the student paper to Editor. We got galley proofs from a local printer produced from his Linotype machine, corrected the galley proofs and did a page layout. Then we walked to the printer and watched him put the Linotype lead slugs in the page layout, and we reviewed a page proof.

I did some searching for "Modern Mathematics" and got — but lost — that in 1961 NSF was starting to provide material for high schools to prepare students for the math needed for computing. Ironically(?) in the summer of 1962 I was a paper grader at Reed on an NSF grant that was helping woefully outdated secondary school math teachers who were catching up to the “new math”. Basically, they were learning that far more than geometry is subject to proof.

DW:     What happened after high school?

RM:     When I graduated from Red Bank High School I went to Reed College, where my brother preceded me by 5 years. My intent was to major in physics or math. However the physics course was dry, traditional, and rigid, whereas the first math course was taught from Joe Roberts' book The Real Number System in an Algebraic Setting. This went through topology in one dimension, from an axiomatic point of view. I was captivated and never looked back. Besides my studies, two things occupied me: playing Go at least once a day and bicycling more than enough to meet the weekly exercise required of all students. On sunny days I could be found on the steps of the Student Union helping people adjust or repair their bikes. But at Reed I also was on the editorial staff of the student newspaper, focused on — again — shepherding the copy through the hot-type commercial print shop.

In my junior year at Reed (1963–64) I applied for and won the only study abroad opportunity. This was in England at Keele University. Keele was a young university modeling itself after Swarthmore and Reed colleges, quite unlike traditional British universities. During term break I had a Cinelli bicycle made in Milan and rode it across France, then by train back to Keele for the rest of the academic year. In the summer, three friends from Reed and I traveled around Europe in an MGB bought by one of us, Celia Hansen, who a year later became my wife and has been ever since.

In 1965, my senior year, Reed's Physics department installed an IBM 1620 which did its arithmetic by table lookup, with addition and multiplication tables located around memory location 400. We experimented by loading tables for finite fields of primes 2, 3, 5, or 7.

On the 1620, Len Shapiro (class of '65), Doug Lind and I extended and corrected the obscure “table of Fibonacci entry points” (a function describing the prime divisors of the Fibonacci numbers) that apparently was published as a pamphlet by the Fibonacci Association of America (see fq.math.ca/Books/Complete/entrypoints1.pdf). Doug was then an undergraduate at the University of Virginia who spent the summer at Reed on an REU (Research Experience for Undergraduates), and is now a prominent algebraist retired from the University of Washington. Len is a computer scientist retired from Portland State University. As for me, the last mathematics I published was in 1980, but my user name on Flickr is “Recovering Algebraist”.

Reed was an inescapable hotbed of calligraphy; however, while there I didn't participate. That said, in later years I began collaboration with Chuck Bigelow about letter forms and whether shape impacted legibility.

DW:     And after Reed?

RM:     By then I was unambiguously headed for grad school in math at Cornell and completed my PhD in algebra under Alex Rosenberg. I was supposed to finish in 1969 but found an error in one of my key theorems. I had already accepted a position at SUNY Albany. This was at the rank of Instructor pending final defense of the thesis. Alas, Rosenberg went on sabbatical to Los Angeles, so during Christmas vacation at SUNY I went to Los Angeles for several weeks. Every evening I spent 4 hours recrafting the thesis. In the morning I discussed what I had done for several hours. After lunch, we spent several hours discussing the morning's discussion. Then back to the evening “homework”. My thesis defense was presented as an invited talk at Queen's University in Kingston, Ontario, and after approval by my committee the degree was awarded in June 1970. Thereupon I was promoted to Assistant Professor at SUNY and spent 4 more years there before taking leave from 1973–75 at the School of Mathematics in the Institute for Advanced Study (IAS) in Princeton,

In Princeton at IAS, was, I believe, the first time I met Dick Palais. By 1980, just at the time he was organizing the effort of the American Mathematical Society (AMS) in what was to be the use of TeX by AMS for its own journals. Around then I began to be more interested in digital typography than algebra and I suppose that is when Dick recruited me to the effort that led to AMS-TeX as well as the birth of TUG.

During the summer that I joined the AMS-TeX birth effort, TeX was being rewritten from Pascal. David Fuchs, a student of Knuth, crashed the Pascal compiler by omitting the handling of a case which seemed silly. I clearly remember him moaning, “Why would anybody ever do that?” I jumped on David saying “Why? Because they can, and your code must handle it.” From that time on, in coding of my students, colleagues, and anyone's whose code I use, including my own, I refer to any form of David's moan as The Forbidden Question.

DW:     It appears that you were involved in the organization's governance for the first few years: Board member, 1980–1982; Secretary, 1980–1982; Finance Committee, 1982; VAX/Unix Site Coordinator, 1982; and 8 reports in TUGboat. Then you pulled back from TUG governance.

RM:     Interleaf was founded in 1981, and for two years I was on leave to Interleaf full time from UMASS Boston for Interleaf's birth; I was the fourth (some argue the fifth) employee of Interleaf. This left me with neither time nor need for TeX or its advancement.

My role in the early days at Interleaf focused on two graphics pieces. One was a chart-making system (pie charts, bar charts, etc.). The second was an algorithm, essentially based on elementary calculus, which Bob Watkins (another early 'Leafer) used with which his Interleaf interactive line art subsystem could rotate ellipses dragged by the mouse. My algorithm let Watkins' stuff drag and rotate ellipses with major axes at an arbitrary angle. My vague recollection is that for ten or more years, no competition could drag and rotate line art in real time like Watkins' graphics could.

There was a bit of a connection with TeX in that early Interleaf era. Mark Dionne, who was one of the main programmers at Interleaf working on text formatting, has reminded me that originally Interleaf's line breaking was “simple and obvious”. Then he added an option of using a line-breaking algorithm close to what Knuth used in TeX. Mark also thinks he initially used a simple hyphenation algorithm based on a paper by Knuth (perhaps the TeX78 algorithm), but then “fairly quickly switched to a dictionary from a third party when we realized how hard it would be to maintain algorithms in multiple languages, and we wanted spell-checking”.

There was a speed issue for paragraph line breaking. Interleaf ran on early-days desktop Sun workstations and DEC workstations with approximately 1 MB/sec CPUs; we sought to have paragraph line changes to take place in 250 ms. Normal human vision cannot notice a pattern change that is quicker than that speed.

DW:     Have you read Tracy Kidder's book about Paul English? Both UMB and Interleaf get a lot of mention in that.

RM:     Yes, I've read Kidder's book. I met with him several times. See page 102 about getting Paul recommended to Interleaf. In the years that followed, as Paul's career grew through Interleaf, Intuit, and Kayak, I recommended only a small handful of really strong UMB students. If I recall correctly, he hired each of them.

DW:     I'd like to go back to how you got from IAS to UMB.

RM:     When I finished my stint at IAS I felt that returning to SUNY Albany had small chance of tenure, not because of qualification, but rather because several dozen algebraists there were already tenured. The Math department at University of Oklahoma (UOK) was building a talented new group and I accepted their offer as an untenured Associate Professor in 1975. I had good NSF funding while there, but I also began some work with a Computer Scientist named John C. Thompson and his student Mike O'Dell, who did answer one email. He recalled that I introduced them to Category Theory, which advanced a project of theirs, and we began to work weekly on their problem.

Meanwhile, the UOK sent a clear message that if I didn't publish a few more algebra papers, tenure would be unlikely in the Math Department. By 1978, my wife and I decided that each of us having grown up on coasts — she in Seattle and I on the Jersey shore — we should move. At that point good luck presented itself. I organized at UOK a conference on Umbral Calculus and Hopf Algebras with principal speakers Gian-Carlo Rota and Moss Sweedler. At the time, Sweedler's 1969 book of the title Hopf Algebra was the only general book on the subject. Sweedler came to Cornell as — I think — an Assistant Professor. At the same time, Bodo Pareigis visited Cornell from Munich. With Pareigis I subsequently published several papers. Both connections to Sweedler were vagely connected to modern Galois Theory. Sweedler's book was published by W.A. Benjamin in their Lecture Notes Series of books printed from typescript. Sweedler autographed the book:

The course notes were taken
by Robert Morris whom I
heartily thank. The manuscript
was typed by Celia Morris
whom I heartily thank.
Heartily and Thankfully.
Moss

When I moved to Boston I had numerous social and mathematical contacts with Rota. Under the influence of the 1978 conference's talks and sessions, I submitted and had accepted Frobenius Endomorphism in Umbral Calculus, MIT Studies in Applied Math, 62, 1980, 85–92. This journal was edited by Rota. The paper was heavily influenced not only by the 1978 Conference but also by the Princeton-Harvard-Paris algebraic geometry that I had absorbed at IAS before I came to UOK (cf. https://en.wikipedia.org/wiki/Algebraic_geometry#20th_century). This became the last mathematics paper I published.

DW:     I learned through Chuck Bigelow that you have collaborated with noted vision researcher Gordon Legge. Please tell me a little about that.

RM:     I believe I met Gordon Legge when Hugh Wilson introduced me to some low-vision psychophysics researchers in New York. Legge himself has very low vision, and he is renowned for research in low vision reading. In recent years Bigelow and Legge have been collaborating. The last paper I published about typography was in 2002, with Bigelow and the New York collaborators: Robert A. Morris, Kathy Aquilante, Dean Yager, and Charles Bigelow, Serifs Slow RSVP Reading at Very Small Sizes, but Don't Matter at Larger Sizes, doi.org/10.1889/1.1830242.

DW:     Please say a little bit more about some of the typography research and experiments you did at UMB. I understand they dealt with spectral analysis of different fonts, legibility of grayscale type, and so on.

RM:     Here are some of the papers we wrote:

DW:     You also were a key organizer of the Boston conference on Raster Imaging and Digital Typography II. How did that come about?

RM:     If my recollection is correct, it was that UMB was early into its building itself into a research university, especially in the sciences. Consequently, it was low cost and near a low cost hotel, but convenient to transit.

DW:     I am curious about your thoughts on the continued relevance of the research on raster imaging that was presented at the four RIDT conferences. For instance, have high resolution displays (or some other newer technology) superseded what was being studied circa 1991?

RM:     My guess is yes. Healthy adult human vision can resolve 1 minute of visual angle separating a pair of vertical or horizontal lines but less than that at 45 degrees (informally, retinal resolution of 60 pixels per degree of separation). Chuck and Gordon would know the formal current status and its impact on reading (see http://jov.arvojournals.org/article.aspx?articleid=2191906).

For what it's worth, the issue may be more interesting for virtual reality goggles.

RM:     In an email correspondence several months before we began this interview, you mentioned the four PROTEXT conferences of 1984–87 to me. Do you have a thought on the importance of those meetings and proceedings in the long term scheme of things?

RM:     In one sense, the PROTEXT conferences were ahead of their time. But at http://eprints.nottingham.ac.uk/186/1/compjpaper.pdf is a copy of a paper by David Brailsford (University of Nottingham) and Richard Beach (Xerox PARC) about the journal Electronic Publishing — Origination, Dissemination and Design, also known as “EP-odd”. I recall that the community attending PROTEXT and the other, few document processing conferences, often published in EP-odd and probably EP-odd's papers had more visibility than PROTEXT, if only because John Wiley Ltd. was involved. My guess is that the life of EP-odd was longer than that of PROTEXT. Aside: I recall Brailsford trying — and failing — to use an early Apple handheld gizmo for producing docs.

DW:     Since you were at UMB (or perhaps beginning there), you have been involved with FilteredPush, Kurator, etc. Will you please say a few words about those.

RM:     Roughly, my 10 years of successful NSF funding in the Electronic Field Guide project (EFG, http://efg.cs.umb.edu/efg/, with UMB biologist Rob Stevenson) was the key. Around 2006, two biologists then at the Harvard Museums (the Harvard University Herbaria (HUH) and the Museum of Comparative Zoology (MCZ)) made an attractive proposal to NSF about managing data about museum specimens. They were turned down with the comment that their software engineering was too naive. The NSF program officer suggested that I help them with a more sophistcated proposal. The result was a two year grant beginning in 2007 with me as Subcontract Principal Investigator, on the grant titled “Filtered Push: Community knowledge and quality control in Biodiversity Informatics: A model for more efficient data capture via a distributed Herbarium network”. That expired coincident when I retired from UMB, so I was hired part-time at HUH on further FilteredPush work and then the Kurator project. I still contribute to that, but am no longer compensated. I've been working on a manuscript about FilteredPush, the first draft of which was in pdfLaTeX.

As a consultant to Biodiversity Informatics research at the University of Arizona, mathematical graph theory entered my research world. Once again, I think of myself as a Recovering Algebraist.

DW:     Thank you very much for participating in this interview. Now I need to go read several of your papers.


Interview pages regenerated June 28, 2018;
TUG home page; contact webmaster; (via DuckDuckGo)