Hermann Maurer Professor Maurer was b orn in 1941 in Vienna, Austria. He has got a Ph.D. in Math- ematics from the University of Vienna 1965. Assistant and Asso ciate Professor for Computer Science at the University of Calgary and Professor for Applied Computer Science at the University of Karlsruhe, West Germany. Since 1978 Full Professor at the Graz UniversityofTechnology. Honorary Adjunct Professor at the University of Auck- land, New Zealand since Octob er 1993. Honorary Do ctorate Polytechnical University of St. Petersburg 1992, Foreign Memb er of the Finnish Academy of Sciences 1996. Author of thirteen b o oks, over 400 scienti c contributions, and dozens of multimedia pro ducts. Editor-in-Chief of the journals J.UCS and J.NCA. Chairp erson of steering committee of WebNet and ED-MEDIA Conference series. Pro ject manager of a num- berofmultimillion-dollar undertakings including the development of a colour-graphic micro computer, a distributed CAI-system, multi-media pro jects such as \Images of Austria" Exp o'92 and Exp o'93, resp onsible for the development of the rst second generation Web system Hyp er-G, now Hyp erWave, and various electronic publishing pro jects such as the \PC Library", \Geothek" and \J.UCS" and participation in a num- b er of EU pro jects e.g., LIBERATION. Professor Maurer research and pro ject areas include: networked multimedia/hyp ermedia systems Hyp erWave; electronic publish- ing and applications to university life, exhibitions and museums, Web based learning environments; languages and their applications, data structures and their ecient use, telematic services, computer networks, computer assisted instruction, computer sup- p orted new media, and so cial implications of computers. Not only Theory When trying to summarize myover 35 years of computer science researchI end up with one ma jor very p ersonal lesson that I have learnt, and that might be helpful to some younger researchers: it is such a rewarding and pro ductive 206 Maurer exp erience to work together with other researchers, b e it students or top-notch exp erts that one must not miss it. Of two p ersons no one is ever \b etter" than the other. Rather, each one has weaknesses and strengths: the fun and challenge is to determine the right symbiosis. It to ok me a long time to nd this out. Once I had realised the full imp ortance of this for me, my life changed: research turned from hard work to hard fun. There are certainly scientists who have obtained greater insights working on their own than my brain would ever have allowed me to achieve, and I admire them. However, for me and I b elieve for many researchers the key for success is collab oration. Research can often b e compared to solving puzzles. On your own, you may easily get stuck; it is my rm b elieve that n p ersons together 2 n 5 can solvea problem more than n times faster: research results are \sup er-linear" in the number of researchers. Thus, this pap er fo cuses more on p ersons than on results. It is a thank you to all who haveworked with me and who have b ecome friends one way or another. I have made an attempt to mention all those I haveever co-authored something with. I ap ologize to others that I have met, learnt to appreciate and who have help ed in di erentways: there have b een many, and there would not b e enough ro om to do justice to all. Before going on, let me mention one further p oint: the b o ok containing this pap er is dedicated to the theory of computer science. However, I have sp ent much time also in other areas. For completeness' sake, and since drawing b orderlines is dicult I will also rep ort on non-theory stu , alb eit shorter. The early years 59-71 When entering university in Vienna in 1959 I was set on studying applied physics. However, the rst mathematics class I attended was taughtby Professor Edmund Hlawka - a sup erb teacher and researcher. He turned me around 180 degrees: mathematics it was to b e, henceforth. And \clearly" the most b eautiful areas of all, the theory of numb ers. I to ok the only two computer science = program- ming courses available in those days in Vienna and progressed rapidly with my mathematics coursework. When I happ ened to meet Professor John Peck who later b ecame famous for e.g., his work on Algol 68 from the University of Calgary at the 2nd IFIP congress in Munichin62Iwas ready to accept his o er to go to Canada as graduate assistantforayear or two. While at Calgary, I fell in love with Canada, learnt more ab out computers and computer science and continued mywork in numb er theory on diophantine equations, i.e. equa- tions where one is interested in integer solutions, only. One of those equations 4 4 4 4 u + v = x + y is called \Euler's equation" like a lot of other equations 4 4 4 4 and the smallest non-trivial solution known in 62 was 133 + 134 = 158 +59 . It was op en whether smaller solutions exist. This seemed likeanobvious appli- cation for computers. In a four-fold lo op one would check for all quadruples of values u; v ; x; y with 1 u v 158; 1 x y 158 and u<x whether 4 4 4 4 u + v = x + y . This brute-force approach for the four-fold lo op takes over 4 8 100 =10 computational steps, to o many for the computer in use at Calgary in Not only Theory 207 1962 an IBM 1620. It was then that I discovered the p ower of sorting! Rather 8 4 4 4 than examining 10 quadruples I would calculate some n =10 values u + v for 1 u v 158 and sort them in ascending order as z ;z ;z ;:::. This requires 1 2 3 5 4 4 4 4 an e ort of n log n, i.e. roughly 10 steps. A smaller solution for u + v = x + y would exist clearly if and only if z = z for some i, a test that can b e carried i i+1 4 out in ab out 10 steps. Thus, using sorting, I could cut down the computational 8 5 e ort from some 10 to some 10 steps, quite feasible on a 1962 computer. The result sigh was negative: no non-trivial solution smaller than the one known to Euler exists. Nothing to publish, but a rst lesson for me: computers can help in numb er theory, and sorting is a surprisingly p owerful to ol in many application areas. This realisation would come in handy years later in the study of data structures and geometric algorithms . While continuing researchinnumb er theory and obtaining a few new results 2 2 on the Pellian equation integer solutions for x dy = 1 I joined the computer centre of the governmentof Saskatchewan as \system analyst" May62 - De- cemb er 62. Those short eight months of really down-to-earth computing work would proveinvaluable later for my understanding of applied computer science. Although not at all related to theory I feel the urge to rep ort two anecdotes. As rst job, I was given a huge assembly language program for an IBM 1401/1410 without further explanations and the request \read it so you understand what it do es". After twodays I was totally frustrated: after reading pages of co de I thought I knew what the initial segment of the program would do: nothing but printtwo columns of asterisks inde nitely,until someone would physically stop the printer. After a restart, 132 dashes would b e printed over and over in the same line until again the printer would b e physically stopp ed! When I rep orted this obviously wrong conclusion but I had checked it three times! to my sup er- visor he was delighted: \Yes, this is what it is supp osed to do. This is used to align the forms in the printer appropriately. And the rep eated printing of dashes creates a p erforation so the forms can be torn o easily." As it turned out, I was thrown into the middle of the rst world-wide pro ject to computerize health care. After gruelling months of work, when the team I felt I b elonged to by then had nally completed its job one day at 4 a.m., we drove out into the prairies. Someone in our group knew enough ab out farming, sawa harvesting machine and a rip e wheat eld and b efore we knew it we released our b ent-up tension by harvesting that eld. A surprised but pleased farmer invited us for a hearty breakfast 3 hours later . Back in Vienna, Austria, I continued my Ph.D. thesis work but also needed a job. Werner Kuich, a friend from my freshman years who later b ecame one of the rst computer science professors at the Technical University of Vienna help ed me to get a job with Professor Heinz Zemanek's IBM funded research group whose ro ots go back to Mailufterl, the rst Europ ean transistorised computer built by Zemanek in the late fties. In this research group I started to learn ab out compilers and wrote my rst one [2] and got interested in formal languages and formal description metho ds. I also made my rst scienti c contribution by noticing that one apparently could improve on the O log n p erformance of 208 Maurer binary searching by making use of the structure of the data by interp olating rather than brute-force halving.