## U. of Florida CS department: let it be destroyed by rising sea levels 100 years from now, not reckless administrators today

Update (4/27): A famous joke concerns an airplane delivered to the US Defense Department in the 1950s, which included a punch-card computer on board.  By regulation, the contractor had to provide a list of all the components of the plane—engine, wings, fuselage, etc.—along with the weight of each component.  One item in the list read, “Computer software: 0.0 kg.”

“That must be a mistake—it can’t weigh 0 kg!” exclaimed the government inspector.  “Here, show me where the software is.”  So the contractor pointed to a stack of punched cards.  “OK, fine,” said the government inspector.  “So just weigh those cards, and that’s the weight of the software.”

“No, sir, you don’t understand,” replied the contractor.  “The software is the holes.”

If the Abernathy saga proves anything, it’s the continuing relevance of this joke even in 2012.  Abernathy is the government inspector who hears that software weighs nothing, and concludes that it does nothing—or, at least, that whatever division is responsible for punching the holes in the cards, can simply be folded into the division that cuts the card paper into rectangles.

As many of you have heard by now, Cammy Abernathy, Dean of Engineering at the University of Florida, has targeted her school’s Computer and Information Science and Engineering (CISE) department for disembowelment: moving most faculty to other departments, and shunting any who remain into non-research positions.  Though CISE is by all accounts one of UF’s strongest engineering departments, no other department faces similar cuts, and the move comes just as UF is increasing its sports budget by more than would be saved by killing computer science. (For more, see Lance’s blog, or letters from Eric Grimson and Zvi Galil. Also, click here to add your name to the already 7000+ petitioning UF to reconsider.)

On its face, this decision seems so boneheadedly perverse that it immediately raises the suspicion that the real reasons for it, whatever they are, have not been publicly stated. The closest I could find to a comprehensible rationale came from this comment, which speculates that the UF administration might be sabotaging its CS department as a threat to the Florida State legislature: “see, keep slashing our budget, and this is the sort of thing we’ll be forced to do!”  But I don’t find that theory very plausible; UF must realize that the Republican-controlled legislature’s likely reaction would be “go ahead, knock yourselves out!”

On a personal note, my parents live part-time in beautiful Sarasota, FL, home of the Mote Marine Laboratory, which does amazing work rehabilitating dolphins, manatees, and sea turtles.  Having visited Sarasota just a few weeks ago, I can testify that, despite frequent hurricanes, a proven inability to hold democratic elections, and its reputation as a giant retirement compound, Florida has definite potential as a state.

Academic computer science as a whole will be fine.  As for Florida, may the state prove greater than its Katherine Harrises, Rick Scotts, and Cammy Abernathys.

Update: See this document for more of the backstory on Abernathy’s underhanded tactics in dismantling the UF CISE department.  Based on the evidence presented there, she really does deserve the scorn now being heaped on her by much of the academic world.

Another Update: UF’s president issued a rather mealy-mouthed statement saying that they’re going to set aside their original evisceration proposal and find a compromise, though who knows what the compromise will look like.

In another news, Greg Kuperberg posted a comment that not only says everything I was trying to say more eloquently, but also explains why I and other CS folks care so much about this issue: because what’s really at stake is the concept of Turing-universality itself.  Let me repost Greg’s comment in its entirety.

It looks like Dean Abernathy hasn’t explained herself all that well, which is not surprising if what she is doing makes no sense. Reading the tea leaves, in particular the back-story document that Scott posted, it looks like she had it in for the CS department from the beginning of her tenure as Dean at Florida. In her interview with Stanford when she had just been appointed as dean, she already said then that “we” wanted to bring EE and CS closer together, even though at the time, there had been no discussion and there was no “we”. Then during discussions with the CS department, she refused to take no for an answer, even though she sometimes pretended to, and as time went on the actual plan looked more and more punitive. She appointed an outside chair to the department, and then in the final plan she terminated the graduate program, moved half of the department to EE, and left the other half to do teaching only. The CS department was apparently very concerned about its NRC ranking, but this ranking only came out when Abernathy’s wheels were already in motion. In any case everyone knows that the NRC rankings were notoriously shabby across all disciplines and the US News rankings, although hardly deep, are much less ridiculous.

So what gives? Apparently from Abernathy’s Stanford interview, and from her actions, she simply takes computer science to be a special case of electrical engineering. Ultimately, it’s a rejection of the fundamental concept of Turing universality. In this world view, there is no such thing as an abstract computer, or at best who really cares if there is one; all that really exists is electronic devices.

Scott points out that those departments that are combined EECS are really combined in name only. This is not just empirical happenstance; it comes from Turing universality and the abstract concept of a computer. Yes, in practice modern computers are electronic. However, if someone does research in compilers, much less CS theory, then really nothing at all is said about electricity. To most people in computer science, it’s completely peripheral that computers are electronic. Nor is this just a matter of theoretical vs applied computer science. CS theory may be theoretical, but compiler research isn’t, much less other topics such as user interfaces or digital libraries.

Abernathy herself works in materials engineering and has a PhD from Stanford. I’m left wondering at what point she failed to understand, or began to misunderstand or dismiss, the abstract concept of a computer. If she were dean of letters of sciences, then I could imagine an attempt to dump half of the literature department into a department of paper and printing technology, and leave the other half only to teach grammar. It would be exactly the same mistake.

### 47 Responses to “U. of Florida CS department: let it be destroyed by rising sea levels 100 years from now, not reckless administrators today”

1. Raoul Ohio Says:

The rap I heard is that Cammy last year wanted CS to merge with some other dept to make Computer Engineering, or something like that. The CS dept voted against this plan, and now he/she is showing who is boss.

2. Scott Says:

Thanks, Raoul! I don’t know the full history (and maybe someone else can tell us more) … but in its bizarre specificity in targeting CS and nothing else, its total senselessness even from a crude cost/benefit perspective, and the continued push for it despite vocal, unanimous opposition from the bottom up, this plan certainly shows the external signs of a personal vendetta.

3. Stas Busygin Says:

At the same time they increase the athletic budget by $2 million, which is more than what they are going to save by shutting down CISE. That clearly shows where their priorities are: how Gators play the next season is all they care about. I am a UF PhD graduate myself, and I say there are better places for people who are interested in CS to study and do research. This incident just made everything obvious. 4. Lev R. Says: I am really surprised that one dean has the power to single-handedly dismantle an entire department, especially when there is such opposition to the move. I would think universities need to be structured to have safeguards that prevent such (terrible) decisions from taking effect. 5. Ankit Says: For those interested in more details, the “Radical Agenda…” post at http://saveufcise.wordpress.com/ has a list of links that gives more color about this situation. It seems that Gov. Rick Scott has also recently announced the opening of a new Florida Polytechnic University (http://www.miamiherald.com/2012/04/21/2760476/gov-rick-scott-approves-bill-creating.html). I wonder how this plays into all the state budget cuts. Apparently, Gov. Scott also sent a list of questions to all faculty about “measurable goals” and how their efforts translate into what employers need. It seems the main concern is jobs. No one would disagree that jobs are important but will this make UF graduates more or less attractive to employers? Is integrated research-and-teaching better for training future computer scientists than teaching-only? I can only judge from my own experience that working with and learning from strong researchers teaches you how to think and how to solve new problems, rather than regurgitate solutions to ones already seen. I’d be interested to hear other’s thoughts on this… 6. Adam Wolbach Says: Just an FYI, University of Florida athletics are run by a private non-profit called the University Athletic Association, which is fiscally separate from the university (so, not funded by UF). They have donated over$60 million to the university since 1990 because they can – they are one of the few profitable athletic programs in the NCAA. That part of the Salzberg article was a pretty superficial look at the situation. Here is an example from 2009 in which they gave a $6 million gift to the university to help offset budget cuts by the legislature: http://www.alligator.org/news/campus/article_73239d6b-8421-578b-93a3-673dc5a44b4c.html 7. Scott Says: Adam Wolbach: I agree. While the expansion of athletics at the same time CISE is being eviscerated constitutes a dramatic statement about priorities (why can’t programming competitions fill the stadiums? ), it’s by no means the real problem here. The real problem seems to be a combination of (1) the doofus plan to start a new university in Florida, rather than adequately supporting the existing ones, (2) the concomitant budget cuts (even while Rick Scott has banned the universities from the obvious response, raising their tuitions to make up the shortfall), (3) Cammy Abernathy’s craven exploitation of the budget problems to ramrod her harebrained agenda (she might genuinely believe that CS is superfluous if you already have computer engineering…), and (4) the inability or unwillingness of the rest of the UF administration to rein her in. 8. Adam Wolbach Says: No arguments there. 9. (^.^) Says: as a past grad student from one of the schools which had ics and eecs separate, I can tell, it is a waste of resources. Why cant people follow the MIT or Northwestern model where grad students benefit without the politics of multiple departments which does more or less same research and build on simlar course work? 10. Scott Says: (^.^): It’s a fair question, and indeed Abernathy seems to have used that as her main “argument.” But… (1) I think examination of the details shows that the situations aren’t even remotely comparable. A few schools, like MIT and Berkeley, have large and thriving EE and CS research groups, which are housed together in an “EECS” department for historical reasons, yet which are allowed to operate semi-autonomously in practice (each having its own director, its own course requirements, etc.). By contrast, it looks like UF explicitly rejected a proposal to create a similar EECS organization; instead they want to decimate CS and transfer some of the faculty over to Electrical and Computer Engineering (among other departments), if ECE will take them and if they conform their research and teaching to whatever ECE wants. It’s like the difference between setting up a combined Physics and Mechanical Engineering department—which is merely quixotic!—and making an intellectual judgment that physics is really just a branch of mechanical engineering, and should only be studied to the extent it’s useful for the latter. (2) My personal experience, at both the Berkeley and MIT EECS departments, has convinced me that EE and CS are really different fields. Yes, certainly they’re related, but no more than math is related to physics, chemistry is related to biology, etc.—so should we just merge everything into one mega-department? Indeed, I’d argue that the defining ideas of CS—abstraction from the physical substrate, universality, recursion, the casual mixing together of machines, programs, and data and of syntax and semantics—are precisely ideas that distinguish it sharply from EE. In practice, as well, there tends to be no more intellectual overlap or “duplication” between CS and EE courses than there is between, say, physics and mechanical engineering courses (neither at UF nor anywhere else). For these reasons, I would never put EE and CS together in the same department if I were doing it today. 11. Ben Says: University of Cincinnati College of Engineering (under Dean Montemagno) also tried to cut the CS department last year; half the faculty were moved to the ECE dept., and the remainder would have been used to teach calculus or something like that. It seemed like a done deal, but never actually happened – local businesses complained too loudly (or something else – I’m not really sure why it never happened). 12. (^.^) Says: Professor: Probably you are right. There are some ‘pure’ fields like analog design, control systems, physical layer communications, semiconductor and there are ‘pure’ cs fields like database theory, complexity, cryptography, programming languages. But there is still plenty of overlap. For instance, network communication, video communications, data compression/analysis, image processing, neural networks and so on. However it may be better to have them together since results of complexity is still used in VLSI or coding theory, results in cryptography influence networking and understanding of programming languages at a foundational level will help even EE students take their career to higher levels in the real-world and non-traditional applications of quantum electronics is useful in complexity (for instance if I am correct in your #P proof of permanent), Bayesian Statistics is useful in Signal Processing. Considering the overlap and interdisciplinary research that is possible, it might seem worthwhile to provide opportunity to have more streamlined system where grad students and researchers will have the opportunity to do get quality education and do good research without politics (such as contending which department is superior) and without having to change programs. Thankyou. 13. (^.^) Says: Also considering the way industry is going, it would make a lot of sense for EEs to broaden and ramp up their CS knowhow or will be left behind in many opportunities. 14. Scott Says: (^.^): By all means, the more collaboration between CS and EE the merrier! My own interest in quantum information (and recently, photonics) means that I often talk to MIT’s EE folks myself. But you can easily have such collaboration without CS and EE being the same department, let alone CS being subsumed into EE! I should add that the demand, both from employers and from undergrads, is much stronger for CS than for EE right now—which makes it truly bizarre that anyone would want to subsume CS into EE, rather than (say) vice versa! 15. Scott Says: Ben #11: There must be some disease spreading among engineering deans that makes them allergic to CS! I feel very lucky to have seen no evidence of that disease at my school… 16. Raoul Ohio Says: A lot of these same issues have played out over the last couple hundred years WRT “pure math” and “applied math” in universities. Most of the time the PM crew ran the show and hogged the resources. 17. Anonymous Says: Having spent some time at a couple of CS departments, I did develop some bias towards so called “core” CS. They are not very theoretical compared to maths, nor are they very applied compared to EE. There is in fact some benchmark, a lot of people from industry would generally prefer maths/physics/EE graduates (I mean undergrads, not PhD/postdocs or anything like that) for programming jobs, for the simple reason that most people coming out of CS schools are not reliable enough to write reasonably bug free codes. 18. Greg Kuperberg Says: It looks like Dean Abernathy hasn’t explained herself all that well, which is not surprising if what she is doing makes no sense. Reading the tea leaves, in particular the back-story document that Scott posted, it looks like she had it in for the CS department from the beginning of her tenure as Dean at Florida. In her interview with Stanford when she had just been appointed as dean, she already said then that “we” wanted to bring EE and CS closer together, even though at the time, there had been no discussion and there was no “we”. Then during discussions with the CS department, she refused to take no for an answer, even though she sometimes pretended to, and as time went on the actual plan looked more and more punitive. She appointed an outside chair to the department, and then in the final plan she terminated the graduate program, moved half of the department to EE, and left the other half to do teaching only. The CS department was apparently very concerned about its NRC ranking, but this ranking only came out when Abernathy’s wheels were already in motion. In any case everyone knows that the NRC rankings were notoriously shabby across all disciplines and the US News rankings, although hardly deep, are much less ridiculous. So what gives? Apparently from Abernathy’s Stanford interview, and from her actions, she simply takes computer science to be a special case of electrical engineering. Ultimately, it’s a rejection of the fundamental concept of Turing universality. In this world view, there is no such thing as an abstract computer, or at best who really cares if there is one; all that really exists is electronic devices. Scott points out that those departments that are combined EECS are really combined in name only. This is not just empirical happenstance; it comes from Turing universality and the abstract concept of a computer. Yes, in practice modern computers are electronic. However, if someone does research in compilers, much less CS theory, then really nothing at all is said about electricity. To most people in computer science, it’s completely peripheral that computers are electronic. Nor is this just a matter of theoretical vs applied computer science. CS theory may be theoretical, but compiler research isn’t, much less other topics such as user interfaces or digital libraries. Abernathy herself works in materials engineering and has a PhD from Stanford. I’m left wondering at what point she failed to understand, or began to misunderstand or dismiss, the abstract concept of a computer. If she were dean of letters of sciences, then I could imagine an attempt to dump half of the literature department into a department of paper and printing technology, and leave the other half only to teach grammar. It would be exactly the same mistake. 19. Lev R. Says: It looks like they’re not going to go through with their original dismantling plan: http://news.ufl.edu/2012/04/25/machen-budget-statement/ But who knows what the final organization will look like. 20. Greg Kuperberg Says: A correction to my comment if it is now more prominent: “She appointed an outside chair” should be “She imposed an interim chair”. He is an emeritus from the same department. (In any case, I thank Scott for giving the comment special placement.) 21. Raoul Ohio Says: For a quick break from this sad situation, check out the animated cartoon of how they are looking for the Higgs Boson at CERN, from “PhD Comics”: http://www.phdcomics.com/comics.php?n=1489 22. MM Says: Definitely despicable… But I think one issue is that CS is not “engineering” per-say. Not that I would consider myself an expert on engineering, but I’ve had enough people who work in other engineering fields tell me that what I do is not engineering. The argument Greg made about Turing universality only serves to underscore that assertion. This is not the perspective of someone with an engineering background. Arguably, CS is neither engineering nor science — we are a unique paradigm that is difficult to pigeon-hole into one single category. This of course explains why it is foolish to try to dissolve CS into other departments, but it also explains why someone rooted in Engineering orthodoxy (like perhaps a Dean of Engineering) might be unsympathetic to the very existence of a CS department in her domain. I imagine that a few decades ago, when CS departments first started popping up, these contentions existed with older disciplines. One would think that the objectors would have been silenced a long time ago by the tons of money that CS generates, and especially now (considering the focus on creating jobs in this country). But Abernathy sounds like she comes out of some prehistoric academic era, and has an axe to grind against people with those new-fangled flashy CS degrees. 23. Greg Kuperberg Says: I would like to say that someone like Cammy Abernathy is a throwback to academia in the 1930s. Unfortunately, hers is a song that can be heard at every university. She Ms. Interdisciplinary, which turns out to mean “every department should work in my favorite areas.” A word that sounds like breadth instead represents narrowness. And sometimes it’s an offer that you can’t refuse. Now, usually this type of micromanagement is restricted to recruitment for new positions, and usually it has at least some measure of support from the department that’s being micromanaged. Attempting to quickly destroy a respectable department with zero support from the faculty in it is a more original step. But even that is not unheard of — in 1995 some crazy administrator tried to destroy the Rochester math department. 24. (^.^) Says: Dr. Aaronson: In your enthusiasm you may be missing one point. Isn’t it true that there is much more difference between cs and ee in MIT than say at UF. I don’t see much depth of ‘pure’ cs in ufl. It seems more of a hyper ee. http://www.cise.ufl.edu/people/faculty/area#area2 May be the direction the decision leads makes some sense with respect to ‘this’ department. 25. Scott Says: (^.^): Just looking at the webpage you linked to, Meera Sitharam, Sartaj Sahni, Timothy Davis, Alper Ungor, and possibly others all do stuff that’s clearly, uncontroversially CS theory (approximation algorithms, computational geometry, computational learning theory, etc). And in any case: (1) As Greg correctly pointed out, the distinction between CS and EE is very different from the distinction between theory and practice. Research in programming languages, user interfaces, etc. can be extremely applied, yet has no more to do with electricity than complexity theory does. And in the other direction, there are extremely pure theorists (e.g., coding and information theorists) who consider themselves part of EE rather than CS. If I had to draw a dividing line, I’d indeed say that EE studies all aspects of electronic technologies related to their embodiment in the physical world (including theoretical aspects like error-correction), whereas CS studies what’s possible (both in theory and in practice) once you assume you’re over the threshold of Turing-universality. (2) I strongly reject the idea that any CS department that’s not one of the “top 5 STOC/FOCS hotbeds” should be axed! This is partly because of personal experience: I spent my freshman year at Clarkson University in Potsdam, NY. If there hadn’t been wonderful professors there who could teach me about CS theory, it’s doubtful that I ever would have gotten started in research. What about the students at UF who want to do CS research, or work for Google, or start software companies? None of them would be nearly as well served by an EE department as a CS department. 26. Math Student Says: While I personally agree with Greg’s post about the abstract nature of CS being a huge advantage, I’m not sure that it’s the best argument to make from a PR point of view. “Computer Science….It’s important because it’s not about computers (and not really science either)!!!” doesn’t seem like a winning pitch to me 27. Gil Kalai Says: I also signed the petition some days ago and I hope that the move to merge cs into ee, demote half the cs people into teaching positions and probably drive away most cs researchers away will not take place. Still the language and some arguments given here towards Dean Cammy Abernathy are both incorrect and inappropriate. I also disagree with the proposed solution of raising tuition. The gloomiest aspect of the “cost of knowledge”, with devastating long-term effects on science, is the crazy cost of university education in US universities. 28. Tobias Says: The Idea of having CS as a part of EE has always confused me a bit. Here in Germany the tradition is more one of having CS as part of MATHEMATICS. Especially in the pure teaching colleges, this led to the existence of departments of CS who also do Math as a side issue. The EE ( and Information Technology ) department usually only has a small CS arm. Which is usually focused on Hardware or Networks. What would you American CS people say, if the department in this discussion had been absorbed by the Math department? 29. (^.^) Says: “What about the students at UF who want to do CS research, or work for Google, or start software companies? None of them would be nearly as well served by an EE department as a CS department” That is correct. But what about talented students in EE who could do well with a cs background that is probably needed in today’s job market or EEs who could get better research outlook with a cs background (personal opinion – for instance we have very little knowledge of contemporary approximation algorithms though it would be very valuable in EE research).. I think there is a disconnect in the way EE and CS are arranged in general for a 21st century education. I do not know the answer. 30. (^.^) Says: not that the example I gave could be picked up by reading in the side.. but there would be a difference if somehow some ideas in CS are brought main stream in to EE curriculum. 31. Scott Says: (^.^), let me get this straight: CS should be decimated and the surviving remnants moved over to EE, for the benefit of those EE students who don’t realize that, for their own good, they should be studying CS instead? Something seems a bit backward there… 32. Scott Says: Tobias #28: Yeah, there are also some universities in the US where CS is part of or allied with the math department. Again, as long as there’s actual scope for CS research to thrive, I’d say the formal departmental structures are of secondary importance. On the other hand, I think that given modern economic realities, any arrangement where CS is viewed as just a small, up-and-coming branch of some other field is bound to create tension… 33. Greg Kuperberg Says: Math Student — No, the point is that CS is important because it *is* about computers. With some regularity, some people fail to grasp what a computer really is. It fits the dismal pattern when someone like Abernathy decides from on high that hardware is the king of research and that software is just for teachers. Gil — I’m not sure what criticism of Abernathy expressed here is inappropriate. She wants to damage the careers of dozens of computer scientists. There is the fig leaf that Florida is pressed to find ways to save money. But there is major, maddening evidence against that explanation: the same Cammy Abernathy is granting open positions to other engineering departments. After enough experience at Davis, I have come to appreciate good university administration; I certainly do not think that all administration is bad when it isn’t very bad. Having seen both good deans and bad deans, it seems inescapable that Abernathy is a terrible dean, and that Florida should “restructure” her office rather than the CS department. 34. Sanjay Chawla Says: In Australia too we have similar problems perhaps maybe more acute. Part of the reason was the dilution of the CS curriculum to attract more students. This was ok till the dot com bust but when the students disappeared, most departments were left with a “non-rigorous” curriculum. Instead of CS being treated like a proper science or engineering, it was treated like “IT” and that became its death-knell. EE (who too had suffered in the wake of telecom crash), poured scorn on CS for immediate gains and fought to grab the shrinking pie. The irony is that CS and EE folks should be natural collaborators (“distinct but equal”). Take modern compressed sensing. It has great ideas from both CS and EE (and Statistics) and should be a cause of celebration for both disciplines. 35. (^.^) Says: Professor: I am not intending that. I am just saying the exposure that most EEs get of CS is as modern as Dijkstra’s shortest path algorithm. I think there is something wrong there given the practicality of CS for EEs and the amount the field has changed in 40 years. 36. Greg Kuperberg Says: Actually I have a major pet peeve about most lower division, intro programming CS courses: They are based on C++ and Java as a rigid industry standard. Whereas it is increasingly obvious that Python is both much easier to learn, pedagogically the new BASIC, and a serious industrial job skill rather than just an arthouse programming language. However, no such shortcoming is a good reason to bring in Dr. Jack Kevorkian to solve a CS department’s problems. 37. Raoul Ohio Says: Greg K: The best language to use for teaching intro CS has been debated endlessly. I have been on several sides of the debate myself. If easy to learn is key, everyone would still be using Pascal. Python is a contender that has gotten popular in the last decade or so. Meanwhile Ruby came along and shot past Python, and appears to be fading. There are new contenders every month. While it is tempting to teach with an up to the minute programming language, it is a lot of work to keep switching. It is hard to guess what languages will become popular for new platforms; tablets, smartphones, kiosks, etc. But it is not hard to guess what languages will be used for most large scale applications; C, C++, C#, Objective C, and Java. These and related languages such as D might be called the “C family”, and it is hard to argue that they should not be part of standard programming coursework. Organizations who follow such things, such as the TIOBE index, consistently show the C family filling all the top spots. My suggested path for teaching programming is (1) start with C++, (2) introduce C as a specialized part of C++, and (3) move on to Java or C#. It might be good to use D to introduce the C family in a “student friendly” way. (I bought a D book, but it is now under a Groovy and a Scala book on my “learn real soon now” pile). I think C++ is clearly the right language for data structures and algorithms classes, so you have to get to it at some point. BTW, in promoting a favorite language, it might not be a good idea to compare it to BASIC. 38. rrtucci Says: Wow, CS academics are such altruistic activists! Maybe next they’ll try to lower college tuition in the USA. (Sorry Scott, you know I have foot and mouth disease.) 39. Bram Cohen Says: My own work is in networking, and I never, ever, touch EE, not once in my entire career. Once you’re at the IP layer everything below it is irrelevant. Raoul, Pascal is a terrible, horrible introductory programming language, with way too many types, a zillion basic cases where behavior is undefined and in practice really is bizarre, and a whole lot of utterly pointless restrictions on what you can and can’t do. 40. lylebot Says: I think there might be another aspect to this, which is that when you compare funding (i.e. grant totals) for CS to funding for “true” engineering disciplines, CS comes out looking really bad. That’s just because we can do our research for cheap, not because it has less intellectual merit; CS probably looks great next to math in terms of funding levels. But if I’m an administrator in a College of Engineering, I’m under extreme budget pressure, and I’m looking for extreme solutions, it sort of makes a twisted kind of sense to try to fold the least-productive department (in terms of$) into a more-productive one.

I guess I always just assume that whenever an administrator is talking, they’re talking about money, even if the words they’re saying aren’t obviously related to that.

41. Greg Kuperberg Says:

lylebot – From checking the financial reports at UC Davis, it’s true that CS research funding is a bit lower than average per faculty member as compared with the rest of engineering. However, it’s still an entirely respectable number for research funding. Moreover, one of the important metrics is not research funding, but whether the research pays for itself. Research funding for CS looks favorable by this measure, because most CS labs need less lab space and fancy equipment than the rest of engineering does. I do not know the numbers at Florida, but I imagine that they are comparable.

The CS department also teaches a lot of students, including engineering students across the board, and that is implicitly also a major source of revenue. By contrast fashionable new departments may look big in research, but in teaching some of them are boutique departments without very many students.

I’m suspect that you’re correct that the Abernathy thought of it in terms of money. I have come to appreciate the need for universities to stay solvent. However, even on its own terms, this argument for killing the CS department is somewhere between short-sighted and outright blind.

42. University of Florida CISE Update « Machinations Says:

[...] department at the University of Florida has previously been discussed in the theory blogs (see here and here for [...]

43. Silas Barta Says:

That joke (added 4/27) is obviously made up. Everyone knows the list would have read “0.0 lb”, not “0.0 kg”.

(Heck, today, in 2012, they still report weights in air vehicle systems in pounds.)

44. Nilima Says:

I’m old enough to remember the ‘bad old days’, when Rochester’s Math PhD program was being cut. I recall the mathematical community’s response then. I sincerely hope similar pressure on UFl leads to a positive outcome.

A money quote from Steven Weinberg’s letter (http://www.ams.org/notices/199603/rochester.pdf)

‘I would not advise any prospective undergraduate or
graduate student who wishes to concentrate on
the physical sciences to go to a university that
did not have a graduate program in mathemat-
ics,…”

I wouldn’t recommend my students in the math sciences go to UF if the CS department evaporates. There would be the obvious problems in pedagogical emphasis when CS and EE are conflated. (Does anyone believe a sparse matrix methods course would be a priority in EE?)

Then there is the matter of taste- who on earth respects a university where the ignoramuses have taken over? Yuck. Reminds me of what my wise postdoc advisor once said: the only thing to be avoided more assiduously than a skunk is an incompetent university administrator.

45. Nilima Says:

And now, emerging from internet-hibernation, I realize my previous comment is somewhat irrelevant, and that CS at UF will actually survive. Or so it appears.

46. Timothy Gowers Says:

There’s a meme going around that “ad nauseam” is spelt “ad nauseum”. Alas, with your wonderful and widely read blog you have done a great job in spreading it. I hope this comment can undo a tiny fraction of the damage. (One way of thinking about it is to observe that “nauseam” is the accusative of “nausea”. I think the mistake is a false analogy with “museum”. Also, it should be pronounced naw-zee-am, which makes the spelling more obvious.)

47. Scott Says:

Thanks so much, Timothy! I fixed the errors (which were actually in the Joy Christian posts, not in this one). I pride myself on spelling and was mortified…