Session 1: The Tenure Decision

Maria Klawe, University of British Columbia
Mary Jane Irwin, Penn State University
Maria Klawe
First of all, I want to welcome you here. We are going to be talking about really important things today, but I also hope we'll have a good time. This is a unique opportunity for us to share some things that are a big part of our life and to learn a lot from the people who are here. There are lots of women present who know a great deal about the various things we are going to be talking about. In fact, some of the people who know the most are the youngest people in the room. This is because the academic environment in computer science is changing rapidly. The people finishing their PhDs in computer science this year and next will live in a different environment than the one experienced by those of us who received our degrees 15 years ago. I hope that throughout the day everyone will contribute their perspective on issues.

For anyone in an academic career, or considering an academic career, the question of tenure is of primary importance. Although this session only lasts an hour, almost every other session during the day has something to do with getting tenure. So if your questions about tenure don't happen to get answered in this hour, there will be lots of opportunities to ask them later on.

I'll start out by saying a few words about my background and then pass the mike over to Janie Irwin. After telling you something about herself, Janie will describe the tenure process at Penn State University. She is going to give you a fairly comprehensive overview. Not all departments are the same but it is useful to see a complete picture of how tenure is handled in a typical department.

When Janie finishes I will discuss some of the ways that other institutions differ from Penn State in handling tenure, and give you some of my own perspective on the tenure process. After that we'll open it up for questions. I am confident that virtually everybody in this room is good at asking questions, and I'm going to ask you to demonstrate that when we get to this part. OK, let me say something about myself. Obviously, my name is Maria Klawe. I'm currently the Head of the Department of Computer Science at the University of British Columbia in Vancouver. I used to worry that people did not know where UBC was. Now I don't have this problem since UBC was site of the three Bs summit last month, i.e. Bill, Boris, and Bryan. If you were watching TV you know how beautiful it is there.

I'm coming to the end of my five-year term as head. Before joining the University of British Columbia I spent 8 years at the IBM Almaden Research Center. What this really means is that I didn't go through the tenure process in the usual way - because I came as a tenured full professor. Believe me this is the easy way to do it, --- laughter--- but I don't recommend it for the following reason. I think that my husband, Nick Pippenger, and I caught the end of the wave where it was easy for senior people to transfer from industry into good academic departments. It is my impression that, for a whole number of reasons (one of which is that you young people are so incredibly good), universities aren't nearly as interested in hiring people at senior levels anymore. So even though this method of getting tenure worked fine for me, I don't recommend it as a general strategy.

During my five years at UBC I have been involved in a large number of tenure cases. I have a good sense of what happens in my department and in my Faculty (Science), and how decisions are made and what is important. Four people in my department came up for tenure this year and all were successful; however, there are still 11 untenured faculty members in my department. Thus I have been spending a lot of time over the past few years talking with my untenured faculty members about what's involved in the tenure process. It's wonderful to have all of you here today to talk about this issue. Now I'll turn the session over to Janie.

Janie Irwin
As Maria says, I'm Janie Irwin. I'm the department head of the Computer Science of Penn State University. We are about to become the department of Computer Science and Engineering and we are changing colleges so it's a very interesting and exciting time at Penn State. I graduated from the University of Illinois in 1977. I was the second female PhD to graduate from the University of Illinois. I was promoted to the associate level and tenured in 1983, (six years as an assistant professore); I was promoted to full professor in 1989 (six years as an associate professor). One thing I did not realize at first is that is considered "normal" progress. So when you make that promotion to associate the next promotion you should be working towards, full professor, is six years from then, five years if you are really good, four years if you're a superstar.

Two years ago I volunteered to become department head for a three year term. Of course we are becoming a new department and the department head situation has not been resolved yet. It's getting late but my guess is that I'll be serving another two or three year term as department head for this new department. And for those of you that are still wandering in, there are three handouts which I'm going to introduce and wave around but I'm not going to go into them in detail; they are for you to read at your leisure.

slide
So is promotion and tenure really publish or perish? We will really see there is more to it than that. The first thing you should do when you get to your institution is to track down the promotion and tenure guidelines, P&T for short, RIGHT. There are usually university guidelines--at Penn State they are called PS23 because we have this fat booklet of guidelines and they are all numbered PS (for Penn State) and the guidelines for promotion and tenure are #23. I didn't xerox those because they describe in detail the entire process (which you should find, read and understand when you get to your home institution), and they are not as important for our discussions here as the next level of guidelines, which are the college criteria for promotion and tenure. Those are the criteria that really spell out what are the areas that you need to concentrate on.

slide
How do they evaluate teaching, what's important in teaching, how do they evaluate research, what's important in research, how do they evaluate scholarship, what's important in scholarship, how do they evaluate service, and what's important in service? And that differs from university to university and it also differs from college to college within a university. So what I did was I made copies of the two college criteria that are most relevant to computer scientists, the college criteria from the College of Science of Penn State and the college criteria from the College of Engineering at Penn State. Those are both signed by the Deans. Typically, these criteria are something that the deans and his executive committee, that is the department heads of the various departments, sit down and work up.

So you've got two copies of college criteria, one from a college of science and one from a college of engineering. As you read through those you will find that although there are a lot of similarities there are some not so subtle differences. You should make sure you understand your college criteria because your dean is going to be very important in this promotion and tenure decision process. And then there may also be department criteria, but typically those are the college criteria that are taken and copied with an additional title on the top of the department that you are in. They may be augmented somewhat from the college criteria but typically these two are the same. So you have samples there you should look at. When you get to your institution, the first thing you should do is make sure that you look at university and college criteria for the college that you are in. There are also usually lots of extra handouts that you can collect on the promotion and tenure process. You should familiarize yourself with all of that.

One thing that you have to be aware of in this process is that there are many review levels.

slide
This is the sequence of review levels at Penn State. I think it's similar at most at most universities if not all universities. When you start to collect internal letters, that is letters that go into your file from various committees and people at your home institution, you get a letter of support from the department promotion and tenure committee. This is typically a committee of senior faculty in your department. It may be entirely elected, it may be partially elected and partially appointed. The people who are eligible to sit on this committee of course are tenured, and typically the people who are eligible to vote for who is on this committee are tenured. So you are not part of the process of deciding who is on this committee. They look at the case independently from the department head, so there are two letters that will go into your file, one from the department head, one from the department committee, which are written independently of one another evaluating you on the various criteria.

Then your folder goes to the college level. There is a committee at the college level that consists of senior professors in the college appointed typically by the dean. I don't know any of these committees that are elected by the college but that is a possibility. At Penn State they are all appointed by the dean. They do the same kind of evaluation process, looking at the letters that were supplied by the department committee and the department head, and make a recommendation to the dean. Then the dean takes the folder looking at those three sets of letters plus everything else that's in there. We're going to talk about what you want to start keeping track of for your dossier in a little bit here. He or she does an evaluation, probably he for most of us, and adds an additional letter to the folder and sends that folder to the university level committee. The university level committee is senior faculty from all units within the university. It looks at the folder with those four supporting letters, makes it's own recommendation, and forwards it to the Provost, President, or Chancellor depending on what you happen to have at your own university.

You will find that the criteria at these levels may differ. In particular, at the department and college level where I am, research is very important. You have to have excelled in research in order to go to the next level and the promotion and tenure process can stop at any level. It could stop at that level and the word could be no and it will not go any further. When I was going through this process myself and I was getting nervous because I hadn't heard anything, I'd go to my department head and he would say, "As long as you haven't heard anything, that means it's still in the system, so don't worry." And if you hear early, then that's typically not good news. As long as it moves forward it goes on to the next committee and you don't hear anything until it finally pops out the end and the board of trustees have approved it.

As I was saying, criteria vary at different levels. In particular we are beginning to see more emphasis on teaching, looking back at teaching effectiveness and teaching evaluations. That's becoming more important, particularly at the university level. So now you have to excel not only in research but you have to excel in teaching.

There is another aspect that is important and that is that you get feedback. It's important that you don't sit there for six years not knowing whether you are doing the right thing or the wrong thing. You really need to get feedback. I think written feedback is very important and I found out this does not happen in all universities. The process at Penn State is that every year I sit down with my junior faculty, actually I sit down with all my faculty but with junior faculty in particular, in a formal setting. I meet with them all the time but this is a formal appointment at the end of the year to do a yearly evaluation to talk to them about how they've done on the components that constitute the promotion and tenure file.

At the second and fourth year we have a formal process of letter writing. In the college of science the department committee, the department head, and the dean all evaluate a folder, your dossier, that is your curriculum vite at that point and write a letter saying what you are doing right and what you are doing wrong. These letters are hard to read. They are not always, in fact they are never, all glowing. It's important to give you guidance at this point and so they say explicitly you should pay more attention to this or that. For example, "you should make sure that those papers that you are getting in refereed conference proceedings (which are very rewarding because you get to go and talk to your colleagues and present the paper) get turned around into archival journal papers. Something that a lot of us don't do as regularly as we should.

Those can be two and three page letters. They are long letters, not something that is just short and sweet. They are really evaluation letters. We are almost doing fifth year evaluations on everyone, that is written evaluations, I think it's important to get written evaluations because it's very easy for you to sit down with your department head and hear what you want to hear or for them not to be as perhaps aggressive as they should be in telling you the bad news as well as the good news. If you see it in black and white, you read it, you understand it, you take it to heart, so it's very important to get written feedback in the process. And if you don't have a procedure for that at your university I encourage you to go to your department head and ask for written feedback. Ihe e college of engineering, the college committee participates in the fourth year evaluation so they do a written evaluation as well. So once again there is an example where, even within a university, the process is different from college to college.

Ok, what do you want to put in your dossier? Well, there are four cells at Penn State: teaching, research and scholarship (I don't know how to differentiate the two so I always refer to them in one line), and service. Service counts about one percent where I am. I try not to ask my junior faculty to do service. It's tough because they are usually very good and very eager, very hard working, and the tendency is to let them do it because they do a great job, but it's not fair to them. Teaching has become more important over the past several years and I expect it to continue. I put it in the range of somewhere between 15 and 25 percent as far as how much it counts in the folder, and the rest is research and scholarship.

slide
This is the dossier that we put together for written review: second year review, fourth year review, fifth year review if there happens to be one and the final promotion and tenure year, the sixth year. Although the copy you have is black and white, the dividers that actually go in are in color so we affectionally refer to this as the "rainbow" file. Each of these dividers has a lot of detail on the sort of information that must go into the dossier at Penn State. I'm hoping it will give you inspiration as to those sorts of things that you should start to keep track of from day one, to make sure that you have all of that material collected so that when the time comes and you put together your dossier you have all the information ready. We have someone in the department, a clerical staff, who does this automatically for everyone, but that's not the case everywhere.

slide
The first or, let me see, the orange, actually all of these have official color names and that doesn't look quite orange, but the first one talks about teaching and on the handout that you have it tells what counts as far as teaching and how it's evaluated. This is what counts - teaching and advising - and this is how its evaluated and that's what you should pay attention to. Sometimes you have no input into the process. One thing you can do, if you think there is a student that you've really had a significant impact on, that you've influenced in some way, or an advisee, keep track of that person, keep their address, because at year six it might be possible for your department head to write to them and say, "Can you give me a supporting letter saying that this person was particularly influential in your education?" Those are the kinds of letters that the university level particularly likes to see. So that's the teaching divider.

slide
The next one is the yellow divider. That's research and those of course are the kinds of things we all understand: refereed journals, conferences, books, and all the publications fit in this category. Research funding fits in this category, graduate thesis supervisor, the committees that you have served on. It's certainly nice to have had a PhD student graduate with their PhD by the time you come up for your sixth year, two is even better, but at least some evidence that you are able to supervise PhD research. We put under creative accomplishments software systems and hardware prototypes for computer scientists. There are other things on this that you should spend some time looking at off line.

slide
The next one is scholarship because we divide it into a different cell. For a lot of you it will just collapse into research. For example, it's papers presented at conferences, talks presented at conferences, invited colloquia, or software systems developed (what we put in here are systems that augment your teaching). And honors and awards, participation in professional societies, etc. Once again you have that to look at.

slide
The fourth cell is service to the university. This is committee work on the various committees within the universities. The other thing Penn State has just added, and I'm hoping many universities are starting to look at this, is contributions to the university's program to enhance equal opportunity and cultural diversity. Also, service to government agencies, the kind of things you may be doing for NSF, serving on committees, etc. also counts under service.

slide
The last two dividers have to do with the divider that says here are all the letters, so all the letters from the second year review, from the fourth year review, from the fifth year review if you happen to have a fifth year review and of course the sixth year review go into the folder at this point. Those are letters which at least at Penn State you have copies of . You don't have copies of the sixth year review until the process is completed, but for the second and forth year review you have copies of those letters so you know what is sitting in your folder.

slide
The last divider is the divider that is really the most important. It's the divider that is for external letters. Ok, so let's spend a little bit of time talking about external letters. It's a very important part of the process at Penn State as I suspect it is at most places. In letters of recommendation, people evaluate your research primarily and they talk about the impact of your research. It's very important for them to be able to say that you have made a significant contribution in a particular area and to evaluate that contribution.

slide
So how can you go about getting people primed to write supportive letters for you? One of the things we are now putting in our promotion and tenure document and that we send out when we send out the file to the people that we are asking to write letters is a one-to-two page statement of research: a concise well thought out research statement that shows the evolution of your research and shows how all of the publications (or most of the publications) that you've had and the grants that you've had go together and make up this research program and what your vision for the future is. So it's very important that you begin to put down your vision on paper and think about how all the pieces fit together. It also allows people to project into the future and see that yes, there is a research plan that continues into the future. For such an important commitment as tenure they want to be looking for people who are going to have a career past tenure, who aren't going to stop research at that point; people who are going to keep working, who have a vision for the future, a goal set out for themselves.

Ok, you need to prime likely candidates for recommendation letters. Now I say likely; at least where I am the candidate is not allowed to come up with a list of people that we write to for letters. In fact, they are not even allowed to suggest names, although if I get a piece of email with a some names in it I don't ignore it. That's not true at every institution. At many institutions you're are allowed to suggest names. That doesn't mean that those names cannot be augmented. So, how do you prime likely candidates? Well, you start to meet people and as you meet people who you think are people that might be able to write a supportive letter for you (they should really be at the full professor level; they can be associates but our deans prefer full professors), find out if they're interested in your work. If you do a particularly nice paper, send them a copy of it with a cover letter. Don't flood them every couple of months because they are all busy people. But every once in a while send them something. Or, this is one piece of service that you can do and I try to track all my junior faculty into it. In their fourth year I ask them if they want to be colloquium chair. Colloquium chair means you get to invite all your friends if you want, but if you plan it right what you do is you invite all those likely candidates to come to your home institution to give a talk. You get to spend a day with them, talking to them, and you have your students talk to them, and you get to show them your vision of research and they get to give a talk and th eir way is paid etc. So they end up maybe owing you a favor so that in your fifth year you can call them up and say, "Hey, I'd like to come and give a talk."

So in your fourth and fifth year you start to invite yourself elsewhere and sometimes you have to pay for it out of your own pocket. Sometimes you do it because you happen to be at a conference at a nearby site and you call up (I know it's uncomfortable but we need to do it) someone at a nearby institution and say, "You know, I'm going to be in the area. Would there be a chance that you would want me to come by and give a colloquium presentation?" I'm sure there are other ideas here--these are just ones I jotted down, but something you do need to start thinking about and you need some support from your department head at least in this area so that you have a chance to build up those favors.

What I wanted to do is close with a couple of case studies. I did the happy ending first, but I think I want to talk about the sad ending first so this is my case two. And both of these cases happen to be men. So if I slip and say "he", that's because both of these happen to be men. I haven't dealt with a case at Penn State myself of a woman going through the promotion and tenure process. In fact at the moment I'm the only woman in the department.

slide
This particular person had a pretty good record of undergraduate and graduate teaching and a number of PhD students. He had lots of refereed journal publications. He was sort of the measure that all of the other junior faculty used. They said, "Oops, he's got, you know, six refereed journal publications this year. There is no way I can compete." He was never successful at getting funded, however, and that certainly raised a lot of red flags because he was working in an area where he should have been able to get funded. He had papers in lots of areas, so many that it really made it hard, I think, for the people that were writing letters to see his research focus, to see his research theme, to see where he had come from and where he was going. It sort of defused what he had been doing, and what really hurt was that because he had concentrated in some cases more on quantity than quality, he had a number of relatively weak publications that the community knew about and that came back to hurt him in his external letters, because people said, well, although he has had these good publications these were not so good. And so he had some supportive letters but he also had a few mediocre letters largely because of this.

slide
Next let's look at a happy ending so we can end on happy note. This was a person that although he really cared a lot and worked very hard at it he is probably never going to be a wonderful teacher. He is very enthusiastic and undergraduates respond to that but he is also not terribly organized and the students also respond negatively to that. On the other hand, at the graduate level the graduate students love him. In our interviews with graduate students they said almost to a person, well if you have a research question and you don't know who else to go to to talk to about this question, you can always go to this faculty member, his door is open, he welcomes you in, he spends time with you, and he's always able to help you with your problem. He established his research independence from his thesis advisor very early on. That's something that you really need to pay attention to. You need to cut that umbilical cord as quickly as you can as far as joint publications. Don't cut it as far as contacts, hopefully that thesis advisor can help you with contacts, but you need to be able to establish your research independently. Make sure you stress quality in your publications not quantity. It's very nice if you have external support, in particular support from NSF, because deans know that NSF proposals are peer reviewed, and that means that your peers think your work is good enough that part of a pool of money that they themselves are competing for should go to support your work and that's a pretty strong statement.

Make sure your research publications are significant. This particular person had made contributions in a number of areas, he hadn't just worked in one area, which could have hurt him because once again it made his research theme hard for these external letter writers to define, but he made enough solid, really solid, contributions in these areas that it didn't hurt him. In fact, it helped. And he got very strong letters of recommendation, and the case went through very nicely.

So, Maria, I'm done at that point.

Maria Klawe
That was one of the nicest overviews of the tenure process I've heard. Let me now talk about a few variants on this. The first thing is that universities and departments differ tremendously in what the expectations for tenure are. Thus, one of the very first things you need to do if you already have an academic position and are hoping to get tenure, is to find out what the expectations are. Similarly, if you are looking for a job in an academic department one of the most important things to do is find out the expectations of the departments you are considering.

How can you find out about these expectations? One very good idea, suggested by Mary Vernon, is to ask to see the CVs of people who recently got tenure. Because they were successful these people will probably be quite happy to show their CVs to you, and it will give you a good idea of what is expected.

The other obvious method is to talk to the department chair when you are interviewing. You should definitely make sure that you discuss this at the point when an offer is made. It's always a bit dicey about when and how to first raise the issue--do you want to come across at your interview as being really scared that you might not get tenure? Obviously not. You want to come across as being a realistic person who understands what the expectations of the department and the institution are. So the timing and style is something you should handle carefully.

Let me talk for a moment about how expectations at institutions differ. In particular I'll talk about two models representing how tenure is viewed at universities. In describing the two models I'm going to use some specific names of universities since I think it will make things clearer. I hope that nobody will find what I say offensive. Let me use MIT, Harvard, and Stanford and Princeton as examples of my first model. These are universities where a substantial number of very good people who are hired as assistant professors do not receive tenure. Individuals denied tenure often go on to receive tenure and be promoted at other outstanding universities and do exceptionally well for the rest of their career. I am not suggesting that there is something wrong with MIT and Harvard, etc.; they simply use a model in which a relatively small fraction of tenure-track assistant professors are expected to get tenure. They believe this model builds the kind of institution that they want, and can give you many reasons to justify their strategy.

This model obviously has some negative impact on the climate for junior people, but there are still excellent reasons to accept a position at such a university. There are often tremendous advantages in being at these universities even for a few years. Moreover, if you fail to get tenure it's not a death sentence. If you do get an offer from such a university you need to balance the advantages with the possibility that you may be on the job market again in a few years. You need to think through the consequences of having to move in relation to two-body problems, and for your children if you have them. Finally, while it is common knowledge that it is fairly difficult to receive tenure at the universities I mentioned, there are others with similar tenure policies that are less well-known. So ask!

The second model is that each person hired into a tenure track position is expected to get tenure, so long as they meet the expectations of the university. That happens to be the model at UBC where I'm department chair. It's the model at all Canadian universities as far as I know. It's often the model at many universities in the United States. The important thing is what are those expectations? Janie has been telling you the expectations at Penn State. I'll describe those at UBC for comparison; they are fairly similar.

First, we have the same kinds of levels, but you come up for tenure in your fifth year. Thus it's a shorter process and this makes it harder in some sense. In our university the crunch level is the faculty level, namely at the advisory committee to the dean. This committee is made up of the department heads and associate deans in the faculty. This committee reverses (positive) department decisions perhaps 15-20% of the time. I've never seen the committee reverse a negative department decision. My impression is that it's rare in any institution to see negative decisions at a lower level reversed at a higher level.

At UBC, like Penn State, the critical factors are research, teaching, and service. For research, UBC wants to see clear indications that individuals have established themselves as being in the top two or three among their peers in Canada, preferably as in the top two or three among their peers, in their research area, in North America. Now the whole question is, how big is the research area? If we say Computer Science, that's pretty hard. If you say working on interactive proof systems that's quite a bit easier. In any case, a lot depends on how well you have established the fact that you are a leader in some area. This has to do with what Janie was saying about focusing your research.

UBC looks very hard at the quality of where you publish your research. Thus it's important to concentrate on top journals and conferences. Each department head has to give an accurate description of the quality and prestige of each journal and conference in which a candidate has published, and this information accumulates over the years. Journal publications are particularly important. It is simply not enough to publish in the most prestigious refereed conferences, though such conference publications do count. UBC also looks at the letters of recommendation extremely carefully, as well as who wrote the letters. UBC expects letters from leaders in the field, especially from full professors at top departments.

As far as teaching goes, I've seen a noticable shift over the last few years. Five years ago, an outstanding researcher who was barely competent as a teacher might have gotten tenure without major problems. Today, such an individual would definitely not get tenured. Conversely, a candidate who is a superb teacher and a respectable researcher will probably get tenure today, but not five years ago.

Now I'd like to talk briefly about service since the situation is different in my department from many others. There are people here who know the people in my department. They know that our assistant professors put a huge amount of effort into service. This differs from Janie's department. Why is this? It's clear that service takes time away from research and teaching. There are two reasons. One is that we are building a department and we cannot build the department without everybody working. The technical staff work on it, the secretaries work on it, the students work on it, and so do all the junior people. We are trying to take a department that was quite small and sleepy, a department with some excellent people but not in the top ranks, and just sort of lift it up and make it one of the best departments in North America. That's a lot of work. Many of our junior people are phenominally talented, and we can't achieve our goal without their help.

How do I justify the amount of time my junior people are putting into service? First of all when they interview for the job they hear all about it so they know what they are getting themselves into. Second, I monitor what is happening in their research very carefully and if I see somebody having serious problems, I yank them out of service responsibilities immediately. I also spend a lot of time in educating the dean's advisory committee about our department and the fact that service contributions must be taken seriously. Are they taken seriously? We had four people come up for tenure last year and everybody mentioned it. They all made it through and service was an issue.

Thus it's critical to check out what the situation is in your own department. It's not the same from place to place, and it's not even exactly the same across departments in a single university.

Last but not least, a few more hints on how to get people to know you. Number one, don't wait for your fourth year to go and give talks in other departments. You should be trying to give a talk at each of the top 10 departments in your discipline within your first three years. You don't have to do them all at once. You do one here or there. It's very important for getting funding, it's important for generating those letters of recommendation at tenure time, and it's important for your research. You'll get feedback, you'll learn things, you'll make friends, and you'll attract undergraduate students to come and be graduate students with you. How do you do this? Call up someone you know in the department you'd like to visit and say "I'd really love to give a talk in your department. Could you set it up?" Few people are comfortable doing this the first few times -- but people still do it all the time. You can use this workshop to make it easier. Get to know each other here. You are all either in de artments already or you will be in departments. Invite each other to give talks.

Number two, getting on program committees and editorial boards. At UBC we like to see people serve on very prestigious program committees, and hopefully become a journal editor towards the end of the pre-tenure period. How do you make sure these things happen? The best way to do it is to let some friendly senior person in your field know that you would like to do these things. Senior people in Computer Science who are senior in the field are overloaded and overburdened with program committees and editorships. They simply have too much to do, so if they encounter talented responsible and reliable junior people who say, "I'd like to do this," they are delighted to make sure that it happens for them.

When you take on responsibilities like these it's essential that you do a very good job. Thus you should be very careful not to take on too many such responsibilities. The key to success in life is balance (at which I am a total failure). Nevertheless, this is the key. You need to do enough research, you need to spend enough time on teaching but not too much. You need to be smart about how to teach well and keep the time investment limited (you will hear more about this later on). You need to figure out what's the right balance of service for you and what you can do effectively in a limited amount of time.

Now let me say one more thing. Most of you are going to get tenure. Probably 80% of you, 90% of you, even 95% of you, maybe almost all of you. However, some people will fail to get tenure, and some others of us will fail at other things that are really important in our lives. So I just want to close my part of this panel by saying failure is a really good thing. Failure, even when it hurts and you go home and cry all night uncontrollably, is a really good thing. People who know me well know that I have failed repeatedly, over and over again, at things that matter to me. Why is failure good? Because it means that you are aiming high enough. And it means that you are learning. If you are not failing at things you are not living your life right. However, again BALANCE is the key.

OK, so now we are going to open it up for questions. We have about fifteen minutes. I ask you all not to be shy; we certainly won't be shy in giving you answers back. So, who wants to start this out. OK, please say your name and where you're from.

Question
I'm Adelle Howell. I'm at Colorado State University. Standards have been changing. Those of us who are in the job market recognize this. For those of us who successfully made it through the job market and those who are about to throw themselves into it, my question to you is, How do you deal with a moving target?

Answer
Want to try that one, Janie? Ok, the question is, standards are changing in our discipline. They're going up, and if you are just starting out or if you are just about to start out, how do you deal with a moving target?

That's a really difficult one. Deans are certainly aware that it is a moving target and deans are often the people that must be convinced. I 'm not sure what can be done other than to realize it's a moving target. I recommend that when you are interviewing or when you are considering accepting a job, you should talk about this issue to the department chair and to the dean if you happen to see him or her. In our department, we made a strong commitment that though standards were changing we should, to a large extent, respect the standards and expectations at the time the person was hired. Obviously, it's a complicated issue and departments differ on how they deal with the problem. So again, it's important to find out. The bottom line, however, is no matter what is happening with the standards anywhere, the best that you can do is to do your best. There is a limit on what anyone of us can do. If you strive for excellence in research, for being a good teacher, and use the limited amount of time effectiv ly, that's your best shot. I just don't think there is much else you can do.

Question
Hello, I'm Ruth Silverman, and I'm a full professor at University of Columbia and I'm also a visiting professor on a permanent basis at the University of Maryland, College Park. Maria, I'd like to make a comment. I think this is a wonderful meeting. I think you could enhance it somewhat if you would address that because of the changing times, many of the people here will end up at schools which do not have PhD programs and therefore have to adjust the balance of teaching verses research. I, myself, of course have been many years in the field and have made many accommodations in my life. I really enjoyed hearing you say that if you are not failing at something essentially you are not trying enough new things. If you could, throughout the meeting, address the needs of people who end up taking a job at a school which perhaps does not offer the research opportunities, how they can accommodate their vision of themselves as a total person to the situation, I think that would be very valuable and would enhance it for everyone. Thank you.

Maria Klawe
Thank you very much, Ruth. Ruth knows that I am a failed mathematician in the sense that my first PhD, in fact my only PhD, was in mathematics. I certainly went through an adjustment in my life as a result of the job market in mathematics was much worse than it is in computer science now; obviously I decided to shift to another field, computer science. We all have to make changes in our life to adjust to the needs of society. This particular workshop was put together with the emphasis on academic careers in institutions with a strong research orientation. It is true that many of us may end up being in institutions where we are the first serious researcher in computer science, or the first person or one of the first people who care about research. It is true that if you go to such an institution you will have to adjust how you spend your time. You will have to adjust your priorities. I don't think this means that you have to give up research altogether; I do think it means that you have to understand the priorities of your institution and select the appropriate balance for that.

Question
When you mentioned the letters that we need to in our dossier you didn't say anything about letters from industry. I've made some contacts on committees and at various workshops with some people who are doing research in industry and would think of them first actually to ask for letters, and wonder if that needs to be balanced or if those are just as good or what?

Maria Klawe
At UBC we certainly have people from industry write letters for tenure cases. Typically these writers are from the major industrial research labs: e.g. Xerox PARC, Bell Labs, or IBM T. J. Watson. Though the letter writers are usually in a senior research position, some of our administrators are still suspicious, and the department heads have to do a good job of documenting the credentials of people from industry.

Jane Irwin
It depends on the university. Our department asks for suggestions of at least three people from the candidate. We have to use at least two others. We often use two and use four others to get a broad spectrum, particularly if it's someone who's worked in more than one area. In terms of using people from industry, if it is somebody from one of the outstanding industrial research labs we will certainly use them. But we wouldn't use more than a third of our choices from industry because they do not carry the same weight at higher university levels, so we just can't afford to jeopardize the chances of the candidate. From your position I think having industrial contacts is extremely good for a whole bunch of reasons. And you should think of it as being good, but make sure that that is not all.

Another voice
Make sure you find out how many letters typically are going to be part of your file. It varies college to college. The college of engineering only requires five letters, the college of science wants 10, or 12, so that's a question to ask and find out and then the percentages differ. And it sounded like you only need three in a minimum case. I've seen cases with three but you are really playing Russian Roullette if you work with three because you can get one bad letter and it's death.

Question
I'm curious about the status of stopping the tenure clock if you have children. Number one, do many schools allow that? Number two, is there prejudice against you for doing that?

Maria Klawe
First of all, in in our university, if you take a 6-month leave it stops the tenure clock for a year. Normally if somebody had a child that would happen almost automatically. We are redoing our policies so that each child will add one year to the tenure process automatically. The whole question of what to do with tenure clocks and maternity leaves is very complicated. The right solution is certainly not the same for everybody. One of the things the CRA Committee on Women is doing is to put together a model policy on these kinds of issues. The bottom line right now is that things vary from institution to institution on how such leaves are regarded. I can honestly say that at UBC it would not be an issue, that people would just say yes this is normal, people have children, they should have children, etc. But we have a fairly enlightened university. However, it has only been enlightened for a couple of years, so if you would have asked four or five years ago I would have said, I'm not sure.

We have the concept of serious personal situations such as primary care giver after the birth or adoption of a child, a serious personal illness, provision of care for a seriously ill family member, or any similar situation. You write a letter to your department head, your department head writes a supporting letter to the dean and the dean writes a supporting letter to the provost so it is certainly possible for a years stay. Now as far as down the road, does it have a negative impact? I feel that it does not but I haven't been through the system to see that because it's only recently in place. So I think the verdict is out on that .

We have one more minute before we pass on to the next panel, so we will have one more quick question.

Question
I have a question about interdisciplinary research areas. There are some research problems which are inherently interdisciplinary which require the collaboration of people from other departments than computer science, in my case. Could you briefly comment on the status of such work.

Answer
I will. There is no question that if you are interdisciplinary it is harder to get tenure simply because you need to get more people behind you. You need to get letters of support from more fields. I feel very badly about this because personally, right now my major research focus is interdisciplinary, and I think everyone agrees that the whole direction of science and engineering is towards interdisciplinary work. However, at most universities the tenure process is not friendly to interdisciplinary work. You can get it, it's just harder. If you were playing safe you might wait until you had tenure before you got heavily involved in such things. I wouldn't be doing the project that I'm doing right now if I were not already a tenured full professor.

jmankoff@cs.oberlin.edu