SESSION 4 MAKING CONNECTIONS Judy Goldsmith, University of Manitoba, Joan Feigenbaum, AT&T Bell Laboratories, and Naomi Nishimura, University of Waterloo. One of the hardest things for everybody I've talked to is making the connections, going up to people, getting to know people. You've all been hearing over and over again how important that is, how those letters you get are absolutely critical. There aren't very many people who can get good letters without making personal contacts. This session is coming right before lunch and I hope everyone will take it as an assignment at lunch to try to put some of the advice that they are getting in this panel into practice. I know it's bad pedagogy to give a talk sitting down, but I don't want to deal with those stairs again. [She was on crutches.] I'm Judy Goldsmith, and I'm currently an assistant professor, tenure track, at the University of Manitoba. Next year I will be an assistant professor, tenure track, at the University of Kentucky. I know we are not supposed to be talking about having a real life, but I want to say that I am going there because my partner and I will both be tenure tracked next year. I want to talk a little bit about making connections: how you make connections, why you should make connections, and how to make them work, and then pass it on to Joan Feigenbaum and Naomi Nishimura, who will probably say the same things from different perspectives. Why do you want to talk to people at conferences? Why do you want to get known? People have been very clear on the other panels on the benefits of being known. You tend to get your papers accepted if your work is known; you tend to have a slightly better chance of getting grants if your work is known; you are more liable to get invited to give talks if your work is known. And we've heard numerous people say why giving colloquium talks is good. How do you go about getting known? One of the best ways to do it is, at conferences, just walk up to people and start talking to them. Introduce yourself, start a conversation, listen to them talk about their work. If they are interested, if they ask, talk about your own work. To whom do you talk? Well if you are standing around waiting in line for lunch, talk to the people next to you. If you see somebody in the evening standing around the lobby of the hotel looking hungry, you can go up to them and say, "Do you have dinner plans?" You can also prepare ahead of time by getting lists of the conference program and saying, "I'm interested in this paper, I want to talk to this person. This person is important in my field. I want to make sure he or she knows who I am." So you can do your homework ahead of time by figuring out who is important in your field whom you want to meet, who is giving talks that interest you. They may or may not be the big people in your field, they may be graduate students. That's fine, graduate students are important too. Anybody you meet and you talk to, you can introduce to someone else, and they can introduce people to you. One of the things that can be important if you are feeling shy and you don't know a whole lot of people is to make a deal with somebody. "I'll meet two people tomorrow and I'll introduce you to both of them if you do the same for me." It is a little thing, but little bit by little bit it adds up. Another thing you should do to prepare for a conference if you can --- and I never manage this --- is get some sleep the week before. Why do you want to talk to people at a conference? Not only might it help you get recognition, or a job, or tenure, sometimes when you having a technical conversation with somebody, you get a new idea. And every so often that snap of "OH YES, I could look at it this way," really happens, and for me it happens a lot at conferences. How do you start a conversation? Besides "are you hungry," or, "do you have plans for dinner?" You can go up to a speaker and you can say "I liked your talk, I had a question." You can go up to somebody whose paper you read and say "I really enjoyed your paper. I was wondering why you did this." Or you can turn to to the person sitting next to you after a talk and say, "did you understand that?" -- Laughter -- I've met a lot of people that way, and furthermore, I've actually understood talks by turning to somebody who looks like they did understand and saying, "can you explain that to me?" Dealing with harassment: One of the problems, one of the insecurities I have about walking up to strange men and introducing myself is that every once in a long while one of them is weird. ---Laughter--- I've had good luck at conferences. I like almost all of the men I work with. I've met one or two that were inappropriate. The first time this happened to me at a conference, I was a graduate student and I was kind of flattered and kind of disgusted and I said "you know that's really very sweet, but..." and he kept making the same offer all weekend, all week. The first thing to do when somebody makes an inappropriate suggestion is say "no." It's not up to you to protect his feelings or her feelings. Statistically it will be a him.--Laughter, laughter, laughter--- It's not your responsibility to make him feel good. He is not complimenting you. A sexual offer, a proposition, is simply inappropriate and it's an insult [in this context --JG]. It might be somebody important in your field, but that has not been my experience. My experience of harassment has been from people who aren't really there for the conference. So say "no" firmly, say "no" politely; if he continues, the simplest thing is to embarrass him. Say, in a voice that will carry at least three people beyond, preferably a little bit further, "NO, I DO NOT WANT TO GO BACK TO YOUR ROOM TO LOOK AT YOUR THEOREMS OR ANYTHING ELSE OF YOURS." --LAUGHTER--LAUGHTER--LAUGHTER--. This is a trick I learned when I was 12 years old in New York City, for dealing with people who harass you on the subway. You just embarrass them and they go away. You haven't done anything that they can criticize except raise your voice a bit. You can say, "well, I was agitated." It's much better not to make enemies. With high probability it won't happen. Following up on conversations: Once you talk to people, you can follow up on your conversation. We all have each other's names and addresses; most conferences do that. It's very easy when you meet somebody, to go back to your room that evening and make a little mark on your address list to follow up and send this person e-mail. If they ask you a technical question, and you don't know the answer, write to them and say, "I like that question, it's made me think and here's a possible answer," or, "Here's a step in the right direction. Do you want to work on it together?" If you write a paper that is related to something you talked about, send them the paper. If you really liked talking to them, you thought they presented an idea well, try to get them invited to a colloquium talk. Follow up. Keep the contacts up. It's not just a matter of talking for five minutes at a time, for instance. It's a matter of building ongoing relationships that'll last your professional life. And they do. And it makes the conferences more fun because you know people and you can continue conversations. I have one more point and then I can pass the microphone on. I put in the sentence at the bottom of the blurb on making connections, something about the "young girls' network." A lot of people seemed not to like that phrase. So I want to explain that I don't mean that we're young girls, I mean that there's a women's network that is young. Our network has not been around very long, has not been effective for very long, but it is effective. I've gotten at least one of the jobs I've had through it. I've gotten grants through it by women knowing my work. I'm here on this platform because of women who know me and my work, and it works. The way you work it is by mentoring. When you need to know something, you can ask one of the women here who seems to know a little bit about that area. If you are interested in a job at a place where one of us is, you can send us e-mail and say "Do you know anything?" We may or may not help but you can at least ask. And if you see an application, a job application, in your institution from a woman, you can follow it up, at least make sure that people read the folder because not all folders get read. You can make sure that women whom you hear give good talks get invited for colloquia. You can call women and say, "Hi, I'm going to be in the area, I'd like to give a talk." And the most important thing, the simplest thing when you hear a woman or meet a woman who impresses you, tell other people about her. Talk us up. And that's very simple. You go back to your institution, you met somebody you didn't know before and you say, "I met this really neat person in your field and she was telling me about this really interesting work," and that's a little increment to that other person. Those little increments add up fast. I'm Joan Feigenbaum. I'm with Bell Laboratories -- so I actually have one of those rare basic research jobs in industry that Jill Mesirov mentioned. I'm a theoretical computer scientist, and I work on complexity theory and the theory of cryptology; in the past I've worked on graph algorithms as well. I've been at Bell Laboratories since I got my Ph.D. from Stanford in 1986. I actually have some very concrete suggestions on how to make connections, and in fact I have two of them that you can act on today or a week from today depending on when you get home and on what you're doing tonight. I'm part of the theoretical computer science community, and the SIGACT bigwigs have asked me to invite you to the STOC reception, which is going to take place tonight. The other immediate action item is the database that has been compiled by the Computing Research Association Committee on the Status of Women. The database consists of women in computer science who are either currently Ph.D. students or have Ph.D.s in computer science or computer engineering. I'm on the CRA Committee, and I'm in charge of the database project. The database is actually starting to be used by journal editors, by program chairs, by potential employers, and by lots of people who should have your names. You're not required to be in it if you don't want to. If you are qualified to be in but you're not yet in, you can get in by filling out one of these forms and sending it to the address on the form or by contacting me electronically. Now if you are in the database -- that is, if you have filled out a form and sent it to me or to the CRA in the past -- you should recently have received a copy of your record to proofread, update and send back. If you have not received this copy of your record, you should get in touch with me electronically. Don't fill out another form, because that would result in duplicates. Those are two immediate short-term things you can do. What are some things you can do regularly? As others have said, it's very important to go to conferences -- not to go to all of the talks but rather to go to the talks of the people whose work interests you. These might be famous people, and they might not. Introduce yourselves to these people, and talk to them about their work. You should have a short summary of your own work prepared so that you can say something if they ask you what you work on, but don't take that question to mean that you're supposed to go into a long lecture on what you work on. Just say very briefly what you do. If they ask detailed questions, then you can tell them details; but most of them really want to talk about what they do, so make sure that you ask them technical questions about that. You can go to small conferences and workshops as well as large conferences. If you're having trouble getting your papers accepted to large, prestigious conferences, it helps to publish in smaller, less prestigious conferences and workshops first. I know this both from my own experience and from watching the evolution of various people's careers. Giving talks at small conferences and workshops will help, because there will be some influential people who will hear these talks. After some influential people are familiar with your work, you might have better luck getting things accepted to large, prestigious conferences. For those of you who are in theoretical computer science, DIMACS workshops are extremely useful, both for meeting people and for publicizing your work. Some of them are not competitive -- talks are scheduled more or less on a first-come, first-serve basis. So you should be on the DIMACS mailing list. You should find out who is organizing which workshops. When you see an announcement of a workshop that you want to attend, contact the organizer of that workshop and tell him that you want to attend and give a talk. That usually works. In general, look for appropriate forums in which to publicize your work, and don't be shy about volunteering to speak at those forums. Tell your adviser that he's supposed to introduce you to people. I know it's shocking that there are advisers who don't know that they're supposed to introduce you to people, but believe me there are advisers who don't know anything. My next suggestion is something I have heard conflicting opinions about, but in my experience it's a good thing to do. Circulate preprints of your papers to people that you think would be interested in them, including well-known people in your field. When I say this, I get a lot of crinkled noses and people thinking "Oh, who wants a flood of crappy papers from people I've never heard of?" But, in fact, I've always circulated preprints of what I've done. Exactly once in the past seven years, somebody told me, "Take me off this mailing list." And dozens of times, literally dozens, people said, "Thank you very much for sending me this. I'm really interested in this stuff." So I think that constitutes pretty good evidence that circulating preprints is a good idea. Personally, I would not ignore a good paper that was sent to me unsolicited. You should be aware that some people will ignore your preprints, but I don't think you will make enemies by circulating them, and you might create a community of interest for your work. Give credit to people who give you good ideas! Right now I'll take some time to say that I discussed this question of how to make connections with a bunch of people, including my colleagues. David Johnson, Peter Shor, David Applegate, and Richard Beigel all gave me very good suggestions. In fact, the suggestion I'm about to tell you is something David Johnson told me, and I think it's controversial. Johnson's suggestion is "If you're just getting started doing research, for example if you're still in grad school, one thing you might want to do is introduce yourself to a well-known editor in your field and volunteer to be a referee." Now of course he is an editor of several journals, and I'm sure he's hoping that at the STOC reception tonight, fifty of you will tell him that you're willing to be referees. I'm also a journal editor, and I would like to hear from those of you who want to be referees. I'm sitting here looking at Maria, who's also a journal editor, and I can see she thinks this is the best idea she's heard all day! Nevertheless, I think this is controversial, and I had a strange reaction to the idea when Johnson first suggested it. That's because I don't know how wise it is for someone just getting established to spend a lot of time refereeing. I think your primary responsibility is to publish your own work. But, I agree with him that a limited amount of refereeing can be to your advantage; for example, if you introduce yourself to me and tell me that you are willing to referee papers for the Journal of Algorithms, I am much more likely to remember you than if you tell me "that's a nice dress." I'd like to generalize Johnson's suggestion and say that you are much more likely to be remembered by famous people you meet if you volunteer to do something that they need to get people to do! The general principle you should observe about refereeing and other "service" activities is that the reward you will get should justify the amount of time you spend. Making a good impression on a senior editor in your field can be a significant reward. I think that one of the reasons that I am an editor now is that the people who appointed me knew that I had done a responsible job on various things that I had done for them in the past. You can ask to be on program committees. The first time I was on a program committee was three years after I graduated, and I asked to be on it. Once you've been on one, you tend to be on many. The first time, I asked the program chair directly whether he would put me on the committee, and it turns out that he did. If that's not possible, you could let the SIG president or someone else who's very influential in the community know that you're interested in being on program committees -- or that you're interested in doing whatever it is you're interested in. Some people think that this practice of volunteering, especially when what you're volunteering for is coveted, is taken wrong or is perceived as impolite or pushy. I think it's not such a big deal. It's really done a lot more often by a lot more people than you realize. Next I have another suggestion by David Johnson. You should know what the important people in your field look like. If you don't know what they look like, you may miss an opportunity to meet one of them. Johnson told me that this actually happened to him -- in the dark ages when he was just starting out. He didn't know what people looked like, and he wound up chatting with someone about the movies or something and later realized that this was someone he should have been introducing himself to. I guess it worked out okay for him. What's ironic is that I was at a DIMACS workshop recently where he was chatting about the movies with a graduate student who is working on something directly related to his previous work, and this grad student didn't know who he was. So, this really happens all the time. The moral is that you should know what the important people in your field look like not so that you can bask in their presence but so that you can take an opportunity to introduce yourself if you have a technical thing to say. I think that's it as far as specifics suggestions are concerned. I'll take questions. I'm going to take a slightly different slant on things. Since the opposite of being well connected is feeling isolated, you could view this discussion as being not only about making connections and research contacts but also about avoiding the feeling of isolation. A conference is a very obvious place to try to start making connections, but it's also a very easy place to feel isolated, to feel that you don't belong. So I'll start talking about conferences but I'd also like to talk about how you might make connections outside of conferences. Whom should you meet at conferences? Don't ignore the people who aren't famous. I think a lot of us have had the experience of someone talking to you and looking over your shoulder for somebody more important to talk to. Getting snubbed. The lowly graduate students of today could be the superstars of tomorrow. Don't ignore them. I don't suggest that when you meet somebody, you try to assess what their future will be like and then decide how many minutes to spend talking to them. Go ahead and talk to them in any case. Why should you meet them? The people who are your peers are the people who will form your research community in the future. They might not seem important now, but you'll be with them for a long time. It's also an easier sort of contact to make, at least for me. I feel pretty shy going up to important people but I feel much more at ease talking to people at my level, and there are also more natural topics of conversation that will come up. You're at the same stage; you're going through the same sorts of experiences. Talking to women might be one of the ways in which you can make this easier; talking to women might be easier for you. Now, I've talked about getting research contacts and said you should talk to graduate students because they might be superstars in the future. But, even if they're not in your research area, it's important to have the social connections. Those are valuable as well. You never know what your research might be in the future or what their research might be in the future; there might be overlaps. And even if there never is an overlap, if you make the social connection, then you're part of that person's research-and-social circle, and then can meet other people in that circle. Those people in turn might become research contacts. And even if you don't make any research contacts through that social connection, there's still a person whom you like to talk to at conferences, making conferences seem like a happier, more welcoming place to be, and that's important. You're going to be happier. It will feel more worthwhile to go to conferences. Where do you meet them? Joan already mentioned that you don't only have to go to big conferences and I think that's important, not only because it might be easier to get papers in [to smaller conferences --JG], but because you might feel more relaxed there. There might be fewer people, the conference might be on for a longer time; there may be more opportunities to make the contacts. If you're a graduate student you might consider looking for conferences that are graduate student operated and organized, and for which all speakers are graduate students. There's one of this sort that is run in Buffalo. There might be retreats or local seminars of that sort. There are summer schools, NSF workshops, where you can meet a smaller number of people over a longer period of time. Or consider doing research during the summer in some location other than your institution. That's another good way to meet people. How do you meet them? Well, I'm kind of tongue-tied and I'm not good at going up and asking people whether they have dinner plans or that kind of thing. Assume that a first stage as Joan mentioned was to have your advisor introduce you. That eases things a little bit. Another natural starting point is to meet with graduates of the institution you went to and start meeting people through them. Now this all assumes that you have an advisor who's already well connected. It also assumes that you go to a big enough school that there are lots of graduates out there. That's not necessarily the case. What can you do then? There are social events at a conference as has been mentioned. Lunch can be particularly good, since you're sitting in one spot. You don't have to worry about people looking over their salads for somebody more important to talk to. They're stuck with you for the duration of the lunch. But be a little bit careful. I know that if you want to go to a lunch and get involved in a technical discussion because you want to make research contacts, the worst thing to do is to sit next to people who are having a technical discussion, because they're probably working on a paper for a deadline. They don't want to talk to you. They're just talking to each other. They'll probably ignore you and that's a wasted opportunity; there could have been other people sitting there whom you could have actually met. So, I think it's better to go to a lunch table where people are talking socially. One other aspect that probably isn't brought up that much: another good way of meeting people is to help other people meet people. The few people that you know, you introduce to each other. You may know people who have either social or research reasons that they might get along. Make an effort to do that. Conferences of course might be too expensive or too rare to be the only way to rely on to make contacts. So what can you do if you're not at a conference? If you are reading a paper and you have questions about the paper you can, through the anonymity of e-mail, not feel quite as embarrassed about asking questions about the paper. It's also a good way of paving the way to actually meeting the person in person if you have a chance in the future. It can make you feel more comfortable about that. You can meet visitors to your department, seminar speakers, post-docs, people who are on sabbatical, students of people on sabbatical and then try to keep in touch with them afterwards. If they're people you'd like to meet and you don't feel comfortable going up to them at conferences, if they're people who are looking over your head, you might consider suggesting that they come as speakers. If you're a graduate student and don't have the power to organize a colloquium yourself, see if you can ask the colloquium organizer or your advisor to make that sort of suggestion. You can try to stay in touch with people electronically or even introduce people to each other electronically. It's another option. I'd like to end with advice that has worked very well for me and this is something that's very local and it doesn't sound like it has direct benefits but builds slowly. When I was a graduate student I met every month with a group of women graduate students in computer science and electrical engineering. We talked about all sorts of things. It made us all feel a little bit more comfortable about being in graduate school because we had a community of like-minded people. We could speak openly, we felt. These people of course -- we didn't think about it at the time -- but these people have graduated and have gone to other institutions and have started forming a network of their own. We've learned from each other's experiences about interviews, about writing grants, about all the steps along the way. And it wasn't something which we ever intended, which is the beauty of it. It was just something that we wanted to do, made us feel comfortable and now has grown and blossomed into more of a set of connections. So, the last point I want to make is, don't expect the connections to come overnight. They come slowly and but those are the strongest and most beautiful ones. Don't give up. Thanks. Shall we take a little time for questions? Okay. So we've told you how to start a conversation with the person next to you and how to get onto a program committee, and a few things in between, and now we'll take questions. We'll try to answer really briefly so we can get a couple of questions, if people have them, and then get to lunch where we can practice. I have a question. What are the criteria for being on a program committee? There must be some. There must be something that's a qualification for being on it other than just you're interested. I think that the official criteria for being on a program committee are exactly what you would expect: You're supposed to be a well-known researcher in the field or an up-and-coming researcher in the field. Now by and large that's true. It's rare that people who are not themselves widely published are put on program committees. It's not unheard of, but it's rare. However, it's not the case that everyone who does publish a lot winds up being on program committees. That was why I suggested letting it be known that you want to be on program committees, because there are very qualified people who are passed over. Not for any nefarious reasons. They're just passed over because nobody happens to think of them at the right time. If you make it well known that you're interested in doing this, I think it increases the chances that somebody will think of you when making up a program committee. As far as other criteria are concerned, one thing I've heard program chairs say is that they want to have people who are well connected. They usually say they want them at well known places where there are a lot of other people. This is because they expect program committee members to ask their colleagues for input on the submissions. This might be unique to my community, and perhaps in other communities it's sort of frowned on for committee members to ask other people's opinions. In any case, I think the only thing you can do to increase your chances, aside from publishing your own papers, is to let it be known that you want to do it. And go to the conference! It's a little thing, but conference committees do prefer to have program committees made up of people who attend the conference. There's one conference that the three of us (Judy, Joan, and Naomi) share some interest in -- the Structure in Complexity Theory Conference. I was just on the program committee for Structures '93, and I know that the Structures steering committee is loathe to ask someone to be on the program committee who has not attended the conference, even if the person's research makes him an excellent choice. Alright. I have a question about preprints and sending out preprints to people. What do you do if you're going to cite something in the paper, in the preprint, when it hasn't been published yet and how do you deal with that? O.K. The best way, I think, to deal with that is to make the preliminary version into a technical report so that it can be cited. You wait until you have the technical report before you...? I was assuming that making something into a technical report was an instantaneous process. If you're in a situation in which you have to wait before something comes out as a technical report, then I would say first try to change that procedure at your university, because technical reports should be a time-stamping mechanism -- if there's a long delay, that's not good. But I think it's perfectly ok to use the citation format "author, title, `Manuscript,' date." I've done that, and I've had my own preprints cited that way. If what you're worried about is that people might take your ideas and not cite them, ... well I don't think you should send your preprints to those people. But you can make people in general understand that, if they want to cite your ideas, that's okay, but that they have to at least say that these ideas came from your manuscript and include the manuscript date in the citation. I have a question about the use of business cards. I have business cards and I just had a lengthy discussion with a recent graduate on whether she should put Dr. or Ph.D. Yet I've been to several conferences over the last couple of years, big and small conferences, and I have never seen the distribution of a business card. Therefore I was not comfortable with that. I just wonder about your comments on the use of these things. I've rarely seen the exchange of business cards. I've seen it at the Crypto Conference where there are a lot of government and industry people. I think it's rare among academics, at least when they're talking to each other. On the other hand, I don't think it hurts to have business cards. It can be a convenient way to give your phone number to someone. It's very good when you want to go up to someone who gave a talk and you want to ask them to send you a paper or something... you can hand them a card. Uh-huh. Yeah and you can write on.... I do think you need business cards if you're dealing with industry at all. I just wanted to say I go through scads of business cards. I never admitted it before. ..... I guess that's one more way in which my job is not "real industry." I have a zillion business cards that I've never used for anything. I have a remark about the difficulty of making connections at conferences. I find that in big conferences like this FCRC, it's just really awkward to talk about your scientific work, having dinner with people you don't know or asking people questions after you've listened to their talk for twenty minutes and you barely have understood what the problem is. I mean it's very hard to come up with interesting discussions. And then even social talk is really unsatisfactory. After three days you're really tired of it and you feel you don't get anything out of it so I find that when I go to a big conference what I like is meeting the people I know. And the way I meet new people is to go to small workshops where the talks are for one hour so after the talk if I'm interested I have enough of a basis for an interesting discussion and the people are closer to my field. Often we do things, since we are just a small workshop with 30 or 50 people, like hike together or do some other activities, which make it easier to know people on a more informal, natural basis. I was program chair for the 1990 ACM SIGmetrics Conference. From the perspective of program chairs and how they choose committees, I just wanted to say that I think it is fine if you speak up in certain ways that you'd be interested in being on a program committee, but I think one thing to recognize about the process is that the program chair is the one who takes responsibility for choosing the committee, and is going to try to choose the committee based on a whole lot of criteria. The way I think is the most effective to speak up is to talk to somebody who knows your work well and who thinks very highly of your work and who, at the appropriate time, suggests you as a very capable person and who'll leave it at that and let the process take care of itself. There are lots and lots of very good assistant professors who don't get on committees until after they've been at the business for several years, so I don't think it's something you need to feel that you have to achieve with the first couple of years or two or three years as an assistant professor. It's really good to recognize what the process is about and the responsibility that the chair takes and all of the different criteria the chair tries to take into account in choosing the balance of the committee. So it's just I think pressure can come in the wrong way and can hurt you if you aren't aware of that whole issue. Yeah, okay. I wasn't suggesting high-pressure. I think Mary is very right. I know that Joan and I are in violent agreement on this. I think that Mary is very right, and I'm glad that she mentioned that it's good to have an intermediary put your name forward. Instead of talking directly to the program chair, you might go through a senior person who knows your work well, or you might go through the SIG president or somebody who's regularly asked to suggest names. I guess I should make clear that I'm not suggesting that you nag people or that you put pressure on people -- just a very clear, simple statement to the effect that you're interested in serving will suffice. I'm Joann Ordille from University of Wisconsin. I just wanted to share some advice I got on how to meet people at conferences. And the advice was to ask people you run into who is working on the thing that you're now most interested in, and remember their name. And then when you meet that person, you say, "Oh so-and-so said that you're working on this." And normally that's enough to get them to tell you all about what they're currently working on and that's what you're interested in too. Then it's pretty natural to start up a conversation. And this worked really well for me when I couldn't quite make connections. Hello. I'm Ruth Silverman from University of the District of Columbia. I just wanted to mention that Ian Parberry said at a STOC meeting or FOCS meeting, I've forgotten which, a couple of years ago, that if you had three papers accepted and you haven't refereed one you're not doing your part of your obligation to the community. So, if you look at it that way, that by putting yourself forward you are fulfilling your obligation rather than being pushy, it might affect your outlook. Well I just wanted to tell you about my first experience attending a conference. It was the AMS Annual Meetings. It was in St. Louis. It was, I think, in 1976. You should tell people what "AMS" is. This is the American Math Society and this is a humungous meeting. There are maybe two or three thousand people there and I went to a special session on the geometry of Banach spaces which was an area I was very interested in, and this was going to run all day for two days in a row. The format was that each speaker spoke for twenty minutes and then there was a ten-minute gap while people could move around and then the next talk started for twenty minutes. There were probably about a hundred or a hundred and fifty people in the room and I didn't know a single person. I did know a lot of the names because I'd been taking a graduate special topics course in this area and I actually, frankly, was quite amazed to see what some of the people looked like, having heard their names. I think I was the only woman in the room most of the time. Sometimes there was one other woman there. My experience with the first day was that I sat and I listened to the twenty minutes of talks and I was really excited. And then I squirmed for ten minutes while everybody else got up and talked to their friends and I thought everybody's looking at her and she doesn't know anybody. And I didn't. I went home and I didn't have anyone to go out for dinner with that evening and I went back to the hotel and I said I'm never going to go to another conference in my life. This is terrible. I'm really humiliated by this experience. But I did go back the next day and do you know it was amazing. Person after person came up to me and asked who I was. I mean they had been watching me sitting there all by myself and wondering who is this woman? Where did she come from? Is she interested in the geometry of Banach spaces? And I got to meet all the big names and all the big stars and they introduced me to each other. Now if I'd given up after the first day, which really was truly horrible, it would have been a pity. So, my advice is, sometimes the way you see what's happening is really different from what's really happening and if you just sort of stick with it, it gets a lot better. Is there one more question? Or anecdote? Okay. Well thank you, panel.