One of the things I love about living in Philadelphia is that there are a lot of colleges here, and that means a lot of math talks. I could easily fill at least half my time just attending (and navigating public transportation to get to) talks in the area. Going to good talks makes me feel connected to the world of mathematics in a way that nothing else quite does—seeing “live math” can be a little bit like seeing live music crossed with solving a puzzle, two of my favorite things. In the best of all math talks, I have to remind myself not to interrupt with loud cheers at the good parts. I take this as a sign that I chose my job well.
That said, it happens sometimes that I go to talks and find myself just not engaged for one reason or another. Maybe I had an intense day teaching and spaced out for a few crucial minutes at the beginning. Maybe I followed for the first 20 minutes and then the speaker stepped to a whole other level of specialization. Maybe the speaker is very quiet and I can’t hear what they are saying. I’m assuming this happens to you sometimes, too? However it happened, there are still 40 minutes left and you realize you have no idea what is going on. What do you do (besides play Seminar Bingo)? How do you make the most of those 40 minutes of your life?
Here are some of my strategies:
1) My first priority is not to fall asleep. So I will do anything at all not to fall asleep, even just draw circles on a piece of paper, get up and stand up at the back of the room, or leave if necessary.
2) Sometimes I start working on my own math on paper or on a computer. This can feel productive, but I honestly don’t enjoy this much because it makes me feel self-conscious and somewhat silly. It just feels weird to blatantly ignore someone who is speaking in front of me. Why am I there? If I’m really going to check out to that degree, and the room is large enough that it wouldn’t be noticed, maybe this is okay, but I might as well just slip out of the talk and work somewhere quiet. If it is a fairly small room and it is possible that the speaker can see my lack of attention, I feel like I should keep looking back up and give the appearance of following the talk, which actually feels kind of creepy and ridiculous—I feel like I’m lying if I nod and smile when I really have no clue what is going on. So maybe it is better just to commit to checking out and be obvious (though quiet) about it. Just to be clear, I don’t always think other people are being jerks when they are clearly working on something else during talks. It just feels wrong for me. However, if the room is fairly large it can still feel better than leaving the talk.
3) Sometimes I work on my own math in my head. This feels less productive than working on paper or computer, but also less silly and embarrassing. I just maintain general eye focus on the speaker while my mind wanders.
4) Sometimes I try to take lessons in style from the speaker. If they are a really good speaker, I try to figure out why I enjoy listening to them even if I don’t understand the talk. If the talk seems ineffective, I try to figure out why. Oh, maybe it’s because they never explained any of their abbreviations! Oh, wait, I think I did that in my last talk. Never again! Sometimes this study can get so interesting that I take notes, then find myself a little bit in situation 2 above, and then need to check myself and go back to 3.
5) If the talk feels like it should be accessible, because it is for a general audience or it is close to my area, I just refocus and try to filter something, anything mathematical from the talk that I am watching. I make it a sort of game: I am a detective, what can I figure out about this story given these clues? That makes confusion less frustrating, partial understanding much more rewarding. If I have a lot of energy I will take notes and organize the information as much as I can; if not I will just listen to it as a story, look for the big picture.
Number 5 is the most satisfying, and this practice has really paid off for me a few times in the last two years. One of my projects is implementing an algorithm (mostly devised by many other people) to solve a particular equation in the group of S-units. My collaborator and I had to solve this equation to do another project, and we decided to try to generalize our code. It turns out that solving the S-unit equation shows up in a lot of other people’s work as well. In the last two years, it has happened 4 or 5 times that I find myself a bit lost in a talk, trying to get some big picture idea of what is going on, when all the sudden here comes the S-unit equation. So then I suddenly have a strong connection to this part of the material and tons of questions spring to mind. In some cases I then get to talk with the speaker about our project, and through this conversation manage to fill in the blanks that left me lost in the talk, so I learn a lot of new, very relevant math. Even if I don’t manage to talk to the speaker or totally understand their talk, I can connect my work with another general area and have a new example of how our work is valuable in the larger world! Hooray!
Every time method 5 pays off, it makes reinvesting in talks easier. But I mention 1-4 because I am being realistic—I don’t always have the energy or desire to really wring all the content out of a talk when I feel confused. So, I would really like to know—what are your seminar strategies? Thoughts on getting the most out of talks once you realize you are lost? Is it better to leave the talk or to fall asleep? What do you think about working on other stuff during a talk? Is it worth being there even if you are not paying attention?
At every university I’ve been at (which includes three grad schools and three professorships) #2 is far and away the most common approach.
I have gone to some talks knowing in advance that it’s going to be #4, because I know just how awful the speaker is and I want to experience life to the fullest. This takes a really special speaker, though.
Wow, talks so bad they become adventures. I guess I have been to talks like that, and wished I’d looked at that way going in. I agree that 2 is the most common approach at research universities. Maybe because talks are really frequent and people are super busy with their own research–you couldn’t get anything done if you engaged fully in all the talks in some departments. Maybe it’s worth it to get the first few minutes of the talk and learn a bit, then check out when it gets hard. Or do people just like to show up as a sign of respect, even if they don’t plan to follow carefully? Is it actually more respectful to show up to someone’s talk, then obviously work through it, than to not show up at all?
I think that it is definitely more polite to go to someone’s talk and do other work when you stop following. However, now that I am in a small department, I usually try to think of a question to ask at the end even if the talk is far outside of my area. To me there is nothing more embarrassing than having a colloquium with no questions for the speaker.
Yes–except maybe having nobody show up for colloquium at all. So I think you’re right that going to the talk and working is often kinder to the speaker than not going at all. I guess the “best practices” are probably different depending on the type of talk and size of the department…
One reason I think #2 is perfectly reasonable is that most talks are structured with increasing difficulty, because most audiences are mixed. Put another way, if the talk is structured according to this common (and common-sense) format, then you _should_ get lost somewhere along the way, with “somewhere” dependent on your closeness in specialization. At that point you go to #2.
Conversely, if the speaker has as usual structured things this way, then they’re expecting to lose you somewhere, and so not be offended by #2.
This is reassuring, for those times when I do completely switch to my own work–many speakers will not think this is rude. I am not offended when people are working during my talks, as long as someone still seems to be following and making eye contact.
In my opinion, people attending a seminar/colloquium etc. are making a favor to the speaker by attending the talk. I believe it is the duty of the speaker to inform herself/himself about the audience and its background and structure the talk accordingly. It is hard to give a math talk that will capture the audience from the beginning to the end especially when giving a colloquium to people from various research areas. I do not take it personally if people space out during my talks, but I try to remember where this happens and improve things for the next talk.
The advice I give my students when they give talks is that they should structure it as past (background, history, why is the problem important), present (what did the student do and why it is important), future (what is left to do in the near and far future) and that practice is very important (give as many talks as possible at all possible levels).
Personally, I’ve tried #2 and typically fail because I’m especially bad at tuning out a speaker and/or find it too hard to pay attention to my own work and theirs at the same time. So I started doing number #5 quite a bit, even down to analyzing the layout of their slides. I found this gets old after a while because bad talks are often bad in similar ways.
Nowadays, I think the thing I try and do is get the gist of what they are trying to do… it’s a proof by contradiction, or they’re using circle method, or whatever. I found that over time you start to see patterns in peoples talks in areas you don’t understand; you learn the common objects of interest in this area and the types of things people are trying to learn about them. So even if I can’t follow along with the talk in that I don’t know what exactly they are doing – something about periods of orbital integrals or a Langlands dual, or heaven forbid both – I at least try and get the basics about the objects at work and the bigger picture goals. Over the years I actually feel like I have learned a lot of math this way (or more accurately ‘learned about’ a lot of math). I can surely recognize a lot more math tools, and so I’m more familiar with what I need to learn about when it comes up in my own work. At least I know a little bit more words to say when talking with people about what they do.
Funny enough, I think this came out of fear of sounding dumb. During my post-doc, some of the faculty would nearly always chat about the talk afterwards in the halls walking back to the department, and usually in a ‘same-ole same-ole’ casual kind of way. The thing I found odd was it was never about an area of their specialization – they just seemed to know about everything. I was a new postdoc trying to fit in and so I would try and have something to say about these very hard to follow talks to participate in the conversation, and just commenting that they were hard-to-follow every time was something I couldn’t bring myself to do. So I tried to follow the talks the best I could with the goal of having something casual to say about it afterwards. I quickly realized that it wasn’t actually that hard to get the gist of a talk to make casual chit-chat about it. After going to talks every week, many in similar hard-to-follow areas, I actually started to learn quite a bit about them and the talks became a little less hard to follow. I’m no expert in linear forms in logarithms but I know that if I ever see one I should try to use Baker’s theorem, essentially from trying to make semi-sensible chitchat.
(no offence to anyone who studies the things mentioned above – they are just used as examples of subjects I know about, but don’t actually know much about)
Good story about wanting to be able to talk about the talk! I feel the same way–talks are in some ways more useful as a chance to connect to people than a good way to learn a bunch of math. I would like to be able to talk with the speaker or other people in the audience about the talk, and this is better and more fun the more I understood. Even if I only understood enough to ask a clarifying question.