Monday, February 27, 2012

Don't Fence Me In

This is self-centered and unreasonable, but when I am invited to give a talk, my ideal situation would be to give a vague title (Cool Science Things) that isn't much more informative than a list of keywords (cool, science, things). I don't really believe that people will come to hear me speak no matter what I talk about, so in my sane(r) moments, I do understand that a talk title is necessary to provide some clues as to whether a talk has any hope of being potentially interesting to those considering attending (for those who have a choice).

Even so, I like to give talk titles that are as vague as possible, so that I can give whatever talk I am most in the mood for when the time comes. This vagueness allows me to talk about the coolest (meaning: the hottest) science.

Some departments that bring in invited speakers want not only a title but also an abstract, presumably to provide further information for those who are considering attending, or just for general educational purposes. I always comply, but I don't like providing talk abstracts because I feel that it limits what I talk about, more than just a title.

So, I can write a brief, vague abstract, no doubt annoying those who like an informative abstract, and therefore still keep some freedom as to talk content. Alternatively, I can provide a talk title and abstract, and then talk about something different; perhaps not completely different, but I could add in a topic (or two) not mentioned in the abstract but still broadly covered by the talk title.

And that leads me to my question. Which is more annoying:

- A talk accompanied by a vague title and/or abstract;
- A talk that is only somewhat related to what you thought it would be based on the title and/or abstract; or
- Neither, you are going to be happy or unhappy, depending on how good or bad the talk is; it doesn't matter what the title/abstract were.


34 comments:

Anonymous said...

ummm if it's a conference I want both specificity and accuracy, though I'd take the specific but unrelated one over the vague one. Usually I have to choose between multiple talks, all of which I might be interested in, and would rather go to the one I can get the most out of.

If it's a dept talk I don't care, I only read the abstract enough to know what the super broad field is. Be as vague as you want.

Female Science Professor said...

I was thinking dept talk, but there are some related issues to conference talks and abstracts (but important differences; abstracts are peer-reviewed and published).

sixsnowflakes said...

Specific but unrelated is far better. I've never been disappointed when a speaker has something even more interesting to talk about. Vague abstracts are annoying since they don't help to decided whether the talk is worth going to or not. I've only seen no abstract from poorly organized students who are required at the end of their masters or Ph.D. to give a seminar. I never attend talks where they aren't organized enough to submit an abstract.

Anonymous said...

sometimes the organizers are not organized enough to ask for or post an abstract

Anonymous said...

For dept talks, I just want enough info to judge if the work is vaguely in my field. Eg I study donuts and need to know if the speaker has something relevant to say about muffins or if she works on the unrelated topic of dolphins. If she/he worked on donuts, I'd know who is was anyway.

So a pretty general title / abstract is fine. The only disappointing talk I've been to recently promised something specific, tried to deliver that something and failed.

mOOm said...

In economics the norm is to require a full paper for an invited seminar...

EliRabett said...

Depends where. At a huge conference vague means not on my schedule. At something like a Gordon conference not a problem, even good. For posters, who cares, Eli wanders

Anonymous said...

For a departmental talk, I'm generally in a position of triage: is this talk interesting enough to give up [other commitment X]. So for me, a specific abstract followed by a loosely related talk is much preferred.

This is because I'm usually going to see a speaker more than a subject. Even if the subject of the talk changes, a nice specific abstract gives me lots of clues about how the speaker thinks and whether I'm going to find them interesting or not.

This is especially true since even a "very different" talk is still generally going to be pretty closely related, just because most good researchers' programs are coherent at some level of abstraction.

queenrandom said...

For a seminar, I'm fine with a vague title, as long as I don't feel bait & switched when I get to the talk. That annoys me as much as a bad talk that's on topic.

Anonymous said...

The policy in my department is to not bug the speaker for a title/abstract until a week before their seminar- earlier is of course appreciated, but we understand wanting to wait to choose the topic. FSP, do you think a week is adequately close enough so you'll know what you're going to present?

As for my preferences: I want the title/abstract to indicate which subfield the person is in, and what questions they are working on. I guess that's reasonably detailed. But I do NOT want an abstract that gives away the punchline, unless it's super controversial (in which case I might do some review before going in; I'm a grad student).

Anonymous said...

I am not annoyed at all by a vague title/abstract, or even a lack of abstract. I figure if the dept. invited them, they must have invited them for a reason (e.g. they are doing good work or are engaging speakers or both).

Anonymous said...

This one -- I always go to our departmental seminars (and my students are required to go). So....

Neither, you are going to be happy or unhappy, depending on how good or bad the talk is; it doesn't matter what the title/abstract were.

Anonymous said...

For me, the worst is when I choose to attend a talk based on a semi-informative title or abstract and then the talk isn't what I was expecting. My time is very limited (surprise, surprise), and I therefore choose my talks very carefully. The Science you are in the mood to talk about that day may not be the Science I was interested in hearing about, and I will leave the talk resentful of (from my perspective) wasted time. One small vote in favor of speaking on the topic of the *informative* title or abstract.

mathgirl said...

If the talk is in my field, I definitely don't care. If the talk is outside my field, I think a vague abstract is slightly better. I just need to have a vague idea of the area.

On another topic, what do you think about large conferences where you have to submit a specific abstract (a short abstract, not a paper) several months in advance? I'm thinking of the American Mathematical Society, for example.

Anyway, the point is that having to decide what I want to talk about in 4-5 months drives me crazy. Sometimes I use these submissions to force myself to finish a paper.

Flora said...

A talk only somewhat related to what you thought it would be - mostly (selfishly) just because if I'm making space in my day to come to a departmental talk, there is probably something about the title/abstract that drew me in. But as long as the talk was good, I'd just be a little bummed I didn't get to hear more about *super interesting to me topic*.

Anonymous said...

Recently our Ag department had someone in to talk about the Farm Bill, which has a potential impact on my work. Unfortunately, the talk was on the impact of the currently proposed bill on the budget of a USDA department that has nothing to do with my work, in a dolphins vs donuts way. It would have been nice if the talk notice had been more specific than "so and so from the USDA will be here to discuss the 2012 Farm Bill"

EscapedWestOfTheBigMuddy said...

//Assuming a departmental colloquium or similar...

Pretty near the only Cool Science Things talks that I don't enjoy are those aimed at domain experts outside of my domains.

So an inviting general audience title/abstract applied to a talk on sixth order optical effects wastes my time. (This category seem to include most of those poorly organized student talks that other posters have complained about.)

And sometimes I'm too busy to go to even a Cool General Audience Sciencey Talk outside my domains, so I like for the title and or abstract to clue me in--is this going to be a "for fun" talk or a "important career development" talk.

Beyond those two things I don't care much. Bring it on!

Anonymous said...

I try to go to all department talks regardless of the topic (because we are a small department and I think everyone should make an effort to support the meager seminar program we have), so I would vote (3).

In the departments I've been at, it's been okay to provide a title/abstract a couple of weeks before the talk. Isn't that a win-win situation? You get to hold out on committing to the topic, and anybody who wants to decide whether it's worth their time to go to the talk can check it out a week (or day) ahead of time.

For a conference, I'm happy with a vague title/abstract... preferably something that at least hints to the general topic (which I assume would be tricky if you wanted to bounce between entirely different projects). Actually, my pet peeve is probably when people use titles that sounds like review articles or talks: "Why do dolphins swim? An investigation of dolphin swimming habits?" And then the talk turns out to be an ultra-specific talk vaguely related to the grand-sounding title. I've learned not to fall for those ones as often.

Amy said...

I don't feel that badly about changing my talk if something extra-cool comes up at the last moment, even if it takes me away from my title (which is always vague). By change, I mean that if my title/abstract indicated that I was going to speak about pink fluffy bunnies I wouldn't hesitate to add some awesome new data on blue fuzzy bunnies. I'd avoid changing the topic entirely to discuss the research in our lab on slimy green lizards.

How often have you been disappointed because you went to a talk thinking you'd hear about X and instead, you heard about Y? This has happened to me, but not often. Probably most often at conferences, not in departmental seminars. If it's a good speaker, I often take something useful from the talk no matter what the topic. Unfortunately not everyone is a good speaker...

It's interesting to hear different opinions about this because I was "raised" in a scientific group where we ALWAYS went to the weekly seminar. And the monthly joint seminar at another university across town - all 30 of us would hop on the subway together, it was probably quite a sight. I've kept up this practice with my own group - sometimes it's a waste of time, but usually you can get at least something out of it (tips about how not to give a seminar, if nothing else). And you're sure not to miss an unexpected gem.

Anonymous said...

I'm with an other anon, if I go to a dept talk outside my field I use title and abstract to triage whether the talk promises to be interesting, eg can the speaker grab my attention? people who write broad but cool abstract usually can tell a good story, which is important outside the field. we also use the abstracts to guide the papers we select for students to read before the talk.
so, speakers, please make an effort!

SassySparky said...

I generally never expect the talk to match up to the abstract or the title. I think those are just to put on fliers that advertise the seminar. The most informative abstracts are ones that generally describe the work being done in one's lab, and allow someone to be aware of the subject area of the seminar.
I also don't mind when someone comes to give a seminar and they say, "Hey you know I was going to talk about subject a but we've had some cool results in a different area so is really lie to speak to you about that today." I'd much rather listen to an obviously enthusiastic speaker than someone who feels forced to speak about a prescribed topic.because of an abstract they sent.
I think this definitely applies more to invited seminars in departments or at smal meetings than at the national meeting for a research society, of course!

John Vidale said...

I'm lured in by the combination of a speaker with a good track record and a topic that I haven't already heard in great detail.

If you feel an even newer result is more interesting, by all means go for it, but you'd better be right; little is more unsatisfying that working free time to get to a talk on an important topic, then hearing boring or incomprehensible prattle. Some people tend to get enthused before their message is coherent and vetted against obvious mistakes.

Vague titles give me the impression that the speaker hadn't yet decided what to say when pressed for the title - not a good sign.

For Eli - at least at the Gordon conferences I've chaired, switching in a different talk would be most unwelcome. The program and discussion is planned to highlight new developments of broad interest, and planning took place on several levels - with the chair and co-chair, with the session organizers, and with the speaker.

Anonymous said...

I'd prefer specific. If you want to talk about something else (or something "cooler"), do justice to what you said in the abstract (10-20 min) and then move on to the newer stuff. If an abstract is vague and I haven't heard of the speaker, I most likely will not go.

While I do understand that it is hard to determine exactly what you want to talk about a month before the talk, I am quite sure you would have a fair idea 1-2 weeks before. Things can't change that much in the last minute. I feel you owe it to the audience to be able to give a reasonably correct abstract then. Perhaps, the best way, in my opinion, would be an early vague abstract (or perhaps just title) and then a more specific abstract a week before the talk.

Anonymous said...

Definitely "A talk that is only somewhat related to what you thought it would be based on the title and/or abstract"

We have lots of seminars around campus, so going to what seems like a REALLY relevant talk across campus that ends up COMPLETELY different is a huge waste of ~90 minutes for me.

Anonymous said...

I find it so, so annoying when talk titles aren't accompanied by abstracts. (This seems normal at my postdoc institution, and I don't understand how people stand it.) For interesting titles, I end up having to search the internet to figure out what kind of methods and approaches the speaker uses, which helps me decide if it's worth attending. I also like to know if the talk is simply going to be a rehash of a recent paper.

Anonymous said...

definitely a clear title and specific abstract. going to a talk is invariably in lieu of something else that needs doing!

Besides which, vague abstracts smack of narcissism/disorganizedness (even if this isn't true of the speaker) and strongly disncentivize me from going.

Like the others said, if the speaker then talks about some different closely related topic I won't be too disappointed. hopefully they won't stray too far.

sam said...

I don't mind if people change their talks a bit, but I have been to a few talks where I was very disappointed after the speaker changed the topic drastically. Usually, I find that this is the case when the audience are mostly bench scientists and the speaker a pure theorist or vice versa. As an example, if I go to a talk about a mathematical model that can be applied to problem X (which is vaguely related to my field), then I want to hear about problem X and not problem Y, even though the same model can be applied there as well. I like going to talks that are a bit outside my comfort zone, but if both the methods and the topic are totally new to me then it has to be a truly amazing speaker to pull it off.

Anonymous said...

Depends who the speaker is. I make my decision to attend based on all the information I have about whether it will be interesting. If I know the speaker is a good one whose talks I'm likely to enjoy generally, then I don't much care if the abstract is vague and/or unrelated to the talk. If it's someone I've never heard of, then obviously the title (and abstract if there is one) are all I have to go by so those matter more.

In practice this means the title and abstract are more important when the speaker is someone junior, since then it's more likely I've never heard of them.

Anonymous said...

Really? As a scientist, you can't be troubled to know a week or two in advance what you might have to say that's worth the time of your audience?

A rather specific title and abstract, with high probability of a close match to the talk, are the norm in mathematics, and it's rare that one sees a mathematician who finds this an unreasonable standard.

Are you worried you're going to randomly do a new fascinating experiment the morning of the talk that can't wait another month to be the subject of a lecture?

Female Science Professor said...

To clarify some things:

Some places that invite me to give a talk ask me to give a talk on a particular topic, but most leave it up to me to choose the topic. At any given time, there are probably 4(ish) topics I could talk about with great enthusiasm and with new results. These would all have different talk titles, so I do have to choose to some extent. In some cases, I can guess what might be of most interest to the department that invited me, but in others, I don't know (or I am told that any of my likely topics will be fine). In some cases, the request for a title/abstract comes many months before the actual talk. For those, it's particularly difficult to predict what would be most exciting to talk about when the visit rolls around. So I try to balance being somewhat general but not vague to the point of providing useless and annoying information; hence, this post.

Anonymous said...

For me, I prefer a specific topic & abstract for dept. talks. I have found that "general" titles often end up with a hodge-podge set of slides assembled at the last minute, a speaker rambling on trying to connect them, poorly thought-out logic, and constant interruptions by the audience who are desperately trying to follow what seems to be a random walk through the speaker's mind. I am in biomedical research; however, my background is in mathematics. Perhaps I'm expecting too much logical thought progression in a presentation...

Anonymous said...

I haven't found any correlation between general (even vague) titles and disorganization of the talk, or narcissism of the speaker. I don't think we are really talking about talks with titles as vague as On Science or Cool Stuff I Feel Like Talking About Today. It would probably be more like the Fluffy/Fuzzy Bunny example in one of the comments. I have found that people who give talks with general titles are "bigger thinkers" than those who give long long detailed titles with a colon. You might as well just read the title and skip the talk. This refers to department seminars, not conferences of course.

Anonymous said...

For a conference, I want the title and abstract to be very specific. If it is a departmental seminar, I like to see the major research problems the talk will be based on outlined. I am choosy about which talk I wish to go to, as I like to spend the hour actually paying attention, and always feel frustrated when I have no idea what a talk is going to be about. Maybe this is cultural, I hate walking out in the middle of a talk if the topic is not interesting to me.

Anonymous said...

I definitely prefer a vague abstract over a bait and switch. And it really helps to have the info far ahead of time, because if the speaker is doing work in a relevant area, then I can invite them to visit my research group before they've bought their airplane ticket.