I had lunch with Sarah Greene, the new editor of The Scientist today (so new, the yolk’s still glistening). It’s not name-dropping if I work with her (and anyway, I wrote the press release).
Where was I? Oh yes.
We got to talking about how we want to bring Faculty of 1000 and The Scientist closer together; what each can bring to the table and where, in fact, I fit in. One of the things we considered was videos. Now, I might be a little old-fashioned in many ways, but I’m pretty hip and happening when it comes to the new meeja (thinks: can I persuade Vitek to buy me an iPad?). At least, that’s what it says on my job description. However, I’m feeling more than a little ambivalent about video—at least for scientists and the stuff that they we like to do during daylight hours (once the lights go out, you’re on your own, boys and girls).
You see, when I was in the lab—and even now that I fly a desk—I feel very itchy about watching work-related videos on the firm’s time. If the video is longer than about 2’53” then I’ll either jump to near the end, if it’s like to be important; or (more usually) skip it altogether. I’ve not even watched the f1000 videos I’m supposedly responsible for. Podcasts? Fine! I’ll stick on the headphones and listen to a podcast, or preferably stick it on my iPod and listen to it on the Tube going to or from work (I’ve tried doing the same with videos, but I find watching out for tourists at Green Park a better use of my optical faculties.) I tend not to watch longer internetty videos at home either, because I usually have better things to do (films etc. is a tad different); and besides, it’s work. And the other side of the coin is that if I were a PI I’d probably have a complete benny if my post-docs were sitting round watching telly when they should be winning me a Nobel Prize.
So how do the rest of you feel about science-type videos and podcasts? What do you listen to or watch if you do? Is there room in the busy scientist’s schedule for either or both of these?
Part two of the question very much depends on the answer to this one, so I’ll ask it later.
Unless it is specifically work related like a seminar or something I almost never watch vids on-line or on portable mediaplayers.
Almost anything longer than 2min gets skipped or ignore too.
Sound good, picture bad, pretty much for the reasons you state.
No video at work. On the podcast front, I usually only listen to music podcasts, because I can’t listen to spoken word and concentrate on any higher-level brain functions at the same time. I do listen to some spoken word if I’m doing something that doesn’t use much brain power, e.g. formatting references or CVs etc. But most of my spoken word science podcasts (Nature & LabLit of course, plus This Week in Virology) are reserved for the bus, or walking, or waiting in line at the bank, or eating breakfast etc. Most of my podcasts are either music or comedy though.
Ooh, thanks for the LabLit pod shoutout 🙂
I meant to say ‘gym’ at some point. Stephen runs, of course: does he listen to anything I wonder? (I don’t, and especially not when I’m on my bicycle).
Yes! Gym and running require my favourite comedy podcasts to motivate me. NPR’s Wait Wait Don’t Tell Me and Friday Night Comedy from BBC Radio 4 do the trick, as did Ricky Gervais’ audiobooks before my husband listened to one and banned me from listening to any more of them without him present. We listened to one on a 5 hour bus journey in Cuba, causing the driver and other locals to give us some very funny looks as we howled with laughter with tears running down our faces!
Richard, yes I do – don’t think I could face running without some distraction. The other week, I was listening to Radio 4’s Front Row (as I mentioned elsewhere) but I am also a fan of the LabLit podcast (think you guys are hitting your stride, now that you have ‘barcodes’ behind you!). I also go for Melvyn Bragg’s In our time (and am hoping I might get my stamina up to the point where I can run for its entire 45 min duration).
On the commute I am not averse to video but, as you say, production values are, if anything more important there. I like Robert Llewelleyn’s Carpool (it’s a charming/disarming format). And there are some excellent TED talks out there. But otherwise science videos – I confess I haven’t seen many that are all that good (apart from the guy who blasts anvils in the air just for the hell of it…). Video’s got plenty of potential but it’s very hard work to do well.
The anvils one was really good, wasn’t it? I came across someone clay-pigeon shooting the other day… with a tank. Awesome, but not science.
Thanks for the kind words about the LL pod. That reminds me of another question—what’s the optimum length? Geoff Bilder was telling me that 45 minutes is good, but I feel that’s far too long. It is for me producing the damn thing, anyway.
And what’s wrong with barcodes?
PS Thanks for reminding me about your ‘running’ post, Stephen. I’ll talk to you when I’m more awake.
Length-wise, I would say be flexible. The podcast format has less formal expectations associated with it than professional broadcasting (radio); anywhere between 15 and 30 min is good. Personally, it has to be really good content for me to want to listen longer. I tend to avoid podcasts that waffle on too long and there are some very long ones on iTunes.
Yeah. Our feeling is that if we’re still laughing (at our own jokes!) and are sorry it’s finished we’re doing OK. The most recent could probably have done with a bit more savage editing, but hey, it’s a learning experience.
I agree that 15-30 min is a good length. No more than 10% of the content of any given episode should involve scanning bar codes with iPhones or reading out the cartoons from the paper 🙂
Heh heh heh! I’ll have you know the free papers in London are an endless source of amusement.
I really enjoy watching TED talks. The only other videos I watch are short (1-2 minute) science activity demos. I don’t have video at work, so any watching has to be done at home (commuting is no good, since I drive, and it’s only 5-10 minutes).
I’m exactly the same – I almost never watch scientific or newsy videos online – I would prefer to read a text article or even a transcript if available. Why? I suppose two reasons:
1. It is much quicker to read, and much easier to get the key points quickly. I suppose technically I would have time to spend three minutes watching, but it still feels like a waste of time.
2. I don’t always have headphones plugged in or easily to hand on the laptop. Again, a really trivial thing to stop me, but it is a real discouragement.
Anything of more than 10 minutes is a different category: that I would actively have to set time aside for, and would be much more likely to do, because it feels like a real activity. Personally, I like 30 mins or less for a podcast, simply because that time is easily fitted in while making lunch / washing up / waiting for laundry to be done. 45 mins+ is a more significant chunk of time.
How long are TED talks, Alyssa?
It’s all really interesting, people: thanks. I wonder if we might propose a session on this at SOLO10? Martin?
I’ll pass this on to the household’s resident expert on videos.
TED talks are about 15 minutes. There are also some good science videos online: Nature generally does a good job (last week’s running video was excellent). I’m happy to watch videos that are about 5 mins long, but it depends a bit on what it’s about.
Like others, I only rarely watch videos, but I do listen to about 8h of podcasts every week, e.g., http://bjoern.brembs.net/news.php?item.92
Thanks for the kind words about the LL pod. That reminds me of another question—what’s the optimum length?
What’s the average length of commute? I think that’s a good time to aim for. My 10 minute stroll through ochre walled lanes is a little too short for most spoken word podcasts.
I would second (?third/fourth?) the 15-30 min podcast, any longer and the boss thinks you’re skiving or you miss your stop on the bus because you are giggling – to say nothing of the bemused looks on your fellow travellers’ faces.
I will watch video but prefer it as a secondary source of information; the primary being written word.
I have a short attention span for talk radio and related podcasts, so 30 minutes is just a little bit too long for a thing without music in it. I usually listen to the LabLit podcast in two separate listening sessions. (For example, with the last podcast I listened to the first few minutes, then skipped to Henry’s interview, and went back later in the day to hear the missing bits.)
As for the perfect length of science videos, I’m taking notes. (Not really, just bookmarking this post/thread)
Heh. I look forward to whatever it is you’re planning, Eva.
This from Bill Town on facebook:
–which reminds me, I heartily approve of videos in supplemental information, especially in Cell Biology. (You listening, Jenny?)
I don’t know what to do with short clips. If I want no information I can as well read a blog post ;-p But I sometimes watch recorded seminars (the usual 45-60 mins stuff), where I allow myself the freedom to fast-forward interruptions by the audience and introductions I’ve heard a million times already. I might do that in the evening after I come home from work. The point is that speakers frequently provide in their seminars explanations you do not find in their papers (or blog posts for that matter). If I have a choice between watching a video and reading, I do prefer reading. Unfortunately, our own seminars are so far only very sparsely recorded. At Perimeter Institute otoh, they do meanwhile record pretty much all seminars (unless the speaker objects). This is great when you’re traveling and would otherwise miss a seminar, you can just watch it later.
More constructively: Unless you really have information that cannot be communicated in words, please add a writeup to your video because I’d rather read it. I frankly don’t see the point watching somebody reading a script when I can read it myself.
That’s a very good point, and closed-captioning I think is very important for video. It would be nice to have good transcripts, I wonder if there’s an easy way to do that?
Björn has a nice list.
Like the other posters, I think 45 minutes is the maximum length of time a podcast should be, and then only if you have a lot of exceptional material. Even In Our Time doesn’t stretch to that, and it is usually of an excellent quality. I’ve had a quick look at the podcasts I listen to (BBC4 and Nature), and all are
This is important, then, Nicolas. Is there decent voice recognition & transcription software available for the Mac? My daughter’s played a bit with the built-in stuff (and that’s quite amusing to watch!) but that’s simple recognition, nothing fancy.
It might be important, but I don’t know anyone that does it apart from the BBC on the iPlayer. I’m more in-tune to that issue because I live in a multi-lingual household where even if someone understands a language quite well, it is often not enough to follow a movie. And Korean/Chinese films dubbed in English are just atrocious. An issue with speech recognition software is that I think it has to be trained to your voice, so you can’t really go to a
pubconference and record an interview and have it transcribed automatically. I might be wrong though, maybe this link and this one will help, after that GIYF.Hmm. How do the BBC do it? I’ve a feeling they’ve actually got a touch-typist or thirty.
Well I guess a lot of BBC material is scripted anyway, so they have that as a starting point anyway. They have to do it for the old subtitles for the hard of hearing, and have been doing it for donkeys years, so I guess they either have a department doing it, or a sub-contractor, that may or may not be software-assisted.
Like some others here, I have a short attention span. For scientific content (like those on Jove or SciVee) anything over 10 minutes loses my interest. For the average YouTube or bloggy-type videos, it’s more like 3 or 4 minutes).
I don’t have issues with video at work, but I’m not in an academic environment. As for TED, the iPhone app is great.
Thanks, Ken. Nicolas, I’ve seen it happen on live news. So I guess that’s manual.
A contrarian view from my mate Nige:
Agree with Ian et al. above. Who has time for watching videos?
[you knew I was going to say that, didn’t you?]
Although a picture may be worth a thousand words yadda yadda yadda, I honestly haven’t found any video that conveyed more information than a well-written article – which you can pick up and put down ad libitum (I suppose you can stop and start a video too – but really, now).
Yeah, that’s more or less my feeling.
Don’t have time to read everyone’s comments in detail at the mo so apologies if I’m repeating things – boss probably already thinks I’m slacking!
The one thing I would say in favour of videos is that in our lab since JOVE now has some useful methods stuff our lab seems to think it’s more acceptable to watch science related movies and our PI doesn’t even blink (Could be why he doesn’t have a nobel prize though, hehe)
In response to a few comments I have scanned through whilst you can get a lot out of writing it can be easier to make a movie and show someone what you’re doing. We particularly find this for things like dissections as we don’t have a dual view microscope. I’m not saying this is necessary useful for The Scientist or f1000 but I do think there are instances when video can be useful.
If I’m interested enough in the topic I would definitely watch something of 30 minutes but much longer and you have to ‘find the time’ which means I don’t get around to it – unless it’s a TV program! Cold Spring Harbor makes videos of all the talks at conferences, these are 15 minutes and quite a lot of people watch them.
So you’re saying that video is good for instruction? I think that’s a very valid point, although as you point out, not quite the same thing. Half an hour to watch a video is a long time at work, isn’t it?
This reminds me that while at the LMB we used to have certain symposia streamed internally. I tried doing experiments with the talk playing on the iMac, but even setting up crystal screens took up too much of my concentration. Whether it was any worse than actually falling asleep in the lecture theatre, I don’t know!
I guess I’m really trying to say that I think scientists are becoming more open minded about using and watching videos.
You’re probably right 30 minutes probably is a bit long, I just wouldn’t dismiss video altogether as I do watch them. I also watch the movies that come attached to papers nowadays. I can’t speak for other people but I like visual things for a change from text and listening to people speak – I do seem to be in the minority here though.
Having said that, didn’t think much of the video in the cell article of the future prototype because as other people said, watching someone simply read from a script is not amazingly useful.
Whereas “this is just lovely”http://jcb.rupress.org/cgi/content/full/jcb.200906043/DC1/4.
Thinks: when are journals going to produce these things in proper 3D? We all kept our glasses when we went to the movies, yeah?
Also, didn’t I hear recently that the University of California is starting a project to webcast all the seminars on any of the universities in the group…
but maybe this won’t take off.
Bugger. JCB Link
Sorry Sam, posts crossed. That’s more or less what we were doing at the LMB, back in ’04, possibly earlier. A much smaller scale, admittedly, but people are slow to catch on.
Still, I love the opening salvo:
Thinks: when are journals going to produce these things in proper 3D?
scary stuff, imagine a paper on phagocytosis 😉
Oh yes. This has so got to be done. Right after my GFP tattoo.
As the editor of a methods journal, I’ve taken the position that very short videos can be extremely useful tools. If there’s a physical manipulation in a protocol, it can be difficult to describe, but a 10 second video can show you exactly what you need to know. The same goes for examples of what you can do with an imaging method. But we’ve avoided long form video for the same reasons many above have noted–it’s so much faster to just read the material. Do you really need a video of each step in a protocol? Add NaCl to H2O. Stir. Stir a little more.
10s? That surprises me (but I’m not disagreeing). Really that short? That’s amazing.
I don’t listen to any science podcasts (haven’t really looked for them), but will watch videos, but not at work.
On the transcribing issue:
I know of a blogger that is using Mechanical Turk to transcribe an hour long interview for ~$35. The result is pretty raw due to segments often being repeated, but readable and better than nothing.
The problem with the LabLit podcast is that we have no budget. Work-wise, it might be worth considering. That reminds me I have a blog post about out-sourcing simmering, actually.
David said:
Do you really need a video of each step in a protocol? Add NaCl to H2O. Stir. Stir a little more.
Glad to hear I’m not the only one… 😉
I’d been wondering about transcribing dialogue (for other reasons) and pretty much came to the conclusion that it’s a brutal exercise (very similar in many ways to transcribing music by ear). To do it without going completely mental, I think you’d need foot controls for start/stop/jump back, or at the very least a proper jog/shuttle wheel. Clicking and dragging to rewind would be a horrible chore.
That said, I wonder if some kind of voice recognition software (similar to what Sean mentions maybe? There must bee freeware options for this) could at least get you “in the ballpark”, allowing for manual clean-up that would be less onerous.
I found out today that a transcript for a 45 minute interview can cost around US$300. Hm.
@Richard: A not-so-experienced typist without a foot pedl will take 8-9 times the length of the interview to transcribe it. A very experienced typist using every supportive tool available can manage in 3-5 times the length of the original interview.
Wow. Sooner we replace humans with robots the better.
@ Richards (PG & W): A while ago, I used iListen (since replaced by MacSpeech Dictate) for the Mac. It wasn’t great, but far better than transcribing by hand. I’m told it’s improved substantially, but haven’t tried it recently.
The most problems tend to crop up when you use it with technical or jargon-filled speech. It requires some additional training of the software, but it’s doable.
Two hundred bucks. Based on the Dragon engine. Hmm.
And, I believe (although I can’t verify this), it’s a Cocoa app, not a bad Carbon port of the Windoze version.
If you do try it, use a headset instead of the built-in microphone.
@Richard–
10 seconds may be a bit of an over-reach, but here are two quick examples from our featured articles that are freely available to all.
http://cshprotocols.cshlp.org/cgi/content/full/2007/20/pdb.prot4848
The movie here shows the physical manipulation needed to separate out hippocampal slices, it’s about 10-15 seconds long.
http://cshprotocols.cshlp.org/cgi/content/full/2007/20/pdb.prot4847
Here the movie shows various Drosophila courtship behaviors. It’s a concatenation of 5-10 second movies into one long one, but could just as easily have been a series of shorter movies each highlighting one behavior.
Hey David, fantastic, thanks!