After a lengthy period of faffing about on the internet intensive study, I have invented devised a statistical thingy what the hell is it called matrix to help my fellow NN bloggers predict the kind of comments they are likely to receive on a given post.
-
rENNISance woman by Cath Ennis
Matt Brown said: "You can blog about whatever you wish, as long as it is related to science and research". His wish is my command! Here are some snippets from my life as a cancer research grant wrangler in Vancouver. Mostly the silly bits.
-
The nature of network comments
- Date:
- Friday, 13 Feb ruary 2009 - 14:56 UTC
The most interesting predictions inferred from this
matrixwhatsitmodel are that pedants haunt the internet, ready to pounce on any new post within secunds of publication; and that the first instance of the “cat” or “pun” comment class on a given thread is likely to spawn a sudden avalanche of similar comments.Also, queries about the destination of a given public transportation device are likely to kill the thread.
Please help me to test the Ennis Comments Hypothesis.
Last updated: Friday, 13 Feb 2009 - 14:56 UTC
-
Comments
-
[pedantry] You misspelled “seconds”.[/pedantry]
;-)
Oh my goodness, how could I possibly have let that slip in? I will fix it forthwith.
Quick! Someone post something insightful, before we move on to the movie quotes peak!
Well, that was on-topic, so we’re still on track
No, don’t fix the typo – a secund is so fecund!
Ah!! Too late…the puns are here already
Because Valentine’s Day is nigh, I considered a post in the “Dodgy puns …” category, in which scientific modifications for the goofy phrases on candy hearts would be suggested:
“Stir my heart” becomes “Stir my solution” or “Stir bar”
“Real love” becomes “Real time PCR”
“Top chef” becomes “Top agar”
“Go girl” becomes “Go-Taq”
etc., etc.
Thanks for the thoughtful analysis. Maybe we can go straight to the last category:
I think your bus ran over a cat somewhere along the line, Martin
_ pedants haunt the internet, ready to pounce on any new post within secunds of publication_
Dammit, Kristi beat me to it!!!!
Also, your model, though elegant and delightfully accurate, fails to mention one important class of comment: poems by Henry Gee. I admit that modeling the incidence of these is tricky, though.
Bloody italics tags (again).
Well, not to break the chain of insanity, but I was thinking about doing some similar analysis myself earlier this week. NN is slightly different to the wider blogosphere (sorry) but in general hardcore science posts are rarely commented on. It is the stranger mix of science, society and (in)sanity that scores higher.
She started it!
!http://farm4.static.flickr.com/3278/
3029764624_21a75c5e74_m.jpg!:http://www.flickr.com/photos/12499225@N02/3029764624/
Eh? Is it time for my appearance already?
Incorrectly formatted comments should form their own category, apparently
Katherine, I’ve noticed that before too. A proper analysis would be very interesting, especially if compared to Scienceblogs.com / scientific blogs on blogger etc.
Richard, Henry Gee is an outlier and would throw off any statistical analysis.
Textile FAIL Cat is very mysterious
There’s also the quite specific subcategory of cats reading books written by Nature Network bloggers.
In seriousness, though, a real analysis would be useful, like the one Euan Adie did for PLoS ONE comments recently
O Cath, your model’s splendid, quite the thing!
Though Richard says it lacks the estimation
Of any frequency of recitation
Of rhymes by me, one might recite , or sing.
(And that’s without artistic execration).
All that being said, I rather think it best
To think up some non-parameteric test.
So now I say with no more hesitation -
‘Excuse me, Madam, but does this bus go to the station?’
It’s all gone very quiet.
QED
Poems don’t count
Anyway, I just spent the last 15 minutes running around the house looking for camera, camera cables, and cats, to bring you this:
Oh noes! Tail still there??!!
Dude, don’t do that to me. Also, please get a life.
One important question unasked. Do the stats for photos of dogs chase the stats for photos of cats?
Somewhat related –
My cat keeps sitting on my Wii Fit board if I leave it on the floor… I swear, when I catch a picture of him, it’s getting posted here.
Hmmmm. Dog photos usually come after the angry rant comments, another class that I missed. Happily angry rants are rather rare on Nature Network, although if we are going to do a proper comparison to other platforms then we really should be counting them.
Dog photos also introduce the Gee factor which, as discussed above, tends to interfere with any statistical analysis.
Christie, in the photos above, Saba is sitting on our backgammon board, which is a new favourite (and now covered in fur). It’s a really old board with a velevety texture that she just loves.
Bum. Try again.
Eh? Is it time for my appearance already?
The Beast is also literary
I hope the sequel is better
Although he does prefer non-fiction
The only way the composition could be improved is if I had a beret.
You’re all a load of pinko leftie atheist bastard fuckwits. So there!
Oh no, the bunny of DOOM!
I take it this is a “revise and resubmit” notice for my hypothesis?
I like how Henry felt that his own one line angry insulting comment was worth THREE pet pictures.
I wanted to get my retaliation in first.
Angry rants are usually a subset of the “on-topic” category, though.
I’m glad I’m not the only person who has books and papers in her bed.
I don’t read blogs explaining the latest scientific research, so I don’t comment on them. In an ideal world your model would be able to distinguish between read blogs that are uncommented upon, and unread blogs.
I wonder if because it’s mostly scientists on NN, it would be like a busman’s holiday to read someone else’s digestion of a scientific paper, so we’re less likely to? Like doing journal club for fun in your spare time….brrrrr
In an ideal world, we would get page view stats ;)
There are a number of scientific topics (bird behavior and evolution, plant ecology, comparative vertebrate biology) that interest me, but for which I don’t have the time or expertise to read the primary literature. Therefore, I enjoy reading summary blog posts on these topics; I also find some of the clinical research summaries to be useful for preparing the “clinical correlations” sections of lectures. I rarely comment on such posts, though.
The Beast said The only way the composition could be improved is if I had a beret.
Something like this, maybe….
That’ll do, Graham.
I am very late, but:
P.S. Jenny:
I wonder if because it’s mostly scientists on NN, it would be like a busman’s holiday to read someone else’s digestion of a scientific paper, so we’re less likely to? Like doing journal club for fun in your spare time
I suspect it would just make me angry, actually.
I’m with Kristi on actually wanting to read summaries of recent scientific work, to avoid being denied by the time factor (or in my case, the slacker factor) – I do enjoy reading summaries when they (rarely) appear here. And it could be a chance to get a good discussion about nuts and bolts in papers without having to read the article carefully 3 times myself.
I think there’s also a strong effect of when a post is published. As a first guess, I’d say Friday is a bad day to publish. Many seem to post as the week winds down, and blogs easily get lost from the “Recent” list, disappearing under the avalanche of other end-of-weekers.
Hummm, maybe I’ll write about some literature I’ve been picking carefully through next tuesday and sit back to enjoy the comments stacking up.
Thanks for the tip Mike. Perfect excuse to postpone my next post until Monday.
Definitely, Mike. Although I’ve found that silly posts (like this one) get more comments at the end of the week, whereas I try to post my more serious and scientific posts (hey, it happens occasionally) on Mondays and Tuesdays. I subscribe to my fave NN blogs via RSS feed anyway, so I don’t miss anything that’s not on the front page, but I definitely miss some of the newer or less frequent bloggers if they post on a busy day.