2011-04-13 15:45:28Wording for The Ville's peer-reviewed visualisation
John Cook


Ville's animation is coming along very nicely. Here's the latest version


I'm starting a new thread to discuss wording that needs to go on the animation - we should nail it down in this thread before we ask Ville to put it in the animation. The main goal - that the animation is easy to understand out of context so if someone follows an external link to the animation, it still makes sense. Here's my suggested text. As feedback comes in, I'll amend the text so the "final version" will be here:

The History of Climate Science

This animation shows the number of climate science papers published in the peer-reviewed literature. They are loosely categorised as skeptic, neutral or affirming Anthropogenic Global Warming (proAGW)*. The size of the circles indicate the number of papers published in a certain year - mouseover any circle to see the year and # of papers. Drag the slider at the bottom to see papers up to a certain year.

Any thoughts on when to launch? My original plan was to wait till May 15 to give people time to harvest papers (I'm directing this question primarily to Ari and Rob)

Also, I had this idea of doing a YouTube movie "The History of Climate Science" (or Evolution), which shows the slider moving left to right - but possibly have it pass at certain moments, and add those YouTube captions to highlight pivotal moments. Eg "Tyndal demonstrates greenhouse effect in lab experiment", "Arrhenius predicts future warming from rising CO2 levels", "Angstrom says CO2 effect saturated", "CO2 saturation shown to be based on false assumptions", etc. I thought this might be an interesting animation and would help promote the animation. Two questions - is this worth the trouble and more importantly, anyone here know how to create such a movie?

2011-04-13 16:21:11
Ari Jokimäki


"Any thoughts on when to launch? My original plan was to wait till May 15 to give people time to harvest papers (I'm directing this question primarily to Ari and Rob)"

I'm going to harvest papers after that too. Right now I'm going for the book-prize. That's too good prize to pass. In the long run I would like to have the paper database really to describe climate science history thoroughly. Perhaps we'll make a coordinated effort collecting papers at some point - when the contest is over. ;)

2011-04-13 17:04:35
Paul D



How about adding the pivotal moments to the XML file??
You could just add another entry to the year in question. You don't have to add it to every year.
As the software stands, I don't think the addition will mess it up.

But if I have time I can add the text, so that it appears as the slider is moved to specific years.

Actually, i'll try it out here first, then if it is easy, you can add the texts to the XML.

Added: just thought, extra years might have to be added with no papers.
I'll set up a test XML file here with some test data and see if I can get something working. But I guess it also depends whether John can add the data to the XML, can it be automated??

2011-04-13 17:09:18
Paul D


BTW would it be better to have the explanation (mentioned in the first post) as standard html/css below the graphics??
Then it will appear in search engines. I don't think any of the text rendered in the app will be search engine friendly (its all vector drawn and the source text is embedded in javascript).

2011-04-13 18:26:46Events & HTML
John Cook

Yes, I can easily add events to the XML and even data drive it so the events can be crowd sourced. Tell me the required format and I'll make it happen. Ideally, events are connected to a particular bias (eg - Angstrom event linked to 1900 skeptic year).

Still a fan of the animation starting by scrolling thru years, is that doable?

Agreed re text being HTML. That means you need to remove the heading from your animation.

2011-04-13 19:15:47
Alex C


I don't know much about software, but one way to do an animation might be to set up the slider moving left to right (such as a year every tenth second) and add in extended periods of wait time at certain years.  You could either script a voiceover for the animation and base pause lengths on that, or just have constant (or text-length dependent) pause lengths with text overlays.  How this could be done in code, I don't know, but from what I remember of the Java I took, you can put the slider value through a for loop and have conditionals within for certain year values, and accompanying wait periods.

Am I right in that such a video would be relatively simple?

I like your introduction John, I don't think it needs to be changed much at all.  Maybe "Each paper is loosely categorized as being skeptical, neutral, or affirmative [affirming?  supportive?] toward anthropogenic global warming (AGW)," or if you're going to say "proAGW" then say "pro-AGW to match the label in the app/animation.  Details details.

I also think that, not only should key scientific publications be put down as events, but major social events should be included.  Maybe Hansen's testimony in (1979?), or something of that sort.

I think it's a toss-up too between "history" and "evolution."  I think the former would be a good idea for a title for such an animation that acts almost as a biography for climatology, but the latter also carries the "see how far we've come?" connotation, which is why I'm slightly preferential to it.

2011-04-13 20:50:37
Mark Richardson

I don't like the use of 'skeptical'. They're no more nor less skeptical than other scientists, they just found results that were significantly outside the average.


'anti' or similar wording might be better.



Will the word 'evolution' turn off many of the 'skeptics' we're trying to reach?



Also, where is the methodology and paper list, out of interest?

2011-04-13 23:47:44
Paul D


John: "Still a fan of the animation starting by scrolling thru years, is that doable?"

Not sure what you mean??
Do you mean automated, from first to last??

2011-04-14 00:01:20
Paul D


Well for now I can confirm putting extra entries in the XML has no detrimental effect on the current code. So I suggest adding the new feature and I'll have a go at creating a new version of the code with the history lesson messgages.

Suggested entry:

        <bias type="skeptic"> 0 </bias>
        <bias type="neutral"> 2 </bias>
        <bias type="proagw"> 0 </bias>
        <note> A message of some sort </note>
        <date> 1941 </date>


If there are no papers then just put zero in each bias entry. My current code will checks for zero, it only creates a circle if the entry is greater than zero.
Also if there is no message then just leave that 'note' entry out.

2011-04-14 00:12:43
Paul D


BTW I also suggest keeping the messages to one liners. Otherwise the formating will be complicated I think. Maybe we can add more text later. Goodness, maybe even pictures??
Like when you get to a year with a message, it could display in a window with text and an image.

I think at some point we need to collate all these ideas and think about what is wanted in the next version. Then zero in on a bunch of practical and useful additions.

2011-04-14 01:16:24
Paul D


Actually John, can you hold off making changes to the XML file?

I thought it wouldn't have any impact but it appears that it does.

2011-04-14 04:08:06
Paul D



Is there going to be a press release??

I reckon New Scientist might be interested, at least for an article on their web site if not the magazine.

2011-04-14 07:31:05
Paul D


This pivotal moments texts idea isn't as simple as I thought. Might take some time to sort out.
I've got the XML code working but it isn't so easy getting the text displayed when the slider hits a specific year.
There are two issues to sort out.
I suggest for now assume cluster35 is the current launch version, with the title removed.

I'll see what I can do with the new idea.

2011-04-14 11:59:51pivotal moments
John Cook


My initial thought was not to have the pivotal moments in the actual coded animation but to add it in later using YouTube's little labeling system. So don't consider it a must-have feature.

Yes, good idea re press release. Will have to think of a way to package it up all sexy like, make it worth showing. Hence the need for the YouTube too, to hand it to the media on a platter.

2011-04-14 12:51:38
Peter Miesler

Gosh that looks mighty good, visual eye candy.

"Ville's animation is coming along very nicely. Here's the latest version


My only head scratcher is "ProAGW" there's got to be a better way of putting that. 

When I think of "pro" it's in favor of something wanted.

2011-04-14 14:32:37
Ari Jokimäki


I have also thought that pro-AGW is not correct here. These papers are for or against denier claims, not for or against AGW.

2011-04-14 16:12:17A better term than proAGW
John Cook


The problem with that term is it sounds like you're for anthropogenic global warming. Mmmm, bring on that climate disruption!

But what's a better term?

Denier / Neutral / Accepts Evidence

A little provocative :-)

Suggestions welcome.

2011-04-14 16:27:28



convinced - neutral - unconvinced

Swap graphic to show 'deniers' on the right of image to match the c-n-u sequence above.

(It's a psychology thing. Most languages read left to right.  1st content word has greatest impact.)


I assume that my other suggestion might brush lightly against the wrong side of SkS policy:

convinced - neutral - way to go fossil fuels! rah! rah! rah!

2011-04-14 17:20:15


I use "AGW-concerned".

2011-04-14 17:37:11
Ari Jokimäki


supports claim - neutral - doesn't support claim

2011-04-14 18:14:58
Paul D



"My initial thought was not to have the pivotal moments in the actual coded animation but to add it in later using YouTube's little labeling system. So don't consider it a must-have feature."

I suggest that you go ahead and make the video. If I can work out an algoritm for displaying the text, I'll have a go at it.
The problem is the speed at which the slider can be moved, if it is moved quickly then the event handler will obviously miss years. This isn't a problem with the circles, the circles catch up with the slider position eventually. However it is a problem with the texts, since a more accurate response/visual feedback is appropriate. I have some ideas, whether they work or not, is another thing.

2011-04-14 18:54:31
Mark Richardson

'Evidence for the prosecution', 'Evidence for the defence'? Sounds like we're prosecuting humans, which anti-AGW people will jump on to avoid the point.


'supports human global warming', 'supports alternative theory'?

2011-04-14 21:54:41
Paul D


Actually one solution to the problem stated earlier maybe to force the slider to increment year by year, despite the speed at which the user wants to move it. Not sure what impact this would have, most likely it will be annoying, but it's worth trying out.

2011-04-14 23:39:02
Paul D


BTW, the links posted on this forum regarding this project are not permanent. I will be deleting the files as new ideas are developed.
So it wouldn't be a good idea using them outside this forum. John will be creating a home for it on this site, that will be the one to link to.

2011-04-16 02:35:48
Paul D


Do we want the labels on the clusters to mention the totals for each cluster??
I'm not sure if people wanted it or not?

Actually, just thought, I guess the totals would have to reflect the year/slider setting??
Why do I make work for myself?

2011-04-16 05:12:34
Paul D


Does this improve things??


2011-04-16 06:25:11Yes, like totals
John Cook

Nice addition. Ville, I'm thinking we really should have a link to the list of papers. When I scroll through the history, I see a lone skeptic paper in 1920 and am very curious about it. I'm sure I'm not Robinson Crusoe there. What about if when someone clicks on a year and the box appears with the year and # of papers, that box links to the following page:


I haven't programmed that page yet but can whip it up quickly enough if you agree to add that feature.

2011-04-16 06:40:41
Paul D


So I just generate that string with a different bias and year on the end?

Sounds doable. Probably easier than faffing about with a load of XML.

Shall I stick it in a pop up window?

Actually I need to think about this, to work out how to attach a url to the graphics. Haven't tried that yet. I'll see if there are examples.

2011-04-16 06:44:28Headings


How about these to get away from "pro-AGW" which - as others have stated - is a misnomer as none of us is actually "for AGW" as far as I can tell

skeptical [of AGW theory] - neutral - supporting AGW theory

skeptical [of AGW theory] - neutral - in support of AGW theory

unconvinced [of AGW theory] - neutral - convinced of AGW theory

2011-04-16 07:47:48
Paul D


Just a thought though John, will the years XML file that I use for the graphics, match these pages generated from the database.
I mentioned this issue previously. Can you make sure they are synced?

2011-04-16 08:40:32
Paul D


Done John. All you have to do is generate the pages:


If you need any changes just email me.
BTW, the 'bias' is generated from the XML 'type' value and the date is generated from the XML date value. I haven't stripped any spaces out.

2011-04-16 09:51:36Synced feeds
John Cook


The XML feed is generated dynamically from the database so it's synced at all times. I may switch it to being a static XML file if the server load gets too big but for now, all good to go.

Great work re adding the links - if I get time today (got people coming over so the missus is working me hard cleaning the house), I'll code up the page.

2011-04-16 13:05:49Interesting how there’s no point when the skeptic papers are in front
James Wight

  • 1824-1907: AGW in front
  • 1908-1923: AGW/neutral tied
  • 1924-1978: neutral in front
  • 1979: AGW/neutral tied
  • 1980-1995: AGW in front
  • 1996-1999: neutral in front
  • 2000-2011: AGW in front
2011-04-16 13:28:19
Ari Jokimäki


"1996-1999: neutral in front"

This has to do with me adding lots of papers between this time interval which from the source I have been using are all almost by default neutral. This doesn't represent the real situation.

2011-04-16 21:47:33Still, interesting to watch
John Cook

You would expect neutral to be in front until the late 80s when the global warming trend emerged.
2011-04-16 22:38:41
James Wight


The Charney report was in 1979.

My point is, if there was a global cooling consensus in the 1970s, the skeptic papers should have been in front then. Our database is biased though - I added many of the "warming" 1970s papers but haven't got around to adding the "cooling" ones yet.

2011-04-17 17:54:39
Paul D



"The XML feed is generated dynamically from the database so it's synced at all times. I may switch it to being a static XML file if the server load gets too big but for now, all good to go."

Hi John, the syncing of the XML and the database is only half the problem.
The other issue is that the XML is downloaded when the app is loaded into the browser then at a later time the user can click on a circle, this generates a request for a list of papers. You now have a period of time between the XML file being downloaded and the data being stored by the browser and the list of papers being dynamically being downloaded from the server. In that time (even if it has a low probability) the database could have been updated and the list of papers is out of sync with the copy of the XML held in the users browser.
The issue is of data consistency.

There are two ways of solving this, the third option is to not bother about it.

1. The easiest is for the server to hold a copy of the database data which is only updated say once everyday from the main tables and that is used to generate the XML file and page requests. The data is then synced and consistent.
Could you set up a timer in PHP, that will just transfer the data to a new database table/tables once every day and then generate the XML and page requests from that table??

2. Develop a much more complicated javascript app, that communicates more with the server and may need to reset itself when it finds a discrepency in the data (some sort of serial number or date/time stamp to check consistency). This will take some time to do and I don't think it can be done by 15 May. I'm actually reworking the XML http request code to be generic so that it can handle multiple requests. But it would require much more work I to do a more complex app that will ensure consistency.

3. Don't bother solving it and accept that occasionally the list of papers may have more or less than the tag says on circle when you rollover it.

As it stands, the basic app (cluster35a) with no links and just the graphics doesn't suffer from this potential problem, so that can be used no probs.
But I suggest that the new version with lists of papers is tested on your server. In fact since the database is being updated quite rapidly, it is a very good time to test it like this because it may show the data consistency or inconsistency issue.

2011-04-17 19:59:41
Paul D


I have updated my test copy of the xml and uploaded to my web space.
So the links above should reflect the change.

2011-04-18 08:02:06


Wow, looking at  The Ville's peer-reviewed visualisation makes you wonder if there was any time in the history of modern science that skeptics had more peer-reviewed papers - doesn´t look like it.

2011-04-18 12:11:43history of modern science
John Cook


To illustrate this, another useful visual would be a graph featuring three lines - skeptic/neutral/proAGW. I might do a graph of this as a follow-up after we launch Ville's visualisation. Good to show the hockey stick of proAGW papers :-)

The more I think about this, the more I see Ville's visualisation as a first step - by making this data accessible via animations, graphs, data-feeds, etc, make it as open as possible and encourage more crowd sourcing of submitting data. So go from strength to strength and think of other ways to display the information, repackage it, encourage others to get creative with the data, etc.

2011-04-18 12:16:38Peer-reviewed papers per year now display
John Cook


If you click on Ville's circles, they now show papers in a particular year. Eg:


The first skeptic paper:


Just had the thought - Angstrom's paper doesn't appear in Poptech's list. We're providing more papers for him!!!

BTW, this 1920 skeptic paper:


Surely, this should be neutral. Saying sun affects climate doesn't mean sun caused modern global warming trend (which started after 1920)

2011-04-18 15:17:27
Ari Jokimäki


"BTW, this 1920 skeptic paper:


Surely, this should be neutral. Saying sun affects climate doesn't mean sun caused modern global warming trend (which started after 1920)"

I added this paper to argument "water vapor masks CO2 effect". This was in time when it was generally thought that CO2 wouldn't have much effect because it was masked by water vapor in same absorption frequencies. Here's relevant quote from Abbott's paper:

"The other two absorbents are each confined in their absorbing regions to comparatively narrow ranges of spectrum, but the ozone absorption band, at about 10 microns, occurs in a region where water vapor absorbs scarcely anything while the carbon dioxide absorption band at about 14 microns occurs in a region where water vapor is also powerfully absorbing. The atmospheric proportion of carbon dioxide is sensibly constant, while water vapor and ozone are variable. Accordingly, while water vapor is certainly the most important of the three, probably ozone, although much less plentiful in the atmosphere, and certainly not more powerful as an absorber for the spectrum of a perfect radiator than carbon dioxide, is yet entitled to be regarded as second in importance on account of this peculiar posture of affairs."

It's athough call, as Abbott doesn't say it explicitly but I think it is implied here that CO2's effect is weak. Therefore this paper is in skeptic bin. This is rather difficult as many times papers fall under several arguments and they are not always with same bias for each argument.

2011-04-18 18:32:06Fair enough
John Cook


Interesting actually, seeing the history of skeptic arguments - the water vapor saturation one still gets trotted out now - so it originated in the 1920s. Cool :-)

2011-04-18 20:44:41
Ari Jokimäki


Well, no, the water vapor argument originates from Ångström (1900) if it had not been presented even before that.