RealClimate logo

New Google search function

Filed under: — group @ 24 October 2006

It can be easy to find climate science information on the web, but that information ranges from the excellent to the atrocious – and it can often be hard to tell them apart without some prior expertise. Wouldn’t it be great if someone could vet the information beforehand so that you had some confidence that it wasn’t completely bogus? Well, you need wait no longer!

Some of you may have already noticed that we have updated our search facility to use a new service from Google Co-op which is being launched today. The idea is that the search is restricted to domains and pages that have passed some kind of quality control. RealClimate is one of the demo sites of the new technology and we have started off with a selection of sites (IPCC, goverment labs, research institutes etc. – as well as RealClimate itself of course!) that we know provide quality information about climate science. As we get used to this service, we will be adding sites and pages that we feel are up to the mark. Suggestions for sites that we might not yet have found or have overlooked, will of course be welcome.

Eventually, we hope to have a service that could be an essential resource for the interested public, journalists, and possibly even scientists, that would give a higher quality level of information than is possible now. Let us know if this ends up being useful to you and if you have any suggestions for improving the service.

78 Responses to “New Google search function”

  1. 1
    David Wilson says:

    seems to me that a better objective would be for people to have the tools to sort the wheat from the chaff themselves

  2. 2
    Will Glenn says:

    And about time too. Some of those sites I’ve gone to, are so way off beam that they are off planet in their ‘thinking’ if that’s what you can call it.

  3. 3
    Mark UK says:

    I think it would be great if the site hosted a list of websites being searched?

  4. 4
    Dave Ireland says:

    So this is cherry picking from sites that support your view? Make sure your users clearly know this!

    [Response:It’s not cherry picking, but a kind of quality rubber stamp. Here you can be fairly safe that the contents have some quality, but you can always use other search engines if you want to a wider search. -rasmus]

  5. 5
    Dan says:

    re: 4. Scientific results from the IPCC are in no way whatsoever “a view”.

  6. 6
    Jack Kelly says:

    Re #5 re: 4. Scientific results from the IPCC are in no way whatsoever “a view”.
    How do you keep ‘selected’ data from being a view?

  7. 7

    Provided that you disclose to searchers that the results are only a limited subset of the opinion on the matter, I don’t see any problem with this. I do see some irony though – you are complaining about the growing amount of information on the web from those who do not have bits of paper to show off to other people, but your site is powered by WordPress which is one of the primary innovations responsible for this ‘problem’.

  8. 8
    David Jackson says:

    Isn’t that the point of scientific peer reviews? If only political decisions were peer reviewed by qualified individuals instead of those who just have money or talk loudest!

  9. 9
    Florifulgurator says:

    Re#6: “How do you keep ‘selected’ data from being a view?”
    Common sense helps a lot here:
    Select the data blindly (disregarding any preferred view on the results), according to objective criteria selected before the selection, and look at the outcome afterwards.

  10. 10

    Important studies have been done in situations where it was not possible to have true peer review, especially by scientists working under government supervision.


  11. 11
    Glen Barry says:

    I share concerns that a broader based climate search engine than just the scientific findings are necessary to solve the climate change crises. In fact, such a resource already exists at the Climate Ark. This is the original, all inclusive biocentric climate change search engine.

  12. 12
  13. 13
    Roger Pielke, Jr. says:

    You guys should have a look at Cass Sunstein’s Republic.Com

    An excerpt:

    “In particular, a well-functioning system of free expression must meet two distinctive requirements.

    First, people should be exposed to materials that they would not have chosen in advance. Unplanned, unanticipated encounters are central to democracy itself. Such encounters often involve topics and points of view that people have not sought out and perhaps find quite irritating. They are important partly to ensure against fragmentation and extremism, which are predictable outcomes of any situation in which like-minded people speak only with themselves. I do not suggest that government should force people to see things that they wish to avoid. But I do contend that in a democracy deserving the name, people often come across views and topics that they have not specifically selected.

    Second, many or most citizens should have a range of common experiences. Without shared experiences, a heterogeneous society will have a much more difficult time in addressing social problems. People may even find it hard to understand one another. Common experiences, emphatically including the common experiences made possible by the media, provide a form of social glue. A system of communications that radically diminishes the number of such experiences will create a number of problems, not least because of the increase in social fragmentation. . .

    What I will also suggest is that there are serious dangers in a system in which individuals bypass general interest intermediaries and restrict themselves to opinions and topics of their own choosing. In particular, I will emphasize the risks posed by any situation in which thousands or perhaps millions or even tens of millions of people are mainly listening to louder echoes of their own voices. A situation of this kind is likely to produce far worse than mere fragmentation.”

    And if you want to read more:

    [Response: This is well put and I fully agree – but we are not talking about picking opinions that we happen to like, we are talking about quality control. I read the newspaper every day (I guess that’s what is meant with “general interest intermediaries”) exactly because I want to be confronted with information that I have not picked myself, but that has been selected by an editor. But I read a quality newspaper, because I have limited time and I want to spend it on reading information that is reliable and well-researched. Thus I want the editors of my newspaper to have high professional standards when it comes to fact-checking, selecting what is relevant, etc.

    I also like to read scientific papers because they fulfill certain quality standards, not because their contents agree with my prior belief (in contrast – those that challenge prior beliefs are of course far more interesting and will be much more widely read – if their quality is good). Journal editors and peer review are part of the system to insure those standards.

    The internet is mostly lacking quality control, and as the information grows exponentially, the need to quickly find reliable sources is growing fast. Some people may have time to read industry lobby websites on global warming – and if that’s what they like, that’s fine and they are easy to find. But many people want to know first-hand what the scientists working in this field think, rather than what spin doctors make of it, and it is for these people that we try to provide a service.

    We link to IPCC because IPCC is a high-quality source of information, compiled by leading researchers in the field after thorough discussion and rigorously peer-reviewed – not because IPCC agrees with us. In fact it is the other way round: I agree with IPCC on most (but not all) issues, because the scientific evidence is simply very strong, so after initial scepticism about anthropogenic warming I have become convinced by this evidence. -Stefan]

  14. 14
    Nereo Preto says:

    Just for fun, I typed “hockey stick” in your search window. First three outcomes are a BBC piece, a very-very-skeptic page and, well, a dealer of sport equipments.

    Now I am a little confused. Perhaps I did something wrong? If extreme views (well, at least the title was extreme, I didn’t really read that web page carefully) and non-relevant sites can do it through the search engine, what’s its purpose then?

    To be completely honest, since I don’t know exactly the criteria under your search engine, I won’t use it if I want to build an opinion of my own, but I admit it might be useful for quick, focused searches. The engine at Climate Ark, however, was able to get rid of sport dealers, at least in the topmost results.

  15. 15
    Parikshit says:

    its a very innovative step taken by Google..

  16. 16
    Alan says:

    The google search is a double edge sword. I have on many occasions used an RC query link for the word “myth” as a standard mythbuster for Chriton, volcanos, ect.

    The search now returns some interesting stuff but it’s too broad for use as a general climate mythbuster. I know google can restrict searches by domain name, can we please have two buttons, one for an RC only search and one for the new google search, please, can we, pretty please :)

    [Response: We will try and get this done with buttons – it shouldn’t be too tricky, but html skills are not our strongest suit! In the meantime, you can hack an RC-only searches using“SEARCH TERM”&submit=Search where ‘SEARCH TERM’ should be replaced with you desired search term. -gavin]

  17. 17
    Grant says:

    Re: #14

    I’ll second the request for two search buttons, one for RC only, the other for the new google search.

    Re: #13

    I agree that it’s quite important for people to be exposed to contrary ideas and opinions; otherwise we are in even greater danger from “what we think we know, that just ain’t so.” I’m guessing that the motivation for the new search feature is not to stifle dissent or radical thinking, but simply to eliminate *garbage*. So I hope the RC moderators will be extremely liberal in their inclusion of sites with alternate viewpoints, rejecting suggested sites only when their “junk” status is beyond a reasonable doubt.

  18. 18
    Alan says:

    Sorry to post twice in a row but it just occured to me:

    How can people logically argue that RC might somehow be in danger of falling into a “group think” mentality by broadening the search results? I mean the old search function only returned results from RC, so why all the hand wringing about democracy when the results now include other relevant sites?

  19. 19
    A Fritz says:

    Re 9, 6, 5 and 4: I think the point here is to select “science” and not a “view.” Essentially, weeded out the view and exposing people to sound science and statistics whatever the results may be.

  20. 20

    i like Google’s and others attempts to make the more technical information more readily understandable and available. however, i am pessimistic about being able to do this in any automated way.

    while David Wilson’s “separating the wheat from the chaff” may or may not mean separating scientific results based upon their policy implications, if he does mean separating based upon policy implications, that’s a shabby level of evaluation. indeed, nothing will serve as well as going to the original papers and struggling to understand them, where necessary using their references to do so. it’s standard library research, backed up by a good scientific and mathematical education. there’ no shortcutting that. and ideology is the worst way of doing it if it is done. laziness is a far better criterion.

    IMO there’s too much of an idea that “truth” can somehow be arrived at by collecting a bunch of “expert opinions” or even “opinions” and voting. yeah, i know, i’m saying juries are terrible ways of obtaining determinations of guilt or innocence. i think there’s been more than one assessment of juries from a perspective of statistical decision making that comes to the same conclusion.

    with respect to climate, for policy to be rationally made, it will need to be based upon a consideration of economic risks wedded to probable outcomes. but i somehow think many don’t care about the nuances of reality, preferring to safeguard their own economic interests and those of what they perceive to be the corporate United States.

  21. 21
    Sean D says:

    Might I suggest, or request, my blog for consideration into this search ability? It’s called “Head in a Cloud”, at The discussion is on scientific matters related to clouds, not climate wars. I generally try to distill relevant and contemporary papers in the scientific literature in a way that scientists and interested people can understand the articles without pouring over the gory details.

  22. 22
    pete best says:

    We are all Doomed im telling you

  23. 23
    Lance Olsen says:

    I really hope that Real Climate will add Glen Barry’s ambitious, excellent, info-packed Climate Ark. Glen was ahead of the climate curve when he started Climate Ark years ago, and he makes great effort to keep its info base abreast of the times..o

  24. 24
    lars says:

    Back to the Future.

    Remember Global Cooling?
    Why scientists find climate change so hard to predict.
    Oct. 23, 2006 – In April, 1975, in an issue mostly taken up with stories about the collapse of the American-backed government of South Vietnam, NEWSWEEK published a small back-page article about a very different kind of disaster. Citing “ominous signs that the earth’s weather patterns have begun to change dramatically,” the magazine warned of an impending “drastic decline in food production.” Political disruptions stemming from food shortages could affect “just about every nation on earth.” Scientists urged governments to consider emergency action to head off the terrible threat of . . . well, if you had been following the climate-change debates at the time, you’d have known that the threat was: global cooling.

    [Response: Try reading to the end of the article:

    The point to remember, says Connolley, is that predictions of global cooling never approached the kind of widespread scientific consensus that supports the greenhouse effect today. And for good reason: the tools scientists have at their disposal now—vastly more data, incomparably faster computers and infinitely more sophisticated mathematical models—render any forecasts from 1975 as inoperative as the predictions being made around the same time about the inevitable triumph of communism.

    – gavin]

  25. 25
    Dian Deevey says:

    I think few of the commenters critical of the new search engine can ever have tried to help ordinary citizens find valid information on the internet. I have produced several guidance documents for local citizens as part of a 3-year effort to block a new coal plant in Gainesville, Florida*. Identifying good sites with credible content comprehensible to the non-scientist is extremely time-consuming. Only those who have tried to do it can have any idea of how the “skeptics” dominate the web. The last thing one should do is tell the lay person to do an ordinary google search on global warming.

    My most recent list of useful sites is about a year old, and full of caveats, because so many federal government sites had either trimmed their sails to comply with administration policy, or simply not updated anything since 2001.

    So I strongly applaud RealClimate’s iniative.

    D. Deevey

    *I think we are close to victory locally, though the picture in the whole state is pretty depressing.

  26. 26
    Bolton says:

    I think it is a great idea, as finding reliable information can be very hard.

  27. 27
    Matt says:

    RealClimate, like any contribution to the global warming fight is welcomed. It, like Ecological Internet, uses the principle of a database of scientifically approved sites as the bases of it’s full text searches. This gives a much more targeted search then the “whole web” type Goggle engine. Ecological Internet is still the only engine that combines this with a taxonomic classification of resources links and a data warehouse of public domain news. The volumes of data going back over the last seven years gives results even though the original sources have long since removed the content from their sites and archives. Climate Ark is concerned with policy and advocacy as well as science. A far better search engine aleady exists.

  28. 28
    Babak says:

    As a search engine founder I completely disagree with this notion in which a search engine provides selective search results. The evolution of search specifically on the web is based on its unprecedented broad access to the information either it has scientific basis or not. However the existing ranking logics behind search engines are good enough to avoid a bogus/spam site or page. If we don’t follow this rule then it would be better stop calling this a search function or a search engine.

    Check climate search and its relevant keywords on RSSMicro:

    RSSMicro Founder

  29. 29
    wacki says:

    heh I don’t like this. The last search engine had some serious flaws but those were limited to faulty regular expressions. Google Co-op has wonderful regular expressions but it’s absolutely horrible at sorting the data. For instance the last search engine would separate the hits in the main entries from the comments. Google co-op does not distinguish between the two.

    I did a quick search. It had 8 pages of hits and the realclimate entries are spread out randomly over the 8 pages. So not only do I have to sift through 8 pages of mostly irrelevant crap but I have to figure out which hits are comments and which hits are from an article. Switching over to google co-op is a very very bad idea. It’s a royal pain in the rear and I’m far more likely to use a “xxxxx” in google than this crap.

    Feel free to display both search engines on your site but limiting your users to just this method will really hamper your readership base.

  30. 30
    Ike Solem says:

    It seems rather like the Google News server, which allows one to search news outlets over the past few months. Not a bad idea at all and potentially very useful.

    Still, when it comes to researching a scientific topic, the tried and true method is to find and read the most recent and authoritative reviews in the field and flip through their lists of references. The journals, Science, Nature and Proceedings of the National Academies of Science (PNAS) are good places to start searching, and they often contain very readable perspectives and summaries as well that can point to interesting papers and reviews. However, this can be a difficult and lengthly process, and you’ll probably need to have a few textbooks at hand to make sense of concepts and jargon – but that’s the traditional ‘literature search’.

    For those who want to just search through for a particular topic, such as clouds, you can go to Google and and type in the search box “ clouds radiative forcing”, for example – that comes up with 425 hits.

    If you do the search, “ clouds radiative forcing” you’ll get 193 hits – most of them from the ‘climate skeptic’ perspective (the models are wrong, the data is wrong, “impressive correlations with solar forcing”, “the reliance of the multi-decadal climate predictions to provide accurate forecasts is further shown to be unjustified” etc.)

    Do the same with “ clouds radiative forcing” and you get 48 hits (all claiming that CO2 emmissions are inconsequential – it’s a fossil fuel site). Thus, the “site:” search tool can help you quickly figure out what a given web site is all about.

    Faced with such differences, it’s always good to refer back to the peer-reviewed journals – which tend to support the conclusions.

    [Response: … and there is a very useful tool to quickly find scientific articles on any topic: try You can see how often each paper has been cited (at least approximately), giving you an idea how relevant other scientists find it. And you can search for all papers by a certain author – very good for journalists who want to check in 2 minutes whether an “expert” on something is really an expert, i.e. someone who has a good track record of relevant scientific publications. -stefan]

  31. 31
    John Hunter says:

    I have to admit that the present search facility does give a curiously skewed view of the web. I typed in the evil word “climateaudit” and got ONLY links to realclimate postings. Even if it is believed that climateaudit contains absolutely nothing of value, isn’t the site sufficiently significant for us to (at least) be told where to find it?

  32. 32
    pete best says:

    Bloke from Freinds of the Earth on the Radio last night been grilled by radio presenter. Basically stated that out of the 2 C temp rise that is currently predicted and likely, 1/3 of it is already here (0.7 c) with another 0.7 C in the pipeline due to latencies and the other 0.7 C likely due to no evidence of emission restrictions been seen at present. The radio presenter was incredulous that the only answer that this bloke had (he has the Governments ear on solutions to CO2 emissions) was taxation that victimises the weak and that we would be better to adapt to climate change rather than try and solve it which he sees as being a hopeless quest anyway due to the eastern empires coming online very quickly and burning of coal being the name of the game.

    Trouble is that when it comes solutions and opinions on what do to about climate change there are many opinions and who knows who is right and what solutions can work. We are all just whistling in the wind at the moment. If we take projected ff fuel use into the future then we will have doubled the amount of CO2 in the atmosphere from present levels by 2100. Another 200 billion tonnes to make 1 trillion tonnes of free Co2 in the atmosphere.

    Doomed and the press and media makes no difference because politically we are frozen because technological solutions do not exist presently.

  33. 33
    kristine says:

    Could google new idea in deciding which sites are categorised and which sites aren’t be a further attempt in censoring information? I know that they have made a recent agreement with China to have exclusive rights over what sites are deemed fit for its citizens. The Chinese Govt are using this as a propaganda tool. I think it’s very dangerous and people should continue searching independantly and sharing info with each other on which sites are worth visiting. Come on surely we don’t need google to dictate to us which sites are good or bad. I know it seems like an attractive offer to you at realclimate but your site can attract visitors without this.

    [Response:I think it’s correct (according to NYT) that search engines in China have blocked certain sites/themes, and I agree that is of great concern. But, let’s not mix issues up. In China, I suppose, there are no other sources of information. Our case is different. If I may, I’d use the analogy with an encyclopedia (in the olden days). In the encyclopedia, I’d expect that the editors have quality controlled the information (I suppose Wikipedia is different, but the argument works for the old days…), and filtered out garbage-information. This makes the encyclopedia useful if you want to look up information. But, if the literary world was just limited to this encyclopedia, then that would be sad indeed. Fortunately, there is a wealth of books, and there are other Internet search engines, so this does, in my opinion, therefore not count as censorship. -rasmus]

  34. 34
    SKA says:

    Q1: What websites does RealClimate use in their specialized Google search?
    Q2: Is there a standard way to look up which sites are being searched or are they forever hidden from the user?

    (As for Q1, I *could* just take a look at some of the search results I get from RealClimate’s special Google search. However, (1)this is tiresome and (2) I could miss some websites in my examination of the results.)

  35. 35
    Jim Cross says:

    I like the new search feature. I typed in “brewer dobson” and got many relevant responses. Previously I only got the RC posts which are few (which surprised me!).

    I like the idea of two search buttons that clearly distinguished which sites they are searching and, perhaps, a link to list of sites included in the Google Co-op search.

  36. 36
    Allister says:

    Re: #16
    Google has the ability built in to all its searches to be site specific.

    In the Search field type “myth” and you will get what you are looking for. You can even do it for only top level domains for example “myth site:edu” for eduational sites or “myth site:au” for australian sites.

  37. 37
    TAC says:

    Gavin is quoted on BBC’s website ( with respect to the new Google search capability

    “Unfortunately, since this topical subject has become rather politicised, the quality of information available on the web is very variable.”

    The article goes on to say: ‘The custom engine on [RealClimate]’s website only searches pages that have been scrutinised by climate scientists and are deemed to provide “solid and reliable information”.’

  38. 38
    Jeff Cowell says:

    Of the two sins, vetting sites to improve efficacy is far better than retrieving willy-nilly the babblings of the uninformed.

  39. 39
    Common Sense says:

    Please alter your search facility so that users can choose between 2 options – one option for an unvetted search for intelligent people who want to see both sides of the debate, and alternate hypotheses, and are capable of making up their own minds in a scientific fashion, and another option for people who are so stupid that they would actually want to search for sites that state exactly the same point of view already expressed on this site – and who automatically accept anything they read without question as long as it contains the word “science” – in a fashion that is entirely unscientific.

  40. 40
    Sam Roberts says:

    But what of policy and advocacy issues? Is science alone going to solve climate change?

  41. 41
    Alan says:

    RE #35

    An elegant fix, thanks!

  42. 42
    Coby says:

    Speaking of Common Sense… how about just going straight to google and doing whatever you choose there?

  43. 43
    Kazz says:

    I get Sun Microsystems and Sun newspaper as first results when searching for “Sun”…. These subjects really don’t have anything to do with climate or anything… :|

    [Response: You did not use our search function. The first results are all about sun earth connections… – gavin]

  44. 44
    Andy Mayhew says:

    Wot no!

    Seriously though, looks fine, especially when you’re looking for data and not opinions.

  45. 45
    Grant says:

    Off topic, but perhaps a useful suggestion.

    Regulars here often refer to previous posts by number (Re: #xx). But sometimes the numbers *change* between the original post and the reply. Perhaps there is merit in designing the system so that the numbers don’t change. If a numbered post is deemed inappropriate and therefore removed, it could be simply replaced with a note that “Comment #xx deleted due to …” (ad hominem, spam, incomprehensibility, etc.)

    Just a thought.

    [Response: It’s because some comments get caught by the filters and it takes time for us to get round to screening them through. They keep their place in the number queue so that comments below get shifted down. We’ve thought about how to adjust this (comment threads for instance), but the practicalities seem more confusing than helpful. Sorry.. – gavin]

  46. 46
    Ed Arnold-Berkovits says:

    This is a nice feature. I do enjoy ClimateArk’s search already.

    How about a listing of which site are included in the Search, and then, with some explanation, which site are specifically NOT included in the Search? i.e the scientific reasoning involved?

  47. 47
    Craig says:

    People will gripe about anything. If you feel that you need to see Penis Enlargement pages in your search results, feel free to go to any numbers of search engines and enjoy sorting through the junk.

    If you prefer focused and localized search functionality from a site you respect, you’ll enjoy using the search info created by the Google application.

    The problem isn’t the technology. The problem is the user-end not being able to make their own decisions when it comes to implementing/using said technology. You have tools at hand folks. Use them or leave them. A very easy choice to make.

  48. 48
    SKA says:

    Hey Gavin or Stefan, how about dedicate some of your precious time to answer all the questions about WHICH sites are included in the search? Is it really such a tall order? Please?

    [Response: I’m at a meeting this week and only sneaking looks at this in between talks. Issues like the list of sites and the search choice will be dealt with when I get back (next week sometime). Thanks for all the comments- they are being assimilated…. – gavin]

  49. 49
    Jay Furneaux says:

    I’d like it if I could also search just realclimate. At the very least the RC pages should be grouped at the top of the results. This kind of search is half-way to having a directory, which may be more useful.
    I suspect many people will also use the â??fullâ?? Google. I usually end up having at least two browsers open at a time when on the web. One is a search engine.

  50. 50
    Karl Martino says:

    Following your example I’ve built a custom search for the Philadelphia region based upon our our extensive blogroll:

    I havn’t finished the refinements yet, but it’s very very powerful for what it is as it stands.

    Note that the search bar to use is the one on the left, in the main rail of the site, not the upper right hand corner site search, which I will wire later.