Can the Internet ever be corrected?

Can we rely on travel information we find on the Internet? An exchange on Twitter with entrepreneur and world travelling phenomenon Gary Arndt set me thinking.

@hackneye (me): Being right more of the time is what guidebooks do better than the Web. It therefore makes me sad to see one with horrendous errors.
@EverywhereTrip (Gary): I’d disagree that they get things right more than the web. They are always 1-4 years out of date given the publication cycle.

@hackneye: Yes, they can be, and sometimes that’s important. But they also tend to be researched and fact-checked more carefully.
@EverywhereTrip: So long as the public can edit the info, errors can be corrected quickly online. 1,000′s of people checking instead of 1.

@hackneye: I agree. But the public don’t correct most of the travel content on the Web. Hence the variable quality.
@EverywhereTrip: What is an example of this? Most pages have comments, or at least a way to contact the owner.

@hackneye: An example of what? An uncorrected error on a travel site? Just Google your hometown and dig around.
@EverywhereTrip: That’s the thing. I don’t see much incorrect information. People always give theoretical examples, never concrete ones.

@hackneye: Chelsea’s football ground is the site of a battle that took place 200 miles away: http://bit.ly/agSoZN… Michelangelo’s David is in the Uffizi: http://bit.ly/aq8Vfn [it isn't]. 2 quick searches, 2 highly ranked sites.

@EverywhereTrip: So leave a comment correcting the information :) problem solved.
@hackneye: I admire your idealism. It’s a fine quality. But I suspect ‘correcting the Internet’ is too big a job.

@EverywhereTrip: The internet is a work in progress. If you see an error, correct it. Everyone does a little bit and it adds up…

And so it continued, with a 140-character limit becoming increasingly unsatisfactory for expressing quite complex ideas. It never really occurred to me that my opening gambit was controversial. I’ve worked with guidebooks for years, and have seen (and, yes, made) sloppy errors. Have any made it as far as print? Sure, a few. But I’m certain I’ve never seen the kind of howlers that I discovered with 2 cursory searches on Google. I’ve never seen a guidebook that claims David is in the Uffizi; almost anyone who’s even read about Florence knows that. Nobody needs a guidebook, of course, and Google (heck, even Yahoo!) easily beats one for up-to-date pricing information. But a well crafted guide (like a history book) can enrich a visit; to compare the Blue Guide to Tuscany, say, with a Google search is to commit an obvious category error.

Further, I can think of at least 6 reasons that we’ll never be able to rely on travel information pulled from a search:

1. The Internet’s strength is also its weakness from a travel information point of view. It’s unedited and (partly) democratic. Ten cheers for that. However, there is no democracy of facts, facts need checking, then double-checking… and fact-checking means editors.
2. The Web follows something like a geometric pattern of growth. The number of people willing to spend their time editing it, for free, is more likely to grow arithmetically. There’s an incompatibility there; we could play whack-a-mole with mistakes forever.
3. The kinds of financial returns available online aren’t conducive to thorough destination research, at the moment and for the most part anyway. Patrick Smith has documented the UK launch of Demand Media on his blog, and the pay for experts at leading travel portals isn’t significantly better. (Not because those running such sites are rapacious capitalists; far from it. They’re professional outfits that have done the maths and worked out what a page is worth in traffic and click-through revenue.)
4. Visitors to travel sites are seeking information, inspiration and enlightenment. They don’t know; that’s why they’re there. They aren’t equipped to spot errors (a classic adverse selection problem).
5. When it comes down to it, why should we spend time correcting the errors of content creators who can’t even be bothered to check Wikipedia? (Even to read the Battle of Stamford Bridge, paragraph one.) If it’s your job (or your hobby) to provide quality, well-sourced travel information, then do your job. Visitors to your site aren’t there to correct your errors, and readers are generally not likely to help people with such cavalier attitudes to quality.
6. There’s just far too much information out there, much of it contradictory, much of it abandoned by its creator, and most of it easy to find with the right search term. This is an inevitable consequence of the Web’s DNA.

It strikes me as a truism that the quality of Web information available for free (or nearly-free) will improve. Algorithms will get better, the semantic Web may help push us to better quality sites. And, as Gary points out, crowdsourcing does work. We can agree on that. Individual travel sites will improve (and there are already good ones) and challenge the established names. One way some disruptors are doing so is by becoming editorially more like the traditional players.

I make a significant slice of my income writing on, or about, the Web. I use it all day, every day, and have done for over a decade. To see it as a ‘work in progress’ is to fundamentally misunderstand its nature, in my opinion. (Incidentally, some Marxists and neoconservatives make exactly that mistake with ‘history’… but that’s another subject.) I’m certainly not fixated on the professionally-produced guide as book, app, or whatever. It’s the message that’s key, not the medium. And unless I’m very wrong, the message we receive from the Internet won’t ever be ‘corrected’.

This is only a first draft of my opinion. Please feel free to use the box below to correct me.

Tagged as:

18 Comments

  1. alexandra March 23, 2010 at 2:49 pm #

    As the web increases the value of its content and google increasingly learns to recognize and “crank up” sites of greater value, the information available online will be perfectly reliable and often better, more in detail, and more immediate than anything you can get from a guidebook.

    I like to think that the sites I write contain information as good as a guidebook, art history article, or university lecture, since I have written/given all of these. There are plenty of qualified writers out there – like you Donald – and we write great content… for the web. And occasionally for print.

    That said, sadly there are a lot of SEO-geared sites that rank high on google and that provide awful information, and it’s very hard to correct them (though I try). Just this morning we found someone on twitter who said that Ferrari was manufactured in Tuscany.

  2. hackneye March 23, 2010 at 8:17 am #

    (At least) 6 reasons why travel information pulled from Web search will never be reliable: http://www.donaldstrachan.com/archive/20… (c/o me)

  3. Donald March 23, 2010 at 3:26 pm #

    Hi Alexandra. I can’t really disagree with what you write there. The Web is full of great writers: many of the same writers who appear in print, as well as talented bloggers. A search gives me access to way more than I could fit into ten guidebooks, I agree. My point is more that access to the ‘neat firehose’, which is roughly what a Google search is, will never be reliable. Reliability needs editors and fact-checkers, even for the best writers, and for that reason search (however good it gets) will never qualify as reliable. Over time the algorithms *are* getting better at picking up *exactly* the content I’m looking for; but that’s completely different from saying that that content is accurate. Because lots of it isn’t.

  4. noddyboffin March 23, 2010 at 8:36 am #

    RT @hackneye: (At least) 6 reasons why travel information pulled from Web search will never be reliable: http://www.donaldstrachan.com/archive/20… (c/o me)

  5. Caitlin March 23, 2010 at 4:41 pm #

    Access to the firehose isn’t what I want because then I have to do the editing. What a guidebook offers me, besides hopefully accuracy, is that someone has done the editing and selected the most important stuff for me. I do like to go off piste and wander around a city and make my own discoveries too, but a guidebook really helps me for the basic stuff. I don’t *need* it but it can really help. And information on the web is just as prone to getting out of date as a book – Google returns old pages with authority more than new stuff.

  6. Nic March 24, 2010 at 7:43 am #

    As someone planning to edit a website for one of the two major tourist attractions of a country generally off the beaten track I find this article fascinating.

    In places where tourism, marketing etc are not as highly developed as those on the beaten track, trying to find some of the information that tourists require can be nothing short of impossible.

    For example our local council has a long-standing tendency to announce events one, or if we’re lucky, two days in advance. Cultural events calendars are notoriously difficult to upkeep at the best of times and these are not.

    If local councils are difficult, national tourist boards can be even worse, hence how guide books for some countries arrive at their information is of great interest. One of the most time consuming elements is finding the person responsible for marketing these events in the first place. A travel writer spending a couple of days in any one place is unlikely to be able to do so.

    Currently we see some tourists do their research prior to the trip on the internet then refer to a guidebook during their travels. Surely this is the wrong way of doing things as the book cannot be updated, a site, if maintained correctly, can.

    Perhaps this is a generational thing and as time goes by people will move from books to the net, depending on whether or not places invest in providing these resources.

    Ultimately, a website or a guidebook is only as good as the sources used and when it was published.

  7. hackneye March 24, 2010 at 2:41 am #

    Is the Web a reliable source of travel information? I think not. Will it ever be? No to that, too. Here’s why: http://www.donaldstrachan.com/archive/20

  8. DonaldS March 24, 2010 at 5:43 pm #

    Caitlin, Nic, thanks for stopping by and commenting.

    Caitlin – I agree, basically. Curation, like you describe, is another thing a guide does well, assuming the writer has done her/his job properly. The firehose is just too unwieldy for many aspects of trip research. And you make a good point about Google returning old pages. Hadn’t thought of that.

    Nic – events, at least ones that don’t occur every year on (roughly) the same dates, are tricky for guidebook research. One way round that is for the books to recommend sites that they know will be accurate and kept up to date (local government affiliated sites, local tourism authorities etc.). That’s what I have done. The trick is doing your research properly, so as to know which sites to recommend. Both books and the Web can be no better than the sources they cite and use, like you say.

  9. hackneye March 24, 2010 at 2:51 pm #

    Thx for RTs on “travel + the firehose don’t mix” (http://www.donaldstrachan.com/archive/20…) @Paul_a_smith @creativetourist @oshratn @brianbge @samdaams @M_Hensh

  10. travelfish March 24, 2010 at 4:00 pm #

    RT @hackneye: Is the Web a reliable source of travel information? I think not. Will it ever be? No to that, too. Here’s why: http://bit. …

  11. Nic March 24, 2010 at 11:59 pm #

    Thanks for the reply and tips Donald.

    I concur about guidebook research and admit that a daily events calendar is not really something they can do well, nor should they. Focussing on the major tourist attractions of a country then breaking it down city by city with additional info on smaller places of interest is probably the best.

    This is why it is frustrating to see tourists arrive with only guidebooks as their travel resource. On a day-to-day basis they simply cannot compete with good, local websites, if available. The small city we live in is more of a staging post for other destinations, making local event info critical as exhausting the more touristy destinations can be easily done.

    Finally, on the subject of crowd-sourcing, this is something that is critical for the success of some tourist websites. Especially if it is an ex-pat site in a country where the local language is their second or third one spoken. Or where there are sites run by locals that for all intent and purposes is a tourist site but not directed for tourists.

    Caitlin’s point about Google’s return is fascinating. I can attest that the top 10 for our city has 2-3 results that have been abandoned since 2007-08. I assume this must be down to their URLs, as I was under the impression that Google’s algorithm was directed towards sites that updated on a more frequent basis.

  12. Stuart March 25, 2010 at 7:37 am #

    Speaking from personal experience, and with the exception of forums (which are their own special beast), generally speaking the more you open a site to comments and input by “the crowd” the more precipitous the fall in quality of submissions becomes. We’ve allowed reader submissions in two forms in the past — one moderated, one unmoderated. We’re about to switch the latter to moderated as it is the only means to have any QC at all. I’d also debate the concept that general readers are all that interested in correcting the incorrect — that’s what editors/researchers are for. My feeling is that readers who are “correcting” pages are often doing so only via online/guidebook research themselves which doesn’t really help matters — if you haven’t been there, you haven’t been there.

  13. creativetourist March 25, 2010 at 2:25 am #

    @hackneye A pleasure – fascinating debate & agree with you. Love the internet but never take anything on it as fact unless it’s been edited!

  14. Donald March 31, 2010 at 9:30 am #

    Interested to hear your thoughts. I’ve tapped a quick reply. Thanks for stopping by.

  15. Gary Arndt March 31, 2010 at 10:44 am #

    There are two things being confused: the medium (print vs online) and the information gathering process.

    Print can’t be corrected. Once it is out the door, you can’t very well bring every copy you printed back for correction. It is the nature of the medium. If you do have a correction, you have to wait for a new edition of a guidebook to come out which could take years.

    Content online can easily be corrected. This too is the nature of the medium.

    The question of crowd sourcing information (which LP might be moving towards with some recent comments made by the LP CEO) vs a top-down, authoritative, fact checking process is independent of the medium. You can published information from a guidebook author online and you could print crowd sourced information.

    The fact that you can find “an” example of incorrect information shouldn’t be a strike against the internet any more than a bad guidebook should be a strike against all guidebooks. I had a guidebook in the South Pacific that had nonexistent flights and airlines that went out of business. It wasn’t the fault of the guidebook, it was the nature of the medium. They have a long publishing cycle.

    In the example you gave, it said that David was at the Uffizi. Fair enough. That is wrong. However, if you did a search on Michaelango’s David and looked at the Wikipedia entry, the first search result on Google, you’d find it is located at the Academia Gallery. in fact EVERY search result I checked on the first page of results on Google for “David Michelangelo” had the correct location listed.

    The problem wasn’t in the page you pulled up (yes, that fact was wrong) it was in the search you did. If you did a search for David, you’d easily find the information and could cross check the information with other sites. In fact, the example you gave was one user submitted comment in a collection of 139 user submitted comments.

    Moreover, the page wasn’t one about David, but was about the Uffizi. If you read the rest of the comments just on the first page, you would have found location, operating hours, phone number, URL, discount information, and admission fees?..all submitted by users. The link to the Uffizi site can confirm this information and provide more official data than even a guidebook author could.

    Likewise, in the other example you gave, was a page on football stadiums, not a page about battles. It makes a big difference. If you are interested stadiums, the battle is just an anecdote. If you are interested in battles, you will probably get better information by searing specifically about the battle, not for football stadiums.

    If you conduct a good search, I contend you can find anything you can find in a guidebook. ANYTHING. You can find more online actually because there are limits on how comprehensive a single researcher can be.

    Furthermore, there is an entire category of data which can be inaccurate due being out of date. If you take that data into consideration, much of the information in any guidebook will be incorrect. I’m not sure why errors of that sort should be dismissed.

    I would be willing to put this to the test. I’d easily make a wager regarding my ability to find information online vs a guidebook.

  16. Donald March 31, 2010 at 11:24 am #

    Lots to say, of course, but I’m on the road with a 3 inch screen so I’ll be brief. My point absolutely isn’t whether you can find more or better info on the Internet. Of course you can: the Web is the greatest invention of modern times; it contains a huge chunk of total human knowledge, and fallibility. To compare the “Internet” and a “guidebook” (on paper or otherwise) is about as meaningful as comparing oranges and car ferries. My key point is about the nature of search combined with the amount (and exponential growth) of information out there. Yes, Google “David” and you’ll find out where it is. Obviously. But will I google “battles” if I’m browsing a trusted site to find inspiration to fill an afternoon in London? I think that’s unlikely. Can crowdsourcing improve the relevance and accuracy of professionally gathered content? Absolutely. Am I prepared to take a search algorithm’s word for what constitutes a fact, based on what floats to the top of the page. Sorry, no.

  17. DonaldS March 31, 2010 at 8:42 pm #

    Just occurred to me, Gary, that this point of yours kinda makes mine:

    > In fact, the example you gave was one user submitted comment in a collection of 139 user submitted comments.

    What’s interesting to me is that this comment isn’t just “one user comment”. It’s the review that a well respected travel site sent to the top spot of a search for one of the world’s most famous tourist attractions. I didn’t put it there; Google sent me. If you scroll down to number 2 in their review list, you’ll find the assertion that Leonardo and Michelangelo used to gather at the Uffizi. Despite the fact that the former was dead 40 years before work started on the building, and the latter hadn’t been in Florence for a decade or two by then and died 15 years or so before the Uffizi opened. Those are the first 2 in the list (ordered by quality, not recency, according to the site). I’m sure there is some excellent advice to be mined from those 139 reviews, but unless you already know what the truth is, there’s no way of telling whether your drink from the firehose is elixir or poison. That’s the problem. Not that there isn’t brilliant information out there; there is more information than you could fit into a library, never mind a guidebook. The trouble is that there is also lots and lots of bad, much of it way too easy to find. That’s the Web’s DNA, it’s why the Web is as powerful as it is, and not a problem that we can fact-check or crowdsource away (see Stuart @12, above, for example).

  18. DonaldS March 31, 2010 at 8:53 pm #

    Oh, and for some reason my reply to your comment, thanking you for stopping by (@ 14), seems to have gone back in time and jumped above your initial comment. Weird. Must be a bug in the Android app. But, ta, anyway, Gary.

Leave a Reply