Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

[P]
Articles and Corrections on the Internet

By schporto in Media
Mon Oct 30, 2000 at 10:42:00 AM EST
Tags: Internet (all tags)
Internet

Poking around on the internet, last week I found a article that had some clearly wrong information. I thought it was kinda funny because it was on an "official" site. That night I showed a friend. The information had changed to the correct information. I thought it was odd but not much more untill I stopped and really thought about it.


OK so the full story goes like this. For context I'm an ice hockey fan. I was reading the NHL's official site and specifically an article about Ray Borque and how he just passed Paul Coffey of the Carolina Hurricanes (look no Coffey) for a record. OK but Coffey play's for the Bruins (look he's there) now. I laughed. That evening I went to show a friend. The article had been corrected. That makes sense, I mean the internet is dynamic data that's part of the beauty of it.

But I have no way to prove that they posted something in error. And there is no admision from the site that they posted something wrong. And no way for me to prove this actually happened.

Now really it doesn't make much difference in this case where it's a player on a hockey team. Most people reading the site probably knew they flubbed. And knew what the truth was, but what if it was something more important. Let's say for instance NASA (for example) reported something seriously wrong. Say that an asteroid was going to collide with the Earth on Nov 1, 2000. Some people read this and assume this is an official notice. They panic, pandemonium ensues. Someone at NASA realizes they goofed. They update their article to read they've discovered an asteroid collided with the Earth on Nov 1, 2000 BC. (Yeah I know far fetched but work with me) A simple mistake, but one with serious consequences.

Newspapers have always had a correction section where they print the articles they screwed up. And what the correct information should be. Yeah these are usually small sections that nobody reads, but they are still there. And there is a way to trace errors. Some newspapers are still providing these corrections online. But some aren't, or at least not that I can find. Some others treat this oddly. They provide the correction and link to the corrected article (complete with editorial note that its been corrected), but I can't find the wrong article.

In some ways this may seem trivial, but in other's it is important. I like CNN's way of doing it, if there was a way to read the wrong article. But at the same time I understand that this might seem like they're still publishing the wrong articles.

So how should news sites provide the correction information? Should they just change the information and never admit they were wrong? Change the data admit they were wrong, but never show the errors? Note the errors, provide a link, and continue showing wrong articles? Any other ideas? Do consider this could effect other types of sites too. Like kuro5hin.

Sponsors

Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure

Login

Poll
How should corrections be done online?
o Just update the article. 9%
o Update the article, admit error, remove errors. 57%
o Display errors and corrections with errors highlighted. 14%
o Just put up corrections, don't change articles. 8%
o Other? 2%
o 42 8%

Votes: 84
Results | Other Polls

Related Links
o Kuro5hin
o NHL's official site
o an article about Ray Borque
o Carolina Hurricanes
o Bruins
o NASA
o corrections
o aren't
o others
o correction
o corrected article
o Also by schporto


Display: Sort:
Articles and Corrections on the Internet | 19 comments (19 topical, editorial, 0 hidden)
Panic could still ensue (3.60 / 5) (#1)
by retinaburn on Mon Oct 30, 2000 at 09:52:43 AM EST

One problem I can forsee about leaving the incorrect article available is someone could still browse and become misinformed from reading it. Even if there was a banner headline "This article is incorrect", with a repeated warning after every complete sentence that this article contains incorrect information people would still read the article and go into hysterics. It would be referenced in essays and the like, causing havoc in the education community.

Ok a little far-fetched but we already have problems in people reading joke-news and just blatantly wrong things and taking them for fact. I can't see allowing newspapers to do the same.


I think that we are a young species that often fucks with things we don't know how to unfuck. -- Tycho


I have little sympathy (none / 0) (#7)
by squigly on Tue Oct 31, 2000 at 04:24:10 AM EST

If people are really so stupid as to be unable to spot a notice saying "This is incorrect" then there's not really a lot of hope for them. the only cure is for them to repeatedly make themselves look stupid by failing to notice and hope they learn that way.

--
People who sig other people have nothing intelligent to say for themselves - anonimouse
[ Parent ]
Here is your sign... (none / 0) (#15)
by Luke Scharf on Wed Nov 01, 2000 at 09:41:16 AM EST

If people are really so stupid as to be unable to spot a notice saying "This is incorrect" then there's not really a lot of hope for them. the only cure is for them to repeatedly make themselves look stupid by failing to notice and hope they learn that way.

Ever hear Bill Engvall's comedy routine "here is your sign"? Might be a good idea. :-)

For those of you who don't listen to this kind of comedy, he suggests that we give signs to stupid people so we don't bother asking them questions:

I'm moving. The moving truck is IN THE DRIVEWAY, and we're carrying big boxes out to it.

My buddy drives up, sees the truck and says "you movin'?"

Here is your sign.

:-)



[ Parent ]
New app needed! (2.62 / 8) (#2)
by Signal 11 on Mon Oct 30, 2000 at 10:07:53 AM EST

I think we need a proxy that records changes to html files and saves everything viewed for 72 hours (encrypted, of course), so if you spot something like this, you can tell others. Maybe in the future, a feature could be implimented so it updates a central database so other users of the proxy could also view the "old" data. Maybe even something like CVS to view each change along the way?

Personally, I think it's wrong to correct anything more than typos without telling people for two reasons:

1. Misinformation. One person got told one story, and believes it is correct, another person got told a slightly different version and believes it is correct. Unless there is some way to reconcile those differences or alert the other person, two "memes" exist in those people's minds which are different. This is just plain bad journalism.

2. If there is no explicit correction, people referring to the "old" content as authoritative and correct may make a preventable mistake - for example, engineering documents. Engineers have "Revisions" and most any change increments the revision, in the same fashion that programmers have versions of software. Perhaps the next version of the "web" should have more intensive version checking...


--
Society needs therapy. It's having
trouble accepting itself.

Journalling internet? (4.00 / 1) (#9)
by spiralx on Tue Oct 31, 2000 at 09:55:59 AM EST

Perhaps the next version of the "web" should have more intensive version checking...

That would be a pretty handy feature in some cases, although for dynamic sites it would perhaps be a little storage intensive... At the moment many parts of the internet are far too transient to be useful and there's lots of information that just gets dropped after a period of time. Hopefully storage costs will drop enough that this won't be necessary in the future.


You're doomed, I'm doomed, we're all doomed for ice cream. - Bob Aboey
[ Parent ]

Taking the 1984 Approach (4.00 / 5) (#3)
by teeheehee on Mon Oct 30, 2000 at 10:22:58 AM EST


Looking at this from a more paranoid angle, it seems only right that sites should have the misprinted versions still accessable, with a correction link added noting whatever changes needed to be made. This is why I think this way:

One of the scariest things about history lies in the ability to rewrite it to fit an agenda. In this case the agenda (hopefully) is to tell the truth. Fine. What if that weren't the case? Sure, the Net is dynamic and that makes it almost alive like we are, but if disinformation is able to replace information with the same efficiency as truth can replace untruth, how do we determine if what we're reading is the original or if it is telling the truth?

The problem will exist anyways, I realize, because it's just as easy for someone to take an article which is telling it like it is, add a link saying "Correction: Blah blah blah", and have that replace the information on the real page, but in Orwell's '1984' there was a method of recreating the past to suite the agenda of the government in the present. They would systematically wipe out information which contradicted with what was going on in the present, to make it seem like things are always getting better, when in fact they were getting far worse. If it's impossible to reference what the original said, it's hard to find evidence that there is a dispelling of the truth... this causes mass dilusion, mass control, and other big nasty things.

Does it take much time on the web designer to add a link for corrections? It makes a difference to me! Say you're wrong when you're wrong, and I'll be more prone to believe in you. This is true of politics as well - if politicians just fessed up to being lying thieves who have done many wrongs in their younger years I would be happier to say "I live in a free country, I will vote for you"... ironic, isn't it? It's unbelievable what the power of truth can bring to the hearts and minds of those actually looking for it.

Just one more point to drive my thoughts home: Science. In scientific experiments there are flaws, miscalculations, bad theories, etc. It's necessary, otherwise the better theories may not be developed, or would be at best on par with some of the wrong ones. When scientists admit to having flawed, it's not so much a defemation of character, but progress in the right direction. Now they know what's wrong, they can start to hone in on what's right...


(Discordia) :: Hail Eris!
Everything you've just read was poetry and art - no infringement!

Change history (3.33 / 3) (#4)
by loner on Mon Oct 30, 2000 at 11:14:14 AM EST

What you really want is a "change history" section much like what we usually put in source code files. Hopefully this'll be implemented in a standard way, so that it can be found easily whether you're on well-known-media-outlet.com or local-news-site.com. Perhaps sites can put a link to the change history right below or beside the author's name (byline?), and put the change list either in a separate page or at the end of the article.

This fixes most cases I can think of: New visitors don't care about previous corrections, and the change history link/section is separate from the actual article body, so it doesn't get in the way. Previous visitors can find and read this section to assure themselves that they are indeed not losing their minds. The case not taken care of is a visitor who reads the older article but never returns, maybe that's why nytimes requires registration, to email folks all corrections *grin*.

BUT, beware that in today's society, many media sites may even ban displaying such correction logs, since this will be a public admission of error and will leave them open to litigation *grin wider*.

Searches (4.00 / 1) (#5)
by schporto on Mon Oct 30, 2000 at 11:26:28 AM EST

A problem here comes when searches are done. Most of the search engines go through all articles and will turn up links to the errors. So the search engines would need to be modified too. Just another little wrinkle.

-cpd

[ Parent ]
Does not compute (none / 0) (#10)
by loner on Tue Oct 31, 2000 at 11:33:49 AM EST

So the search engine links to an article based on text in the "change history", so what? That same search engine will probably link to this K5 article because it mentions the erroneous text. It looks to me like this is more a shortcoming of the search engine.

But still I don't see what a big deal that is. Say a user searches for "Coffey and hurricanes", they get a hit on the NHL article, they go there and find no mention of Hurricanes in the article body, then they notice the change log that mentions the correction and understand how the search engine pointed them there to begin with.

Just to be sure you're not misunderstanding me. I'm not advocating keeping older versions of each article, I'm talking about updating the article as required, removing the incorrect versions altogether, and showing a log of changes with the article for posterity. E.g.:

Article Title
Author
(link to change history)

Article Body
...

Change History:
10-23-00 - Corrected the fact that Paul Coffey plays for the Bruins, not the Carolina Hurricanes.


[ Parent ]

1984? (3.33 / 3) (#6)
by kjeldar on Tue Oct 31, 2000 at 12:33:25 AM EST

Is this Orwellian? Is history being rewritten by major news sites?

No. You were perfectly free to save a copy of the erroneous page on NHL.com for yourself. That would have probably been adequate proof of the error and its subsequent correction.

Reverse? (none / 0) (#8)
by schporto on Tue Oct 31, 2000 at 08:24:18 AM EST

OK so if I do save a copy would you (or more importantly people in general) believe me? And if so what's to prevent me from just tampering stuff to make other's look bad (like I could be with this story :).

Let me try another example of where I think this starts getting sticky. How many times do you surf the net and forget where you've read something? You know like "Company A announced they will deliver wiz-bang product in 1 week." You go "Cool, they're gonna make a mint off that." You go and buy their stock cheap before the product comes out. Next week you go to get the wiz-banger, and find its not out yet. The site you had read goofed and ment next week of next year. Now you can't even find that info again because you can't remember where you read it. And it doesn't show up in any searches. They all have the current data. I mean it was a typo after all. Yet you invested money on this info.

I don't know maybe its me, but I think this might be a problem that some people just haven't thought of yet.

-cpd

[ Parent ]
Pick better examples (none / 0) (#13)
by dubious on Tue Oct 31, 2000 at 03:13:21 PM EST

If you buy a stock based on a news story without doing due diligence you deserve the returns that you receive from that investment regardless of whether or not the news story was a typo or not.

[ Parent ]
Won't change a thing. (3.50 / 2) (#11)
by Dr Fau5tus on Tue Oct 31, 2000 at 12:26:55 PM EST

The problem with the example is that it clearly states the mistake was made by NASA. If this was the case then it would be all over the more conventional news as well, this would cause the panic not some web news service. A website would have no need to change their content, e.g. 'NASA announces end of World', it wasn't their mistake. As for the immediate dangers of a wrong or dishonest posting on a news services: compare the meteorite example above and Orson Welles' infamous 'Martian Invasion' radio broadcast. The credible mob that ran for the hills back in 1938 at least had the basis of an excuse, they were listening to their primary, and till then largely unsubverted, source of 'up to date' information. No-one reading a website has even that excuse and anyone who relies on the internet as it currently stands as their primary and definitive source of news is a muppet, I'm sorry. If they rely on an individual site and subsequently find themselves up a mountain after giving away all their possessions and making love to their spouse 'one last time' waiting for a meteorite that doesn't turn up, well...:-)

Schporto raises an interesting point, how to give the mutable internet some kind of 'memory' to prevent news services merely changing their content as if the previous version never happened. Some kind of monitoring bodies, independent of the news service and each other, which regularly record and timestamp the output from a site seems to me to be the only solution that can provide anything resembling accountability. News services which voluntarily submit to this and make revisions explicit on site will naturally gain the most credibility.

The problem is that none of this will guarantee veracity for newsites any more than it really does for TV or print based news. They will suffer from all the same problems: bias, parochialism, a tendency to present themselves as definitive as well as glaring sins of omission which tend to lead even well intentioned people to a consensus view if not outright propaganda. The internet didn't invent lying to people and certainly didn't invent the practice of allowing people to forget a news services mistakes. Newspapers often have to be dragged kicking and screaming to retractions and then hide them away where large numbers of people will not notice them. Here in the UK, for example, our tabloids are often shameless in changing stance, wallpapering over issues and speculation that they paraded around as 'facts' only weeks before following the populist pound as public perception and attitudes change (sadly people generally prefer to have their prejudices confirmed rather than challenged). Our more respectable broadsheets are not always much better. Concerns about the past being 'changed' are missing the point. People are perfectly capable of forgetting all by themselves.
_________________________________________________ I'm not censoring you, I'm censuring you. Stop complaining.

Yay for crypto (none / 0) (#12)
by royh on Tue Oct 31, 2000 at 02:19:11 PM EST

Schporto raises an interesting point, how to give the mutable internet some kind of 'memory' to prevent news services merely changing their content as if the previous version never happened.

Cryptography, in this case public key, is our friend. Websites (well, anyone), can sign their content, thus allowing anyone with an article from that source to prove that the article is legitimate.

This allows everyone to become a monitoring body, as you suggest:

Some kind of monitoring bodies, independent of the news service and each other, which regularly record and timestamp the output from a site seems to me to be the only solution that can provide anything resembling accountability.

[ Parent ]

Technicality (none / 0) (#18)
by mcelrath on Fri Nov 10, 2000 at 11:23:43 AM EST

This is a great idea. A web site that simply signed other people's web pages on a semi-regular basis so you could see when, if, and maybe how they changed. Such a site could also detect changes and list them for users. You'd only really have to monitor news organizations, significantly cutting down on the number of pages you'd have to keep track of. This is a logical extension of Google's webcache, which can be wrangled into performing this.

One technicality though...most news service pages are dynamically generated, and the ads change and rotate. How do you algorithmically tell whether the news site changed the article, or is just changing ads, or layout? (Of course you could strip the ads using something like FilterProxy, but this is not necessarily deterministic, and won't help you if they change their layout.

--Bob
1^2=1; (-1)^2=1; 1^2=(-1)^2; 1=-1; 2=0; 1=0.
[ Parent ]

Technical Solutions (2.00 / 1) (#19)
by royh on Mon Nov 13, 2000 at 09:21:51 PM EST

This would be a fairly nasty problem right now. I was thinking longer term, maybe five to ten years from now, when cryptography will be commonplace. Think of XML, where you could enclose the real meat of an article with special tags so that it would be easy to just grab the plaintext, without worrying about ads, or changes in format.

But even if you weren't able to do this, you could grab the whole page, and show it to people. It would be impossible to tell the difference between a format change and a content change, but you can still save the whole thing.



[ Parent ]

The Right Thing is to provide revision history (2.00 / 1) (#14)
by Paul Crowley on Wed Nov 01, 2000 at 09:11:41 AM EST

I've wondered about this with regard to sites like this. I sometimes see articles with minor errors in (like URL errors) and correcting responses from the same author, and think "if only you could correct after posting!" But this would be bad in a flamewar, because you could make those who follow you up look silly.

I think there's an easy solution: provide the old versions. Something like this:

This response has been revised 4 times. Previous responses: -3- -2- -1- It is in response to an older version of the article above.
--
Paul Crowley aka ciphergoth. Crypto and sex politics. Diary.

Along the Orewllian Lines (5.00 / 1) (#16)
by Forum on Fri Nov 03, 2000 at 12:13:36 PM EST

Sure, it was just an NHL Article. The fact remains, however that the article, any article can be changed, and no one would know except the person who originally wrote it, and the person who changed it. Subversion of information attacks have already occurred, and are likely to occur again. These are, in my opinion, the most dangerous attacks ever, because instead of a stupid "w3 own y0u" page, the netadmins see the same page they are used too, with only a few words changed. I used to have the link to an article about a hacked web page where the only thing changed were a few words detailing the arrest of an alleged computer criminal. My point is this, the entire value of a news article lies in the details. When the details are corrupted, the entire article is.

-forum

-- "When I walk down the street and only 3 or 4 shots are fired at me, I find it hard to stay awake." -HC
Dynamic Content (none / 0) (#17)
by Wah on Mon Nov 06, 2000 at 03:51:49 PM EST

This is a curious side effect of dynamic content. I had a small personal example with this effect, covered here.

One thing that I've taken to doing for some of the stories I mention is saving the full webpage (IE has a fully function "save as" feature) and then mirroring it on my site. Yea, this probably causes some legal problems, but I think it might be defensible as long as you don't change anything. I didn't do that for that particular link and any "proof" I might have had disappeared into the ether.

I also do version numbering and a changelog for my own stuff, but I don't really cover what exactly has been changed. For someone going for full credit as a reliable resource, any significant changes dealing with facts or presentation of facts should be fully documented. Fully documented and easily accessible in the manner of a good linked document.
--
Fail to Obey?

Articles and Corrections on the Internet | 19 comments (19 topical, 0 editorial, 0 hidden)
Display: Sort:

kuro5hin.org

[XML]
All trademarks and copyrights on this page are owned by their respective companies. The Rest 2000 - Present Kuro5hin.org Inc.
See our legalese page for copyright policies. Please also read our Privacy Policy.
Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
Need some help? Email help@kuro5hin.org.
My heart's the long stairs.

Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!