April 06, 2003

Has the New York Times Shot Itself in the Foot?

Richard Gayle thinks that the New York Times has just lost an enormous amount of long-run intellectual mindshare by trying to make a little bit more money from its archive. I tend to agree. I think that (a) people are going to be surprised at how rapidly their links to the New York Times succumb to linkrot, and that then (b) they will start linking to other news and opinion sources in preference.

I know I will.

Richard Gayle's Weblog:

The New York Times Changes Access to Old Content. The New York Times has just changed their archival policy so that all links we've used in the bIPlog that are more than 30 days old will redirect to a page requesting that you purchase the article for $2.95. Links have worked before now, even though articles were months or years old. Vin Crosbie, President of Digital Deliverance, talked last week at the JSchool about news online, and the mechanisms and logic that publishers use... [bIPlog]

This could be the end of the Times as a source of links in the Internet. I will no longer link to any of their pages since no one would be able to see anything after 30 days. Why tell anyone else about something interesting if they will have to pay 3 bucks to read it? For anyone to have to pay $2.95 to read a single article that is over 30 days old is ludicrous. You can get a great deal by paying $25.95 for access to 25 articles. Wow!! That is so nice of them. Plus you get 6 months to use this. Actually, all you buy is access to the article, not the article itself. You can look at the article (without pictures or graphs) for 90 days. Then you have to pay again. And, apparently, this holds true even if you are a subscriber. At least there is nothing on the payment page that says differentlyWhat is funny about this is that scientific journals are going exactly the opposite direction. It costs money to read the current issue but many are making all their work open to everyone after a period of time. PNAS, for instance, allows open access to anything 6 months older or more. So, at least here there is a benefit to having a subscription. You can get access to 6 months of material at a reasonable price. But, if you can wait a while, you will eventually get older material. Plus, if you do not want to buy a subscription to get those 6 months of access, you can access all the articles at their site for for seven days for $15. Now, scientific papers may age more rapidly than newspaper articles. But I am skeptical. The NYT just lost its premier role in links and moved itself way down the list of places to read. (I wonder if the writers get any of that $3 as a residual?) [A Man with a Ph.D. - Richard Gayle's Weblog]

Posted by DeLong at April 6, 2003 09:52 PM | TrackBack


I'd say they have. Reuters keeps their links, they'll get my traffic...and I'm sure others.

Posted by: Zack Lynch on April 6, 2003 10:26 PM

The Times would do much to better itself by firing fuckhead supreme Krugman. Anyone expecting a jeering "cakewalk" column this week?

Posted by: John on April 6, 2003 10:32 PM

Forget the effect on the Times--what about the effect on us poor Times writers who don't own the rights to our own articles? (When you're the New York Times, you have a lot of market power.)

Posted by: Virginia Postrel on April 6, 2003 10:39 PM

The New York Times has already squandered its editorial reputation. I personally cancelled my subscription only a few weeks ago. This decision, though, permits me to direct more tip money to bloggers like Instapundit and Virginia Postrel!

Posted by: David Thomson on April 6, 2003 11:44 PM

I wish some micropayment system would work. I think it could be reasonable to ask people to contribute a tiny sum/article.

Virginia's comment reminded me of the Tasini case (NYTimes vs Tasini) but I see: "Publications such as The New York Times now require permission for electronic republication of works by freelance authors, but this was not standard industry practice until fairly recently."
from: http://www.arl.org/info/frn/copy/taslib.html

Posted by: Eszter on April 6, 2003 11:52 PM

Virginia, what are you complaining about?
You took their money, you signed your rights away.

Wasn't it a free market transaction?

Posted by: Barry on April 7, 2003 05:50 AM

I had the distinctly untimely pleasure of speaking with one of the folks at The New York Times Digital a few weeks ago. This individual stressed to me the great business importance of the Times' "business to business" archive revenue stream (their cut of the per-seat licensing fees paid to Lexis-Nexis by corporate librarians and such). They do not wish to jeopardize that by giving the archives away for free on the web.

I suggested a web-based "subcription-to-the-archives" for which I, at least, would gladly pay. They no longer have me as a print subscriber, and that just seems awfully silly as I still read the "paper" as intently as I did when it was on paper -- so maybe that's a way to capture some value from me?

The response there was that I should just buy the 25-pack of archive access. They are very afraid of someone getting an archives "subscription" and then scraping the whole thing and posting it to the web for free -- a reasonable fear, I suppose.

So, I agree that they are shooting themselves in the foot and this will one day make a great Clayton Christensen case study. In the meanwhile, I am left wondering what will become of the business model underpinning the regular writers of the New York Times who I enjoy reading: Morgensen, Norris, Friedman, Ben Ratliff on jazz, and the food section. Their employer doesn't get a dime from me I'm afraid.


P.S.: The Times is actual very commendable in at least one respect -- they make it easy to email the full text of an article to oneself or to a friend (as opposed to just a link). I have taken to emailing myself interesting things and I am thereby building up my own personal archive of Times material....

Posted by: Motts McGregor on April 7, 2003 06:43 AM

The NYT site is already bad actor since they use a method for putting pop-under windows that circumvents Mozilla's preference setting. One could excuse this at first as due to ignorance or overeager marketing-- but to continue doing it is clearly policy.

Posted by: Matt on April 7, 2003 07:18 AM

What this means is that, as Motts is doing, you will simply create your own digital clippings file. If you had paid for a copy of the NYT, you could clip the paper version for yourself. Indeed, you could make photocopies to preserve it, in case the paper medium itself started to wear out.

This is in general a good idea anyway to prevent linkrot for articles you value: save the page yourself (at least, the text).

So now the NYT electronic archives will be essentially off-limits and uncheckable. Of course, microfiche or other physical records in public or university libraries will still exist, and be open for free. It seems odd that the paper of record is thus only a record on paper.

Posted by: Ethan on April 7, 2003 08:02 AM

Several other abservations:

If you make a habit if saving all of the interesting online articles you read as text files, they are quite small. That means:

A) you can have a very large archive that is rapidly searchable. Thus, a complete record of publications and information you have read can be indexed by computer - a massive expansion of your own memory and recall.

B) unformated text files aren't very large (3K per page or less) - you can take them on your PDA! All the good stuff you've read in the last year, whenever you want it, without even needing internet access.

C) you never worry about linkrot. If you need to get a fresh link, you can find it because you've got the original text and can search on Google for it.

D) if you need a hard copy, you can make one at any time. Digital information is still not very long-lived, compared to paper.

E) as formats and document standards change (or as websites go under), you can ensure that the documents you care about are always accesible without relying on others to maintain their archives. When finally Salon goes belly up, how good are your links going to be when the servers are all sold?

Even if you read or skimmed 200 pages a day, and wanted to save it all, you'd be archiving less than 1 MB per day.

Posted by: Ethan on April 7, 2003 08:25 AM

Interesting comments, Ethan. It implies that any given, reasonably popular source will have dozens to thousands of private archives. Not for profit (for long).

So the next wave in net-searching technology should be focused on such archives.

Will we soon see a peer-to-peer Gnutella equivalent for text articles? That would give accessibility to large bodies of work, while making legal measures more difficult.

Posted by: Barry on April 7, 2003 09:08 AM

Ethan is right!

We simply save the text file or print important articles. We also save the daily paper for several months. The Times is wonderful.

Posted by: jd on April 7, 2003 10:20 AM

Dear Brad

Please do remove the horrid comment about Paul Krugman.


Posted by: bill on April 7, 2003 10:21 AM

In response to Barry, of course it's a free market transaction. I calculated costs and benefits and decided I was better off with the contract they were offering. If their links disappear, however, I may recalculate.

Based on my experience writing for the Times and other papers before and after Tasini, that decision has had terrible effects.

The good news is that the Times apparently isn't pursuing this policy after all. All my links are still working fine, even the ones that are several years old.

Posted by: Virginia Postrel on April 7, 2003 11:12 AM


We do love the NYTimes and you!

Anne & Co.

Posted by: anne on April 7, 2003 11:36 AM

Looks like the Times Archive is back

Posted by: Kimon on April 7, 2003 12:26 PM

There is no problem with the NYTimes archive. A superb resource.

Posted by: lise on April 7, 2003 01:27 PM

Ethan, your model assumes that everything that one day may be interesting for us to read we will have read the day it was out for public view. But this method doesn't really work if you're doing research and an article which you never archived pops up as relevant and you don't have access to it b/c you hadn't seen it back when and didn't save a copy.

Posted by: Eszter on April 7, 2003 05:46 PM

So instead of linking to the article on the NYT's site, bloggers will quote as much as they think they can get away with and are willing to reproduce.

Research is the more serious issue. I've done hardcopy research in closed-stack library systems. I hate it. They dribble out three books at a time, and you have to take their catalog's word on everything. You get inferior results, and it takes you all day to do what you could have done in twenty minutes if you'd had access to the stacks.

Doing research via articles you have to pay for in advance is every bit as objectionable, for remarkably similar reasons. I've looked at the NYT archives. You can search on a word or phrase, but all it tells you is whether or not your search string appears in a given article. There's no context. "Wheat" could be a fashionable color, an allergy, part of a flower arrangment, part of an irrelevant tossed-off line, a recipe, an invidious comparison, a crop, a commodity, the name of a street, or the subject of a perfectly fascinating anecdote buried three-quarters of the way down the article.

The sample openings are inadequate. Here's as much as you can see about a piece by Erika Kinetz, called "Muffled Voices", that appeared on March 23rd:

"The phone lines to Baghdad were jammed last week. But not much ever got said over these wires, which have been more a tool of surveillance than of communication since Saddam Hussein's Baath Party took power in 1968. Conversations between Iraqis in New York and Iraqis in Baghdad have typically gone something like this:

"New York: 'How are you?'"

That's it. That's all. One has to assume that further themes are introduced and developed in the course of what the NYT archive says is a 2980-word article.

And of course, there's no guarantee that an article that started out on one subject might not switch to another a few paragraphs later. Likewise, the fact that an article's opening paragraphs don't seem to have anything to do with your subject is no guarantee that it doesn't switch to it later on.

Three bucks a pop is too much to pay for that much uncertainty. It would also be too much to pay if the search process were less uncertain. Newspapers are raw data. It can take a lot of news stories to establish one point. Law firms and political consultants can afford those prices, but many researchers can't.

If the New York Times wants to get out from under the burden of being the newspaper of record, this is a good way to do it.

I'm not terribly impressed by their fear that someone might copy off their entire archive and post it to the web. That's been happening to novels for years. When it does, we sic our legal department on them.

Anything that can be read can be pirated. We have laws against such piracy, but we haven't been in the habit of enforcing them, because the convenient difficulty of reproducing text made small-scale piracy more trouble than it was worth. Yes, you could type out a copy of a book; but even laying aside the cost of your own time and labor, the price of the typing paper and typewriter ribbons would probably come to more than you'd pay for a printed copy of the work. The same goes for newspapers; not to mention they'd be out of date by the time you finished typing. Law enforcement was there to stop gross plagiarism and full-scale commercial piracy. The natural refractoriness of the physical universe took care of the rest.

Microfilm changed that a little, and copiers changed it a lot. Same goes for the advent of the internet. Advances in scanning and OCR technology will change it even more. In the long run, strategies that attempt to enforce copyright laws by purely technological means are going to fail. The only document that can't be pirated is one your customers can't read.

Posted by: Teresa Nielsen Hayden on April 8, 2003 12:23 PM

Folks, you're all wrong. This move by the NYT is the absolute right thing to do.

First, assume I'm wrong. Let's say their page views go into a death spiral and their ad revenue plummets. Guess what? They can just reverse the policy. It will be like bringing back Classic Coke. There's not a lot of risk here.

Second, this has a huge potential upside. Most people do not care about articles over 30 days old. They will continue to read the current NYT. Remember the NYT is a newspaper. It's a journal of what is current. The archives are useful primarily to researchers.

Third, for researchers and other interested parties this is just about the best policy we can hope for until someone reinvents sliced bread for the Internet.

Let's start with the URL (Internet address) used for NYT articles. The NYT URL design is the best one out there. From the URL, you can figure out the day and section of an article. For example:


This is an opinion article that is in NYT of 9 April 2003. The URL will never change. (Or at least it better not.)

With the URL alone it would be easy to get the article from microfilm, if you ever have to do that.

Unfortunately, the actual page number in the newspaper is not provided, as on the Washington Post's web site. Small loss, though.

Now let's say you run into one of those "expired" articles. A great thing happens. You still get a summary of the article. That may be enough for many purposes. If you need the whole article, you can actually get it by paying for it. That's because at the NYT site, the URL does not expire. The URL is permament. As long as you have an URL, you can buy the article. I don't believe the same holds true with the Washington Post. To get one of their archived articles, you'll need more info than just the URL, no matter how much money you give them. You'll have to go hunting through stacks and stacks of unwieldy search results just to find one stupid article. The NYT solution is the elegant one.

If you need a lot of old NYT articles, then you should buy a Nexis account or an NYT subscription, or head to your local library (it's the building with all the books inside, remember?).

Now, is this perfect? No. Ideally there would be such a thing as a free lunch and unlimited information on the net for free, but this is the real world, and the NYT believes they can make money in this business. If they can do so, we will all be better off for it. The NYT presence on the web will continue in force, and other sites might also gain profitability.

So let's cheer on the NYT for doing the right thing when it comes to URLs and archives. If they just added page number references inside the HTML code, it would be perfect.

The last point is the NYT credibility. A lot of people are angry with them for their leftward lurch, but they haven't gone as far as the Baghdad Broadcasting Corporation. The NYT is still a very strong paper. They need to hire some more good national security correspondents, and make some other changes, I think, but otherwise, they're still top notch. I would be the first to say that the Wa Po has had better coverage of the war.

Posted by: Andrew Hagen on April 8, 2003 11:54 PM
Post a comment