Slashdot Mirror
New York Times Wipes Journalist's Online Corpus
thefickler writes "Reading about Peter Wayner and his problems with book piracy reminded me of another writer, Thomas Crampton, who has the opposite problem — a lot of his work has been wiped from the Internet. Thomas Crampton has worked for the New York Times (NYT) and the International Herald Tribune (IHT) for about a decade, but when the websites of the two newspapers were merged two months ago, a lot of Crampton's work disappeared into the ether. Links to the old stories are simply hitting generic pages. Crampton wrote a letter to Arthur Sulzberger, the publisher of the NYT, pleading for his work to be put back online. The hilarious part: according to one analysis, the NYT is throwing away at least $100,000 for every month that the links remain broken."
the NYT is throwing away at least $100,000 for every month that the links remain broken."
now how much would it cost to fix all those links...
no wonder newspapers are not doing well
Its not my fault, someone put a wall in my way.
Yet another reason why locking up content is wrong. Let it be freely copied, and then ANYONE who finds the work valuable can potentially become a caretaker of the work and keep it accessible online. Then the only way a work would disappear is if nobody has interest or time to preserve it.
And it's got unlimited space. Strangely enough, some people are adamant about keeping their works out of this library. And I say they have the right to insure the internet forgets about them when they die. This poor soul seems to understand what's going on.
My work here is dung.
Whenever I redesign my site, I try hard to avoid changing and URLs. But if I do have to change a URL, I always make sure that there is a redirect (preferably a HTTP/301 permanent redirect) that points from the old URL to the new URL. Updating links is not enough, because you will always have links that come from external sites that you don't control, user bookmarks, links found in "Hey, check this article out" e-mails, etc.
This is one of those basic principles of the web that the W3C (and for those who don't pay attention to them, you can substitute that with "plain old common sense" here) strongly recommends.
It means that users can always find and view content. It means that you still retain your ad revenue. It means that you still keep your PageRank for external sites that link. It means less bitrot and a more useful web...
I feel for the guy and his lost articles, but I am wondering why he did not keep backups of everything? The stories seem to be gone forever, or else his letter would be about to re-publishing. his stories on his own website.... That is a rather bad case of negligence on the publisher's side , but more so on the part of Mr. Crampton. For comparison: I work with a professional fotojournalist and this guy has been working for 50 years now and has archived everything (more than 1.5 million pictures) like a mad squirrel. If you ask him about an article he wrote in 1961, it takes him about five minutes to find a copy of the article and the raw materials. Everything analog but nonetheless... That makes you wonder if -while embracing digital media and the blogosphere - many journalists have not brought with them the necessary tools to manage and archive their digital assets.
In the digital age, wiping out thousands of volumes of material takes mere seconds. Permanently. Gone. Poof.
We have books, printed books, which go back hundreds and hundreds of years (well, written material; the printing press is a fairly recent invention).
We don't even have a record of some newspaper articles that came out 5 years ago. We're LOSING our history, not retaining it, because we lack sufficient "printing" to always keep a copy in circulation. Witness the Avism.com debacle and hundreds of other cases where this has happened.
Until we can have a hard-copy of digital media which can NOT be changed, edited, altered or redacted... we're lost.
When we all have "Kindle DX2" devices in the classroom for digital copies of our textbooks... what is stopping them from "gently changing" some of the wording over time, over a few years, to permanently alter the way our youth views the history of times they never lived through?
How can you compare one version of a website today, with the one that was there last week? Was anything changed? Was article content "censored" in any subtle way?
We're heading down a very slippery slope, when digital information can't remain static enough to hold through the years, and be validated and verified to be unchanged, with sufficient copies in enough hands, to ensure survivability. The Internet is not the place to "store" things you want to keep for years and decades.
My company links to articles on a lot of magazine websites, and I'm just amazed at how often the links become broken. Sites get redesigned and they don't bother redirecting the old URLs to the corresponding new locations. Or, even worse, they just discard all of the old articles, or random articles disappear or come up blank or mangled. Does it not occur to them that websites, search engines, and blogs are left with broken links? Do they not realize that people bookmark the articles?
It's hard to tell from the linked article (yeah, I read it) but it doesn't seem like Crampton has no copies of the articles (surely he would keep of his own stuff) but that they're just not accessible on the Internet. All the links that should point to them from the NYT and the IHT went kablammo when the two sites merged.
There's no way a back up on his end could fix this problem.
"There is no time, sir, at which ties do not matter," Jeeves, (Jeeves and the Impending Doom)
You do realize that it's sample code and thus is used ot illustrate the API in question? As it's illustrative only, it will be missing a lot of essential code in the name of clarity. Stuff like error handling (the docs will tell you what it returns and how it returns it), parameter/return checks on associated API calls, or even input checks. After all, most people want to see how to use the API in a few lines of code, not deal with a 1000-line program because the author decided to check every return value (even the ones to printf()) and abort gracefully in every potential instance. That's not sample code, and extracting the "how do I use this API?" information from it is quite difficult because of the extraneous code.
It's assumed a halfway competent programmer would realize that, and use the API properly with proper error checking and input sanitization. Alas, that isn't the case most of the time, and you'll find the sample code copied-and-pasted into production code by codemonkeys who don't appear to think. If I was a particularly vicious developer at Microsoft, I might code the sample code with known security holes (but any halfway decent programmer would fix since it would be obvious) and then check applications for those holes later...
Comment removed based on user account deletion