NewsIsFree

» see newsisfree.com as it lived, on the wayback machine

~ the last words ~

I was online for every single year this thing was running and I never once heard its name. NewsIsFree started in 1999, when I was a year into dial-up and still counting kilobytes, and it was still indexing in January 2018, by which point my job was machine vision work in Ankara. Eighteen years of overlap and not one point of contact. That turns out to be the site's own complaint about itself, which I will get to.

What it was: a news portal built out of syndication. You made a free account and picked your sources. What came back was pages of headlines that refreshed themselves. In July 2001 it was pulling from 1,651 sites. By the end it carried 12,609. The about page promised updates as often as every fifteen minutes, with every single source touched at least once a day, across more than 25 languages.

Two people ran it. The 2001 about page names them and describes the operation with a candor that has not aged a day: "NewsIsFree is run by Mike Krus and Pino Calzo. It is a voluntary effort supported by banner advertising (as if that could still support anything), and voluntary donations." It credits a site called Geekboys as the original inspiration, already gone by the time that sentence was written, and names Userland and a news service called Moreover as the places it took ideas from. A commercial arm, NewsKnowledge, sold custom feeds to companies. That is presumably what paid for eighteen years of servers.

~ the record ~

~ how it made the feeds ~

The claim that matters sits in one line of the about page: "Sources are gathered either from their syndication feed or by analysing the actual web pages." In 2001 most news sites had no RSS whatsoever, so NewsIsFree read their HTML and published the feed on their behalf. It was manufacturing syndication for publishers who had never agreed to it and in a lot of cases had no idea. That is what separates it from a directory like Syndic8, which listed feeds other people had already made.

It also kept building things nobody remembers it for. In 2006 it shipped an AJAX headline reader and source browser, back when AJAX was barely a year old as a word, along with a river-of-news page called "What people are reading" and personalized NewsMaps that the site news pitched as an industry first. Mobile and SMS alerts sat behind a premium tier. There was an XML-RPC interface for firing a headline straight into a Manila or Blogger weblog. That interface is how Mike Krus ended up on Dave Winer's syndication mailing list in March 2001, patiently debugging a double-escaping bug in his own PHP.

~ the last counter ~

On the last day it was ever archived alive, the line across the bottom of the homepage read "271411 news added last 24 hours, across 12609 sources and 364 categories." The copyright underneath said 1999-2018 NewsKnowledge. Nothing on the page suggests anyone had decided to stop.

~ the hole ~

The Internet Archive had been collecting that homepage roughly once a week for years, usually two captures at a time, one for the apex and one for the www host. The last of them is 31 January 2018 at 09:28 UTC. Then the whole domain goes silent until 19 July, every host and every path. A crawler that comes back every week for five and a half months and records nothing is not finding a page that failed to change, it is finding no server.

What eventually answers, on 23 July 2018, is a brand new WordPress install in French with a single post two days old, titled "Bonjour tout le monde !". Within a week the domain was an affiliate site comparing VPNs and web hosting, and that is what it stayed for years. Nobody bought the brand. The registration lapsed and somebody took the name.

~ what sits there now ~

The squatter is gone too. newsisfree.com no longer resolves on either scheme, and newsknowledge.com, the commercial half, forwards to an unrelated press-release marketplace. What is left is the archive copies and one line the site wrote about itself in March 2006, after TechCrunch ran a roundup of online feed readers and did not include it: "TechCrunch reviews online readers and forgets NewsIsFree (as most do)."

@}-->--

              )
             (_)
            .-'-.
            |   |
            |   |
            |   |
            |   |
          __|   |__   .-.
       .-'  |   |  `-:   :
      :     `---'     :-'
       `-._       _.-'
           '""""""

« previous  ·  random  ·  next »