edocr

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

~ the eulogy ~

edocr did one thing. You uploaded a document, it rendered the pages in a browser without asking anyone to download anything, and it handed you a URL you could paste into an email. In 2007 that was not a trivial trick. PDFs lived as attachments or as links that fired up a plugin and hung the tab, and a service that turned a file into a readable web page solved a problem almost every small business had without knowing how to describe it.

What turned it from a hobby into a business was lead capture. A company could put its catalog or its white paper on edocr, leave it open to the world, and find out who had read it. That pitch aimed at marketing departments rather than at readers, which is why edocr never got the consumer recognition Scribd did while quietly serving documents into something close to two hundred countries.

Manoj Ranaweera ran it out of Manchester and bootstrapped the whole thing. There was no funding round and no board, which meant there was no burn rate to outrun either. In March 2015 Accusoft, a document-imaging company in Tampa, bought it for cash on undisclosed terms, and Ranaweera did not go with it.

I never had an account. What I had, repeatedly, was the experience of searching for a datasheet or a manual in the late 2000s and landing on a site exactly like this one, with somebody else's uploaded PDF rendered in a viewer and a sign-up prompt in the corner. edocr sat in the small rotation of addresses search kept handing me. You read the page, you closed the tab, and you never thought about who was running it.

~ the dossier ~

~ bootstrapped out of manchester ~

The bootstrapping explains the shape of everything that came after. Scribd raised money and chased scale. Docstoc raised money and got bought by Intuit, then closed in 2015. edocr did neither, which meant it survived both of them by a decade and arrived at its own acquisition with nobody to answer to except the buyer.

~ what accusoft did with it ~

Accusoft's plan, stated at the time by chief executive Jack Berlin, was to flip edocr from paid to free and use it as a shop window for Accusoft Cloud Services. Free accounts got a thousand documents, the lead data, the analytics. The viewer underneath was swapped for Accusoft's own rendering technology. On paper that is a sensible use for a small acquisition.

In practice it meant edocr stopped being anybody's priority. A free document host with no moderation budget attracts exactly one kind of uploader, and the archive had filled up with it well before the 2020s were half over. Crawl the catalog in late 2025 and the document titles read like a spam folder, search-engine bait for sleeping pills sitting next to printer support hotlines. Eighteen years of corporate catalogs and university handouts sat underneath, undeleted and unfindable.

That is the quiet way a document host dies. Nobody shuts it down for being spammy. It just becomes a cost line that returns nothing, and some year the owner looks at the line.

~ what was left to crawl ~

Ranaweera posted about it on his own blog on 25 September 2025 under the heading that it was a sad day, noting that edocr.com would be shutting down permanently on 29 October. He had not owned it for ten years. He wrote the obituary anyway, which tells you something about what building a thing for eight years does to you.

The Internet Archive's record matches the announcement almost to the day. Individual document pages were already returning 404 through the middle of October 2025 while the category listings still answered, so the library was being taken apart from the inside before the front door closed. The last capture of the home page is 25 October 2025, 994 bytes. After the 29th the record simply stops, since nothing remained on the other end to crawl.

@}-->--

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

« previous  ·  random  ·  next »