Thanks yourself – and a web metrics question

Apologies: This is a blind post out of necessity–but it’s not at all controversial.

I published the final 2006 issue of Cites & Insights Wednesday evening–a few days earlier than I’d been planning, but it was as done as it was going to be.

As with many of the issues these days, I wasn’t sure it had the “right mix” of stuff–but since the highest readership and greatest impact pretty consistently comes from big chunky essays, I didn’t worry too much about it. Some issues flourish, some just sit there, some move along slowly over time.

Yesterday morning, I had two pieces of email, each one thanking me for something in the new issue. I was touched; those two emails, both distinctly personal (which is why this is a blind item), were quite enough to make me happy about the issue as a whole.

This morning, I had a third piece of email thanking me for a third piece of the issue.

I responded privately in each case. Publicly, I’ll just say “You’re welcome, and thanks yourselves–you’re letting me know this is worth doing.”

That’s it. Nothing terribly important (except to me). Call it a six-day-early bit of Thanksgiving.

Oh, and a question for web metrics gurus out there:

Urchin (the stats package LISHost uses) tells me that, for the four months now that C&I has been on its own domain (on LISHost), the first issue published there–the “Great Middle” issue–shows up as:

  • A requested page 21,066 times (through yesterday)
  • A downloaded file 3,352 times (through yesterday)

Suggestions as to how you request a PDF without downloading it? What should I consider the readership of that issue to be–an astounding 21,000 or a great (but not astounding) 3,352?

Advice welcome.

Update 11/20: Blake answered the direct metrics question, but it turns out the answer is “neither.” I’d forgotten the other piece, v6i10a.htm–the HTML version of the essay. So the “direct readership” through 11/17 turns out to be just over 5,000 (adding the HTML copies), which is great.

4 Responses to “Thanks yourself – and a web metrics question”

  1. Blake says:

    Why are Page Views higher than Downloads for the same file?

    It is common for your page views reports to include a higher number of hits than your downloads reports for the same file. For example, many customers report to us that specific PDF files display more hits under the Requested Pages report than they do in the Downloads report.

    Here’s why:

    In the case of a PDF file, many browsers include a plug-in that displays PDF files one page at a time. The initial page request results in a 200 status code from the web server. All subsequent pages result in a 206 (partial download) status code since the browser plugin only downloads the page you are viewing and not the entire document. So, in the case where you view 5 PDF pages from the same document, the web server will record one hit with a 200 status code and 4 hits with a 206 status code, all logged as the same PDF document.

    Here’s how Urchin Reports interpret and display this information:

    Downloads: A download hit requires a status code of 200, 302 or 304, so the report will show only one download for this PDF file – e.g. the single hit with the 200 status code.

    Requested Pages: The report will show all successful file requests where the status code is 2XX, 302 or 304 and where the file is matches the proper MIME type to be considered a page view. By default, Urchin 5 treats all MIME types as page views except for files with these suffixes: gif,jpg,jpeg,png,js,css,cur,ico,ida.

    So, the 5 PDF requests in the example above would all be counted as hits to this document in the Requested Pages report page views, since each hit contained either a 200 or 206 status code.

    The choice was made to count each partial download hit (206 status) as a discrete page view in the Requested Pages report since each hit more or less corresponds to a single page within a multi-page PDF document. This is consistent with the philosophy of the Requested Pages report representing actual page impressions seen by a visitor.

  2. walt says:

    So the answer is 3552, some portion of which involved page-by-page reading. Fascinating. Now I know!

    (And, Blake, I do still check spam before it goes away…although I can’t for the life of me imagine why this would be flagged as spam!)

  3. Iris says:

    It was so interesting to read about your writing habits in this C&I. You’re right, they wouldn’t work for me, but it’s the way I always think *should* work for me. 🙂 Me, I sit sideways on the couch with my laptop on my lap, my cat on my legs, and listening to music or watching TV. It would drive you batty. A good cup of tea in the evening or coffee in the morning never hurts, either.

    But I think I need to get a little bit more deliberate about my “other” writing and research, and setting aside regular time will help with that. I also find that I have to do that kind of writing sitting at the table (I don’t have a study). I don’t find that writing fun, so any distraction (other than music) is too much. What’s more, because it isn’t fun, if I don’t plan time for it I always end up filling my time with activities that are more interesting. I’m just so tired of clicking through web sites, collecting their metadata, and entering that into my growing database. And the analysis me and my co-author do isn’t interesting enough to make two years of boring data collection seem worth it. Ah well. The sooner I get started the sooner I’ll finish, and then I’ll say “no” to any invitations to continue on with similar research.

  4. walt says:

    As noted in the update to the post, “So the answer is 3552” is wrong. “The answer” for direct readership is just over 5,000. No complaints here.