4 ms·

Getting completely off topic here, but I found it interesting that there were only 27 actual PDF files in the first 10 pages of search results using "allinurl:.

Getting completely off topic here, but I found it interesting that there were only 27 actual PDF files in the first 10 pages of search results using "allinurl:.pdf". So I tried using "filetype:pdf", and none of the first 10 pages was an actual PDF file. So then I tried: https://www.google.com/search?q=filetype:pdf+allinurl:.pdf https://www.google.com/search?q=filetype:pdf+allinurl:.pdf ...and only got 34,900 results. What?

grkvlt 10y ago
That query searches for files of type PDF that contain the text "allinurl:.pdf" because 'allinurl' appears not to work any more. The following query, however, generates 1,410 million results, all of which appear to be PDF type files, and the URLs contain '.pdf' https://www.google.co.uk/search?q=filetype:pdf+inurl:.pdf https://www.google.co.uk/search?q=filetype:pdf+inurl:.pdf It's an interestingly eclectic bunch of results. I particularly liked the 'International Chronostratigraphic Chart' I found on the second page, immediately followed by 'A Visual Guide To Achieving LinkedIn Profile Perfection' ... http://www.stratigraphy.org/icschart/chronostratchart2015-01.pdf http://www.stratigraphy.org/icschart/chronostratchart2015-01...
- GregBuchholz 10y ago
  I can't seem to get beyond 59 pages of results. Is this the expected result? https://www.google.com/search?q=filetype:pdf+inurl:.pdf&biw=1072&bih=598&noj=1&ei=UhYAWKiCHsv0vATSsYboDg&start=580&sa=N&filter=0 https://www.google.com/search?q=filetype:pdf+inurl:.pdf&biw=...
  - grkvlt 10y ago
    Yeah, I found that, too. Sad... It seems to be what happens with any Google search that returns millions of pages; I think the total is an estimate, and they don't actually compute all the results, to save CPU or something? Maybe using the 'advanced query' page, or the programmatic search API would allow paging through the entire result set?
  - saalweachter 10y ago
    "Why isn't my search engine an efficient file browser?" Google isn't really a library; you can't just walk up and down the aisles until something pops out at you. Which raises the question: What is the equivalent on the internet? If I want to browse random books until something pops out at me, where do I go on the internet?
    - bitJericho 10y ago
      Piratebay?
  - barrkel 10y ago
    It's the expected result for Google, yes; they do not support full pagination over their entire estimated result set. To get more results, you need to narrow your query. (Did you only just find this out now? I'm finding it hard to take you seriously.)
    - GregBuchholz 10y ago
      That sounds like an interesting AskHN: Have you ever clicked beyond the second page of search results?
      - grkvlt 10y ago
        It's a standard research technique for me - find a useful query and go through N pages (2 < N < 20) opening links that look interesting in background tabs, then review the tabs for useful content. I had actually forgotten about Google's limit on rendered result pages; most queries I have are specific enough that the total number of pages is less than that, often only single digits.
    - obmelvin 10y ago
      Why do you think it's weird someone hasn't tried to visit the 60th page of search results before?

grkvlt 10y ago

That query searches for files of type PDF that contain the text "allinurl:.pdf" because 'allinurl' appears not to work any more. The following query, however, generates 1,410 million results, all of which appear to be PDF type files, and the URLs contain '.pdf' https://www.google.co.uk/search?q=filetype:pdf+inurl:.pdf https://www.google.co.uk/search?q=filetype:pdf+inurl:.pdf It's an interestingly eclectic bunch of results. I particularly liked the 'International Chronostratigraphic Chart' I found on the second page, immediately followed by 'A Visual Guide To Achieving LinkedIn Profile Perfection' ... http://www.stratigraphy.org/icschart/chronostratchart2015-01.pdf http://www.stratigraphy.org/icschart/chronostratchart2015-01...

GregBuchholz 10y ago
I can't seem to get beyond 59 pages of results. Is this the expected result? https://www.google.com/search?q=filetype:pdf+inurl:.pdf&biw=1072&bih=598&noj=1&ei=UhYAWKiCHsv0vATSsYboDg&start=580&sa=N&filter=0 https://www.google.com/search?q=filetype:pdf+inurl:.pdf&biw=...
- grkvlt 10y ago
  Yeah, I found that, too. Sad... It seems to be what happens with any Google search that returns millions of pages; I think the total is an estimate, and they don't actually compute all the results, to save CPU or something? Maybe using the 'advanced query' page, or the programmatic search API would allow paging through the entire result set?
- saalweachter 10y ago
  "Why isn't my search engine an efficient file browser?" Google isn't really a library; you can't just walk up and down the aisles until something pops out at you. Which raises the question: What is the equivalent on the internet? If I want to browse random books until something pops out at me, where do I go on the internet?
  - bitJericho 10y ago
    Piratebay?
- barrkel 10y ago
  It's the expected result for Google, yes; they do not support full pagination over their entire estimated result set. To get more results, you need to narrow your query. (Did you only just find this out now? I'm finding it hard to take you seriously.)
  - GregBuchholz 10y ago
    That sounds like an interesting AskHN: Have you ever clicked beyond the second page of search results?
    - grkvlt 10y ago
      It's a standard research technique for me - find a useful query and go through N pages (2 < N < 20) opening links that look interesting in background tabs, then review the tabs for useful content. I had actually forgotten about Google's limit on rendered result pages; most queries I have are specific enough that the total number of pages is less than that, often only single digits.
  - obmelvin 10y ago
    Why do you think it's weird someone hasn't tried to visit the 60th page of search results before?