In this (very good) answer, the answerer linked to a PDF that is on a dropbox site:

Overhauling Shimano Dynamo Hubs

There's no problem with the answer itself, but in comments/discussion, we noted that it'd be best if it were hosted on a more stable site (perhaps like the images on SE are hosted on imgur).


| |
  • 1
    The issue there is that the linking user doesn't want to mirror the content, but we really have to for stability. The issue is that pdf is a final output format, so converting it to something more usable is generally difficult. Screenshots would work here, but they're a poor solution. Annoyingly, this is an archive of a broken web page, but at least the text inside is copyable. Ideally someone would re-create the web page in that answer. – Móż Jan 14 '16 at 3:07
  • Ideally someone would re-create the web page in that answer (which isn't hard, but the user might revert it since they explicitly want not to do that). – Móż Jan 14 '16 at 3:12
  • 2
    Copyright comes into it too - trinity bicycles seems to be gone, or at least their web site has left the internet. The content sans pictures is available here web.archive.org/web/20090107023114/http://trinitybicycles.com/… but the pictures are pretty important. I'm tempted to edit the answer and copy/paste the entire relevant content of the PDF, so that its searchable. And provide attribution to the original sources of course. This is a specific answer about that PDF. Would a Shimano tech ref page be any different? I think not. – Criggie Jan 14 '16 at 4:03
  • 1
    I've got halfway through but can't do the images yet - will finish it off in 6 hours. – Criggie Jan 15 '16 at 0:33
  • Completed - That's a lot of work. If the PDFs were hosted like images, would the search engine find text in them? – Criggie Jan 15 '16 at 7:03
  • PDF is not specific to bicycle. Meta.stackexchange.com is a more appropriated site for this question. Not single SE supports posting of PDF. There would be a myriad of problems. Yes pretty much everyone has a PDF viewing application installed but it is not guaranteed. You cannot even use HTML in any SE site. Some sites support Latex or Math Jax but that is a much simpler and stable. – paparazzo Jan 18 '16 at 20:22

I guess my actions on this would speak as an answer?

Answer If the content is worth saving, then copy/paste the relevant text and then insert the images inline. Full accreditation of the source should be given as well as a link to where the PDF came from, regardless of whether the PDF link is stale or working.

Method Details I copied and pasted the text into the answer, and then edited it to remove the repeating header/footer.

To get the images I opened the PDF in GIMP and imported (rasterised) to a resolution of 150 DPI. Then I manually selected each relevant image, cut it out and pasted it into a new file. An Autocrop removed extra whitespace, and then the image was saved as a PNG.

One image was blurry, a "sharpen" helped a little. Another image had jpegged text in the image, so I cleared that and retyped it. At least once, two images were taken together because they were side by side.

Once that was done I inserted each images in the correct place in the text. I was careful to keep all the pictures and text in the same relative position and the only thing to change would be text wrapping at the end of each line.

| |

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .