[R] Broken links on CRAN

[(Ted Harding)]

On 05-Dec-05 Doran, Harold wrote:
> Sorry, didn't think about that. The mirror I used was
> 
> http://lib.stat.cmu.edu/R/CRAN/
> 
> I checked other mirrors and they did work fine. 

Mysterious! I've just been to that mirror, and tried the PDF files
from a few of the packages chosen "randomly" (don't ask ... ),
namely akima, ash, CircStats, dyn, Api and geoR, in each case
sinply clicking on the respective "*.pdf" link, without once
seeing the "damaged file" message.

Since you had trouble only with that mirror, it can't be put
down simply to a Reader problem on your local machine since that
should show the problem on all mirrors.

The indicated conclusion is perhaps that there is something about
what happens when *your* machine tries to download from *that*
mirror which leads to file corruption during the transfer.

As a further diagnostic test, can I suggest that (maybe just you
and me, Harold, unless anyone else feels like joining in):

1. You find a PDF file on that mirror which gives the problem
   when you click on it to open it in Acrobat Reader. This should
   preferably be one of the smaller ones (like "ash.pdf" though
   of course ensure that it exhibits the problem).

2. Having verified that it gives the problem as in (1), do whatever
   it is you do in Explorer to download the file to your own machine.

3. Having done that, mail the copy *stored on your machine* to me
   so that I can compare it with the copy I download myself from
   the same source. *Don't* use any resource which your browser
   may offer for you to email the link!

4. Assuming, of course, that no-one else has found the definitive 
   solution in the meantime!

Best wishes,
Ted.


--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding at nessie.mcc.ac.uk>
Fax-to-email: +44 (0)870 094 0861
Date: 05-Dec-05                                       Time: 16:20:19
------------------------------ XFMail ------------------------------