[tex-live] TeX Live on usb drive

[Manuel Pégourié-Gonnard]

Philip TAYLOR a écrit :
> No, nor did Robin suggest that it was CPU usage
> that concerned him; he specifically referred to
> the file size increasing as a result of zipping.
> 
> But I would genuinely have thought that he (and
> Cambridge) would have preferred one 1Gb transfer
> to 5890 piddly little ones.  What is the (CPU or
> otherwise) overhead of 5890 FTP transfers compared
> to one ?
> 
I don't know, but I guess Robin as more experience than me in the administration
of a CTAN node :-)

Anyway, I agree that from the user's point of view, downloading one large zip
file is less likely to cause problems, and if it can be faster, I'm sure a lot
of users will appreciate. The question is, is it reasonable from the server's
point of view? On this point I have no idea.

> I did not suggest that even in passing; I merely told
> Jerónimo how I had tested my hypothesis that on-the-
> fly zipping might be possible.

Ok, I didn't understand it was a proof of concept.

>  I don't think any
> reasonable person would have expected him (or anyone
> else) to use "ftp.tex.ac.uk" just because I cited it
> as an example.  Should we all start using "Ctan.Example.Org"
> in future correspondence ?
> 
Well, given how random posts are presented out of any context by google, better
be careful :-) My point was I'm not sure many mirrors besides the main nodes do
provide this feature. If there are enough of them, ok. If not...

Manuel.