R-alpha: Re: Separation of Program & Doc ...

[Paul Gilbert]

>    PaulG> Also, for users always connected to the Internet, it is not
>    PaulG> really clear that they always need and should be forced to
have
 >   PaulG> the help ...
>This maybe practical in a future internet (or in today's net within
>North-America?) where bandwidth of small files
>like these is practially zero.   Today, even getting only a thousand
bytes can
>take many seconds, which is too long.

I am familiar with the low bandwidth problem and I appreciate that lots
of people will want to install help files locally. For many people in
North America the bandwidth is not so low anymore, and I am simply
pointing out that the logic of forcing everyone to install help locally
assumes a set of circumstances which may not exist for everyone.
(Although, for convenience I can certainly appreciate that doing things
one way may be a lot simpler.)

On a related, but slightly longer term issue, it is worth at least
thinking about the possibility that some code libraries could be loaded
on the fly over a high speed Internet. This is a small bandwidth
requirement compared to video, which seems to be coming more quickly
than most people thought. The advantage of this is that development can
be done more easily, continuously and quickly.  As long as the installed
base portion remains stable, newer libraries could be put in place
simply by pointing at them over the network rather than doing an
"install" in the fashion with which we are familiar.

All this is simply to point out that install strategies should probably
be considered to be for the short term rather than the long term.

Paul

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=