Unix Version 5, Release 3:
Anyone found a workaround for the erroneous behaviour when reading
.S.chapter, which does not attempt to read the chapter $HOME/S (as in
the example in Programming with Data) but instead attempts (and fails)
to attach $SHOME/library/$HOME/S - clearly rubbish. If the file was
called .S.library I might perhaps find this excusable.
Also, why is failing to find my intended every-session chapters such a
disaster? The subsequent appearance of a prompt offers false hope: not
enough of S is loaded to actually do anything in the session. It might
as well generate a fatal error. (An indication of what's causing the
error might be more than useful too -- suppose I've moved a library but
forgotten to update .S.chapter; I could spend hours looking for the
problem ... )
More importantly, why are we idiots paying a grand or two for software
that is so much less than fully ready for release? Why do we not get
paid for our beta testing? How dare a company release a product with an
actual admission/warning that it will not even do what it says in the
manual (Programming with Data) !!?? Why does it have LESS functionality
than previous releases (I'm thinking of the absence of help.start)?
If I didn't have to try to support other users I would only ever use R
now, on principle.
Simon Fear
Statistician
Roy Castle ICLCR
Liverpool L3 9TA, UK
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news
|