> On 2004-07-23, rpriceis@hotmail.com <rpriceis@hotmail.com> wrote:
> > Thanks Aaron,
> >
> > Clearly I have messed the installation up. I think starting afresh is best
> > option also.
> >
John Duncan <agley@optusnet.com.au> wrote:
> Just a thought but do you have the C-shell ( or tcsh ) and
> the Xfree86-dev rpm. I think that poplink_cmnd needs a C-shell
> and the headers etc. from the Xfree86 development libraries
> may be needed for linking.
>
Yes, on making an additional instalation from *.tar [I had lost
the *.tar.gz] on a separate partition, to investigate RCLIB and
the SIM_AGENT, I found that I had to AGAIN hack-around
before I could get poplog/xved to work - on Mandrake 9.
Again, the [unnecessary IMO] problem was the same as some years
back when initially installing under RH6.2 [still used]: the C-shell.
Isn't bash the standard for linux ?!
We must avoid band-aid patches and 'build universally applicable
methods':-
While reading:
http://www.cs.bham.ac.uk/research/poplog/packages/simagent.html
* I was again reminded that I'm trying to emulate A. Sloman's
documentation style/format [if I lived long enough].
* there seems to be ONE typo near the file end:-
suggestions from users), while every effort is maded to retain...
* I'm hoping that RCLIB and the SIM_AGENT can shed some light on
'usability': i.e. where only the valid choices are selectable at each
state of the user's session, instead of pretending to communicate
with 'some one in side the box by "composing" a dialog'.
i.e the menu/spread-sheet paradigm vs. the author/poet.
Running poplog-ved after months of other intense work, I found
AGAIN that I would need to search for my notes to find out how to
'zoom' the double window to one. These are IMO the reasons why
poplog doesn't move out of 'home base' to general useage.
By contrast if you/I want to use a spread-sheet after years of
familiarity-run-down, we only have to remember how to start it.
* As an admitted reductionist, I wonder why the HCI specialists don't
admit that:
1. use recognition instead of memory,
2. base the system on a small number of orthogonal principles;
are proven succesfull methods.
== Chris Glur.
PS.
> > I could format my hardrive and re-install everything choosing to include
> > everything possible. Although drastic it may solve the problem!
>
>No -- that's far too drastic. Adding tcsh from the rpm file should be a
>matter of seconds.
>
Again the band-aid/short-term view is bad:
you can't do a start-from-scratch every time [in life or] in computing;
successive refinement, by small steps, and moving back to the last
'good' state is a proven/winning method.
If you don't learn what you did wrong, you won't be able to fix
the many other problems that await you.
It's never 'right', just better managed chaos.
|