[Info-vax] BASIC compiler in the hobbyist distribution

[Paul Sture]

On 2015-05-16, John Reagan <xyzzy1959 at gmail.com> wrote:
> On Saturday, May 16, 2015 at 10:04:21 AM UTC-4, Phillip Helbig
(undress to reply) wrote:

>> Has anyone noticed that the ZIP file for the BASIC compiler contains 
>> BASIC017_IGUIDE.TXT which, however, is actually
>> 
>>                     DECnet-Plus_for_OpenVMS_____________________________
>>                     Installation and Basic Configuration
>> 
>> OK, "Basic" configuration.  :-|
>
> LOL.  I'll double check my files to make sure that isn't checked-in.
>
>
>> 
>> However, if one answers "1", then a) one is asked about some others and
>> b) the default is sometimes no.  So, it is not clear what a "complete
>> installation" means.  Is there any harm in always answering yes?
>
> I suspect the defaults of "NO" for the "Full Install" was based back
> in the days of TU58s or TK50s.  Reading the next saveset takes
> time....  Today, it doesn't make much difference with DVDs.

The V3 and V4 versions of VMSINSTAL.COM would read the entire tape with
DCL READ statements to verify that the saveset you had given on the
command line was indeed there, and if you omitted the product name from
the command line it would do the same to present you a list of the
products found on tape.

The fastest way to proceed was to copy the lot to disk first.  There was
also the persistent myth that you needed to put copied savesets into
[000000] so that you could give VMSINSTAL just the device name.  That
simply wasn't true (except for the occasional kit which included code to
insist on that restriction) and I performed many successful
installations by feeding a device plus directory to VMSINSTAL, in true
VMS style.


> I doubt anybody uses the BASIC Translator anymore.  
>
> One of my plans is to switch the remaining compilers that use
> VMSINSTAL (BASIC and Pascal) over to using PCSI to be consistent with
> the other compilers (I don't want to start another discussion on
> whether PCSI is good/bad/modern/etc.). I'll change the defaults and/or
> just eliminate some of the questions.

Can you please, please, please change the defaults for the relevant
compilers to Yes for the system header files.  In my experience, even
skilled system managers took the default of No here, so when you needed
them for a program which called a system service you were stuck.  I cannot
emphasise enough how tedious it could be to locate the installation tape
(usually in a locked safe somewhere else), check for later patches just
in case you were going to clobber them *and* get authorisation to
perform the addition of these header files at a later date.

Pretty please :-)

-- 
Nobody has demonstrated a HTTP/2.0 implementation that approached
contemporary wire speeds.                    -- Poul-Henning Kamp
                     <http://queue.acm.org/detail.cfm?id=2716278>