[Info-vax] BASIC compiler in the hobbyist distribution

[johnwallace4 at yahoo.co.uk]

On Friday, 29 May 2015 14:54:37 UTC+1, Stephen Hoffman  wrote:
> On 2015-05-29 09:13:20 +0000, johnwallace4 at yahoo.co.uk said:
> 
> > On Friday, 29 May 2015 00:25:10 UTC+1, David Froble  wrote:
> >> 
> >> I'm thinking that VSI knows who their customers are.  Check the map  :-)
> >> 
> >> Their best bet is to service their customers.  A really bad bet would 
> >> be> to go after IBM, again, which is part of how we got to where we are 
> >> now.
> > 
> > VSI know who yesterday's VMS customers were.
> 
> Those customers are also today's and tomorrow's customers.   This for 
> the next several years and probably out past the availability of the 
> x86-64 port; when the VSI developers can really be turned loose on 
> enhancements and improvements and updates.  (Now where the computing 
> market will be in ~2020?  Donno.)
> 
> > DEC knew a bit about who yesterday's customers were too. DEC didn't 
> > know (or didn't properly think about) who their customers (not just VMS 
> > customers) were going to be in (say) three years time. The HQ message 
> > seemed to be 'same customers, bigger value sales' (added to a bit of 
> > bravura about taking on IBM), ignoring the reality that what was good 
> > enough for lots of people was getting cheaper each year.
> 
> If you're funding and running a large sales force -- whether salaried or 
> particularly commissioned -- that's the only direction you can try to 
> expand.  Going the other direction is a bigger change, and one that 
> various of your existing (big) customers won't like.   (Simply having 
> your customers call up a sales representative to buy a US$100 software 
> license and get a paper package printed and mailed and then typed into 
> some registration mechanisms just isn't cost-effective for anybody 
> involved, but I digress.)  Sure, your existing customers will like the 
> lower prices, but can you make up those losses with increased sales 
> volumes?  Cannibalizing your own products and your own services is 
> tough, and it also usually requires you to have relatively few cost 
> centers.
> 
> > VSI have a bit more focus than DEC/CPQ/HQ.
> 
> Vastly more.  Though with less access to the hardware platform teams -- 
> Nemonix may well change that -- and to other corporate-internal software 
> and tools, projects and expertise.
> 
> > VSI still need to think about not just yesterday's customers, but
> > tomorrow's as well. Who those customers might be, and (as you
> > rightly point out) who they probably are not going to be.
> > 
> > Fingers crossed they're in a good position to do so.
> 
> Ayup; whether VSI can expand beyond the current OpenVMS market.   That 
> will take a whole lot of work and a whole lot of insight into where the 
> market is headed, more than a little patience, and a willingness to 
> take on more than a little financial risk.
> 
> > ps
> > in a couple of years time it will be VMS's 40th birthday (taking 
> > release date as birth date). VMS will be middle aged.
> > 
> > Some people seem to think new is inherently improved; that young is hip 
> > and cool. After all, that tactic works so well for cheap and cheerful 
> > consumer goods and for fashion victims, so much of
> > the IT market has adopted the same tactics.
> > 
> > On the other hand, in the field of real engineering, incremental 
> > improvement over time is usually a *virtue*.
> 
> Some of that new stuff is junk.  Some of it is not.  Some of the new 
> stuff is massively useful.   Some is marketing.
> 
> But assume that the marketing means there's a poor or questionable 
> product involved -- or that a "less good" product won't sell in massive 
> quantities and thus undercut your products -- at your peril.
> 
> This is how more than a few existing markets have gotten torpedoed from below.
> 
> Some of the old environments -- OpenVMS included -- are much harder to 
> work on and develop for than some of the newer environments, too.  
> That's based on my experience, and my own familiarity with how OpenVMS 
> works.  For better or worse, large walls of documentation and dozens of 
> PDF files just aren't the sorts of things that many folks are 
> interested or willing or able to read through -- developers have work to 
> get done.  As often as I've pointed folks to the documentation, that's 
> increasingly not something that folks are able to invest the time and 
> effort necessary to read through.  This feeds back into tool designs, 
> massively easier and faster search tools, examples and a whole host of 
> other details that can allow a programmer to work more quickly.  Also 
> into the sorts of user interfaces and the application frameworks and 
> libraries and abstractions that are available, too.
> 
> 
> 
> 
> -- 
> Pure Personal Opinion | HoffmanLabs LLC

Wrt
"Simply having your customers call up a sales representative to buy
a US$100 software license and get a paper package printed and mailed
and then typed into some registration mechanisms just isn't
cost-effective for anybody involved, but I digress."

DECdirect UK, and reportedly in Europe too, managed to do pretty
much that. Together with manufacturing in Ayr (Scotland) and Software
Supply in Galway (Ireland), they had LEAN systems before anyone even
knew what LEAN was. End user orders a MicroPDP/VAX one week, Ayr custom
build it, customer gets it the next week. Software and documentation
produced on demand, not from stock. 

DECdirect UK even did outbound marketing, e.g. via the DECdirect
catalogues for high volume hardware and software (which looked
nothing like the DECdirect US catalogue, fwiw). They contained
articles for stuff that *could* be high volume, if people only
knew about it. DECdirect UK software revenue went from zero to
$USM100+/yr in the space of 18 months.

Lots of people knew what needed to be done. Some got a chance to
show how it could be done. HQ weren't listening.

HQ decided that stuff was mostly better done by their golf mates in 
"the channel", who hadn't been doing any of it before DECdirect
started, and quickly stopped doing it when they were handed
DECdirect's business on a plate. (Maybe a slight exaggeration,
but only slight).


VSI need to do a lot better than DEC generally did, and hopefully
they will. But there may be stuff to learn from too. Some of the
stuff to learn from may not be in the US either. VSI's approach
to the world outside the USA will be interesting (bearing in mind
where Boot Camp attendees typically come from, for example).


"the sorts of user interfaces and the application frameworks and
libraries and abstractions that are available, too."

Build GNV and the rest will follow? Do I need half a smiley here?

Have a lot of fun.