[Info-vax] The best VMS features, was: Re: openvms renaming file

[Jan-Erik Söderholm]

Den 2018-05-31 kl. 20:38, skrev Dave Froble:
> On 5/31/2018 11:55 AM, Jan-Erik Söderholm wrote:
>> Den 2018-05-31 kl. 17:51, skrev Dave Froble:
>>> On 5/31/2018 7:38 AM, Bill Gunshannon wrote:
>>> 
>>>> My point was that there is little difference between desktop and 
>>>> server systems today and writing either off is a bad decision.
>>>> Can it really be that hard  to write a device driver for a limited
>>>> set of graphics cards?  Remember, we are not looking at taking
>>>> over the gaming market (VMS could not do that graphics support or
>>>> not!!)
>>> 
>>> In the past it has been rather hard to write the graphics device 
>>> drivers. In no small part to the secrecy practiced by some graphics
>>> card mfgs.  I've read that some of that secrecy is less of a
>>> problem recently.
>>> 
>>> Also, remember past practices.  If the graphics mfgs wanted to sell
>>> into the weendoze market, they, not Microsoft, wrote the drivers.
>>> It appears they have no such desire for VMS.
>>> 
>>> Didn't help that a new device would be produced every couple of
>>> months.
>>> 
>>> I've read about some issues with Linux at times, with poor or no
>>> support for certain graphics devices.
>>> 
>>> So, yes, it just might be that hard to produce graphics device
>>> drivers for VMS, even a limited set.
>>> 
>>> One might wish it was otherwise, but, it is what it is.
>>> 
>> 
>> Right, I have to ask...
>> 
>> Exacelly *what* would you use a graphic driver for in VMS?
> 
> VSI has mentioned that there would be enough graphics to drive a
> monitor. That is a graphics driver.  Perhaps not for the latest nvidia
> or AMD have to offer, but, still graphics.
> 
> Now, I know nothing, but, there has been use of the compute engines in 
> graphics cards to do floating point stuff.  It can greatly enhance the 
> performance of a computer.  Might be nice to have, huh?
> 
>> Why would anyone (apart from hobbyists) want that?
> 
> Workstations for CAD and such sure need good graphics.  Not saying it's
> a market for VMS.  Lots of reasons for graphics, unless you're stuck on
> a serial port and a terminal.
> 

20 years ago there were only text terminals *or* "graphics".
But since the whole industry has moved to "the web".

>> Why whould VSI spend resourses on this?
> 
> I don't think they have much choice.  Just don't look for them to
> support every graphics device available.
> 

For install/boot/config maybe. But that is also more and more
built into the server environment and VM tools.

>> Development for VMS will in the future not be done on VMS anyway. Not
>> if you like modern and full-featured tools anyway...
> 
> Somehow I don't understand that concept.  I'm assuming there will still
> be compilers and the linker.  Maybe I'm wrong.  But it doesn't make much
> sense to write cross compilers when a VMS system is much simpler.
> 

You run your front end including editing the actual source codes on your
normal desktop environment (most propably Windows or Linux based) and then
have a backend on VMS with compilers, linker, debugger and other tools.

> I've always maintained that VMS has had a rather rich development 
> environment, with many languages.  Perhaps not so many anymore, but, my
> opinion is that it's still a good development environment.
> 

Its not about the (programming) languages and compilers, it is about
the environment where you edit your sources.

> Bells and whistles do not produce good software.  Good design and 
> implementation is what's required.  VMS still allows such.
> 

Yes, for many parts. Not so much for the code editing environment.
Like an MS Visual Studio frontend for VMS programming. Or some of the
open/free alternatives.

Editing code in EDT/TPU/LSE/whatever is not the future.

That has absolutely nothing to do with the design of the code, build
and general application environment as such.