[Info-vax] Should VSI create a modern day VMS applications book ?

[Arne Vajhøj]

On 8/9/2018 9:50 AM, IanD wrote:
> On Wednesday, August 8, 2018 at 9:15:35 AM UTC+10, Arne Vajhøj wrote:
> 
> <snip>
> 
>> * a good chunk of what is available is free open source
> 
> ^ This
> 
> Open source doesn't mean uncontrolled either. Typically the version and patch levels are frozen and stored internally in the organisation and can only be installed from that internal source

I agree.

I am sure the demand for information about available open source exists.

My concerns was about the supply of information about available open source.

For a commercial product there is a bunch of marketing/sales/business
people that will see a purpose of getting their product in such a catalog.

Most open source developers will not want to spend time on such an
activity.

> More and more of the interfacing applications are open source and the shrinking core is mainframe or core applications that undergo little or no change and that core functionality is whittled away bit by bit over time

I totally agree.

> In one place I worked, a logistics company that built it's whole enterprise on VMS was bought out and instead of continually enhancing the 30 year old extensively complex and all optioned application, they have decided to move the data to a large enterprise offering based on SAP Hanna and offer a much reduced subset of their logistics offering.
> 
> The complexity offered and built up over the years won less and less business as time marched on. Sell standard offerings and let the customers repackage and onsell so your more of a wholesaler than anything else seems to be the flavor of the day.
> 
> In this regard, VMS has nothing extra to offer over a cheap Linux or Windows server in the businesses eyes
> 
> The fluffy human interface stuff is a dime a dozen and increasingly going to mobile or web. The core is the data and that's sitting on things like Oracle or MySQL or even db2 on older systems and a lot of the newer stuff is being developed on nosql offerings because they scale.
> Some of the new funky stuff like neo4j is slowly making inroads as people look to more flexible DB schemas. Neo4j is heading towards real time machine learning in nature as well
> 
> I'm my last 3 roles (2 with VMS and 1 with Linux & Windows), there really is no role for VMS as an OS anymore, at least in its current form

It is no big secret that VMS is not selling millions of copies per year.

> It doesn't run the latest major DB's, it doesn't run most of the latest open source applications or tool chains. System management and Application management isn't even close to being what most places want in terms of DevOPS. That's why in all 3 of my last places it's earmarked for decommissioning, one of which had already removed VMS and the other 2 have active projects underway

Yes. The applications are important. Companies usually pick the
application they need and then the OS that can run that that
application.

> The flow of modernisation of the IT landscape has become a torrent now and I'm left wondering if the ploy to modernise VMS has come just too late, especially when we are looking at late 2019 or 2020 before it's production robust and ready
> 
> Someone cheer me up with some valid VMS data to the contrary please, I'm feeling somewhat dispondant about VMS :-(

VMS would have been far better off if VSI had taken over 5 years
earlier.

And I can not point to any huge VMS success stories.

But there are some positives in the application space.

Todays applications are typical way less OS dependent than
they were 30 years ago.

If VMS get the platform right:
* uptodate C compiler and RTL
* uptodate C++ compiler and RTL
* uptodate Java
* uptodate PHP
* uptodate Python

Then there are tons of software that will run on VMS. Simply because
they are not OS specific - just require a certain platform (in current
version).

Arne