[Info-vax] OpenVMS in the future, Open sourced or Closed? Intent is to keep it...

[IanD]

On Saturday, May 2, 2015 at 3:40:34 PM UTC+10, JF Mezei wrote:
> On 15-05-02 00:42, IanD wrote:
> 
> > It would be interesting to know if VSI's intent is to open source OpenVMS where possible 
> 
> At the moment, it is my understanding that JP keeps rights to VMS and
> has licensed VSI to play with the source code. If this is correct, then
> VSI doesn't have the rights to decide to open source it.
> 
> My guess is that eventually, HP might just give VSI the kit, especially
> at some point where VSI has modified enough of VMS that it isn't so
> clear that HP owns the rights to that source code anymore. Not sure what
> sort of precedents exist to decide at what point VSI gets rights over a
> product after having changed it significantly from the owner's code.
> 
> With regards to the IP stack, it would not surprise me if VSI struck a
> deal with Process.
> 
> 

I figured this

It would be 'interesting' to know however from the horses mouth if open source is indeed a goal for VSI or not

I know it's early days in the piece but there are a number of young programmers out there who will not even look at projects unless they are open. Sure it may have no consequence at the moment because the announcement to keep OpenVMS alive has only just happened but a future direction on it's intent may be important later on

> > That being, I'm wondering if VSI will be open sourcing the new stack or keeping it closed
> 
> If VMS engineering remains in touch with the community, this may work
> far better than open sourcing it and having the code move left and
> right.  Having the engineering team communicate with end users/system
> managers makes them responsive to needs and can shape the product and
> make minor upgrades that really benefit the customers.
> 

Totally agree!!

This is partly why I raised a topic before about future topics for OpenVMS so that people would put their ideas forward so that VSI could hear from the community and take them into account (well, that was the hope at least)

At the end of the day if the OS doesn't fit people's needs, then it's dead in the water, so adapting it to meet the needs of people is extremely important and the only way to do this is to get feedback from the masses

Open source is one way to get rapid development happening

Hell, I'd like to see things like crowd funding happen to expedite development faster. I'd throw a few bucks of my own money into seeing things like Python 3 and Postgresql get onto OpenVMS quicker, but they would need to come under VSI's authorship and oversight, much like how linus still has final oversight and sign-off authority on linux kernel code

I really want to see this type of framework put in place by VSI

> I recall Guy Peleg coming here before 8.3 was released and asking for
> ideas for minor changes to DCL and backup which made it to 8.3 and was a
> big improvement for customers.
> 

Really? I didn't know that. This is certainly what is needed

> On the other hand, for portions of VMS that VSI/HP do not intend to
> update anymore it makes sense to let the community try to make updates
> to it and if VSI finds the updates worthy, included it in distribution.

Yeah, VSI cannot do everything and it's naive of people to think that they can

I'm of the opinion that with the current changes going on in the IT world, that OpenVMS users cannot afford to sit back and wait for VSI to develop everything. 
I work in the Telco space and have close associates in the banking space. In the DC's I have access to, the number of servers are in the 1000's to 10's 1000's and the big driver for change is automation and standardization, on all 3 fronts. 
On the OS front (linux is current flavor), on the hardware front Oracle seems to be the winner here and on the application front, again it's Oracle applications in the driving seat at present, OpenVMS is being shunted out the door :-(
Windows is doing well in the banking sector as banks modernise their applications, moving from mainframe apps. 
The commonwealth bank in Australia committed to spend a huge amount of money on software development, revamping it's aging mainframe based systems. 200M in 2013!

http://www.lifehacker.com.au/2015/02/how-commonwealth-bank-spends-its-it-budget/

In the Telco space, the OpenVMS systems that still lurk in the shadows are being shown the door because they are one stop shop solutions and tend to not play easily with other systems and they don't integrate well from an enterprise application aspect

I happen to think OpenVMS needs more than VSI to make it viable in the very near future and open source may, read 'may', provide another method