[Info-vax] Updated HPE/VSI OpenVMS V8.4-2L1 Marketing Brochures

[Stephen Hoffman]

On 2016-10-05 03:25:02 +0000, David Froble said:

> IanD wrote:
> 
>> Where is Linux today compared to OpenVMS? Years ahead and the gap is growing
> 
> Linux has some large organizations doing development.  It's not just 
> individuals.

Yes, Linux is years ahead, in many ways.   Linux involves many more 
developers and many more organizations.  There are parts and ideas 
worth using and worth borrowing in other operating systems and tools, 
and there are areas that are problems.   Same as OpenVMS.   Same as any 
other operating system or tool, for that matter.

>> Where is Windows today compared to OpenVMS? Years ahead and the gap is growing
> 
> Weendoze has a large organization doing development.  It's never individuals.

Ayup.   Operating systems are huge projects.  Microsoft too is much 
larger than VSI, and — in many ways — much further ahead, and — within 
the constrains of their installed base — moving far faster.   Far more 
in-house and ISVs and affiliated developers, and with much better 
development tools, too.

Outside of catering to the installed base, it'll be interesting to see 
where VSI starts heading.   But I digress.

>> Please enlighten us on the following...
>> 
>> How do you think VMS is going to catch up or stop being left behind? 
>> Doing a redhat perhaps? Redhat usurped open source indirectly by 
>> piggybacking on Linux and open source and monetised it for their own 
>> profit. I'm merely advocating VMS look to do something similar and 
>> piggyback on what is clearly working
> 
> What I'd like to see is some details on what's leaving VMS behind.  
> Yes, there is vast room for improvement.  But strictly for an OS, I 
> tend to doubt these claims of "vast differences".

If the goal is 1990s applications and 1990s-era code, OpenVMS works 
quite well.  It will very likely continue to do so, too.

Where OpenVMS falls over — badly — has been discussed here.   
Repeatedly.   TLS integration, IP integration, LDAP integration, ease 
of distributed management, ease of deployment, simpler management.

To attract newer applications and newer deployments?   For the 2020s, 
that likely involves getting back to what made OpenVMS a fine choice as 
an operating system in the 1980s — ease of use, consistency, security — 
and that involves performing redesigns of some parts, large overhauls 
of other parts, and substantial updates, vastly improved development 
tools, and a whole host of other details.

This gets back to experience with other tools and platforms.    And 
then there's the current pricing and related expectations in the 
commercial software market.   Start out your business or your prototype 
or your small project for free with Centos, then migrate (easily) to 
RHEL for projects that want or need to purchase the ability to blame an 
outside organization.

>> Apart from existing customers, how do you propose to attract new 
>> customers to VMS? Through organic growth or tapping into an existing 
>> open source code base out there?

> Which one do you think will create the largest growth and bring the 
> quickest dividends?

There aren't any obvious candidates.  Any of those candidates are 
already running on other platforms, and usually better integrated into 
at least some of those other platforms.

> When looking at open source stuff, you need to ask, does it have any 
> benefit for VMS?  Perhaps some of it might not.

Chunks of what-should-be-part-of-OpenVMS are open source now.  OpenSSL, 
Apache, Python, Perl, LLVM, libxml2, libjson, libarchive or zip and 
unzip, and various other giblets.  These or equivalents are available 
on most platforms.

>> How do you propose to attract new developers to VMS? Do you have 
>> contacts in the education sector? I do and I can tell you with

> 100% certainly that open source virtually totally dominates here. 
> Perhaps your snide remark about brainwashing might have been better 
> focused pointing out what the education sector is doing to young minds 
> when it comes to pushing ideas and ideals because these are the people 
> you need to attract to VMS

Maybe the open source works well enough for their requirements, at a 
price they can afford?  Linux, Windows, iOS and other platforms have 
followed variations of that approach as part of establishing and 
increasing their business, too.

> AS I've pointed out in the past, being exposed to VMS in education 
> would be nice, but not essential.

All businesses perform some sort of explicit or informal training for 
their new folks.   More than a few businesses have effectively 
outsourced chunks of their employee training programs to schools, and 
with the costs paid by the schools and by the students.   Most 
organizations that are using servers involve Linux and Windows, as well 
as hosted services.  This situation is a problem for other businesses, 
particularly if the schools aren't teaching the tools the business 
needs.  These businesses now have to fund (more of) that training, and 
a smaller pool of potential staff.

>> As for computing languages, the one's attracting the largest base of 
>> current coders and future are in fact open sourced based one's or 
>> moving towards it
> 
> But, are those "current coders" doing anything worthwhile?

Enough are.    Some of these current or future coders are also the 
folks that will be replacing each and every one of us.   If we don't 
succeed in somehow getting rid of our own jobs first.

>> Java will go some of the way to helping code being ported over the VMS 
>> but it's hardly going to drive new innovation to be developed on the 
>> VMS platform.  VMS will be a target for deployment not the home for 
>> software development for large scale projects other than specific 
>> projects targeting VMS itself and I think that number will be tiny on 
>> actual VMS
>> 
>> Browse GitHub and see what's coming down the pipeline and then 
>> enlighten us how VMS is going to be playing in these future arena's and 
>> working with even a fraction of these endeavours without an open source 
>> presence and/or open source focus or embracing open source?
> 
> Quantity is not the same as quality.  How many of those things do you 
> have a use for?

Not much, but that's like asking somebody which computer games they 
play.   There are lots of projects, and lots of games.   But of what 
Github and open source code I am using?   It works, and it's useful.   
So is VSI.   VSI is centrally dependent on some of that same 
Github-based code.

>> VMS is closed source and it appears the licence agreement will keep it 
>> that way, that's ok but it doesn't mean that going forward open source 
>> cannot be looked at for newer aspects of the OS or do you somehow think 
>> that VSI can keep up with Linux with it's 10's of 1000's of 
>> contributors who are moving it's innovation further along at an 
>> accelerating pace.
> 
> If there are 10s of thousands of coders for the Linux OS, how many 
> lines of code are allocated to each, 2-4, maybe 5?
> 
> :-)
> 
> No, the core stuff is being done by a small core of people, probably 
> most of them being paid for their work.

Yes, Linux has a lot of folks involved.   Many are getting paid.   This 
from 2015:   
https://www.linux.com/publications/linux-kernel-development-how-fast-it-going-who-doing-it-what-they-are-doing-and-who 


As for getting paid...  I've gotten paid to write code that's been open 
sourced.   I've had projects to port code to OpenVMS, too.   Payments 
are how you can get folks to do what you want, after all.

As for project scale, BSD is seemingly moving forward faster than 
OpenVMS.   At least in terms of development and features.

>> Do you think VSI will keep pace with what's coming down the pipeline in Linux?

No.  I don't.  VSI just doesn't have the scale.

> It's going to take VSI 2 - 3 years more just to move VMS to x86. What 
> is Linux going to bring to IT in this time?
> 
> That's a scary thought.  We can hope not too much ....

Linux will bring quite a lot, I expect.   Some ideas worth borrowing, 
some not.    Here's a quick list of what arrived in Linux 4.8:  
http://www.phoronix.com/scan.php?page=article&item=linux-48-features&num=1 
  ARM, POWER, graphics hardware, file systems, work on security well 
past what OpenVMS offers around ASLR, improved and integrated 
cryptographic random number generation, etc.   The live kernel patching 
that merged into Linux 4.0 is something that doesn't exist on OpenVMS, 
too.  
http://phoronix.com/scan.php?page=news_item&px=Linux-4.0-Kernel-Big-Features 



> 
>> On Sunday, October 2, 2016 at 5:40:59 PM UTC+11, Phillip Helbig 
>> (undress to reply) wrote:
>>> In article <27c6e2e0-e540-4eca-bd63-78af2b48c780 at googlegroups.com>,
>>> clairgrant71 at gmail.com writes:
>>>> The VSI/HPE agreement does not allow us to make VMS open source.
>>> And a good thing that is too.

Out of curiousity, why?  Are you aiming for not having to deal with the 
platform — akin to folks using Microsoft Windows or macOS now — or 
because you perceive open source as being problematic, and 
closed-source as somehow better?

>>> For the first time since sometime BEFORE the demise of DEC, VMS is in 
>>> the hands of people who know it well and care about it.  While there is 
>>> a fear that it might be too little too late, let's give them a chance 
>>> and see what happens.
>> 
>> DEC failed, it's as simple as that
> 
> DEC failed because they could not trim the huge work force that the 
> very expensive early computers supported.

IMHO, DEC failed because it didn't adapt to changing markets and 
products quickly enough, and actively avoided trying to cannibalize its 
own products and services.   For a company that was founded and that 
worked extensively and directly with ISVs and skilled customers on 
early hardware integration and on software, that DEC utterly missed the 
lower-cost shifts and the open source transition and missed the 
ascendence of software was and remains unfathomable.   DEC missed out 
on new (huge) markets (PCs, mobile, hosted) and on new products, and 
folks move on to Windows and Linux and other platforms.  Folks migrated 
as DEC and Compaq and HP had suggested.  But I digress.  Having too 
many folks on staff was secondary to missing these and other market 
changes.  Once you miss these sorts of shifts within and underneath 
your own businesses, you're in deep sneakers, too.


-- 
Pure Personal Opinion | HoffmanLabs LLC