[Info-vax] For sale: VAXstation 4000/90 128MB Fully Working and Tested

[Dave Froble]

On 6/30/2022 9:58 AM, Arne Vajhøj wrote:
> On 6/29/2022 10:01 PM, Dave Froble wrote:
>> Erik (the boss) is 80, I'm 76, Bill in early 70s, and Niel is late 60s, and
>> really want to retire.
>>
>> We made a pitch to the customers.  Buy the software package, and we will give
>> you a year or whatever we can to train new people.  They said they didn't want
>> to be in the software business.  (Almost every business for the last 30-40
>> years has been in the software business, in some way.)
>>
>> Largest customer, now being run by original owner's children, had their
>> auditors come in, and were told:
>>
>> You can't use that ancient OS (VMS) ...
>> You can't continue with that ancient language (Basic) ...
>> You need a relational database ...
>> And a few more ridiculous claims ...
>> Above all, "you can't count on these old geezers" ...
>>
>> They refused to give them a successful audit ...
>>
>> Can't purchase business insurance without successful audit ...
>
>> We have continually offered to help.  Other customers are looking at the
>> clusterfuck, and are re-thinking about being in the software business.
>> Codis does all they need to run their companies.  The main problem is
>> the age of the people involved.  We'd really like to do something about
>> that, but if we could, we'd be rich and in the medical field, not software.
>>
>> Erik has told the customers, the company is shutting down at the end of
>> 2023. We'd like to continue, but we can't make any promises.  We will do
>> what we can. But if the beer truck gets us ...
>>
>> We have told the customers:
>>
>> VSI is currently supporting VMS, and porting it to x86 ...
>> VSI is currently (I hope John) supporting Basic ...
>> The auditing firm is a bunch of crooks ...
>
> There are some reasons for concerns:
> - niche OS (VMS)
> - niche language (VMS Basic)
> - obsolete language (Macro-32)
> - probably non-optimal persistence technology (RMS index-sequential files aka
> NoSQL Key Value Store)
> - the team getting close to retirement
>
> But it is not an urgent problem:
> - all the software is supported and can be expected to be
>   supported for many years
> - the team has not retired yet
>
> What you need is a roadmap showing a long term viable future.

We tried that.  Did you miss the part about the company now being run by the 
founder's children?  Don't know what's required, so count on outside help, and 
the outside help turns out to be a bunch of crooks.

> Obviously I don't know the company or the product that well, so this is
> pure speculation, but something like:
> - keep VMS but switch to VMS x86-64,

which has been the plan

> it is supported and even though
>   Linux is server OS king today then everybody does not need to run
>   Linux, working with VMS is a learnable skill (someone that know
>   Linux should be able to learn VMS in weeks)

> - keep VMS Basic, it is supported and a rewrite would cost a lot of
>   money and create a lot of bugs that would need to be fixed, and
>   programming in VMS Basic is a learnable skill (someone that
>   know VB.NET/Delphi/C should be able to learn VMS Basic in a few
>   months)

Try to port it, and just as well start from scratch.  We've done evaluations. 
Don't see any good in that direction.

> - get rid of the Macro-32, rewrite to VMS Basic or C, too difficult to
>   get people onboarded with Macro-32

The only Macro-32 is the database.

> - change to a relational database, besides looking better in the audit
>   report you will actually get some benefits in the form of less code

We looked at such a project.  Again, might as well start over.  We at least 
implemented moving all the live data to a WEENDOZE based database so the 
WEENDOZE weenies could pull out data they might want.  But they cannot touch the 
live data.  And that is a VERY GOOD THING.

> And hire somebody new. Somebody at least below 50, preferably below 40.
> To give some assurance that there will be a team around for a long time.
>
> Maybe do a gradual phase in of more new people:
> 2022-2023: +1 new
> 2024-2024: +another new, let 2 of the old switch to part time
> 2025-2026: +another new, let 2 other old switch to part time

That is what we invited the customers to do.

>> Then they pointed out a company with a cloud based solution for the customer
>> to talk to.  Too bad they didn't mention that the company was a wholely owned
>> subsidiary of the auditing firm.
>>
>> :-)
>>
>> The app uses a single connection to the cloud.  There is inadequate
>> throughput. It is slow.  Every time the customer asks for a feature in Codis,
>> they are told that cannot be done.  After several postponements, the customer
>> went live on the cloud solution.  They are missing many features they counted
>> on.  The system is inadequate.  For example, (Bill likes to gather
>> statistics), max orders from Amazon have hit 100 per minute.  The new system
>> is lucky to handle 5 Amazon orders per minute.  Most people aware of this are
>> predicting that they're going to crater.  There is also the penalties from
>> Amazon if a vendor does not perform as agreed.
>>
>> This customer might have spent a million over the last 30 years on Codis.
>> Already they have spent over 5 million trying to get the new cloud based
>> working, and it isn't doing so well.
>
> Auditing and other services should not be mixed.
>
> I think that was one of the lessons from Enron!

You got that right.


-- 
David Froble                       Tel: 724-529-0450
Dave Froble Enterprises, Inc.      E-Mail: davef at tsoft-inc.com
DFE Ultralights, Inc.
170 Grimplin Road
Vanderbilt, PA  15486