[Info-vax] Python on VMS

[Stephen Hoffman]

On 2019-01-17 14:42:51 +0000, g	rard Calliet said:

> Le 17/01/2019 à 04:19, Stephen Hoffman a écrit :
>> The balances that were struck have led OpenVMS to its present position 
>> in the computing market, too.
> I don't think the balance between compatibility and new things at the 
> expense of lack of compatibility is the principal cause of the place 
> VMS has got in the market.
> 
> The big reasons seem more: structural problem of the business (DEC 
> empire falling), deafness to the emergent markets (deskstop, internet), 
> failure of itanium.
> 
> On the opposite the very good compatibility qualities is one of the 
> causes which explain VMS didn't completely die.
> 
> Gérard Calliet


It is simply not possible to fit thirty-two-byte hashes into eight-byte 
fields without breaking app compatibility.

Not given the API designs common on OpenVMS.

This is not the only case where the old designs are broken.

The first storage volume past two tebibytes is going to break apps, too.

Not breaking those APIs was the easiest answer.  Less effort. Good 
short-term strategy, too.

Now there are ways to make future changes more transparent to 
apps—itemlists are not it—and those changes are a large and longer-term 
investment.

As you're fond of arcane references, let's try an easy one.  DEC is the 
Ancien Régime here.

The business and economic and hardware and management environment that 
DEC grew within and operated within then no longer exists.

What happened then cannot be changed.  What happened then has led to 
now.  Both good and bad.

The current and future choices at VSI are what matter to where OpenVMS 
might go.   What DEC chose, less so.

Upward-compatibility is most definitely very desirable and very 
important.  But in very specific and selective cases, app compatibility 
has to be broken to make forward progress.

Because you cannot fit thirty-two bytes in an eight-byte field.




-- 
Pure Personal Opinion | HoffmanLabs LLC