[Info-vax] OpenVMS Development Annoyances

Arne Vajhøj arne at vajhoej.dk
Sun May 5 20:42:08 EDT 2019


On 5/5/2019 6:41 PM, Dave Froble wrote:
> On 5/5/2019 4:18 PM, seasoned_geek wrote:
>> On Monday, April 29, 2019 at 10:43:32 AM UTC-5, Stephen Hoffman wrote:
>>> As was mentioned else-thread, ACMS integration with SYSUAF and logical
>>> names?  Yeah, okay, but that's not selling this for me.  SYSUAF and
>>> logical names and ilk are among the more problematic OpenVMS features.

>> If you don't like the high quality things which make OpenVMS, VMS 
>> (common SYSUAF, logical name tables with logicals actually in them, 
>> and file versioning) why don't you just go develop on Linux? <Grin>

>> I love all those things you hate. If they go away there is no reason 
>> for VMS to even exist.
> 
> For me, there is a whole bunch of reality in that last statement.
> 
> DLM

The API is horrible.

But it is a nice feature.

It is just not unique for VMS - only the tight integration with the OS 
is relative unique.

If you like the API then the *nix libdlm has the exact awful API.

:-)

> Logicals

Logicals are OK.

But other platforms has found different solutions for the same problems.

> Basic

:-)

> Common calling std ( well, except for things such as C, C++, et;al

Very nice.

But also limited to procedural (not-OO).

JVM and CLR platforms can do the same with OO.

> BACKUP

For general backup VMS BACKUP is extremely primitive.

And even decades ago more advanced backup products where
available.

> Stand Alone BACKUP !!!!!

Also primitive, but I like it too.

It solves one specific problem in a simple easy to
use way.

And one need to restore then one really appreciate simple and safe.

> Oh, Ok, VMS clustering

Everybody does clustering today.

Not the same way as VMS, but ...

> RMS (yeah, it's old, but at least it's there)

The extensive set of file attributes can be nice and provide
some options.

But if you think index-sequential files then there are a ton
of those out there (if they are old they are called iSAM, if they
are new they are called NoSQL key-value store).

> As for passwords, no matter what is done, it will never be secure.  If 
> anyone get a copy of your SYSUAF or whatever replaces it.

Long passwords and a good (read: slow) password hash algorithm
can ensure that it takes a few million years to brute force it.

Arne



More information about the Info-vax mailing list