[Info-vax] CLI editing, was: Re: VMS - Virtual Terminals - A security risk way back yonder OR was that an Old Wives Tale ?

[David Froble]

Johnny Billquist wrote:
> On 2016-02-12 21:56, Stephen Hoffman wrote:
>> On 2016-02-12 19:53:13 +0000, Simon Clubley said:
>>
>>> On 2016-02-12, Hein RMS van den Heuvel <heinvandenheuvel at gmail.com>
>>> wrote:
>>>>
>>>> The other thing Windows is also infinitely better than OpenVMS is for
>>>> command line recall and editing. As much as I love OpenVMS and DCL,
>>>> the command line editing is a lame joke in DCL.
>>>
>>> If you think Windows CLI editing is better than VMS (and it is), then
>>> try bash on Linux. It leaves Windows standing in the dust.
>>
>> Linux, BSD, Unix, illumos, OS X, etc, all have far better support for
>> the command line than does OpenVMS.
> 
> To be fair, Linux, BSD, UNix, illumos, OS X, etc, would seem to all 
> actually be Unix, and what you are referring to is bash, which have a 
> better support for the command line than does OpenVMS.
> 
> Unix (by any of these names) do in fact not have better support, but 
> worse. The system functionality for reading and editing from a terminal 
> is way more primitivt that what VMS have.
> 
> It is bash (and a library called readline), which offers they very nice 
> experience.
> 
> Curious about bash, does not bash retain this functionality in the VMS 
> port?
> 
>>>> Shameful, hurtful, NIH syndrome. "we know what's good for you".. no
>>>> you don't.
>>>
>>> I have expressed my opinions _many_ times on this in comp.os.vms and
>>> even submitted formal enhancement requests in the past (with no
>>> response to the latter).
>>
>> Ayup.   No one wants to get near the terminal driver for any significant
>> work, absent sufficient schedule and staff for something between an
>> overhaul and wholesale replacement.   And that will break some existing
>> applications   There's decades of accreted and legacy cruft lurking
>> under that particular cairn of code, but — as soon as you start turning
>> over any rocks to see what's underneath and to rework and replace the
>> worst of it — the COMPATIBILITY! COMPATIBILITY! COMPATIBILITY! crowd
>> arrives.    As with more than a few other places in OpenVMS, not
>> breaking stuff effectively precludes fixing and enhancing and replacing
>> code.   But y'all have heard this rant before, of course.
> 
> In a way I find this morbidly interesting. The terminal driver in RSX is 
> probably the largest and most insane can of worms there is. I would have 
> hoped that it was cleaned in up VMS, and made more sane. But by your 
> description, it sounds almost identical to the RSX code...
> 
>     Johnny
> 

To be a bit fair, there have been numerous methods for connecting terminals to 
VAX systems.  I omit mentioning Alpha and IA64 because by the time they came out 
most terminal access was LAT or TelNet or other network connections.

RS232
Current loop
LAT
TCP/IP

and perhaps more.

So yeah, it's understandable that terminal driver has been set up to handle lots 
of different connection methods.

I'm going to guess that any serial lines still in use would not be for 
terminals, but other devices.  But what do I know?

With VMS in a server mode, the existing stuff works for me.