[Info-vax] Programming languages on VMS

[Johnny Billquist]

On 2018-02-07 18:56, Stephen Hoffman wrote:
> On 2018-02-02 21:20:11 +0000, DaveFroble said:
> 
>> Stephen Hoffman wrote:
>>> On 2018-02-01 17:29:24 +0000, DaveFroble said:
>>>
>>>> Stephen Hoffman wrote:
>>>>> TL;DR: RMS file versions are a pain to use, require the user or the 
>>>>> system manager explicitly manage them, and don't particularly solve 
>>>>> the problems that we even say that RMS version files solve.   Not 
>>>>> well.  If at all.
>>>>
>>>> Maybe I'm just too old school, but, I understand versions, and I 
>>>> understand what they can do for me.  And so, I use them.
>>>
>>> There's no reason not to use them.  They are, however, a complete 
>>> pain in the rump, and don't really solve the problems we use them 
>>> for. Combining versions and backups, and direct integration with 
>>> source code control tools for developers really makes this clear, too.
>>
>> It really depends on how such is used.  Most definitely not for backup 
>> and such.
>>
>> I've seen systems where an inadvertent PURGE would be a disaster. 
>> Blamy that on people, not versions.
>>
>> Anything I might want to save will NEVER be a prior version.
> 
> For developers editing code, file versions are backups.  Transient 
> backups the way they're currently implemented, but fundamentally they're 
> used as backups.  And they're not at all integrated with the "actual" 
> backups.  If this isn't clear, then ponder what you'd do without file 
> version support, but with a backup system that collected a backup on 
> every file close, and automatically thinned out the more frequent 
> backups into less-often and longer-term archives; into backups that are 
> kept longer.  This isn't a hypothetical design, either. This 
> implementation exists.

First of all, having the file versions are much more lightweight when 
you want to go back and check something from a previous version, than to 
go through the backup system and restore a previous version, not to 
mention the simple ability of being able to compare them side to side 
without having to start doing renames or other tricks so that the 
restored version from the backup don't actually overwrite the current 
version with the same name.

Second, I'm not aware of any backup system that does a snapshot at every 
file close. The closest to what you describe that I know of is the Time 
Machine from Apple. But that only takes a snapshot every hour. I can 
create many versions in one hour when I'm developing code.

And there is still that issue of having to go into the time machine 
interface and restore if I want to get an older version back, and of 
course the name collision if I also want to retain the current version, 
and maybe also want to make comparisons or checks against 7 other recent 
versions in short order.

But hey, I certainly do recognize that I'm not a typical user, and my 
patterns are probably both odd and archaic. I don't really believe in 
IDEs either, but I do believe in version control systems. But I do not 
consider version control systems to really overlap my use of file 
versions either.

   Johnny

-- 
Johnny Billquist                  || "I'm on a bus
                                   ||  on a psychedelic trip
email: bqt at softjar.se             ||  Reading murder books
pdp is alive!                     ||  tryin' to stay hip" - B. Idol