Re: Execute flag for database files

[Vern Hamberg <vhamberg@xxxxxxxxxxxxxxx>]

Hi Patrik

I'm glad you tested my theory, I would never have thought to try it 
until this discussion. But I have long thought of the library/object 
system as a shallow file system, only 2 levels except for some kinds of 
files with members. As I mentioned, I got the feeling when I started to 
work with FTP. This 3-layer structure or 3-part naming has existed from 
the beginning, I believe, as reflected in low-level names of objects, 
which are 30 long and are made up of 3 10-character segments.Many object 
types use only 2 of those segments.

*Best Regards*

*Vern Hamberg*

IBM Champion 2025 <cid:part1.HnTE0vnS.Fp7PspZO@centurylink.net> CAAC 
(COMMON Americas Advisory Council) IBM Influencer 2023

On 4/11/2025 4:55 AM, Patrik Schindler wrote:
> Hello Vern,
>
> Am 10.04.2025 um 19:12 schrieb Vern Hamberg via MIDRANGE-L<midrange-l@xxxxxxxxxxxxxxxxxx>:
>
> > Patrik, I would say it is probably not ignored. When you say database files, those have at least 1 member - that member is "contained" in the file object, so it seems like 1 more layer in the library/object structure - library/object/member - we see that even in naming format 0 in FTP, I believe.
> >
> > X in the IFS - you need that authority on a directory to see what is in it, is that not correct? If my analogy is valid, then if you remover X from a PF, you should not be able to display its member (or members, as in the case of source PFs, perhaps)
> You are absolutely correct! I have verified this claim just now — on V4R5, but I presume the behavior is not different for newer releases.
>
> Thanks for pointing out!
>
> Sometimes it's good to ask, discuss, get new ideas, try them out instead of assuming. Such assumptions tend to become mental fossils over time, because nobody ever challenged them in the first place. :-)
>
> :wq! PoC