Community, Please share your experience and insight on this question:
When working with a large application under source control, what is the REAL value of placing each PBL in its own folder?
I observed this benefits of multi vs single folder when working with TFS
Refresh and get status operations are per folder. If all code is in one folder, get status returns the status of ALL code objects. In a large application with many thousands of code objects, time lag is noticeable. Expanding a PBL causes a report on all code objects in all PBLs.
But when code is split into subfolders, the IDE only gets the status of the code in the requrested PBL folder not all code.
However moving code objects between PBLs requires the same procedure, Checkout and edit the PBGs and move via ORCAScript, or the library painter in a non-SCC workspace
True PBL per folder is cleaner because files with the SCC repository match the PBL structure. Also should humpty dumpty fall off the wall it might take less than all the kings horses and all the kings men to put humpty dumpty back to together again (Figuring out what source goes into what PBL is a 1 : 1 relationship. vs reading potentially corrupt PBG files to determine what goes where)
Are there other benefits that make it a best practice?
And the $64000 question is 'Is it worth the effort to reorganize a large project from single folder (all pbls in one) to multi-folder (one pbl per folder) structure?
Thanks for sharing your thoughts
Yakov