Experimental Physics and
Industrial Control System

"J. Lewis Muir" <[email protected]> · Tue, 11 Aug 2009 09:51:20 -0500

On 8/11/09 8:24 AM, John Dobbins wrote:
> Mark, Andrew, et al,
> 
> I applied the updated patch to osdTime.cpp and all my problems are
> solved! (Static build on multiprocessor Windows machine now works.)
> 
> I guess the joke is on me since I could of avoided this trouble by
> reading the documentation more carefully. (Why didn't Aesop have a fable
> about reading the manual?) Though I still think the errors encountered
> were bizarre.

Hi, John.

I suggest the joke is only slightly on you.  I have observed that what
happened to you happens often to many people.  I think the real lesson
here is that if there is a known problem for an EPICS release, a new bug
fix release should be made that includes the fix, rather than keeping a
list of patches on a Known Problems page.  Most people expect the latest
release of a software project to compile out of the box and include any
known bug fixes.

I think the main reason the EPICS developers do not make bug fix
releases is a feeling that the testing required would be too much.
While I understand this view, I still think making a bug fix release is
worthwhile.

One could keep the Known Problems list of patches so people could apply
individual patches as needed just like now, but for many people I think
they want the version of EPICS that "just works."  Whenever I build a
new version of EPICS, I download the EPICS base archive, download and
apply all the Known Problems patches, and then build EPICS base.  If
there were an EPICS base release that already had the patches applied,
that would be great.  I think both EPICS developers and users would benefit.

Lewis

-- 
J. Lewis Muir
Software Engineer
CARS/IMCA-CAT

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System