Experimental Physics and
Industrial Control System

"Jeff Hill" <[email protected]> · Thu, 1 Apr 2010 11:17:51 -0600

> Changing a well-established name would require 
> a strong justification, which I don't see here.

First, some comments:
1) IMHO, this is software, and if we don?t adapt to future trends we are out
of the game. So we must always have the short term strategy and the long
term strategy.
2) I am going to guess that most of the users outside of base don?t
currently use the C++ interfaces so namespace issues don?t impact them.
3) In R3.14, noone of this is currently in namespaces so we aren?t changing
anything for existing users even if they use the c++ interfaces.

Second, a strategy for the future:
1) Decide what the c++ namespaces will be called. That?s the aim of this
email exchange. We don?t have any c++ namespace now so we are not
constrained by other than our best judgement.
2) Put new functions in the proper namespace.
3) Eventually put old functions in the new namespace, create global inline
functions with the "well-established name" in the global namespace which
calls the new namespace qualified functions.
4) Eventually (this is the long term part), move the "well-established name"
in the global namespace to the "epics :: deprecated" namespace so that an
old code that moves to EPICS R3.14+nnn will need to add "using namespace
epics :: deprecated" in order to compile.

> ps. I really like "open source imitative" - you should 
> trademark this one ... ;-)

The Microsoft context guessing, and auto amending, spell checker is a wicked
taskmaster for dyslexic spellers such as myself. One time, it changed
"inconvienent: sp into incontinence :-)

Thanks for your consideration of my ideas,

Jeff
______________________________________________________
Jeffrey O. Hill           Email        [email protected]
LANL MS H820              Voice        505 665 1831
Los Alamos NM 87545 USA   FAX          505 665 5107

Message content: TSPA

> -----Original Message-----
> From: Ralph Lange [mailto:[email protected]]
> Sent: Thursday, April 01, 2010 7:15 AM
> To: Jeff Hill
> Cc: 'EPICS core-talk'; Janet Anderson; 'Davidsaver, Michael'
> Subject: Re: naming the operating system independence layer
> 
> Changing a well-established name would require a strong justification,
> which I don't see here.
> 
> Ralph
> 
> ps. I really like "open source imitative" - you should trademark this
> one ... ;-)
> 
> 
> On Wed 31 Mar 2010 17:47:50 Jeff Hill wrote:
> > All,
> >
> > As I create new operating system independent interfaces it would be
> best for
> > them to be placed in a C++ namespace. Currently I have, nested inside
> the
> > namespace "epics", a namespace "osi" standing for "operating system
> > independent". However, perhaps there is some dismay related to the name
> > "osi", as an acronym for "operating system independent", because it is
> of
> > course similar to other acronyms; for example there are of course "open
> > system interconnect" and "open source imitative" in common use. The
> name
> > "osd" is also commonly in use today in the EPICS distribution standing
> for
> > "operating system dependent".
> >
> > Other options:
> > o epics :: osl - standing for "operating system liberated"
> >
> > At this juncture, this issue remains quite easy to fix in the source
> code as
> > the new namespace is minimally populated.
> >
> > Comments?
> >
> > Jeff
> > ______________________________________________________
> > Jeffrey O. Hill           Email        [email protected]
> > LANL MS H820              Voice        505 665 1831
> > Los Alamos NM 87545 USA   FAX          505 665 5107
> >
> > Message content: TSPA
> >
> >
> >
> >

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System