Re: Making a case for cases

["Mark Davis" <davism50@msu.edu>]

While not explicitly stated, I assume the names we are talking about are 
Channel names in the control system (?)

While including all that info in each channel name may help insure unique 
names, it also makes them VERY long.  We might  want to consider a more 
compressed version for the channel name itself, with an expanded version in 
the DESC field (if there are record types without a DESC field, we can 
always add it).

For example:
  Name:  DTL_D:BLM130:MTLR
  Name.DESC:  DTL_Diag : BLM130 : MPS Trip Limit Rad

Not ideal, of course, as you need to agree on (and hopefully enforce) 
standard acronyms for use in the names to avoid overlap, but the idea how 
long the names would be without something like this is a bit disconcerting.

Mark Davis
Control Systems Software Engineer, NSCL


----- Original Message ----- 
From: "Pete R. Jemian" <prjemian@gmail.com>
To: <tech-talk@aps.anl.gov>
Sent: Thursday, August 05, 2010 12:54 PM
Subject: Re: Making a case for cases



Mathias:

Alexander's observation is important.  You seem to want to use "_" as a
context delimiter.  But you see it is /also/ popular as a word
separator.  Be careful how you pick and use context delimiters.

Pete



On 8/5/2010 11:29 AM, Zhukov, Alexander P. wrote:
> I think mixed case is the best.
>
> You want to ensure proper notation that is strict. E.g. what to use
> instead of a space? Just skip it or use underscore?
>
> If everyone follows it then there will be no ambiguities or problems
> with search
>
> Unfortunately we at SNS didn’t follow it very well and now we have PV
> names like:
>
> DTL_Diag:BLM130:MPS_Trip_limit_Rad
>
> Instead of
>
> DTL_Diag:BLM130:MPSTripLimitRad
>
> or
>
> DTL_Diag:BLM130:MPS_Trip_Limit_Rad
>
> Alexander Zhukov
>
> SNS/ORNL
>
> Beam Instrumentation Group
>
> *From:* tech-talk-bounces@aps.anl.gov
> [mailto:tech-talk-bounces@aps.anl.gov] *On Behalf Of *Steiner, Mathias
> *Sent:* Thursday, August 05, 2010 11:49 AM
> *To:* tech-talk@aps.anl.gov
> *Subject:* Making a case for cases
>
> Learned Friends,
>
> The planned Facility for Rare Isotope Beams, a.k.a. FRIB, is looking to
> finalize the naming convention.
>
> The proposed system is an evolution of the SNS convention. I'll list the
> format to give an idea of how many uninterrupted characters one might
> encounter:
>
> SSSS_BBBB:DDDD_QIIII:TTTTIIIIXXXX
>
> System_subsystem:Devic_Qualifier:Signal Name
>
> One of the remaining questions concerns letter case.
>
> Option 1: Uppercase only.
>
> It's simple; it's unambiguous; it's what we do now.
>
> On the downside, it is hard to read.
>
> Option 2: Lowercase only.
>
> It's as simple as uppercase, but easier to read.
>
> Option 3: Mixed case.
>
> This is best option for legibility.
>
> But it creates ambiguities and can make searching difficult.
>
> Is there something like a community consensus on this?
>
> Are there any pitfalls we should be aware of?
>
> Thanks in advance,
>
> -Mathias
>
> Mathias Steiner
>
> Staff Physicist
>
> NSCL/Michigan State Cyclotron
>
> East Lansing, MI
>
> voice 517-908-7423