[osis-core] mID vs mS and mE

[Patrick Durusau]

Todd,

Todd Tillinghast wrote:
<snip>

>  
>
>>to insert the mType values. Would be vastly more accurate and less
>>    
>>
>time
>  
>
>>consuming than having users do it by hand.
>>    
>>
>
>I would agree that if something is time consuming for creators that the
>process should be automated.  However, we have no control over what
>actually gets automated.  What we can control is the standard.  If there
>is going to automation then let the creation/authoring side of things
>make sure that the user properly uses mS and mE.  Something (human or
>machine) has to make sure that the mID values or the mS and mE values
>are the same anyway.  For the user it would seem LESS error prone if it
>were clear which was the start and which one is the end.  This way no
>scripts have to be run on the creation/editing or the consumer side.
>
>I don't see any reason to add a requirement for specialized scripts or
>software when they can be avoided.  With mS and mE no scripts are
>required.
>  
>
Sigh. I don't disagree that we only control the standard and the 
solution proposed would allow users to insert the information you want, 
albeit without abusing an attribute name to carry what should be 
attribute data. No scripts are required by Troy's solution either, I was 
just suggesting that as a side benefit.

Isn't it true that with the mS and mE solution that the user has to know 
both attribute names in order to use the milestoneable elements? And not 
only that, but use them in the correct order? Seems to me that is more 
work han simply using mID and possibly the mType attribute with value. 
Note that they are not required to use the mType attribute, which, if 
desired, can be more accurately inserted by a script (although that is 
not required).

While we certainly control the standard, I don't think it is useful to 
attempt to force users to do markup "the right way" when we have 
alternatives (like scripts) that make their lives easier and less error 
prone.

I see Troy's solution as giving you the information you say is needed so 
I don't really understand your objection, unless overloading the 
attribute name, that is mS and mE is of some other significance.


<snip>

>>>      
>>>
>>Actually we only have to tell them to use the same mID for both ends.
>>    
>>
>As
>  
>
>>noted above, the rest should be a processing issue, on Windows and
>>    
>>
>Linux
>  
>
>>anyway, don't know about Macs.
>>
>>    
>>
>
>What does the operating system have to do with the processing of mID?
>Are you presuming a specific software package?
>  
>
Not presuming a software package. Demonstrating ignorance of Mac OS X 
and earlier versions. Don't know if they run scripts at all, suspect so 
with OS X but simply don't know.

Hope everyone is having a great day!

Patrick

-- 
Patrick Durusau
Director of Research and Development
Society of Biblical Literature
Patrick.Durusau@sbl-site.org
Co-Editor, ISO 13250, Topic Maps -- Reference Model