Re: SMEs : was RE: Party XML Schema Defintions

[Sandy Klausner <klausner@coretalk.net>]

Title: Re: SMEs : was RE: Party XML Schema Defintions

Thought the ebXML crowd would be interested in a parallel naming convention discussion taking place in Doug Englebart's Bootstrap project.
Sandy

----------
From: SNighthawk@aol.com
Reply-To: ohs-dev@bootstrap.org
Date: Fri, 9 Feb 2001 01:40:50 EST
To: ohs-dev@bootstrap.org
Subject: Re3: Use Cases, requirements, etc. posted

In a message dated 2/8/01 8:57:14 PM, ncarroll@hastingsresearch.com writes:

>On further poking around, I'm concluding that if one has an established
>alphanumeric naming convention for a given DKR (like the OAD) -- AND an
>intelligent mix of letters with the numbers, including the judicious use
>of upper and lower case letters -- users will be able to remember hundreds
>of nodes in their own DKRs.

Actually, this was the way our phone nomenclature system was originally 
systematized, after much research in memory at Bell Labs (remember The Bell 
Telephone Company).  They found that the optimum amount of easily remembered 
numbers was around 7, and that this was with the 3/4 grouping.  It was 
increased to 10 with the 3 digit area code prefix, figuring that the area 
code was seldom used anyway (at that time).

Also, most phone systems decades ago used a combination of letters and 
numbers in subscriber identification (phone numbers).  It was common to see 
phone "exchanges" such as "Liberty 2" (LI 2) or Walnut 5 (WA 5).  Instead of 
all numbers, a phone number would be expressed as WA 5-3491.

This gave a sing-song, almost rhythm or lilt to the way a phone number was 
said.  Often, advertisers used this phenonomenon in jingles, which were also 
very popular in years past.  Phone numbers were sung along with the rest of 
the lyric for client identification, and it lent itself very well to memory.  
Businesses would pay extra for phone numbers that rhymed or otherwise went 
well with this effect.

It has only been in the past 25 years or so that all-numeric phone numbers 
have replaced the combination alpha-numeric system.  Even then, it took many 
years to phase in the "new" system.  Now, in some parts of the country, it is 
necessary to "dial" the area code as well as the 7 digit numeric sequence to 
get your connection, making a total of 10 numbers "dialed".

As with the term "dial", much of our nomenclature is inherited from times 
when the words actually meant something, as with the telephone dial, 
something which has all but disappeared.  Even now, most telephones have a 
switch to select between tone and pulse "dialing", or switching.

And not to stray too far from the topic, the 7 digit standard was 
incorporated into many intelligence tests.  For example, in the WAIS, or 
Wechsler Adult Intelligence Scale, this was considered about "average" in the 
Digit Span subtest, and given a statistical weighting of such.

Thanks,

---Sheldon



> From: martin.me.roberts@bt.com
> Date: Fri, 09 Feb 2001 09:15:27 +0000
> To: ebxml-core@lists.ebxml.org
> Subject: RE: SMEs : was RE: Party XML Schema Defintions
> 
> Dear all,
> The data seperation from UI is good but I get a bit worried when you
> end up with IDs being the main way of distinguishing an construct.  Would
> these be numbers? who would issue them? etc become issue.  If however you
> said that the ID was the initial semantic bearing name such as ( at this
> point I dive into semantic soup and drown! perhaps ID are a good idea after
> all.!!!:)
> 
> Martin M.E. Roberts
> xml designer, BTexaCT
> 01473 643775
> martin.me.roberts@bt.com
> 
> 
> -----Original Message-----
> From: Gregory, Arofan [mailto:arofan.gregory@commerceone.com]
> Sent: 08 February 2001 21:12
> To: 'Joaquin Miller'; ebxml-core@lists.ebxml.org
> Subject: RE: SMEs : was RE: Party XML Schema Defintions
> 
> 
> Folks:
> 
> At the Orlando meeting, Harmut suggested that we have non-semantic-bearing
> IDs for each construct, and then "official", language-specific
> human-readable tags for each language, with an English-language version
> being one of the first outputs, but with others constrcuted along the same
> lines. This approach lets you build applications to either the unique IDs
> (for multi-language support) or to build applications that support only a
> single, language-specific version. Presumably, the human-readable tags would
> have their own consistent rules about providing names, but these rules
> themselves might be prone to localization(!)
> 
> This is a good, flexible approach that would seem to offer the most in terms
> of handling localization problems, and I suggest we follow it.
> 
> (Harmut - sorry if I have mangled your suggestion - if I am wrong, please
> feel free to correct).
> 
> Cheers,
> 
> Arofan Gregory
> 
> -----Original Message-----
> From: Joaquin Miller [mailto:miller@joaquin.net]
> Sent: Thursday, February 08, 2001 12:48 PM
> To: ebxml-core@lists.ebxml.org
> Subject: RE: SMEs : was RE: Party XML Schema Defintions
> 
> 
> At 12:09 PM 2/8/2001 -0800, Hayes, Brian wrote:
> 
> 
> It is general considered to be good architecture to seperate user interface
> issues from data issues: Your user friendly user interface sould not be
> displaying XML tag and attribute names.
> 
> 
> Absolutely!  I could not agree more.
> 
> It's why i wrote: "Everyone can use software to display the data with field
> names they can read and to provide explanations for what those field names
> mean."
> 
> I side with those who suggest using identifiers.  I feel it is a much better
> way than hanging our hat on natural language tags, whether English or
> so-called "Foreign."  I don't want to repeat here all the traffic about the
> problems with dependence on exclusive use of natural language tags.
> 
> 
> 
> 
> -----Original Message-----
> From: Joaquin Miller [ mailto:miller@joaquin.net <mailto:miller@joaquin.net>
> ]
> STEPHENIE COOPER wrote:
> 
> ebXML's goal should be to help make that happen.
> 
> Yes.  Let's go for the whole planet.  Not just our company's customers or
> partners.  And let's go for the people who speak Foreign, too.  Everyone can
> use software to display the data with field names they can read and to
> provide explanations for what those field names mean.  There has been a lot
> of recent eMail traffic about how to define ebXML in order to make that easy
> for everyone.  Let's do it.
> 
> 
> 
> Cordially,
> 
> Joaquin 
> 
> .................................................
> Joaquin Miller
> Chief Architect
> Financial Systems Architects
> 
> mailto:joaquin@acm.org <mailto:joaquin@acm.org> 
> 
> San Francisco
> phone: +1 (510) 336-2545
> fax:   +1 (510) 336-2546
> PGP Fingerprint:
> CA23 6BCA ACAB 6006 E3C3 0E79 2122 94B4 E5FD 42C3
>