[sc34wg3] Individual contribution on the U.S. N.B. position onthe progress ion of Topic Map standards

[Jan Algermissen]

Dmitry wrote:

> If I were to develop "merging" theory I would not go RM route.

Was do you mean by "not go RM route"?

Why would you not do that?

Please forgive me the sloppy language below, no intention to be rude!

> Firstly, I would introduce  TMDM Relaxed (TMDM-R). TMDM-R is TMDM with relaxed merging rules.
> 
> I would introduce "identity tag".  

Congratulations! You have just defined an SIDP ;-)


> Each TMDM-R construct can have several identity tags.
> 
> I would introduce main identity axiom:
> 
> "If two TMDM-R constructs have at least one identity tag in common these constructs are identical."

Ah...a perfect RM conformant merging rule!

> 
> I would allow defining identity identification rules. Each identity rule transforms TMDM-R instance into new TMDM-R instance with
> new identity tags.
> 
> And I would allow defining merging implementation rules. Each merging implementation rule transforms TMDM-R instance into new TMDM-R
> instance. Implementation rule defines what it means to be merged for a specific type of TMDM constructs.

Hmm...what do you need these rules for? Why not simply say: when merging two subject surrogates
(topics in RM sense) just unite all characteristics.

> 
> These transformation can be described in a very formal manner.
> 
> Using these mechanisms I can (informally) describe "not relaxed" TMDM as following:
> 
> Identity Identification Rules:
> 
> IDRule 1:
> If topic X1 has Subject Identifier S
> And topic X2 has Subject Identifier S
> Then assign X1 the same identity tag as X2
> attach justification based on this rule to identity tag
> 
> IDRule 2:
> If topic X1 has Address A
> And topic X2 has Address A
> Then assign X1 the same identity tag as X2
> attach justification based on this rule to identity tag
> 
> IDRule 3:
> If topic X1 has SourceLocator L
> And topic X2 has SourceLocator L
> Then assign X1 the same identity tag as X2
> attach justification based on this rule to identity tag
> 
> IDRule 4:
> If topic X1 has base name N1 in scope S
> And topic X1 has base name N2 in scope S
> And value of N1 is equal value of N2
> Then assign N1 the same identity tag as N2
> attach justification based on this rule to identity tag

Hmm...why is this superiour to what the TMDM already says?
The effect is the same.


Jan


 
> ...... <Other rules>
> 
> Merging Implementations Rules (they create new TM instance) :
> 
> IMRule1:
> 
> If  X1 is a topic in TM with identity tag T
> and X2 is a topic in TM with identity tag T
> then
> retract topic X1
> retract topic X2
> add new topic T into TM
> copy all constructs related to X1 to T
> copy all constructs related to X2 to T
> record fact of merging and snapshot of information about X1 and X2 for unmerging
> 
> IMRule2:
> 
> If  N1 is a base name in TM with topic X , scope S,  value V and identity tag T
> and N2 is base name in TM with topic X, scope S,  value V and identity tag T
> then
> retract name N1
> retract name N2
> add new name N into TM with  scope S and value V
> record fact of merging and snapshot of information about N1 and N2 for unmerging
> 
> .......<Other rules>
> 
> Dmitry
> 
> _______________________________________________
> sc34wg3 mailing list
> sc34wg3@isotopicmaps.org
> http://www.isotopicmaps.org/mailman/listinfo/sc34wg3

-- 
Jan Algermissen                           http://www.topicmapping.com
Consultant & Programmer	                  http://www.gooseworks.org