IWSM 2018 Forums Sizing Sizing – FPA Entity change in enhancement projects

This topic contains 2 replies, has 2 voices, and was last updated by  Martin Jacobs 3 years, 11 months ago.

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
  • #2265

    Ye olde Phorum

    A key-key entity is enriched with (functional) attributes and therefore changes from a dependent entity to an independent logical file. So this change results in a new ILF/EIF. Does this only affect the new ILF/EIF or do the ILF/EIF that have been linked by this entity?
    According to section the referring attributes of the key-key entity should be counted with the linked logical files. Now the key-key entity has become an ILF/EIF these DET should no longer be part of the linked ILF/EIF. Should the linked ILF/EIF be counted as changed due to the change of the key-key entity?

    Is the answer different for an estimated count (where we don’t pay attention to individual DET)?

    Originally posted in Dutch by Andreas Schuderer
    Een sleutel-sleutel-entiteit wordt met (functionele) attributen aangevuld en verandert daardoor van een niet bestandszelfstandige entiteit naar een bestandszelfstandige entiteit. Hierdoor ontstaat dus een ILGV/KGV. Wordt hier alleen maar een ILGV/KGV toegevoegd of wijzigen daardoor ook de ILGV/KGV die aan die de sleutel-sleutel-entiteit gekoppeld zijn geweest? Volgens de telrichtlijn moeten namelijk de verwijzende attributen van de sleutel-sleutel-entiteit bij de verbonden gegevensverzamelingen worden geteld. Door het zelfstandig worden van de sleutel-sleutel-entiteit leidt de telrichtlijn tot het verdwijnen van 1 DET van ieder verbonden ILGV/KGV. Moeten dus naast de toegevoegde ILGV/KGV ook twee gewijzigde ILGV/KGV geteld worden? En geldt hetzelfde bij globale tellingen (waar meestal niet op de DET gelet wordt)?


    Martin Jacobs

    Let’s analyse this situation.
    The two original entities A and B are independent, so we have optionality at both sides of the n:m relation (so in fact it is A (n) : (m) B). we have two ILFs/EIFs: A and B.
    Assumption: enriching the key-key entity does not change this and there are no business rules stating otherwise.
    So yes, a new independent ILF/EIF C emerged and should be counted. This is not the issue however.
    Now let’s have a closer look at A and B.
    A is related to C and not to B directly anymore. The same goes for B. One relation was replaced by another one.
    If A would get a new independent child, then we would not count A as changed as in fact there is no change. In this situation it doesn’t matter whether this child is new or existing.
    In my opinion, in case of a key-key entity enrichment, the parents should be counted as changed, because of the counting guideline
    the paragraph 4.21.2 is also used to identify changes, not just to decide on complexity.
    Other views invited.


    Martin Jacobs

    Originally posted in Dutch by Andreas Schuderer

    Hello Andreas,

    It has been some time…
    Did my answer satisfy?


Viewing 3 posts - 1 through 3 (of 3 total)

You must be logged in to reply to this topic.