[an error occurred while processing this directive] [an error occurred while processing this directive][an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive] (none) [an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive][an error occurred while processing this directive] [an error occurred while processing this directive][an error occurred while processing this directive] [an error occurred while processing this directive][an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive] (none) [an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive][an error occurred while processing this directive]
 
[an error occurred while processing this directive] [an error occurred while processing this directive]
Skåne Sjælland Linux User Group - http://www.sslug.dk Home   Subscribe   Mail Archive   Forum   Calendar   Search
MhonArc Date: [Date Prev] [Date Index] [Date Next]   Thread: [Date Prev] [Thread Index] [Date Next]   MhonArc
 

Re: [LOCALE] Test-database design



>> På http://speling.chin.dk/dbtables.html er mit første
>> udkast til indhold i databasen.
> Der er mig bekendt ikke noget sprog med sprogkoden "uk".
> Der er "da", "da_DK", "en", "en_UK" (eller er det "en_GB"),
> "en_US", m.fl.

Jeps, kunne ikke lige huske det... :-)

> All words
> =========
>
>  1) Technically the field name "word" is inappropriate,
>     since we should also be able to keep information about
>     possible misspellings in the database.  Or should they
>     be kept elsewhere?  Where?

Jeg mener at det må være i en af "stemme"-tabellerne (editions/changes). Når
der er kommet tilstrækkeligt med stemmer mod et ord, skal det vel bare
fjernes?

>  2) With the field "word_class" you assume that one from the
>     first time a word is added to the database has certain
>     knowledge of which word class a given word (string)
>     belongs to.  If we add "misspelling" and "unknown" to
>     the table "word_classes" one can argue that it is okay.
>     But it removes the option of letting the reviewers vote
>     for or aginst the classification.  Or is it your idea
>     that each reviewer should add the word (string) to the
>     table with the classification that he finds correct?

OK, jeg tror at vores opfattelse af reviewers er lidt forskellig. Jeg mener at
der skal være to ting man skal kunne:
1) Tilføje nye oplysninger om hvert ord
2) Stemme om de givne oplysninger er korrekte

Jeg regner ikke med at man fra start af ved noget om ordklassen. Det skal man
tilføje senere hen. Og dernæst kan man stemme om det.

>  3) What kind(s) of operations will there be on this table
>     when reviewers add or update information in the
>     database?  And what kind(s) of operations will be batch
>     oriented?

Der er intet i denne tabel om adds/updates. Alt sådant vil komme ned i
editions/changes der lige nu ikke er særlig godt beskrevet i min beskrivelse,
det vil jeg gøre når jeg har et system oppe og køre... :-)

Som det er nu skal der være én "change" række for hver addition. Dermed vil en
batch komme til at fylde N rækker, hvor N er antallet af additions. Lidt dumt
måske, men som jeg skriver i starten er det udelukkende for at noget op at
køre, det vil være relativt nemt senere at konvertere forhenværende batches
til mere simple strukturer, da hvert changes-række peger på den samme edition
for én batch (faktisk er det ikke præcis sådan nu, men jeg sætter
editions.word_key-feltet ned i changes.word_key i stedet, så passer det :-)).

> Descriptions
> ============
>
>  1) Should every reviewer add his preferred description for
>     each word?  Or how can the reviewers vote for a
>     description for a word?  And how can reviewers vote
>     against a description for a word?

Dette havde jeg også tænkt på, jeg havde regnet med at alle kunne lave en
beskrivelse, og at alle kunne stemme på alle beskrivelser. Så vil man i et
orbogs-udtræk eller andet tage den beskrivelse der har fået flest positive
stemmer. Men alle forslag er velkomne.

>  2) Should all strings (including possibly misspelled words)
>     be allowed to have descriptions associated with them?

Det er jo op til os alle. Hvad synes I andre? Man kunne jo sagtens sige at det
ikke var tilladt at tilføje nye oplysninger før et ord havde fået et vist
antal "ok"-votes?

>     Since a description can be added before a reviewer
>     notices that a word is incorrect, the answer to this
>     question is most likely "yes".

Dette har jo ikke noget med databasen at gøre som jeg ser det, det er bare
logik i grænsefladen...

> Synonyms
> ========
>
>  1) Why are reverse (or "redundant") synonym links not
>     allowed in the table?

Ingen grund til at have det. Hvis synonymet allerede er indsat er der jo ikke
nogen grund til at indsætte det igen??? Eller har jeg misforstået dig?

>  2) The problem of allowing reviewers to vote for or against
>     information in the table has been ignored.

Sådanne ting skal være i editions/changes-tabellerne.

> Language identifiers
> ====================
>
>  1) "dk" is a country code, not a language code.

Merci... :-)

>  2) Why not handle "international name" by looking up the
>     name of a language in the relevant dictionary?  If you
>     need the Italian (it) name for Danish (da), then you
>     look up what Danish is called in Danish in a language
>     information database and find the word "dansk".  In the
>     same database you can also look up a reference to a
>     Danish dictionary (for example "da.speling.org").  You
>     can then look up the Italian (it) translation of the
>     word "dansk" in the dictionary "da.speling.org" and find
>     the word "danese".

Det kræver at vi har alle sprogenes databaser. Det synes jeg er lidt overkill
for disse få oplysninger.

>  3) Is it really necessary to use an integer and not the
>     language identifier itself as the key in this table?

Nej. Det går faktisk imod mit formål med maksimalt informationsindhold med
minimalt antal parametre... :-) Jeg retter det.

> Det lyder fint nok.  Men husk på at korrekturlæserne ikke
> bare retter i ordlisten.  De stemmer bare på oplysningerne.

Hmmm.... Skal det så ikke være muligt at tilføje ord, ordklasser, osv for
"almindelige" brugere?

Med venlig hilsen Preben

-- 
Don't take life too seriously. You'll never get out of it alive.




 
Home   Subscribe   Mail Archive   Index   Calendar   Search

 
 
Questions about the web-pages to <www_admin>. Last modified 2005-08-10, 20:53 CEST [an error occurred while processing this directive]
This page is maintained by [an error occurred while processing this directive]MHonArc [an error occurred while processing this directive] # [an error occurred while processing this directive] *