2008/6/26 Benoit St-Pierre <[EMAIL PROTECTED]>:
>
> PGN tags are handled as strings. With this layout, each game will have an
>> UID which will cost around 20 chars. So the sg3 file is increased by 60 MB
>> for a 3 M games DB. So the average number of games in each block will
>> decrease, leading to an overall search penalty (more I/O). Note that events,
>> white and black players tags for example are stored as references in the
>> file sn3, because there are some repetitions. Of course UID does not repeat
>> ... By experience I saw that games are around 100 bytes (often less), adding
>> UID, even in binary compressed form, would increase the average size of
>> games by a non neglictable percentage.
>
>
> That means that Ids should be treated like a flag, considering Scid's way
> of storing things. Or that that the indexed information should build up the
> Id.
>
Technically, it is not feasible.
[...]
>
> Moreover I prefer the reference "Spassky Fischer 1972, round 3" than "game
>> my_huge_ref_db:v12-12345678", because if I copy a game from a base to
>> another "Spassky Fischer 1972, round 3" is still valid. Not "game
>> my_huge_ref_db:v12-12345678".
>>
>
> What do you do if you have the Gligoric annotated game, and the Kasparov
> one, your Togofied one, and all the drafts about the game ? What do you do
> if you have ten snapshots (fragments) from this game ?
>
This is a good example demonstrating UIDs are useless : everybody would
prefer to set an "Annotator" tag and even comments, far easier to understand
and manage than a game with same tags but different UID.
[...]
Pascal
-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php
_______________________________________________
Scid-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scid-users
