Hello
First example:
I have two databases A and B, a game list window and a tree window for A.
I create a new game in A and move 1.b2-b4. The games are filtered and the
statistics
for possible responses is shown. Now I switch to B, create there also a new
game and
move 1.g2-g4. The tree window is updated, the game list window still shows
all games
for B. Now I switch back to database A and then on the board there is still
1.g2-g4
and my first new game is gone and nobody asked me before. Ok. here it was
only one
move, but it could be more...
Second example:
I have a database and a tree window for it.
I put in some moves and see the games with this position. Now I want to see
the Kasparov
games with this position from 1980 to 1985. I can use Find Text at the
bottom of the game
list but this finds only his first game. So I can set up a header filter and
find the games.
There are 20 so I do an additional move on the board. Now the header filter
is lost and I
have to create it again...
Now think about the third example combining the first two. I guess there are
not many users
still knowing what is going on. I think mixing up the header filter with the
board positions
creates unpredictable situations.
So my idea is:
1. If you have a game loaded in database A and B, you switch between these
games when you
switch the database. It is not possible to see the game of A, if B is the
actual database.
No data loss anymore.
2. The game listing of each database, each tree window has its dedicated
filter. For the game window
The filter consists of the same parts as today. Moving a peace on the
board does only update the
position filter if activated and keeps the others. For the tree window
the filter can be a just a header
filter. So if the user switches the database from A to B the game
listing shows the games with the filter
of B. No global filter anymore.
So I can set up a header filter for my game list window to view only
Kasparov games between
1980 and 1985 and then I can move on the board and see still only his
matching games. If I have
a tree window open, I can decide whether the games in the game window are
filtered also.
I can set up a header filter for the tree window to limit the statistics to
players with
a rating of 2400 or more. Today I maintain a top database with these games
to get a
reasonable statistics for a position. At the same time I have to maintain a
complete
database of 4.500.000 games for preparation of potential opponents. So I can
do both with
the same database and the statistics calculation would ignore 70% of its
games and still
be fast. The current Find text window would no longer be needed and could be
replaced by
a button to open the filter. The resulting behavior of the system is
predictable.
Each browser has its own filter, resets are not done sometimes somewhere in
the background
but only by the user himself.
If there is an agreement to go for this and somebody updates the UI parts,
I will try to
update the C++ parts.
Comments are wellcome
Gerd Lorscheid
------------------------------------------------------------------------------
This SF.net Dev2Dev email is sponsored by:
Show off your parallel programming skills.
Enter the Intel(R) Threading Challenge 2010.
http://p.sf.net/sfu/intel-thread-sfd
_______________________________________________
Scid-users mailing list
Scid-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scid-users
