How deep?

[david_c]

I'm wondering what the aim is for these standard system files. How many sequences are they supposed to describe? I can see three three different options:

 

1. Aim to define everything. The file will be constantly added to as more and more sequences are defined. If we come across a call where there is some doubt about what it means, then we will have a discussion about it, decide on the proper meaning, and put that in.

 

2. Aim to define only the sequences that all players of the system would agree on. This would include sequences which are deep enough that no-one would ever have discussed them with their partner, but which have only one logical meaning.

 

3. Only define those sequences which are an intrinsic part of the system (i.e. those which are defined in the system notes.)

 

My preference is for the third option: if the idea is that these files will be distributed to all BBO users, then I don't think it should claim that they have an agreement when they don't. But on the other hand, producing a detailed system file would be useful for other purposes.

[fred]

I'm wondering what the aim is for these standard system files. How many sequences are they supposed to describe? I can see three three different options:

 

1. Aim to define everything. The file will be constantly added to as more and more sequences are defined. If we come across a call where there is some doubt about what it means, then we will have a discussion about it, decide on the proper meaning, and put that in.

 

2. Aim to define only the sequences that all players of the system would agree on. This would include sequences which are deep enough that no-one would ever have discussed them with their partner, but which have only one logical meaning.

 

3. Only define those sequences which are an intrinsic part of the system (i.e. those which are defined in the system notes.)

 

My preference is for the third option: if the idea is that these files will be distributed to all BBO users, then I don't think it should claim that they have an agreement when they don't. But on the other hand, producing a detailed system file would be useful for other purposes.

I think (and hope) that these files will continue to grow and evolve, but I agree with you that achieving 3) is a reasonable goal to begin with.

 

Another reasonable approach would be to try to define all constructive auctions through responder's rebid and through (say) 3NT. This sounds like a managable amount of work, even for one person. Furthermore, I suspect that our "average member" could learn plenty if such a file were available for a system like SAYC.

 

Fred Gitelman

Bridge Base Inc.

www.bridgebase.com

[Gerben42]

Agree that having them as large as possible is great.

I will start to get WJ2005 up to (3). After that, let's see.

 

Then there are of course some changes when you look at 2000 and 2005. What does "merge" do in case of conflicts? Is it possible for example to have a file for 2005 and then write a "differences" file for 2000, overwrite and presto! WJ2000.bss?

[david_c]

Well, to take an example, there was a discussion recently about the sequence 1♣:1♦,1♥:1NT,2♥:3♦ in WJ05. I think it's fair to say that there wasn't any definite conclusion about what this sequence meant. If the files are to keep growing, then eventually you're going to have to look at sequences like this. But personally I think that "no agreement" is the best explanation. Even for someone who is reading the file in order to learn the system, they ought to be aware of which bids are clearly defined and which aren't.

 

How about producing two different types of file, say for example:

WJ05 - basic system.

WJ05 - detailed agreements discussed on BBF.

[badderzboy]

Merge overwrites the old records with the new ones so specific new auctions take preference.

 

This means with the ACOL one I've been playing with you create the Benji bids as a seperate file and it will overwrite the old one including the weak2s this is great as it allows a modular build and the introductions of conventions to certain bids like puppet stayman / Jacoby 2NT / Ogust / Feature Asking.

 

The same goes with correcting system files if people post a correction file you can merge it into the old one and correct the bids in one easy step!

 

You then need to think how deep to go after conventional responses...

 

One thing I've learnt is how much I've never thought what certain sequences should mean and I've learnt things on route!

 

Cheers

 

Steve

[Chamaco]

Another reasonable approach would be to try to define all constructive auctions through responder's rebid and through (say) 3NT. This sounds like a managable amount of work, even for one person. Furthermore, I suspect that our "average member" could learn plenty if such a file were available for a system like SAYC.

This would become something very similar (and likely better) to Alan Truscott's "Bidding dictionary"

[hrothgar]

I think that its appropriate to distinquish between two different types of covention files.

 

The first set includes the personal convention files created and maintained by individual players. Players should feel to document these file to whatever depth that they feel is appropriate. Recall: The primary purpose of Full Disclosure is to make our lives easier by automating Alert Strings and having tools to teach new partners.

 

The second set of convention files includes those for "Official" systems like Bridge World Standard. This is where life starts to get complicated... I firmly believe that we would all be best served if there was one and only one Convention File for each official system. Consider whats happened with SAYC. SAYC has transitioned from a reasonable well defined system to a generic phrase meaning "I play 5 card majors and Stayman". No one has a bleeding clue what bread and butter sequences like "1m - (P) - 2N" or, god forbid "1♦ - (P) - 2♣ - (P) - 2N". I'd hate to repeat this same mistake with the new convention files.

 

In an ideal world, official responsibility for maintaining official convention files should fall on those organizations that "own" the actual systems. Case in point: The Bridge World has assumed responsibility for designing and maintaining Bridge Wworld Standard. Long term, I believe that they are the only ones who should be able to create a convention file labelled Bridge World Standard . In a similar fashion, I would hope that the ACBL would maintain a file for "Standard American", the EBU would create Standard English, etc. (Its also possible that these organizations would decide to delegate responsibility, but thats another thread)

 

I certainly believe that players should be able to create their own customized versions of individual bidding systems. Case in point, the MOSCITO variant that I prefer differs quite a lot from the one that Paul Martson currently plays. Accordingly, when I label the file I should probably refer to it as "Willey MOSCITO" or some such.

 

For anyone who still cares, I'm (essentially) suggesting that basic concepts such as Trademark are potentially appropriate in this arena...