[hrothgar]

inquiry, on Nov 22 2005, 04:38 PM, said:

I guess this as good a time as any to discuss a "sticky" issue for volunteers who are working on FD cards for us.

Just because you or a small group of people working with you develop a completed FD card and submit it, even if we begged you to do so, doesn't mean we will accept it as written.

What distinquishes FD files that are being developed by BBO from "random" files that are being developed privately?

From my perspective said differentiation could be based on a number of different criteria:

"Official" files are hosted centrally on BBO's server. "Rogue" files reside on local clients or third party web sites.

"Official" files comply with some presumed standard: This standard could be related to nomenclature or other descriptive rules. Alternatively, the standard could be based on some subjective opinions regarding the "quality" of the submission.

For what its worth, I submitted a MOSCITO FD file because Fred expressed interest in getting some example code. I have no interest in jumping through hoops to make sure that said file is officially blessed by BBO. As long as there is some way that the opponents are able to access the FD file that I have resident on my machine, I'm relatively content.

For what its worth, I think that the FD file process is being approached in the wrong way.

1. As I've noted many times in the past, centralized systems don't scale well. I'm very much confused why BBO seems intent on creating one.

2. I understand what you are trying to do with the Conventions directory. Modular design is a great thing. With this said and done, modular design is typically associated with relatively mature products. I don't think that FD or the FD Editor is nearly powerful enough to build processes which assume modular system design.

3. I think that its too early to be focusing so much on "Process". There is certainly a time and place for process. Lack of process / planning can be incredibly expensive. With this said and done, the amount of Process that you have is also a function of the maturity of the product.

At this point in time, we need FD files out there being used. Yes, we want them to be "good". However, imposing early standardization based on your own assumptions abot the right way to describe a system strikes me as a major mistake.

In short:

I think that you should be spending a lot less time worrying about other people submissions and lot more writing some FD files. If / When you do decide to write up "BBO Basic", don't present this as a modular system. Your going to have enough trouble introducing FD without building in another layer of complexity. Instead,

Provide a single difinitive version of the system.
Make sure that the FD file is accompanied by VERY good notes
Make sure that the major teaching programs have bought into the system

Let a Thousand Flowers Boom.

[inquiry]

Well, richard, I am busy writing a very vanilla version of "standard american" myself, since no one has stepped up for this version. The one I am working on uses Game Forcing jump raises (not limits, etc). This is, for lack of a better term, the one my mother plays. This will be for people who don't read books, don't play in sectional tournments etc.

Clearly Wayne and I can't be writting all the FD files for a lot of reasons. One, as much as we think we know, we don't "know" French Standard, Moscito (well Wayne has experience with that one), and others. I think with all the Polish club players, polish club is necessary. I think Acol is necessary, I think SAYC is necessary, and as I said, at least two versions of 2/1. Not to insult our Australian friends, I am not so sure that Moscito is necessary. I certainly will not lose sleep if it is not part of the standard "systems", however, if a version can be created that fits with views of a group of moscito players (who are not parnters), then including it (or any other system) would be fine. I mean, if someone wrote up ultimate club in all its glory, as complex as that is I would have no objections to including it either.

But, necessarily, our goal is to get FD files for systems that are used most commonly: clearly this is SAYC, 2/1, Polish club, and precision. There will never be agreement between large groups of players what should and should not be included in any one of these systems. Should we go with simple Polish club? WJ2000, WJ2005? etc. In the case of moscito, your version is different from Paul Marston's, and no doubt other peoples. Opening this up for a voting contest simply will not work. Wayne and I are going to do the best we can to try see that a playable, basic version of each system gets approved.

Let me address some of your issues...

Quote

1. As I've noted many times in the past, centralized systems don't scale well. I'm very much confused why BBO seems intent on creating one.

Creating FD files is a time-consuming, laborous process. Users didn't even bother to fill out the current "easy" convention cards. There is no reason to suspect they would fill out an FD card. By providing completed FD in the most common systems, a few things happen, it makes it easy for users to pick their system (even if the card has some differneces in it), it makes it easy for them to modify those differences or add conventions from the pick list. Ideally, most people will be able to fairly quickly get a reasonably accurate version of what they are playing.

Quote

2. I understand what you are trying to do with the Conventions directory. Modular design is a great thing. With this said and done, modular design is typically associated with relatively mature products. I don't think that FD or the FD Editor is nearly powerful enough to build processes which assume modular system design.

The idea is, of course, to include options. A user could choose Bergen Raises for example, and one thing it should change is 1M-3M to "preemptive" or "weak" as well as the meaning of 1M-3C and 1M-3D and all the subsequent auction on both 1M-3M and 1M-3m (not just describe what 1M-3m means).

Quote

3. I think that its too early to be focusing so much on "Process". There is certainly a time and place for process. Lack of process / planning can be incredibly expensive. With this said and done, the amount of Process that you have is also a function of the maturity of the product.

On this we disagree. The next major release of BBO client willl be fully FD aware. To make it useful, there has to be standard "system" FD cards people can use. Since these cards are on the system, we can (I assume) correct any glaring mistakes we make in those over time. But we NEED these CC before the next release, at least IMHO, and more importantly in FRED's opinion. IF you have played at tables where the FD auto alerts are done, you will have found how much better that is, how much less typing you need to do, and how much faster and pleasant the game is. Having said that, if the FD system alerts your bid as one thing, and you have a completely different agreement, problems will exist. This we know.

[hrothgar]

inquiry, on Nov 22 2005, 06:28 PM, said:

Not to insult our Australian friends, I am not so sure that Moscito is necessary. I certainly will not lose sleep if it is not part of the standard "systems", however, if a version can be created that fits with views of a group of moscito players (who are not parnters), then including it (or any other system) would be fine. I mean, if someone wrote up ultimate club in all its glory, as complex as that is I would have no objections to including it either.

MOSCITO is obviously a fringe bididng system. 99% of BBO members couldn't care less if there was a MOSCITO FD file available. .9% think that the system should be banned. Then there is the .1% who play it or might want to experiment with it. I'm certainly not suggesting that there is any need for BBO to formally support the system.

Where I get confused is all this talk about "Including" MOSCITO. Is there any formal definition regarding what it means to be "Included"?

More importantly: Assume for the moment that MOSCITO is not "Included", but I have a complete FD file documenting MOSCITO sitting on my hard drive. Can this file be used to automate alerts?

[hrothgar]

inquiry, on Nov 22 2005, 06:28 PM, said:

The idea is, of course, to include options. A user could choose Bergen Raises for example, and one thing it should change is 1M-3M to "preemptive" or "weak" as well as the meaning of 1M-3C and 1M-3D and all the subsequent auction on both 1M-3M and 1M-3m (not just describe what 1M-3m means).

As I said, I understand the idea behind a modular structure. However, I think that it would be a major mistake to introduce modular FD files on the first pass.

Modular designs are more complex and take more time to implement.
Neither feature is particularly desirable during an initial roll out.

Even if the underlying achitecture is modular, I'd strongly recommend suppressing these options in order to simplify the presentation of new material

[inquiry]

hrothgar, on Nov 22 2005, 10:42 AM, said:

More importantly: Assume for the moment that MOSCITO is not "Included", but I have a complete FD file documenting MOSCITO sitting on my hard drive. Can this file be used to automate alerts?

Yes. When the next client is released, you can load your own FD files (that means, they don't have to be part of the "standard" ones on line). The system will use your file and alert just fine. This works today for those of us using the latest beta...but if our opponents lack the beta version, they will not see the alerts.

I happen to ahve a full FD version of my system, complete with MisIry, New minor forcing by opener, a fully populated ETM Vicotry over !NT (complex version). None of this will ever make it to BBO Stanard, either as system or convention, as there simply isn't enough interest in it. However, I will load it up when I play with the wacky few willing to use my prefereed bidding agreements.

[cherdano]

hrothgar, on Nov 22 2005, 05:42 PM, said:

Where I get confused is all this talk about "Including" MOSCITO. Is there any formal definition regarding what it means to be "Included"?

More importantly: Assume for the moment that MOSCITO is not "Included", but I have a complete FD file documenting MOSCITO sitting on my hard drive. Can this file be used to automate alerts?

Certainly. Not being "included" just means that it won't be part of the system files downloaded with every version of BBO, and presented as a ready made option when choosing a system file.
Everybody will still be able to download Wiley-Moscito.bss from some place on the internet and use it for play on BBO.

Arend

[inquiry]

cherdano, on Nov 22 2005, 10:49 AM, said:

hrothgar, on Nov 22 2005, 05:42 PM, said:

Where I get confused is all this talk about "Including" MOSCITO.  Is there any formal definition regarding what it means to be "Included"?

More importantly:  Assume for the moment that MOSCITO is not "Included", but I have a complete FD file documenting MOSCITO sitting on my hard drive.  Can this file be used to automate alerts?

Certainly. Not being "included" just means that it won't be part of the system files downloaded with every version of BBO, and presented as a ready made option when choosing a system file.
Everybody will still be able to download Wiley-Moscito.bss from some place on the internet and use it for play on BBO.

Arend

I think you can save your opponents convention cards.

[luke warm]

i understand what richard is saying re: modular development, but i don't understand why it is a bad thing, even at this stage of the game... for example, i'm almost through with a game forcing 2♦ stayman convention, not that i think many will use it but because i think *some* will (and more will want to experiment with it, given the number of weak nt players)...

i've focused more on conventions, myself... this is because there isn't a lot of agreement as to what should make up a complete system (ben's sayc example comes to mind)... that way, it seems to me, a pair can agree 2/1 in principle and add bergen raises, etc if they want

of course i could be way off base with my understanding of what's wanted/needed

[hrothgar]

luke warm, on Nov 23 2005, 02:58 AM, said:

i understand what richard is saying re: modular development, but i don't understand why it is a bad thing, even at this stage of the game... for example, i'm almost through with a game forcing 2♦ stayman convention, not that i think many will use it but because i think *some* will (and more will want to experiment with it, given the number of weak nt players)...

I can picture an "ideal" world in which players can build a logically consistant system by assembling a series of seperate modules. For example:

I would start by selecting one of a variety of chaisses:

Option 1: Strong Club, 5 card majors, 13-15 NT
Option 2: Strong 2♣, 5 card majors, 12-14 NT
Option 3: Three way 1♣, 5 card majors, 15-17 NT
Option 4; Strong club, transfer openings, 11+ - 14 HCP NT

After which, I start adding on a few crucial conventions

Keri over 1NT openings
2/1 Game Force + Forcing NT
Gazilli
Transfer advances after overcalls
Yada, yada, yada

Stir it all together, add in a bit of love, and voila...
A complete Full Disclosure file is available

This is a wonderful extrapolation of what COULD be done using FD type technology. However, for now its a pipe dream. The chaisses aren't available. The convention modules aren't there. The GUI does have any of the necessary heirarchies build it. (How do we explain that 2 way Stayman, Keri and Scanian are all NT response strutures, but Bergen raises are used over Major suit openings?) I suspect that that for this type of system to work, the FD file format needs to become a lot more complex. (You might want to include vaiables such that ranges change based on whether people are playing a weak or a strong NT).

In short: It's not ready for prime time... I think that wide spread deployment of this type of complex functionality would be a recipe for disaster. I have no problem if "Power Users" want to experiment with modular system. However, the success of FD depends on widespread adoption by the great unwashed masses... (You know: The folks who find it too difficult to load the current Convention Cards). Providing these people with a complex system doesn't benefit anyone.


"There is nothing more difficult to take in hand, more perilous to conduct, or more uncertain in its success, than to take the lead in the introduction of a new order of things. " - Machiavelli