Poker Hand XML

[TSchauenberg]

[ QUOTE ]

* check = call with amount = 0

I kindda like having a raise and bet be two different elements but I have changed it back and forth in earlier drafts.


[/ QUOTE ]

Also, actions like "completes" and incomplete bets and raises that don't reopen the betting probably need to be handled in a somewhat rule-independent way since sites use different rules to decide their own handling of this.

Terence.

[OrcaDK]

[ QUOTE ]
You'll never create a useful standard database as there are too many size/speed/usage tradeoffs for the "not in your wildest dreams" kind of tools.

You really have to bind the database to a specific solution.

[/ QUOTE ]

I agree completely. It is way out of scope to create a standardized database format, not to mention that it is impossible for the very reasons rvg Phil153 mentioned.

[Tickner]

The metadata should NOT be included in PokerHandXML. While it does have something to do with the hand I think the focus of PokerHandXML should be on the history of what actually took place.

However, I know that LegoPoker for ex will be needing metadata in some of the submitted hand histories and perhaps a new schema can be developed that EXTENDS on PokerHandXML which includes additional information that doesnt really relate to the actual hand history.

BTW, LegoPoker/HM integration will likley include this aspect, and the idea of having the stats appear next to the players/stacks is amazing.

[TSchauenberg]

[ QUOTE ]
The metadata should NOT be included in PokerHandXML. While it does have something to do with the hand I think the focus of PokerHandXML should be on the history of what actually took place.

However, I know that LegoPoker for ex will be needing metadata in some of the submitted hand histories and perhaps a new schema can be developed that EXTENDS on PokerHandXML which includes additional information that doesnt really relate to the actual hand history.

BTW, LegoPoker/HM integration will likley include this aspect, and the idea of having the stats appear next to the players/stacks is amazing.

[/ QUOTE ]

I'm fine with this as approach since then we explicitly have a basic hand definition everyone can use and then we can work on the other definition later.

[advis0r]

i agree that player stats should not be included right now.

regarding the win/loss info:
this information is present in EVERY handhistory, the only difference between sites is how it is encoded.

some have a summary section that shows each win/loss, others dont have this section but still indirectly contain the information.

so i dont see a problem with including the win/loss info in the standard, because you dont want to throw away this information if the site gives it to you and its included in every HH anyway.
(the parsing and correctness of this info is just moving up a level from the 3rd party applications to the XML creator)

i agree that there should be something like a street tag with an attribute indicating the order and maybe an optional attribute indicating the name (i.e. Flop/Turn/River for holdem games).
this makes it easier to support stud and draw games.

[rvg72]

My 2 cents

[ QUOTE ]
I disagree with this, we should include the common elements in all hand histories, which allows us to reconstruct the hand itself.

[/ QUOTE ]

I agree with the disagreement - the site where the hand was played should not be important beyond identifying the site. No custom processing should be required based on site.

[ QUOTE ]

<tournament>
<tournamentcontext>
.. buyin, vig, etc..
</tournamentcontext>
<pokerhands>
.. a bunch of pokerhands
</pokerhands>
<tournamentresult>
.. stuff like how much $$ we won and what place we came
</tournamentresult>
</tournament>


[/ QUOTE ]

I don't like this because the hand, when removed from the tournament element loses its ability to properly reflect the hand. I think a PokerHand needs to be able to be isolated and still be fully descriptive. If there are built in intuitive ways to deal with this then I'd be ok with it.

[ QUOTE ]
I believe including the metadata is a big error. The metadata is not logically related to the hand itself, it's related to the players. And the reason for this format is not to share hands among players, it's for applications to more easily support several sites as there will hopefully be a community effort to keep the converts up-to-date. Reads should be for the implementing applications to add.

[/ QUOTE ]

I understand the concern and know it doesn't fit as a "real" part of the hand. I do think we need a solution for this though since applications are going to be passing these hands around and will require this information. For example, someone right clicks a hand in HM and the xml gets converted on LegoPoker complete with some key player stats which may or may not be supported by the converter. The player then takes that and fires it into his PokerHand xml compatible replayer and now he can display the original stats that were used. So, we do need a solution since I think we can agree that this type of interoperability is the goal. What that solution is I am not sure - as long as we have something defined and standardized to allow the passing of hands with stats or comments / reads etc.

[ QUOTE ]
I disagree with all three points.

[/ QUOTE ]

Yeah, I disagree as well. PokerHand xml needs to be consistent across sites.

rvg

[OrcaDK]

[ QUOTE ]
If we want a simple-to-query format, then we definitely want, a won/loss summary for each player. Sorry if I keep banging on about this, but I haven't seen it in any of the drafts yet.

[/ QUOTE ]

I would like to include some summary information as well. Not a whole summary section, just a net win/loss for each player at the end, also recapping the current stack.

[OrcaDK]

[ QUOTE ]
1. Possible action types should be "call, raise, fold". A check is just calling a bet of 0 chips and a bet is just raising the 0 chip bet (this is out of The Theory of Poker by Sklansky). Converters etc will obv put this in modern terms but in a strict poker hand history definition these are all we need to indicate what is happening (and actually its more techincally correct). Though there might be an argument that this reduces readability because we are all so used to the common poker terms, maybe if we are querying it there would be some logic involved, ie if we want to count how many "checks" there were before a bet or something we would need to look at the actions where type="call" AND amount="0" but it seems like a trivial issue to me at first glace.

[/ QUOTE ]

I'm not too fond about having check = call(0). Technically you are correct, a check is a call of 0 chips. Semantically though, there is a difference between a check and a call, and most application will behave differently on a check compared to a call. I see no reason not to have all the common actions, check, fold, call and raise.

[Tickner]

Orca, I agree I change my mind. Your reasons are good. Basically every program that uses this will just have to change a call(0) or a bet(bigger) into a "check" and "raise".

[clowntable]

Sorry, I only read the first 5 posts but has the possibility of using an XML database instead of SQL been discussed?
Could be interesting.