Many a Slip 'tween GIBBO and Bidding Blurb

[virgosrock]

[hv=http://www.bridgebase.com/tools/handviewer.html?lin=pn%7CVirgosRock%2CNOT_BOT%2CRobotN%2CRobotE%7Cst%7C%7Cmd%7C4S37JAH3478QD27JCJ%2CS248H59TD8C2567TQ%2CS9KH2KAD569TKC34A%2C%7Crh%7C%7Cah%7CBoard+2%7Csv%7Cn%7Cmb%7C1D%7Can%7CMinor+suit+opening+--+3%2B+%21D%3B+11-21+HCP%3B+12-22+total+points+%7Cmb%7Cp%7Cmb%7Cp%7Cmb%7Cd%7Can%7C3%2B+%21C%3B+3%2B+%21H%3B+3%2B+%21S%3B+9%2B+HCP+%7Cmb%7Cp%7Cmb%7C2D%7Can%7C11-16+total+points%3B+forcing%7Cmb%7Cp%7Cmb%7C2N%7Can%7CInvitational+to+3NT+game+--+2-5+%21C%3B+2-5+%21D%3B+2-3+%21H%3B+2-3+%21S%3B+12%2B+total+points%3B+stop+in+%21D+%7Cmb%7Cp%7Cmb%7Cp%7Cmb%7Cp%7Cpg%7C%7Cpc%7CHJ%7Cpc%7CH3%7Cpc%7CH5%7Cpc%7CHA%7Cpg%7C%7Cpc%7CHK%7Cpc%7CH6%7Cpc%7CH7%7Cpc%7CHT%7Cpg%7C%7Cpc%7CH2%7Cpc%7CD4%7Cpc%7CHQ%7Cpc%7CH9%7Cpg%7C%7Cpc%7CH8%7Cpc%7CC2%7Cpc%7CD5%7Cpc%7CD3%7Cpg%7C%7Cpc%7CH4%7Cpc%7CC5%7Cpc%7CD6%7Cpc%7CS6%7Cpg%7C%7Cpc%7CDJ%7Cpc%7CD8%7Cpc%7CD9%7Cpc%7CDQ%7Cpg%7C%7Cpc%7CDA%7Cpc%7CD7%7Cpc%7CS2%7Cpc%7CDK%7Cpg%7C%7Cpc%7CC8%7Cpc%7CCJ%7Cpc%7CCQ%7Cpc%7CCA%7Cpg%7C%7Cpc%7CDT%7Cpc%7CST%7Cpc%7CD2%7Cpc%7CC6%7Cpg%7C%7Cpc%7CSK%7Cpc%7CS5%7Cpc%7CS3%7Cpc%7CS4%7Cpg%7C%7Cpc%7CS9%7Cpc%7CSQ%7Cpc%7CSA%7Cpc%7CS8%7Cpg%7C%7Cpc%7CSJ%7Cpc%7CC7%7Cpc%7CC3%7Cpc%7CC9%7Cpg%7C%7Cpc%7CS7%7Cpc%7CCT%7Cpc%7CC4%7Cpc%7CCK%7Cpg%7C%7C%0A%0A]399|300[/hv]

 

Blurb said 2NT is 12 total points stop in D and there the matter rested. Vul game missed @0.5 cents a point. Do the Math :)

GIBBO has 17 HCP. Somebody figure out the Rule. I can't. There seems to be no consonance between GIBBO bid and the blurb.

For instance a modicum of Artificial Intelligence would say "I have 17 HCP, My blurb says 12+ points"

Maybe that is the problem. No consonance between Blurb and Bid. I think I said this 6 years ago too as "GIBBO does not read its own blurbs".

 

vrock Just another $4

[Stephen Tu]

The description could be improved here certainly, but you have to give the programmers a certain amount of leeway for a very difficult task. It's really hard to programatically generate accurate descriptions for every single auction under the sun. You write a bunch of rules, attach descriptions to the rules, then as the rules are triggered you combine the descriptions in some way and hope to spit out something intelligible.

 

For one, the description said 12+ total points, not 12- total points. That means 12 or more, not 12 exactly. You misread if you thought the top end was limited.

 

I have to blame this result on you though, you should have been able to work it out:

- you could have overcalled to begin with

- cue bid in response to double really ought to promise a rebid

- logically if North had no 4 cd major and 11-14 hcp, stopper in diamond, should have balanced 1nt not double so must be stronger than that. Interpreting as 12 total points max makes no sense.

 

It would be more accurate if the description said 15+, but the bid did match its description here.

[virgosrock]

The description could be improved here certainly, but you have to give the programmers a certain amount of leeway for a very difficult task. It's really hard to programatically generate accurate descriptions for every single auction under the sun. You write a bunch of rules, attach descriptions to the rules, then as the rules are triggered you combine the descriptions in some way and hope to spit out something intelligible.

 

For one, the description said 12+ total points, not 12- total points. That means 12 or more, not 12 exactly. You misread if you thought the top end was limited.

 

I have to blame this result on you though, you should have been able to work it out:

- you could have overcalled to begin with

- cue bid in response to double really ought to promise a rebid

- logically if North had no 4 cd major and 11-14 hcp, stopper in diamond, should have balanced 1nt not double so must be stronger than that. Interpreting as 12 total points max makes no sense.

 

It would be more accurate if the description said 15+, but the bid did match its description here.

 

Before I respond, are you, Stephen, part of the programming team?

 

vrock

[Stephen Tu]

No I'm not.

[johnu]

Before I respond, are you, Stephen, part of the programming team?

 

vrock

 

Georgi apparently does most of the GIB patches. In the past, he has posted that he would be working on specific bugs brought up in the forum. He hasn't posted in months.

 

Barmar seems to poke around in several areas of BBO code but I don't know if he does any bug fixing in the GIB code.

[Bbradley62]

Georgi apparently does most of the GIB patches. In the past, he has posted that he would be working on specific bugs brought up in the forum. He hasn't posted in months.

 

Barmar seems to poke around in several areas of BBO code but I don't know if he does any bug fixing in the GIB code.

 

30+ Months? I created a thread on 4/20/16 asking if Georgi was still around, since it was the one-year anniversary of his last post. Georgi replied that he was still around and working on GIB. It appears that this was his most recent post.

[virgosrock]

30+ Months? I created a thread on 4/20/16 asking if Georgi was still around, since it was the one-year anniversary of his last post. Georgi replied that he was still around and working on GIB. It appears that this was his most recent post.

 

Seriously! just two programmers working occasionally to fix bugs. The problem is gigantic!!

 

vrock