[Cs5500] [Cs4800] Fwd: Request Function problem

Fri Nov 11 21:06:38 EST 2011

James,

The baby avatar that was part of yesterday's release ran to completion and
was never kicked out of the tournament.
However I cannot guarantee your avatar will run to completion if you have
used only the provide method of the baby avatar and rest of the
intelligence of your own avatar.
It may not be enough to substitute the provide method with the baby's, you
may also need to make sure it is compatible and consistent with your
propose, solve and oppose methods.

On Fri, Nov 11, 2011 at 3:05 PM, James Antonius <antonius.j at husky.neu.edu>wrote:

> Hmm... At first I thought the problem was on my avatar side, but why it
> still gave the same problem when I used the same "provide" function that
> the BabyAvatar used (I copied the same provide function that the BabyAvatar
> has)?
>
> For example, the provide function that I copied from BabyAvatar:
>
>
> /** providing instance - in HSR this is trivial as the instanceSet is
> singleton**/
> public InstanceI provide(Claim claimToBeProvided){
> HSRInstanceSet hsrInstanceSet = (HSRInstanceSet)
> claimToBeProvided.getInstanceSet();
> HSRInstance hsrInstance = new
> HSRInstance(hsrInstanceSet.getSingleton().getN(),
> hsrInstanceSet.getSingleton().getK());
> return hsrInstance;
> }
>
>
> But my avatar still didn't respond when the provide function was requested.
>
>
>
> Thanks,
>
> James Antonius
>
>
> On Fri, Nov 11, 2011 at 9:49 AM, Madhuvanthi Balasubramanian <
> balasubramanian.m at husky.neu.edu> wrote:
>
>
>> Hello Everyone,
>>
>>  The problems reported below are on the avatars' side.
>> When an oppose action is sent to them and they fail to respond, they are
>> kicked out. That was what was happening to James' avatar.
>> I promised him that we would make the kick-out messages more descriptive,
>> as in how many responses were expected and how many were actually sent from
>> his avatar. That might help.
>>
>> As to Andrew's bug-report, we encountered the issue yesterday while
>> testing, but did not have enough time to analyze it. As far as we know,
>> this behavior is spotted in the spill-over rounds, when the number of
>> rounds is complete, but the protocol steps are not compete.
>> We will address this issue soon. Thanks to Andrew and James for bringing
>> these up.
>>
>> Here is a list of yesterday's fixes:
>>
>> 1.StrengthenP method was fixed to return true in the context of
>> ForAllExistsMin. This took care of how strengthening was printed out in the
>> smart history files.
>> 2.Name switching was fixed, so all oppose actions look consistent across
>> history files. Earlier, the history files looked as if the avatar that was
>> refuting solved the instance and the other was providing the instance. It
>> was also reversed in strengthening. This issue is fixed now.
>> 3.Identity HashMaps were used in place of HashMaps in some classes so
>> every object was treated as a different object and there were no overwrites
>> or overlaps.
>> 4.GenerateRandomAllowedProtocol method in the baby avatar was removed,
>> because this is one possible place that generates instance of protocols
>> other than ForAllExistsMin. We also made sure every claim proposed by the
>> baby avatar proposes instances of ForAllExistsMin only.
>> 5. We also ran a series of regression tests to make sure these fixes
>> don't conflict each other and that the results are as expected.
>>
>> Thanks to each and everyone of you for reporting such issues and bugs. We
>> would not have been able to improve our playground so well without your
>> valuable feedback. Keep such bug-reports coming and we will strive to solve
>> every issue, thereby making this entire experience pleasant and fun for all
>> of us !
>>
>>
>> On Fri, Nov 11, 2011 at 6:07 AM, Karl Lieberherr <lieber at ccs.neu.edu>wrote:
>>
>>
>>> Hello class:
>>>
>>> the tournament last night indicates that SCG Court, thanks to your
>>> efforts, is much healthier now. I would like to thank the current
>>> management team: Madhu and Sandeep, and all their supporters, for properly
>>> integrating all the bug fixes
>>> and for handling the communication with the users.
>>>
>>> There are still histories that are puzzles: see the messages below.
>>>
>>> Please can Madhu and Karan compose a message to the undergraduate class
>>> explaining at a high level what kind of problems you had to solve to repair
>>> strengthening. The IdentityHashMap is one issue.
>>>
>>> Do we need a new release for the warm-up tournament tonight? A release
>>> that repairs the problem below?
>>>
>>> -- Karl
>>>
>>>
>>> ---------- Forwarded message ----------
>>> From: Andrew Ziehl <ziehla at gmail.com>
>>> Date: Fri, Nov 11, 2011 at 1:09 AM
>>> Subject: Re: [Cs4800] Request Function problem
>>> To: James Antonius <antonius.j at husky.neu.edu>
>>> Cc: CS4800 Mailing List <cs4800 at lists.ccs.neu.edu>
>>>
>>>
>>>
>>> I saw something like this with a number of my games in the warm-up
>>> (nobackbone avatar) where a strengthening or refuting action was made, with
>>> no response and no winner.
>>>
>>>   On Fri, Nov 11, 2011 at 12:02 AM, James Antonius <
>>> antonius.j at husky.neu.edu> wrote:
>>>
>>>
>>>>  Hi everyone,
>>>>
>>>> Does any of you have encountered a problem where our function doesn't
>>>> give respond when it's requested?
>>>>
>>>>  For example:
>>>>
>>>>    - When it requested my "provide", my function didn't respond:
>>>>
>>>>
>>>>    claim hsr.HSRInstanceSet {{   HSR(733,607)   }} scg.protocol.ForAllExistsMin {{      }} 0.013642564802182811 1.0
>>>>    proposer  {{ TeenAvatar }}
>>>>    opposer  {{ BabyAvatar }}
>>>>    action strengthening 0.013642563802182811
>>>>    responses  winner  {{  }}
>>>>    pointsWon 0.0
>>>>
>>>>    I think here it should request for my provide function after the
>>>>    BabyAvatar strengthens it, but my avatar doesn't respond to it. (Note: I
>>>>    even tried to copy the exact "provide" function from the BabyAvatar code,
>>>>    but it still gave the same error.)
>>>>
>>>>
>>>>    - Same thing happen when it requested my "solve" function:
>>>>
>>>>
>>>>    claim hsr.HSRInstanceSet {{   HSR(54,5)   }} scg.protocol.ForAllExistsMin {{      }} 0.1111111111111111 1.0
>>>>    proposer  {{ TeenAvatar }}
>>>>    opposer  {{ BabyAvatar }}
>>>>    action refuting
>>>>    responses provider  {{ BabyAvatar }}  pr provide hsr.HSRInstance {{  HSR(54,5)  }}
>>>>      winner  {{  }}
>>>>    pointsWon 0.0
>>>>
>>>>    I think here it should request for my solve function after the BabyAvatar provides the instance, but my avatar doesn't respond to it.
>>>>
>>>>
>>>>
>>>> Any idea how to solve this?
>>>>
>>>>
>>>>
>>>> Thanks,
>>>>
>>>> James Antonius
>>>>
>>>> _______________________________________________
>>>> Cs4800 mailing list
>>>> Cs4800 at lists.ccs.neu.edu
>>>> https://lists.ccs.neu.edu/bin/listinfo/cs4800
>>>>
>>>>
>>>>
>>>
>>> _______________________________________________
>>> Cs4800 mailing list
>>> Cs4800 at lists.ccs.neu.edu
>>> https://lists.ccs.neu.edu/bin/listinfo/cs4800
>>>
>>>
>>>
>>> _______________________________________________
>>> Cs5500 mailing list
>>> Cs5500 at lists.ccs.neu.edu
>>> https://lists.ccs.neu.edu/bin/listinfo/cs5500
>>>
>>>
>>>
>>
>>
>>
>> --
>>  Thanks
>> - Madhu
>>
>>
> _______________________________________________
> Cs4800 mailing list
> Cs4800 at lists.ccs.neu.edu
> https://lists.ccs.neu.edu/bin/listinfo/cs4800
>
>
>
>

-- 
 Thanks
- Madhu
-------------- next part --------------
HTML attachment scrubbed and removed