[Scarabin]

Tried but could not find or access Fred's chat. Perhaps I just got the time wrong but apart from that possibility can anyone help, please?

[barmar]

it was in Vugraph from 3-4 pm ET (it ended about 1.5 hours before you posted).

A rough transcript has been posted: http://webutil.bridg...tch.php?id=1481

[Scarabin]

Thanks for prompt reply and link. I have read transcript which compensates for missing talk. I guess I got the time out by an hour and tried to join just after the session ended. My fault, apologies and thanks.

[1eyedjack]

One question that was raised early in the chat was whether GIB double dummy solver would ever be made available to kibitzers in the new, ie web-based interface.

Hitherto, I had always understood that the absence of this feature was due to technical programming issues. I now understand from Fred's chat that the limiting factor is one of cost.

Given that BBO is so keen for members to migrate from the windows version to the web version, and on the current understanding that there are no technical limitations, it would make sense to me if BBO were to disable the facility currently available in the Windows download client to run GIB DD solver as kibitzer, and then use the cost saving to enable that facility to kibitzers using the web version.

[paulg]

1eyedjack, on 2013-March-02, 08:51, said:

One question that was raised early in the chat was whether GIB double dummy solver would ever be made available to kibitzers in the new, ie web-based interface.

Hitherto, I had always understood that the absence of this feature was due to technical programming issues. I now understand from Fred's chat that the limiting factor is one of cost.

Given that BBO is so keen for members to migrate from the windows version to the web version, and on the current understanding that there are no technical limitations, it would make sense to me if BBO were to disable the facility currently available in the Windows download client to run GIB DD solver as kibitzer, and then use the cost saving to enable that facility to kibitzers using the web version.

When the GIB engine is used in the web-based interface, the GIB program runs on the BBO infrastructure and consumes processing (and other) resources. This means it costs BBO money because they would need more infrastructure if everyone used GIB.

On the downloaded client, the GIB engine runs on the user's personal computer, not the BBO infrastructure. So there is no BBO cost saving if it were disabled.

[FM75]

paulg, on 2013-March-02, 09:03, said:

When the GIB engine is used in the web-based interface, the GIB program runs on the BBO infrastructure and consumes processing (and other) resources. This means it costs BBO money because they would need more infrastructure if everyone used GIB.

On the downloaded client, the GIB engine runs on the user's personal computer, not the BBO infrastructure. So there is no BBO cost saving if it were disabled.

LOL - Really just kidding here. But if BBO rewrote GIB in javascript it could run on the web user's computer - except that they use flash instead of javascript.

Really, really, really, just kidding!! Rewrite GIB in ActionScript - so that it runs on the user's computer.

One problem facing BBO is harder. For the long term, should they bet on the local platform as we have looked at it up to now (mainly PC, or Mac/linux with a browser) continuing? Or should they be planning for tablets, phones, etc. and the GP personal computer simply vanishing? Regardless of their longer term projection, should they try to webify so that it meets tablet/phone/computer platforms?

[1eyedjack]

As I understand it, the Windows client code was going down some dark alley from which recovery was not possible without a total rewrite. And so a total rewrite was called for. And with that decision came an opportunity to reconsider such fundamental issues as whether to migrate to a cloud-based engine. I am not a programmer but I can see some of the advantages. No need for the end user to download software and maintain upgrades etc No need to concern yourself with the multitude of operating systems that users have, including national and language variations even within a single OS.
But in so doing there are also disadvantages: You give up on exploiting the massive amount of processing power and storage available on end user's machines, and due to security concerns the chosen engine (Flash) absolutely forbids such exploitation. So you can't save hands locally as they are played, you can't save chat locally, and now we find that we can't use GIB when kibitzing, all of which would be cost-free if the engine were local.

All I can say is that personally I regret the fact that we cannot have the best of both worlds, and it now appears that these limitations will be with us for ever and a day. And I sort of gather that Adobe reckons that Flash is a bit of a dinosaur, so presumably Fred has to devote time writing separate apps anyway for the mobile consumer.

Oh well, progress I suppose.

[woefuwabit]

1eyedjack, on 2013-March-03, 04:15, said:

All I can say is that personally I regret the fact that we cannot have the best of both worlds

We can now, and it's called HTML5. Not completely secure, but definitely moreso than Flash. You can save stuff using using the HTML5 File API. And you can run the GIB engine locally by compiling it using clang and emscripten into asm.js, probably leveraging on HTML5 Web Workers too