From nobody@digitalkingdom.org Wed Mar 22 07:16:45 2006 Received: with ECARTIS (v1.0.0; list lojban-list); Wed, 22 Mar 2006 07:16:46 -0800 (PST) Received: from nobody by chain.digitalkingdom.org with local (Exim 4.60) (envelope-from ) id 1FM54V-0004lf-KZ for lojban-list-real@lojban.org; Wed, 22 Mar 2006 07:16:27 -0800 Received: from zproxy.gmail.com ([64.233.162.200]) by chain.digitalkingdom.org with esmtp (Exim 4.60) (envelope-from ) id 1FM54T-0004lY-MP for lojban-list@lojban.org; Wed, 22 Mar 2006 07:16:27 -0800 Received: by zproxy.gmail.com with SMTP id x7so179909nzc for ; Wed, 22 Mar 2006 07:16:24 -0800 (PST) DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=PJdPXmxr/Erd5r7RcJScDEHMGinHNDhtYJeW2Flv9AzL3hIMeHKLokZ+35gTIDaQmCKHmL6A1TVUuCeRqmhylMe5JgAEXWLAAzc0EmUbJNCbkPOhZrkBrT1oNyBO76Dw112wmIBYIl7CVvL81ASxz2xpqhjMcHZ0j5NctgzOmoI= Received: by 10.36.148.1 with SMTP id v1mr572415nzd; Wed, 22 Mar 2006 07:16:23 -0800 (PST) Received: by 10.37.20.27 with HTTP; Wed, 22 Mar 2006 07:16:23 -0800 (PST) Message-ID: Date: Wed, 22 Mar 2006 10:16:23 -0500 From: "Matt Arnold" To: lojban-list@lojban.org Subject: [lojban] Re: Suggestions for jboselkei In-Reply-To: <20060322153201.L33225@mail.sksys.net> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_17230_1519168.1143040583600" References: <1073788385.20060322143818@mail.ru> <20060322153201.L33225@mail.sksys.net> X-Spam-Score: -1.8 (-) X-archive-position: 11211 X-ecartis-version: Ecartis v1.0.0 Sender: lojban-list-bounce@lojban.org Errors-to: lojban-list-bounce@lojban.org X-original-sender: matt.mattarn@gmail.com Precedence: bulk Reply-to: lojban-list@lojban.org X-list: lojban-list ------=_Part_17230_1519168.1143040583600 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline On 3/22/06, Cyril Slobin wrote: > > > 3) Computing proficiency in Lojban as: selecting only those translation= s > from the set of the > > translations made by the user that were reviewed by at least three > > reviewers, then find an average value of these translations. > > (An value of a translation is a product of its estimated > > difficulty and its estimated quality.) This is, however, much more > difficult to realize in > > real-time comparing with computing ratings themselves. > > Not very good. The good stimulus is to recive rating shortly after work > is done, and this suggestion makes waiting for revard too long. I am not sure that I agree, Cyril. It might be a worthwhile trade-off. The more meaningful the rating is, the more incentive it is. Yanis, how long of a wait would it be? Considering that computers work quit= e fast, perhaps it would be only a matter of minutes at most. Psychologically= , that would still be a stimulus. I approve of the titles and the work rewards as well. {stati}, {certi}, and {selsinma} come to mind. Lujvo might be even better, but my lujvo creation is not yet strong, and I'll leave that task to those who are good at it. In the translation screen, offer a field where the translator can put nonce lujvo that are being invented, along with suggested place structure. A fiel= d for the English gloss, a field for x1, x2, etc. Then output a webpage of al= l nonce lujvo that the jboselkei players have proposed as suggestions for inclusion on jbovlaste and let them make comments for improvement. Ultimately it would be wonderful for BPFK committee members to be able to search the entire database of jboselkei for usage examples of the cmavo which they are shepherding. -epkat ------=_Part_17230_1519168.1143040583600 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline On 3/22/06, Cyril Slobin <slobin@sksys.net> wrote:
> 3) Computing proficiency in Lojban as: selecting only those translatio= ns from the set of the
> translations made by the user that were revi= ewed by at least three
> reviewers, then find an average value of the= se translations.
> (An value of a translation is a product of its estimated
> d= ifficulty and its estimated quality.) This is, however, much more difficult= to realize in
> real-time comparing with computing ratings themselve= s.

Not very good. The good stimulus is to recive rating shortly after = work
is done, and this suggestion makes waiting for revard too long.

I am not sure that I agree, Cyril. It might be a worthwh= ile trade-off. The more meaningful the rating is, the more incentive it is.= =20

Yanis, how long of a wait would it be? Considering that computers w= ork quite fast, perhaps it would be only a matter of minutes at most. Psych= ologically, that would still be a stimulus.

I approve of the titles = and the work rewards as well. {stati}, {certi}, and {selsinma} come to mind= . Lujvo might be even better, but my lujvo creation is not yet strong, and = I'll leave that task to those who are good at it.

In the translation screen, offer a field where the translator can p= ut nonce lujvo that are being invented, along with suggested place structur= e. A field for the English gloss, a field for x1, x2, etc. Then output a we= bpage of all nonce lujvo that the jboselkei players have proposed as sugges= tions for inclusion on jbovlaste and let them make comments for improvement= .

Ultimately it would be wonderful for BPFK committee members to be a= ble to search the entire database of jboselkei for usage examples of the cm= avo which they are shepherding.

-epkat
------=_Part_17230_1519168.1143040583600-- To unsubscribe from this list, send mail to lojban-list-request@lojban.org with the subject unsubscribe, or go to http://www.lojban.org/lsg2/, or if you're really stuck, send mail to secretary@lojban.org for help.