[Babase] Important update to Ranker Tweaking
Lacey Maryott
lacey.maryott at duke.edu
Mon Jul 14 14:56:08 EDT 2008
Jun Yang wrote:
> Hi Lacey:
>
> I just got a chance to look at the problem, and I have a few questions:
>
> * Can you clarify what ranker output you are seeing? Is this 1) a
> ranking previously stored in Babase, 2) the "default" ranking
> generated by the ranker when there is no existing ranking in Babase,
> or 3) the result produced by automatically incorporating a ranking for
> a period period into the default ranking? In other words, what's the
> sequence of responses/clicks in ranker that led you to this output?
>
> * If it is case 3 above, the result you are seeing could be explained
> by the current rule implemented by the ranker for incorporating an
> earlier ranking. Basically, it will first list the individuals from
> the earlier ranking (in that order), and then followed by "new"
> individuals that did not appear in the earlier ranking (according to
> the default order). I chose this simple rule because by putting all
> "new" individuals at the end of the result ranking, it is easier for
> the user to notice them. We could implement a different rule, but I
> need a clearer specification. For example, suppose there are four
> adult males and the default ranking (by age) is (apple, banana,
> cherry, date). However, the previous ranking turns out to be (date,
> banana, apple). Where would cherry be in the result?
>
I am using the 'please use last known ranking in babase' option. I
actually found that it isn't a big deal to just identify these
individuals and slide them up, that is what I have been doing to fix
this years matrices. I think we can probably work around it.
A potentially bigger problem, that I discovered the hard way today, is
that there is no 'stupid box' when you (accidentally) click on save
ranking to babase, to ask you if you are absolutely positively sure you
would like to save the ranking. I accidentally clicked on it and my
ranking got saved. It isn't terrible to fix, but just means more mussing
in the data that could lead to accidental deletions, etc. How hard is it
to incorporate a pop-up asking if we are ABSOLUTELY sure we want to save
before saving rankings to babase?
Tabby suggested maybe even moving the 'save' prompt to its own drop-down
if this is easier.
Alternatively, being able to save to babase-test would eliminate the
problem, as we know that test is for messing up things. For some reason,
it isn't letting us use test right now.
Which of these solutions would be easiest?
thanks!!
Lacey
> --- Jun Y.
>
> On Mon, Jun 30, 2008 at 2:11 PM, Lacey Maryott <lacey.maryott at duke.edu> wrote:
>
>> Hi All,
>>
>> I just started pulling together the ALM ranks for Susan for 2007. One
>> possible issue I am seeing in the ranker output is that immigrant males for
>> any given month get put in all the way at the end of the subadults and
>> juveniles, rather than at the top, right under the last male. Susan, I'm not
>> sure if this is something you will see as a big deal, but it seems as though
>> it will make it moderately harder to fit them in and carry them through
>> month to month. I am going through the matrices as I make them and
>> highlighting any new comers, etc. I just didn't know how important it is to
>> you that those new immigrant males be nearer the top, under ADM's but above
>> juveniles and SAMs.
>>
>> I also don't know how much would be involved in changing this. Jun, any
>> idea how tough it would be to distinguish between newborn boys being put at
>> the end of the list versus immigrant males being put in the middle? This
>> also begs the question is Jun still our go-to for these types of changes and
>> where are we on that transition? I have lost track.
>>
>> I'll proceed for now creating the soft-copy matrices for 2007.
>>
>> Thanks!
>> Lacey
>>
>> --
>> Lacey Maryott
>> Alberts Lab
>> Department of Biology
>> Duke University
>> ph: 919-660-7306
>> fax: 919-660-7293
>> Lacey.Maryott at duke.edu
>> _______________________________________________
>> Babase mailing list
>> Babase at www.eco.princeton.edu
>> http://www.eco.princeton.edu/mailman/listinfo/babase
>>
>>
> _______________________________________________
> Babase mailing list
> Babase at www.eco.princeton.edu
> http://www.eco.princeton.edu/mailman/listinfo/babase
>
>
--
Lacey Maryott
Alberts Lab
Department of Biology
Duke University
ph: 919-660-7306
fax: 919-660-7293
Lacey.Maryott at duke.edu
More information about the Babase
mailing list