[Babase] Comments on social interactions
Catherine Markham (amarkham at Princeton.EDU)
amarkham at Princeton.EDU
Fri Feb 3 13:24:58 EST 2006
Hi Karl,
Thanks for the feedback. Short notes inserted below...
Catherine
----- Original Message -----
From: "Karl O. Pinc" <kop at meme.com>
Date: Wednesday, February 1, 2006 3:15 pm
Subject: Re: [Babase] Comments on social interactions
To: The Baboon Database Project <babase at eeblistserv.Princeton.EDU>
> Census. No. There are rows in CENSUS that are only there
> because of demography notes.
Not sure I follow this one... my comment in that Word document just had to do with the description of the CENSUS table. I had thought another option for describing CENSUS might be "... a day each individual is censused preent or absent in a group" instead of "day each individual is (or is not) observed in a group." Was just a thought - not important.
> Demog. We obviously didn't review the table summary. I'll take
> one more stab at it, but as I'm not familiar with the entry
> protocols you'll have to suggest revisions. Note that the point
> is _not_ to reproduce the entry protocols but to describe the
> software's capabilities.
Ok.
> Re: vividity. That was the best I could come up with that would
> fit in a table.
Ok.
> Re: it's. Oops. That's what my fingers type. I obviously have
> not gone back through the document to check for this. So, I just
> did but you can expect them to creep back in until we get a final
> version.
I get stuff like that confused all the time - fresh eyes definitely help. :)
> Re: Sexual cycles ER diagram. Each CYCPOINTS row represents a
> single Ddate, Tdate, etc. So, PREGS.Conceive does point to a date.
> (Same reason why the label from CYCLES.Sname says "Female", because
> it points to a BIOGRAPH row that represents a female.) CYCPOINTS is
> new, so it is a change from the old system. They are Cpids, not Cids.
Cool, I got it now.
> Re INTERACT_DATA "lower date limits as to when" does not scan as well
> to me as "lower limits as to when". Is it not clear that "when" means
> "date"?
No big deal - looks ok as is.
> (What does "Strikethrough" mean?)
The strickthrough was just that I had crossed out the "for that year" portion of "For these data, all dates are entered as the first day of the month for that year". Not important.
> INTERACT_DATA.Date: No they would not change the month but would
> change the date. (As usual, only usually the change would be to the
> first day of the month.) The original docs said to make the date
> the first day during the month when both individuals were in the
> group. Usually this was the first day of the month.
This still doesn't make sense to me, but maybe I just never knew this rule with entry before. I would think the data entry person would discard the interaction if it was written, say, for July of 2004 but the two baboons were not together in the group that month. I know I wouldn't have back-tracked to see that the two participants were in the group in maybe May and assumed the data was for that month. Is that what you're suggesting?
> I don't know what the story is with multi-party interactions. The
> sentence was written before they were recording multi-party
> interactions and, apparently, at the time it seemed clear that
> multi-party interactions would be (were?) "disassembled" into
> pairwise interactions. Now that there are multi-party interactions
> we need to revisit the database design. (I think only in terms
> of the ACTS table, but we'll see.)
My fault - sorry about that. Jeanne just explained it to me, and I'll check the monitoring guide too. Ignore my original comment.
> Re: Recording grooming and agonism on infants. I don't care,
> the doc describes the software's capabilities.
Ok.
> Re: submissive. I could just delete the parenthetical clause?
Perhaps this is just me being too picky or not knowing the exact protocols. Definitely don't base edits soley on my comments.
> Re: POINT_DATA ejaculations. The protocol says nothing about
> ejaculations. Anybody researching ejaculations will have
> to investigate what was recorded. How else should I write this?
I actually like what you just said better than the original ("However, the protocol is silent regards ejaculations occurring during all-occurrences sampling so the researcher will need to investigate this further")... something along those lines reads much easier to me.
> Re: INTERACT View. I capitalized Julian everywhere. This is
> probably right.
Ok.
> Re: cpm (Startcpm) Cpm should always be spm. Got some typo in there
> somehow.
Ok.
> Re: ACTOR_ACTEES Normally, you can't have an INTERACT_DATA without
> PARTS, but you can while you're in the process of entering the data
> (for technical reasons summarized by saying that something has to
> get into the database first.) This allows you to see the "incomplete"
> entries and fix them.
Ok.
More information about the Babase
mailing list