[Babase] [Fwd: Re: more information that I need]
Leah Gerber
lgerber at duke.edu
Wed Aug 30 11:18:45 EDT 2006
-------- Original Message --------
Subject: Re: more information that I need
Date: Fri, 25 Aug 2006 15:04:54 -0400
From: Leah Gerber <lgerber at duke.edu>
To: patchu at itsyourturn.com
References: <HNEJJMICBFCMONFLODLEIEFBIEAC.patchu at itsyourturn.com>
This does make sense, thanks for the clarification. One reason why we
are not too concerned about checks on the 3-letter short name is
because the majority of the errors are valid short names but the animal
has been dead for a while. The field team sometimes get confused when
individuals look similar to old individuals. There are also some other
checks, like self grooming or self agonism which are not allowed and
checks on the sex of the individual for particular interactions. These
errors should get picked up by the triggers so we will have errors to
deal with anyway even if the unknown 3-letter codes are picked out right
away. If you can see a reason why we should do one check while comparing
the data before the rest of the errors get picked out with the triggers
then we are open to discussion. We just don't see an advantage at this
point.
We have no plans for students to enter data at this point so instead of
making you wait indefinitely I entered two files. I have attached A and
B copies of a consort file and a grooming file. Let me know if you need
some more and I will enter more.
Leah
Patrick Chu wrote:
> I'm sorry that I didn't explain this properly before.
>
> The verification checks will be driven from the same exact BIOGRAPH table
> that you're already maintaining. As long as this table is up to date, then
> the checks will be accurate. There's no additional work on your part, and no
> code needs to be rewritten. As soon as you put in new data into BIOGRAPH,
> the verification will immediately be updated to reflect the new information
> (it will simply do a join against this table and see which 3-letter codes
> are missing).
>
> No additional work on your part, no reprogramming is necessary.
>
> Does this new information change anything?
>
>
> -----Original Message-----
> From: Leah Gerber [mailto:lgerber at duke.edu]
> Sent: Friday, August 25, 2006 11:44 AM
> To: patchu at itsyourturn.com
> Subject: Re: more information that I need
>
> It is a good idea, I just worry because this list will change every 6
> months.
>
> Patrick Chu wrote:
>
>> The reason for a list of 3-letter codes is that I can flag an invalid
>> 3-letter code as a "mistake" before the insert attempt is make. But it's
>> probably unlikely that TWO people would enter the same mistake? Still, it
>> might save the verifier some time if we were to show the baboon's full
>>
> name
>
>> next to the 3-letter codes, right on the verification page.
>>
>> Just a thought, not strictly necessary.
>>
>>
>> -----Original Message-----
>> From: Leah Gerber [mailto:lgerber at duke.edu]
>> Sent: Friday, August 25, 2006 10:44 AM
>> To: patchu at itsyourturn.com; alberts >> Susan Alberts
>> Subject: Re: more information that I need
>>
>>
>> Patrick Chu wrote:
>>
>>
>>> Leah,
>>>
>>> If those other files come in, please email them to me. You can email me
>>>
>>>
>> old
>>
>>
>>> files too if you have them around.
>>>
>>>
>>>
>> I will try to enter a few for you. We delete the B copy when we are done
>> with them so no old ones are lying around.
>>
>>
>>> Is there a list of all valid three-letter codes for each baboon somewhere
>>>
>>>
>> in
>>
>>
>>> the database?
>>>
>>>
>>>
>> There is no specific list of three letter names. The BIOGRAPH table has
>> one row for every baboon so that is the easiest way to pull out all of
>> the names.
>> Just curious why you would need this. This list changes every few
>> months, so we would not want this set in the program. I can send you a
>> semi recent list if you want it. Also, for the acts, I am wondering why
>> we would need this information in the compare procedure when the
>> triggers are all set up in the database already. I exported the ACTS
>> support table from the database and attached it as an excel file. This
>> one is unlikely to change. I just thought of one thing, the grooming
>> file I gave you does not have a column for act "G". We had planned on
>> incorporating this but Karl's import program needed to be changed before
>> we could. So, if it is easier for you we can easily add an act column to
>> grooming. All of the rows will be "G". Let me know if you need more
>> information. I will gladly extract the baboon names if it is necessary
>> and makes sense to use them.
>>
>>
>>> Also, could you summarize the input formats of the 3 types of Excel
>>>
> files,
>
>>> and where that information gets inserted into which tables? I have some
>>> sketchy notes here.
>>>
>>> I have the letters A, G, E, M, and C. Can you tell me the labels for each
>>> letter? I know the A = agonisms and G = grooming.
>>>
>>> Also, can you give me a list of the codes for the possible acts? Is this
>>>
>>>
>> in
>>
>>
>>> a database table somewhere?
>>>
>>> Can you tell me if the act codes and baboon IDs are in the database
>>> somewhere? If so, is it possible to get a full or partial copy of those
>>> tables? Exporting it to a text file (any format) is sufficient, although
>>>
>>>
>> an
>>
>>
>>> Excel file would be even better.
>>>
>>> Thanks for your help.
>>>
>>>
>>>
>>>
>>
>>
>
>
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: IC0406B.xls
Type: application/vnd.ms-excel
Size: 13824 bytes
Desc: not available
Url : http://www.eco.princeton.edu/pipermail/babase/attachments/20060830/4164d027/IC0406B-0001.xls
-------------- next part --------------
A non-text attachment was scrubbed...
Name: IG0406B.xls
Type: application/vnd.ms-excel
Size: 20992 bytes
Desc: not available
Url : http://www.eco.princeton.edu/pipermail/babase/attachments/20060830/4164d027/IG0406B-0001.xls
-------------- next part --------------
A non-text attachment was scrubbed...
Name: IC0406a.xls
Type: application/vnd.ms-excel
Size: 13824 bytes
Desc: not available
Url : http://www.eco.princeton.edu/pipermail/babase/attachments/20060830/4164d027/IC0406a-0001.xls
-------------- next part --------------
A non-text attachment was scrubbed...
Name: IG0406a.xls
Type: application/vnd.ms-excel
Size: 20992 bytes
Desc: not available
Url : http://www.eco.princeton.edu/pipermail/babase/attachments/20060830/4164d027/IG0406a-0001.xls
More information about the Babase
mailing list