Difference in number of unique surnames shown by different gramplets.

classic Classic list List threaded Threaded
10 messages Options
Reply | Threaded
Open this post in threaded view
|

Difference in number of unique surnames shown by different gramplets.

John W. Kitz-3
All,

I noticed that there is a difference in the number of unique surnames shown
by the statistics and top surnames gramplets, as follows:

On my dashboard page the statistics gramplet shows the following:

Family Information
----------------------------
Number of families: XXXXXXX
Unique surnames: 439

While the top surnames gramplet shows:

Total unique surnames: 397
Total people: XXXXXX

I.e. a difference of 42 surnames between the two gramplets using the same
data.

Is this a known bug or worth entering one?

Regards, Jk.


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Difference in number of unique surnames shown by different gramplets.

Paul Smith
It one using soundex or similar to match 'mole, moule, moles', all different
aliases for the same family in my own tree so is this '3' or '1' unique
surnames?

Paul DS.

-----Original Message-----
From: John W. Kitz [mailto:[hidden email]]
Sent: 25 January 2018 10:34
To: [hidden email]
Subject: [Gramps-users] Difference in number of unique surnames shown by
different gramplets.

All,

I noticed that there is a difference in the number of unique surnames shown
by the statistics and top surnames gramplets, as follows:

On my dashboard page the statistics gramplet shows the following:

Family Information
----------------------------
Number of families: XXXXXXX
Unique surnames: 439

While the top surnames gramplet shows:

Total unique surnames: 397
Total people: XXXXXX

I.e. a difference of 42 surnames between the two gramplets using the same
data.

Is this a known bug or worth entering one?

Regards, Jk.


----------------------------------------------------------------------------
--
Check out the vibrant tech community on one of the world's most engaging
tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Difference in number of unique surnames shown by different gramplets.

John Paton
In reply to this post by John W. Kitz-3
On 25/01/18 10:33, John W. Kitz wrote:

> All,
>
> I noticed that there is a difference in the number of unique surnames shown
> by the statistics and top surnames gramplets, as follows:
>
> On my dashboard page the statistics gramplet shows the following:
>
> Family Information
> ----------------------------
> Number of families: XXXXXXX
> Unique surnames: 439
>
> While the top surnames gramplet shows:
>
> Total unique surnames: 397
> Total people: XXXXXX
>
> I.e. a difference of 42 surnames between the two gramplets using the same
> data.
>
> Is this a known bug or worth entering one?
>
> Regards, Jk.
>
>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> Gramps-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gramps-users
> https://gramps-project.org
>
I was a bit bored so I thought I would check my own much more modest
database to see if this applied. I found a difference of 1 between the
surname totals in the two gramplets. I assumed a bug therefore and
intended to await a response from someone who knows more about the program.

However I also noticed in the 'Statistics' gramplet that I had 3
individuals with incomplete names. When I had tracked these down and
fixed them I noticed that the 'Statistics' gramplet and the 'Top
Surnames' gramplet now agreed on the total number of unique surnames
(46), which was also the number of Surnames I had physically counted.

I have no idea as to which of my 3 incomplete names was the important
one so I can give no more useful advice on what to look for other than
check if you have incomplete names in the 'Statistics' gramplet. In all
3 cases the individuals were female and in each case it was the
*married* name which was the problem (although not the same problem in
all 3)

--
Time flies like an arrow. Fruit flies like a banana.

John Paton

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Difference in number of unique surnames shown by different gramplets.

Paul Smith
Sounds like there's a niche market for a name checking tool :-).

-----Original Message-----
From: John Paton [mailto:[hidden email]]
Sent: 25 January 2018 11:41
To: [hidden email]
Subject: Re: [Gramps-users] Difference in number of unique surnames shown by
different gramplets.

On 25/01/18 10:33, John W. Kitz wrote:

> All,
>
> I noticed that there is a difference in the number of unique surnames
> shown by the statistics and top surnames gramplets, as follows:
>
> On my dashboard page the statistics gramplet shows the following:
>
> Family Information
> ----------------------------
> Number of families: XXXXXXX
> Unique surnames: 439
>
> While the top surnames gramplet shows:
>
> Total unique surnames: 397
> Total people: XXXXXX
>
> I.e. a difference of 42 surnames between the two gramplets using the
> same data.
>
> Is this a known bug or worth entering one?
>
> Regards, Jk.
>
>
> ----------------------------------------------------------------------
> -------- Check out the vibrant tech community on one of the world's
> most engaging tech sites, Slashdot.org! http://sdm.link/slashdot 
> _______________________________________________
> Gramps-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gramps-users
> https://gramps-project.org
>
I was a bit bored so I thought I would check my own much more modest
database to see if this applied. I found a difference of 1 between the
surname totals in the two gramplets. I assumed a bug therefore and intended
to await a response from someone who knows more about the program.

However I also noticed in the 'Statistics' gramplet that I had 3 individuals
with incomplete names. When I had tracked these down and fixed them I
noticed that the 'Statistics' gramplet and the 'Top Surnames' gramplet now
agreed on the total number of unique surnames (46), which was also the
number of Surnames I had physically counted.

I have no idea as to which of my 3 incomplete names was the important one so
I can give no more useful advice on what to look for other than check if you
have incomplete names in the 'Statistics' gramplet. In all
3 cases the individuals were female and in each case it was the
*married* name which was the problem (although not the same problem in all
3)

--
Time flies like an arrow. Fruit flies like a banana.

John Paton

----------------------------------------------------------------------------
--
Check out the vibrant tech community on one of the world's most engaging
tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Difference in number of unique surnames shown by different gramplets.

prculley
The Statistics Gramplet uses some metadata in the db that is only updated when Gramps closes cleanly.  Unfortunately that means that a crash, and probably some other issues, can leave the metadata out of sync with the other parts of the db.

The 'Tools/Family Tree Repair/Rebuild Gender Statistics' can be used to correct this situation.

If the two Gramplets remain out of sync after running this tool, the developers would be interested in looking over the tree to see if there is a bug.  So it would be appreciated if you would file a bug report https://gramps-project.org/bugs/bug_report_page.php and attach an XML export of tree.

Paul C.

On Thu, Jan 25, 2018 at 6:44 AM, Paul Smith <[hidden email]> wrote:
Sounds like there's a niche market for a name checking tool :-).

-----Original Message-----
From: John Paton [mailto:[hidden email]]
Sent: 25 January 2018 11:41
To: [hidden email]
Subject: Re: [Gramps-users] Difference in number of unique surnames shown by
different gramplets.

On 25/01/18 10:33, John W. Kitz wrote:
> All,
>
> I noticed that there is a difference in the number of unique surnames
> shown by the statistics and top surnames gramplets, as follows:
>
> On my dashboard page the statistics gramplet shows the following:
>
> Family Information
> ----------------------------
> Number of families: XXXXXXX
> Unique surnames: 439
>
> While the top surnames gramplet shows:
>
> Total unique surnames: 397
> Total people: XXXXXX
>
> I.e. a difference of 42 surnames between the two gramplets using the
> same data.
>
> Is this a known bug or worth entering one?
>
> Regards, Jk.
>
>
> ----------------------------------------------------------------------
> -------- Check out the vibrant tech community on one of the world's
> most engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> Gramps-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gramps-users
> https://gramps-project.org
>
I was a bit bored so I thought I would check my own much more modest
database to see if this applied. I found a difference of 1 between the
surname totals in the two gramplets. I assumed a bug therefore and intended
to await a response from someone who knows more about the program.

However I also noticed in the 'Statistics' gramplet that I had 3 individuals
with incomplete names. When I had tracked these down and fixed them I
noticed that the 'Statistics' gramplet and the 'Top Surnames' gramplet now
agreed on the total number of unique surnames (46), which was also the
number of Surnames I had physically counted.

I have no idea as to which of my 3 incomplete names was the important one so
I can give no more useful advice on what to look for other than check if you
have incomplete names in the 'Statistics' gramplet. In all
3 cases the individuals were female and in each case it was the
*married* name which was the problem (although not the same problem in all
3)

--
Time flies like an arrow. Fruit flies like a banana.

John Paton

----------------------------------------------------------------------------
--
Check out the vibrant tech community on one of the world's most engaging
tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Difference in number of unique surnames shown by different gramplets.

John W. Kitz-3
In reply to this post by John Paton
John,

On 2018-01-25 12:40, John Paton wrote:

> On 25/01/18 10:33, John W. Kitz wrote:
>> All,
>>
>> I noticed that there is a difference in the number of unique surnames
>> shown
>> by the statistics and top surnames gramplets, as follows:
>>
>> On my dashboard page the statistics gramplet shows the following:
>>
>> Family Information
>> ----------------------------
>> Number of families: X
>> Unique surnames: 439
>>
>> While the top surnames gramplet shows:
>>
>> Total unique surnames: 397
>> Total people: X
>>
>> I.e. a difference of 42 surnames between the two gramplets using the
>> same
>> data.
>>
>> Is this a known bug or worth entering one?
>>
>> Regards, Jk.
>>
> I was a bit bored so I thought I would check my own much more modest
> database to see if this applied. I found a difference of 1 between the
> surname totals in the two gramplets. I assumed a bug therefore and
> intended to await a response from someone who knows more about the
> program.
>
> However I also noticed in the 'Statistics' gramplet that I had 3
> individuals with incomplete names.

I would expect an empty surname field be counted as one unique surname.

My statistics gramplet shows I have 2 incomplete names, i.e. individuals
without a surname. Interestingly when I look at the data itself it in
fact contains 7 individuals without a surname. Whether the correct
number of individuals is 2, 7 or some other number, for which I at this
time have no explanation, it is not equal to the difference of 42
between the number of unique surnames shown by the statistics gramplet
and the top surnames gramplet.

> When I had tracked these down and
> fixed them I noticed that the 'Statistics' gramplet and the 'Top
> Surnames' gramplet now agreed on the total number of unique surnames
> (46), which was also the number of Surnames I had physically counted.
>
> I have no idea as to which of my 3 incomplete names was the important
> one so I can give no more useful advice on what to look for other than
> check if you have incomplete names in the 'Statistics' gramplet. In
> all 3 cases the individuals were female and in each case it was the
> *married* name which was the problem (although not the same problem in
> all 3)

All considering it looks like this warrants entering a bug report.

Thanks for taking the time to respond, regards, Jk.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Difference in number of unique surnames shown by different gramplets.

John W. Kitz-3
In reply to this post by prculley
Paul,

On 2018-01-25 15:56, Paul Culley wrote:
> The Statistics Gramplet uses some metadata in the db that is only
> updated when Gramps closes cleanly.  Unfortunately that means that a
> crash, and probably some other issues, can leave the metadata out of
> sync with the other parts of the db.
>
> The 'Tools/Family Tree Repair/Rebuild Gender Statistics' can be used
> to correct this situation.

Ran "Rebuild Gender Statistics" per your suggestion, but the issue
remains.

> If the two Gramplets remain out of sync after running this tool, the
> developers would be interested in looking over the tree to see if
> there is a bug.  So it would be appreciated if you would file a bug
> report https://gramps-project.org/bugs/bug_report_page.php and attach
> an XML export of tree.

I'll raise a bug report in due course, but for obvious reasons don't
expect me to attach an XML export of all data.

>
> Paul C.
>

Thanks and regards, Jk.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Difference in number of unique surnames shown by different gramplets.

John Paton
On 25/01/18 15:09, John W. Kitz wrote:

> Paul,
>
> On 2018-01-25 15:56, Paul Culley wrote:
>> The Statistics Gramplet uses some metadata in the db that is only
>> updated when Gramps closes cleanly.  Unfortunately that means that a
>> crash, and probably some other issues, can leave the metadata out of
>> sync with the other parts of the db.
>>
>> The 'Tools/Family Tree Repair/Rebuild Gender Statistics' can be used
>> to correct this situation.
>
> Ran "Rebuild Gender Statistics" per your suggestion, but the issue remains.
>
>> If the two Gramplets remain out of sync after running this tool, the
>> developers would be interested in looking over the tree to see if
>> there is a bug.  So it would be appreciated if you would file a bug
>> report https://gramps-project.org/bugs/bug_report_page.php and attach
>> an XML export of tree.
>
> I'll raise a bug report in due course, but for obvious reasons don't
> expect me to attach an XML export of all data.
>
>>
>> Paul C.
>>
>
> Thanks and regards, Jk.
>
> ------------------------------------------------------------------------------
>
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> Gramps-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gramps-users
> https://gramps-project.org
>
I have done some more testing - for two of the individuals with
incomplete names the married surname had been entered in the prefix
field in the additional name editor instead of the surname field. This
was not obvious until the name editor was actually opened - elsewhere
the names displayed as expected. However in effect for these two
individuals they had a " " for their married surname.

The Top Surnames gramplet showed 47 unique surname while the Statistics
gramplet showed 46. After the corrections both gramplets showed 46
unique surnames.

In other words gramps treated " " as a unique surname. There were two
individuals with this unique surname so both of these individuals had to
be corrected in order to eliminate the " ". Once both were corrected the
number of unique surnames was reduced by one.

Why the two gramplets originally showed different totals is no clearer
however.

--
Time flies like an arrow. Fruit flies like a banana (or perhaps a melon).

John Paton

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Difference in number of unique surnames shown by different gramplets.

John W. Kitz-3
John,

On 2018-01-26 00:41, John Paton wrote:

> I have done some more testing - for two of the individuals with
> incomplete names the married surname had been entered in the prefix
> field in the additional name editor instead of the surname field. This
> was not obvious until the name editor was actually opened - elsewhere
> the names displayed as expected. However in effect for these two
> individuals they had a " " for their married surname.
>
> The Top Surnames gramplet showed 47 unique surname while the
> Statistics gramplet showed 46. After the corrections both gramplets
> showed 46 unique surnames.
>
> In other words gramps treated " " as a unique surname.

I obviously can't speak on behalf of the developers of Gramps and its
gramplets, but generally speaking a field intended to be used to enter
data does not necessarily have to be empty even though it appears empty
to the human eye. The simplest form being a field that contains a space
character.

> There were two
> individuals with this unique surname so both of these individuals had
> to be corrected in order to eliminate the " ". Once both were
> corrected the number of unique surnames was reduced by one.
>
> Why the two gramplets originally showed different totals is no clearer
> however.

Having looked at it in a bit more detail myself, and given the
difference of 7 between both counts, I wonder if the discrepancy may
actually result from the top surname gramplet counting families rather
than unique surnames as the statistics gramplet reportedly does.

Regards, Jk.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Difference in number of unique surnames shown by different gramplets.

Nick Hall
On 26/01/18 14:50, John W. Kitz wrote:
Why the two gramplets originally showed different totals is no clearer however.

Having looked at it in a bit more detail myself, and given the difference of 7 between both counts, I wonder if the discrepancy may actually result from the top surname gramplet counting families rather than unique surnames as the statistics gramplet reportedly does.

The statistics gramplet only counts surnames in primary names.  All surname parts are used except for patronymic and matronymic.

The surname cloud gramplet also counts surnames in alternate names.  In addition it uses the "group as" where defined and strips off leading and trailing spaces.

Nick.



------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org