Show Posts
1
Europe / Re: German Pork Butchers' Database
« on: Wednesday 26 August 15 20:26 BST (UK) »
SwissGill,
Thanks for the vote of confidence for the work limehouse79 and I are doing as data coordinators, researchers and contributors. It is appreciated, and sorry for the long dissertation on duplicates below.
The data coordinators have over 2700 records to work through, many with date of birth and abode fields blank (although many have this data in the notes section), and this makes it more difficult to identify duplicates. Further, we do not have, nor would we accept general record deletion privileges, but we are working with other contributors to try and wean the duplicates through consensus. It is a slow process that sometimes takes a few PMs to resolve for each and every duplicate.
To better identify duplicates we have been editing records to copying information from the notes sections to the appropriate birth, death, marriage, immigration, naturalisation and parents data fields (I leave the data in the Notes Section and do not delete it), so that gradually the database will become more searchable and sortable. One other edit I perform is selecting / filling in country and county in the Abode Fields where they are blank, but a city or town has been referenced in the notes. I am not adding information to the records of other contributors as I am simply copying existing information to the appropriate data field. This is an agreed process between the data coordinators and Berlin-Bob, the GPB database Guru.
By this process I have begun to identify certain records as duplicates, and I have been approaching other contributors about the deletion of a duplicate, many of which have just “skeleton” information. This process is random as I work through some of my own research, as I do like to keep my research interests going.
A more difficult duplication occurs with the surname entry, but we are gradually working away on this as well. There are many anglicised entries input by contributors unaware of the German surname. Another problem with this is the fact that numerous German surnames have letters with umlauts, and the search/sort listing do not always sort anglicised and German surnames together. For example, do a quick search on GOLTENBOTH/GÖLTENBOTH. You will find the umlauted version sorts at the top as GOELTENBOTH is assumed, followed by a GOLDHAMMER, 2 GOLLERS, and then the anglicised GOLTENBOTHS. This is an unavoidable quirk in the database and how umlauts can be sorted, but not all contributors search the full list for other possibilities before adding a record of their own.
Removing duplicates will be a slow process, but work is being done behind the scenes to reduce duplication. So far it seems to require numerous PMs and much elapsed time between identification and completion, and not everyone responds. If you are aware of these please feel free to send me, limehouse79 or Berlin-Bob a PM identifying the issue/duplicate.
Finally, we have many, many “orphan” records left by well-meaning contributors whose last log in and/or contribution is measured in years, not months or days.
As a general comment, I attempt to lead by example. If a record already exists I add Readers Comments and send a PM to the owner advising them of the new information. If I create a record and find that someone else has a familial relationship, where I do not, and they wish to have ownership of the record, I suggest that they create a new record, copy the information I have and then request I delete my entry … no problem. Both these methodologies rely heavily on contributors using the PM system to notify other contributors of issue, so here is hoping this will help to spur on more dialogue among the GPB contributors.
Thanks for your comments and confidence. Cheers … arky101
Thanks for the vote of confidence for the work limehouse79 and I are doing as data coordinators, researchers and contributors. It is appreciated, and sorry for the long dissertation on duplicates below.
The data coordinators have over 2700 records to work through, many with date of birth and abode fields blank (although many have this data in the notes section), and this makes it more difficult to identify duplicates. Further, we do not have, nor would we accept general record deletion privileges, but we are working with other contributors to try and wean the duplicates through consensus. It is a slow process that sometimes takes a few PMs to resolve for each and every duplicate.
To better identify duplicates we have been editing records to copying information from the notes sections to the appropriate birth, death, marriage, immigration, naturalisation and parents data fields (I leave the data in the Notes Section and do not delete it), so that gradually the database will become more searchable and sortable. One other edit I perform is selecting / filling in country and county in the Abode Fields where they are blank, but a city or town has been referenced in the notes. I am not adding information to the records of other contributors as I am simply copying existing information to the appropriate data field. This is an agreed process between the data coordinators and Berlin-Bob, the GPB database Guru.
By this process I have begun to identify certain records as duplicates, and I have been approaching other contributors about the deletion of a duplicate, many of which have just “skeleton” information. This process is random as I work through some of my own research, as I do like to keep my research interests going.
A more difficult duplication occurs with the surname entry, but we are gradually working away on this as well. There are many anglicised entries input by contributors unaware of the German surname. Another problem with this is the fact that numerous German surnames have letters with umlauts, and the search/sort listing do not always sort anglicised and German surnames together. For example, do a quick search on GOLTENBOTH/GÖLTENBOTH. You will find the umlauted version sorts at the top as GOELTENBOTH is assumed, followed by a GOLDHAMMER, 2 GOLLERS, and then the anglicised GOLTENBOTHS. This is an unavoidable quirk in the database and how umlauts can be sorted, but not all contributors search the full list for other possibilities before adding a record of their own.
Removing duplicates will be a slow process, but work is being done behind the scenes to reduce duplication. So far it seems to require numerous PMs and much elapsed time between identification and completion, and not everyone responds. If you are aware of these please feel free to send me, limehouse79 or Berlin-Bob a PM identifying the issue/duplicate.
Finally, we have many, many “orphan” records left by well-meaning contributors whose last log in and/or contribution is measured in years, not months or days.
As a general comment, I attempt to lead by example. If a record already exists I add Readers Comments and send a PM to the owner advising them of the new information. If I create a record and find that someone else has a familial relationship, where I do not, and they wish to have ownership of the record, I suggest that they create a new record, copy the information I have and then request I delete my entry … no problem. Both these methodologies rely heavily on contributors using the PM system to notify other contributors of issue, so here is hoping this will help to spur on more dialogue among the GPB contributors.
Thanks for your comments and confidence. Cheers … arky101
In other words, who knows what combination of keystrokes and mouse clicks were used to get me from formatted to un-formulated ... perhaps the easiest task is to let me reformat the text, keep an eye on it, and move on.