From: Frederico M. <fs...@gm...> - 2009-04-06 14:52:26
|
Greetings, This is again about one of the areas which I simultaneous like the most in GRAMPS (compared to others) and also the one with which I struggle the most: source management, in particular the source and source reference components. I follow a single source -> multiple source references approach, which means that I have a single Baptism Records of Parish Foobar and multiple source references pointing there, with the log data and number if applicable. I find this clean and logical. One problem however is that it can sometimes be hard to pin exactly which source reference is the one that actually includes a certain information. A small example will make this clearer: Generally civil birth records around here contain the age of both parents. This means that this birth record is also a source for the parents birth event. Sometimes the same happens with Church records. So for a single person I can end up having around 4 different sources for the birth event, including the most important one that refers specifically to the individual. When using source references all sources look similar, i.e. it is hard to look at them and see "this is the birth record of X". After a while this is a bit of a bore since I have to hunt down exactly what the source reference actually means. As a way to mitigate this I have been using the source reference note field to add a note saying "This is the birth certificate of John Doe". This at least makes it possible to know what the source refers to, even if it still needs a lot of clicking to get there. This is made worse if I need to add something to the source reference *after* having already copied it to all the relevant events (say, a note as above or the transcript of the source reference) given that all source references are copies (understandably). I must find out every similar reference and share the same note. While I have no idea on how to make this last problem go away I was thinking that it could perhaps be possible (or not, this is just an idea) to show in the sources listing of an event an extra field to be able to know that "Baptism Records of the Parish of Argh, Fls. 212, n. 45, Year 1820" is Johan Stewart Whatsisname birt record. I'm not sure on how this could be made in terms of GEDCOM compatibility, or even if at all. Alternatively maybe this is not an issue to most, and maybe I'm doing something wrong, in which case corrections would be welcomed ;) Cheers, Frederico Muñoz |
From: Benny M. <ben...@gm...> - 2009-04-06 15:24:17
|
2009/4/6 Frederico Muñoz <fs...@gm...> > Greetings, > > This is again about one of the areas which I simultaneous like the > most in GRAMPS (compared to others) and also the one with which I > struggle the most: source management, in particular the source and > source reference components. > > I follow a single source -> multiple source references approach, which > means that I have a single Baptism Records of Parish Foobar and > multiple source references pointing there, with the log data and > number if applicable. I find this clean and logical. Yes. Source reference is about how to fine the info of the object in the source. So you should put there everything you need to find your info in the source: page, log date, share a note of the source with the sourceref, ... At the moment, the list of sources in an object only lists the page of the sourceref. You should interpret page broadly in my view, you can use it to list more info depending on your needs. > One problem however is that it can sometimes be hard to pin exactly > which source reference is the one that actually includes a certain > information. A small example will make this clearer: > > Generally civil birth records around here contain the age of both > parents. This means that this birth record is also a source for the > parents birth event. Sometimes the same happens with Church records. > So for a single person I can end up having around 4 different sources > for the birth event, including the most important one that refers > specifically to the individual. > When using source references all sources look similar, i.e. it is hard > to look at them and see "this is the birth record of X". After a while > this is a bit of a bore since I have to hunt down exactly what the > source reference actually means. > I believe you say that sourceref needs a description. A running text to make clear what it references in the source. Do a feature request. You can misuse the page field to achieve it somewhat: page 2 - section 4.1 descendants of John Doe or page 2 - birth record John Doe or page 2 - note 19090401: birth John Doe The last in the idea that you organize the source with notes, every notes beginning with a line of the log date and the person it is about. My point is just that the sourceref should contain the data needed to find info in the source. It is hard however to agree on what is needed to achieve that. At the moment page and date are given. The discussion should go about what extra is needed. We could try to be intelligent: if a note is present that is present also in the source, then display the first line of that note too. That looks like a difficult scheme however. Adding a description field the user uses as he likes is probably easier. Benny > istinfo/gramps-users<https://lists.sourceforge.net/lists/listinfo/gramps-users> > |
From: Frederico M. <fs...@gm...> - 2009-04-06 15:44:25
|
Hi, On Mon, Apr 6, 2009 at 4:24 PM, Benny Malengier <ben...@gm...> wrote: > Yes. Source reference is about how to fine the info of the object in the > source. So you should put there everything you need to find your info in the > source: page, log date, share a note of the source with the sourceref, ... > At the moment, the list of sources in an object only lists the page of the > sourceref. You should interpret page broadly in my view, you can use it to > list more info depending on your needs. Yes, I do that, I add whatever info I have, be it page, registry number, folios, etc. > I believe you say that sourceref needs a description. Ehe, yes, that is an apt short version of my entire rambling ;) > A running text to make > clear what it references in the source. Do a feature request. You can misuse > the page field to achieve it somewhat: > page 2 - section 4.1 descendants of John Doe > or > page 2 - birth record John Doe > or > page 2 - note 19090401: birth John Doe > I will make a feature request then, I didn't do it yet since I wanted to see if I was looking at it the wrong way. While possible I think that adding that text to the page field is a bit "hackerish". > My point is just that the sourceref should contain the data needed to find > info in the source. It is hard however to agree on what is needed to achieve > that. At the moment page and date are given. The discussion should go about > what extra is needed. We could try to be intelligent: if a note is present > that is present also in the source, then display the first line of that note > too. That looks like a difficult scheme however. Adding a description field > the user uses as he likes is probably easier. I actually think that the source reference fields are fine as they presently are in terms of the above purpose: finding the information on the source. I don't have a need for anything new in that regard, merely in terms of a more human-readable description to be able to tell apart source references that end up being very similar. So it is more of a cosmetic issue in the end, and a description field would work perfectly, so I'll put that in the feature request. I'll also search the existing feature request concerning a way to find similar (i.e. "equal" in terms of actual reference information) source references and perhaps "sync" them in some way. As I said adding a transcript at a latter date is non-trivial given the possible proliferation of the source reference. Regards, Frederico |
From: Jérôme <rom...@ya...> - 2009-04-06 18:51:00
|
>> I believe you say that sourceref needs a description. > > Ehe, yes, that is an apt short version of my entire rambling ;) I sometimes use "Publication information" for adding the place, date of the source (or others informations). http://www.gramps-project.org/wiki/index.php?title=Image:Add_source_2.png http://www.gramps-project.org/wiki/index.php?title=Image:Add_sourceref_2.png http://www.gramps-project.org/wiki/index.php?title=Image:Add_repository_2.png It is OK for scanned or numerical sources, and I only use sourceref for books (currently, often scanned). Maybe this was not the case 10 years ago (GEDCOM 5.5) ? Jérôme Frederico Muñoz a écrit : > Hi, > > On Mon, Apr 6, 2009 at 4:24 PM, Benny Malengier > <ben...@gm...> wrote: > >> Yes. Source reference is about how to fine the info of the object in the >> source. So you should put there everything you need to find your info in the >> source: page, log date, share a note of the source with the sourceref, ... >> At the moment, the list of sources in an object only lists the page of the >> sourceref. You should interpret page broadly in my view, you can use it to >> list more info depending on your needs. > > Yes, I do that, I add whatever info I have, be it page, registry > number, folios, etc. > >> I believe you say that sourceref needs a description. > > Ehe, yes, that is an apt short version of my entire rambling ;) > >> A running text to make >> clear what it references in the source. Do a feature request. You can misuse >> the page field to achieve it somewhat: >> page 2 - section 4.1 descendants of John Doe >> or >> page 2 - birth record John Doe >> or >> page 2 - note 19090401: birth John Doe >> > > I will make a feature request then, I didn't do it yet since I wanted > to see if I was looking at it the wrong way. While possible I think > that adding that text to the page field is a bit "hackerish". > >> My point is just that the sourceref should contain the data needed to find >> info in the source. It is hard however to agree on what is needed to achieve >> that. At the moment page and date are given. The discussion should go about >> what extra is needed. We could try to be intelligent: if a note is present >> that is present also in the source, then display the first line of that note >> too. That looks like a difficult scheme however. Adding a description field >> the user uses as he likes is probably easier. > > I actually think that the source reference fields are fine as they > presently are in terms of the above purpose: finding the information > on the source. I don't have a need for anything new in that regard, > merely in terms of a more human-readable description to be able to > tell apart source references that end up being very similar. So it is > more of a cosmetic issue in the end, and a description field would > work perfectly, so I'll put that in the feature request. > > I'll also search the existing feature request concerning a way to find > similar (i.e. "equal" in terms of actual reference information) source > references and perhaps "sync" them in some way. As I said adding a > transcript at a latter date is non-trivial given the possible > proliferation of the source reference. > > Regards, > > Frederico > > ------------------------------------------------------------------------------ > _______________________________________________ > Gramps-users mailing list > Gra...@li... > https://lists.sourceforge.net/lists/listinfo/gramps-users > |
From: Frederico M. <fs...@gm...> - 2009-04-06 19:04:24
|
Hi, On Mon, Apr 6, 2009 at 8:08 PM, Jérôme <rom...@ya...> wrote: > I sometimes use "Publication information" for adding the place, date of the > source (or others informations). > http://www.gramps-project.org/wiki/index.php?title=Image:Add_source_2.png > http://www.gramps-project.org/wiki/index.php?title=Image:Add_sourceref_2.png > http://www.gramps-project.org/wiki/index.php?title=Image:Add_repository_2.png > > It is OK for scanned or numerical sources, and I only use sourceref for > books (currently, often scanned). Maybe this was not the case 10 years ago > (GEDCOM 5.5) ? Well, the main difference here is that we primarily use two different ways of dealing with sources (as per another previous conversation that we had): you mostly use the one-source-per-record approach, while I use the one source reference per record, linking to the same source. Mainly this mean some kind of a book, and 90% of my sources are books in a way or another (Parish Records, etc). The Public Information field is tied to the Source, and what I'm lacking is something that applies to the different source references. Regards, Frederico |
From: Jérôme <rom...@ya...> - 2009-04-07 08:18:59
|
True, as source reference is unique between one source and (event, person, media, place, attribute, association, name, address, etc ...). Sourceref related to event will be well displayed on reports using EndNotes, but there is still a sourceref management issue on other primary or secondary objects ! I had added simple filter rules for matching Event having sourceref. Maybe this could help you to manage your source references on Events : http://www.gramps-project.org/bugs/view.php?id=2831 Jérôme Frederico Muñoz a écrit : > Hi, > > On Mon, Apr 6, 2009 at 8:08 PM, Jérôme <rom...@ya...> wrote: >> I sometimes use "Publication information" for adding the place, date of the >> source (or others informations). >> http://www.gramps-project.org/wiki/index.php?title=Image:Add_source_2.png >> http://www.gramps-project.org/wiki/index.php?title=Image:Add_sourceref_2.png >> http://www.gramps-project.org/wiki/index.php?title=Image:Add_repository_2.png >> >> It is OK for scanned or numerical sources, and I only use sourceref for >> books (currently, often scanned). Maybe this was not the case 10 years ago >> (GEDCOM 5.5) ? > > Well, the main difference here is that we primarily use two different > ways of dealing with sources (as per another previous conversation > that we had): you mostly use the one-source-per-record approach, while > I use the one source reference per record, linking to the same source. > Mainly this mean some kind of a book, and 90% of my sources are books > in a way or another (Parish Records, etc). > > The Public Information field is tied to the Source, and what I'm > lacking is something that applies to the different source references. > > Regards, > > Frederico > > ------------------------------------------------------------------------------ > This SF.net email is sponsored by: > High Quality Requirements in a Collaborative Environment. > Download a free trial of Rational Requirements Composer Now! > http://p.sf.net/sfu/www-ibm-com > _______________________________________________ > Gramps-users mailing list > Gra...@li... > https://lists.sourceforge.net/lists/listinfo/gramps-users > |
From: Frederico M. <fs...@gm...> - 2009-04-07 20:44:05
|
Hi, On Tue, Apr 7, 2009 at 9:34 AM, Jérôme <rom...@ya...> wrote: (...) > I had added simple filter rules for matching Event having sourceref. Maybe > this could help you to manage your source references on Events : > http://www.gramps-project.org/bugs/view.php?id=2831 Excellent, extremely useful indeed, especially given my extensive usage of source references. Many thanks! Frederico |
From: Frederico M. <fs...@gm...> - 2009-04-09 23:06:27
|
Hi, Sorry to answer twice in a row. On Tue, Apr 7, 2009 at 9:34 AM, Jérôme <rom...@ya...> wrote: > > True, as source reference is unique between one source and (event, person, media, place, attribute, association, name, address, etc ...). Sourceref related to event will > be well displayed on reports using EndNotes, but there is still a sourceref management issue on other primary or secondary objects ! I've been thinking and experimenting and I don't really see a way out of it, I'm probably have to stop using my sourceref approach and just put everything as a source, which I find clumsy and lacking in elegance. Since sourcerefs can't be shared, only copied, it is extremely difficult to keep track of them. I've just started adding some transcriptions and I've been spending almost the same amout of time writing the text as I do trying to find every single instance of the source reference to share the note with the transcription. Very error prone, very annoying and not worth the time in the end since every time I need to correct something or add a note I have to repeat the process again. As I said in a genealogy review site I have actually tried some other genealogy software and I think that GRAMPS has the best source management, so this is not some "it sucks!" kind of comment :) Maybe one way would be to adapt the filter that you wrote in a way that I could find specific instances of a source reference, ideally via a contextual menu of some kind. This could also be used as the basis for some kind of "sync every source reference" action. This are just wild ideas, I'm not sure if something better and more elegant is being implemented. But as it is using source references extensively (GRAMPS has made me very obsessed with adding the correct sources and citations to everything) is something that doesn't work very well... this sort of adds to my months-old thread concerning images and sourcereferences: the "correct" way of doing things (which I would like to use, since I find it superior in theory) finds some obstacles in actual usage that hinder it. I will of course file feature reports, I'm even looking through the code and trying to come with something simple based on filters, but I still think that this warrants some more in-depth discussion. I've seen this two issues come up as the first "real" problems in at least two different persons to which I introduced GRAMPS, and which are very happy with it ("real" in the sense that they are not due to lack of familiarity... they are actually the opposite, since GRAMPS induces good practices in terms of source management which make them more apparent) Regards, and a Happy Easter to all, Frederico |
From: Jérôme <rom...@ya...> - 2009-04-10 06:32:13
|
Hi, As said before, source references on events are useful for publications, books, papers, which will be like "bibliography references" on Reports. Note, Gramps just deals with GEDCOM specifications ... http://www.gramps-project.org/wiki/index.php?title=Sources_in_GRAMPS Happy Easter, Jérôme R. Frederico Muñoz a écrit : > Hi, > > Sorry to answer twice in a row. > > On Tue, Apr 7, 2009 at 9:34 AM, Jérôme <rom...@ya...> wrote: >> True, as source reference is unique between one source and (event, person, media, place, attribute, association, name, address, etc ...). Sourceref related to event will > be well displayed on reports using EndNotes, but there is still a sourceref management issue on other primary or secondary objects ! > > I've been thinking and experimenting and I don't really see a way out > of it, I'm probably have to stop using my sourceref approach and just > put everything as a source, which I find clumsy and lacking in > elegance. Since sourcerefs can't be shared, only copied, it is > extremely difficult to keep track of them. I've just started adding > some transcriptions and I've been spending almost the same amout of > time writing the text as I do trying to find every single instance of > the source reference to share the note with the transcription. Very > error prone, very annoying and not worth the time in the end since > every time I need to correct something or add a note I have to repeat > the process again. > > As I said in a genealogy review site I have actually tried some other > genealogy software and I think that GRAMPS has the best source > management, so this is not some "it sucks!" kind of comment :) > > Maybe one way would be to adapt the filter that you wrote in a way > that I could find specific instances of a source reference, ideally > via a contextual menu of some kind. This could also be used as the > basis for some kind of "sync every source reference" action. > > This are just wild ideas, I'm not sure if something better and more > elegant is being implemented. But as it is using source references > extensively (GRAMPS has made me very obsessed with adding the correct > sources and citations to everything) is something that doesn't work > very well... this sort of adds to my months-old thread concerning > images and sourcereferences: the "correct" way of doing things (which > I would like to use, since I find it superior in theory) finds some > obstacles in actual usage that hinder it. > > I will of course file feature reports, I'm even looking through the > code and trying to come with something simple based on filters, but I > still think that this warrants some more in-depth discussion. I've > seen this two issues come up as the first "real" problems in at least > two different persons to which I introduced GRAMPS, and which are very > happy with it ("real" in the sense that they are not due to lack of > familiarity... they are actually the opposite, since GRAMPS induces > good practices in terms of source management which make them more > apparent) > > Regards, and a Happy Easter to all, > > Frederico > > ------------------------------------------------------------------------------ > This SF.net email is sponsored by: > High Quality Requirements in a Collaborative Environment. > Download a free trial of Rational Requirements Composer Now! > http://p.sf.net/sfu/www-ibm-com > _______________________________________________ > Gramps-users mailing list > Gra...@li... > https://lists.sourceforge.net/lists/listinfo/gramps-users > |
From: Frederico M. <fs...@gm...> - 2009-04-10 13:52:27
|
Hi, On Fri, Apr 10, 2009 at 7:38 AM, Jérôme <rom...@ya...> wrote: > Hi, > > As said before, source references on events are useful for publications, > books, papers, which will be like "bibliography references" on Reports. I see, but bear in mind that just as the example in the GRAMPS page (a death certificate) books of some kind are in some countries the backbone of almost all the information regarding events and the same reference can be the source of severeal events. Which is why I will start to migrate from my way of doing things to stop using source references altogether (as it seems most people do anyway) given the limitations I described. > Note, Gramps just deals with GEDCOM specifications ... > http://www.gramps-project.org/wiki/index.php?title=Sources_in_GRAMPS Of course, and to be honest the little experiments I made with some other programs to see how they managed this were even worse. But GEDCOM specifications wouldn't be I think an obstacle to having something in place that would facilitate the "syncing" of source references since in the end the GEDCOM output would not be affected, merely the way the information in GRAMPS is treated. Since this is not apparently something that bothers people I'm probably the one doing something wrong, so as said I'll just stop insisting on using the source reference approach and "promote" all my source references to sources. It will look odd in reports but at least the infrastructure is there to easily manage transcripts, notes, images, etc. Regards, Frederico |
From: Benny M. <ben...@gm...> - 2009-04-10 14:35:25
|
2009/4/10 Frederico Muñoz <fs...@gm...> > Hi, > > On Fri, Apr 10, 2009 at 7:38 AM, Jérôme <rom...@ya...> wrote: > > Hi, > > > > As said before, source references on events are useful for publications, > > books, papers, which will be like "bibliography references" on Reports. > > I see, but bear in mind that just as the example in the GRAMPS page (a > death certificate) books of some kind are in some countries the > backbone of almost all the information regarding events and the same > reference can be the source of severeal events. Which is why I will > start to migrate from my way of doing things to stop using source > references altogether (as it seems most people do anyway) given the > limitations I described. > > > Note, Gramps just deals with GEDCOM specifications ... > > http://www.gramps-project.org/wiki/index.php?title=Sources_in_GRAMPS > > Of course, and to be honest the little experiments I made with some > other programs to see how they managed this were even worse. But > GEDCOM specifications wouldn't be I think an obstacle to having > something in place that would facilitate the "syncing" of source > references since in the end the GEDCOM output would not be affected, > merely the way the information in GRAMPS is treated. > > Since this is not apparently something that bothers people I'm > probably the one doing something wrong, so as said I'll just stop > insisting on using the source reference approach and "promote" all my > source references to sources. It will look odd in reports but at least > the infrastructure is there to easily manage transcripts, notes, > images, etc. In my view, the point is that source refs are unique and sources are shared. There is no other way to do this. However, the consequence is that you should not add data to the sourceref that is supposed to be shared. The sourceref should be used only to indicate where in the source the data concerning the person/event/.... can be found. So if your source is a parish registry, you can make the subentries sources, or you can make them notes in your source using a numbering scheme you like. You should not use the sourcereferences however to hold data of the source entries. This data _must_ be present in the source. If you use notes to hold a birth entry, it is easy to share this note also in all sourceref. Alternatively, if a source is a log entry (as birth certificates), then the source ref only needs to contain the log date, nothing else, and the information in the source must be done in such a way that this log entry can be easily found in the source. So the media object contains this date in the title, the Note has as first line this date, .... Clearly, many people create a specific source per birth certificate and not for the entire parish registry. This is due to the fact that GRAMPS does not allow a source-subsource structure. That is, a scheme where a source has a field: parent source, indicating the source it is part of. Benny > > > Regards, > > Frederico > > > ------------------------------------------------------------------------------ > This SF.net email is sponsored by: > High Quality Requirements in a Collaborative Environment. > Download a free trial of Rational Requirements Composer Now! > http://p.sf.net/sfu/www-ibm-com > _______________________________________________ > Gramps-users mailing list > Gra...@li... > https://lists.sourceforge.net/lists/listinfo/gramps-users > |
From: Frederico M. <fs...@gm...> - 2009-04-10 16:14:13
|
Hi, Many thanks for the feedback. On Fri, Apr 10, 2009 at 3:35 PM, Benny Malengier <ben...@gm...> wrote: > In my view, the point is that source refs are unique and sources are shared. > There is no other way to do this. However, the consequence is that you > should not add data to the sourceref that is supposed to be shared. The > sourceref should be used only to indicate where in the source the data > concerning the person/event/.... can be found. What mislead me then is that there is the possibility of adding things like "transcript" and "source text" in the source reference - probably something that is allowed since they are standard fields but not really appropriate for a source reference. However I should point out that other programs out there have two specific fields in the source reference that seem to contradict this best-practice: they have a "Source Text" and "Image" fields in the Source citation. > So if your source is a parish registry, you can make the subentries sources, > or you can make them notes in your source using a numbering scheme you like. > You should not use the sourcereferences however to hold data of the source > entries. This data _must_ be present in the source. If you use notes to hold > a birth entry, it is easy to share this note also in all sourceref. Well, easy is relative... I would have to find out every specific reference made to share it. It is here that I mainly have problems since it is a bit tedious. Any kind of sharing of notes is a manual process in terms of finding the source references. Point taken on the location of Source Text though, I'll start adding them to the Sources. And then simply leave the source reference "as is", without any note sharing or additions since that would mean that if I at a latter date decided to add something I would have to track down every source reference. > Alternatively, if a source is a log entry (as birth certificates), then the > source ref only needs to contain the log date, nothing else, and the > information in the source must be done in such a way that this log entry can > be easily found in the source. So the media object contains this date in the > title, the Note has as first line this date, .... I sort of had that, then I decided to move things (apparently wrongly so) to the source reference to make it easier to find. I think I will revert to it then, as it is now I have more problems than before :) > Clearly, many people create a specific source per birth certificate and not > for the entire parish registry. This is due to the fact that GRAMPS does not > allow a source-subsource structure. That is, a scheme where a source has a > field: parent source, indicating the source it is part of. GRAMPS or any of the other programs I tried, which I stress were not at all "better" than GRAMPS so this isn't some kind of specific limitation. What I will do is to try to readjust what I have according to some of the suggestions, I will do the following: - Leave the Source References alone in terms of content, no notes apart from a description-like note to indicate what they mean (as I said in another thread this is useful for me to quickly relate a reference to something more informational as "Birth Certificate of John Doe"). - Add the source text as notes in the Source itself - Add the images in the Source itself In addition *maybe* share the Source Text notes with each source reference... but this means that I would have to track them all down, which is a bore and not really well supported. The only problem with the above approach is that going from an Event to, say, the scanned image that references involves several chained windows and some clicking (Event->Source Reference->Source->Finding the image "manually" by finding it in the Source Gallery), but it is better than having to worry about having inconsistent data scattered by the different instances of the same references. Ideally I think that a program such as GRAMPS should simplify data entering and as much as possible avoid the need of manually checking consistency. This is, I see it now, more possible when adding everything to the source and leave the source references as empty as possible. I will try to write something for the wiki about this. Regards, Frederico |
From: Benny M. <ben...@gm...> - 2009-04-11 07:35:01
|
2009/4/10, Frederico Muñoz <fs...@gm...>: > which is a bore and not really well supported. The only problem with > the above approach is that going from an Event to, say, the scanned > image that references involves several chained windows and some > clicking (Event->Source Reference->Source->Finding the image > "manually" by finding it in the Source Gallery), but it is better than > having to worry about having inconsistent data scattered by the > different instances of the same references. You do not give the gramps programmers their due credit :-D It is not Event->Source Reference->Source->Finding the image as source ref and source are shown at the same time, so as to avoid this extra step. I have always wanted to implement some way to indicate in the source what relates to the present sourceref a person is looking at (eg bolder note/media/data). But I have not been able to think of a good scheme that would not be too manual. My ideas more go to implementing a subsource object, so that a source can have subsources, and a sourceref can be to a source or to a subsource then. The sourceview would then be as stacked view as the person view is (source can be expanded to see subsources). If ideas around this, please share. Benny |
From: Frederico M. <fs...@gm...> - 2009-04-11 14:54:41
|
Hi, On Sat, Apr 11, 2009 at 8:34 AM, Benny Malengier <ben...@gm...> wrote: > You do not give the gramps programmers their due credit :-D > It is not > Event->Source Reference->Source->Finding the image > as source ref and source are shown at the same time, so as to avoid > this extra step. Apologies, you are of course correct. And while I know you were joking do believe me when I give GRAMPS programmers all the credit that they deserve, given that almost all my genealogy work is only possible due to their effort. > I have always wanted to implement some way to indicate in the source > what relates to the present sourceref a person is looking at (eg > bolder note/media/data). But I have not been able to think of a good > scheme that would not be too manual. This is exactly what I would like. I'm happy now, given that I know that what I was referring to is something that people are aware of as something that could be useful, regardless of being implemented or note due to difficulties in finding a good method. > My ideas more go to implementing a subsource object, so that a source > can have subsources, and a sourceref can be to a source or to a > subsource then. The sourceview would then be as stacked view as the > person view is (source can be expanded to see subsources). If ideas > around this, please share. I can only visualize it by example. It sounds a nightmare in terms of GEDCOM though :) To mitigate one of the problems I described - the need to track down all identical sourcerefs to fix an error or add a note - one possibility that wouldn't introduce any change to the program logic is as I mentioned using filters to find the identical source references and perhaps a Tool that doe a macro substitution of all of them. This is I think perhaps the most simple way, and I'm trying to see if I can work up something to that end. That, and the feature request I opened concerning a description field in the sourcerefs :) Regards, Frederico |
From: Frederico M. <fs...@gm...> - 2009-04-11 23:42:59
|
Hi, Just another quick comment that have occurred to me: On Sat, Apr 11, 2009 at 8:34 AM, Benny Malengier <ben...@gm...> wrote: > 2009/4/10, Frederico Muñoz <fs...@gm...>: (...) > I have always wanted to implement some way to indicate in the source > what relates to the present sourceref a person is looking at (eg > bolder note/media/data). But I have not been able to think of a good > scheme that would not be too manual. One thing that makes some sense to me is that assuming that all the information is in the Source it would be nice (not saying possible...) if the sourceref had a view of the parts of the Source that are applicable to the reference. When I say "view" I mean to say that it could be something that is displayed by GRAMPS but that doesn't necessarily translate into an actual relation, say, in terms of GEDCOM. Taking you own example as a starting point: - Source Text notes follow some well-defined criteria that allow one to pinpoint them to a sourceref - The same with the images *If* there was a way to "codify" this "well-defined criteria" then in the sourceref view the relevant material would be displayed. The sourceref would act as a "filter" that would only show the relevant information present in the source. The main problem with this idea is that both Notes (including Source Text) and the Gallery are non-keyed and have a free format. One would either need to add some sort of "key-value" attributes for this to work (even if merely discarded when converting from/to GEDCOM) or rely on some sort of regexp matching on the notes and Media names. The advantage would be that every problem associated with relating (and maintaining the relation) sourcerefs with parts of the Source would be solved; the disadvantage is that this is either hard to implement or goes against the way things are done. Just an idea though, even if very abstract and honestly lacking in terms of knowing what it entails in terms of GRAMPS internals. Regards, Frederico |
From: Frederico M. <fs...@gm...> - 2009-04-13 01:06:48
|
> *If* there was a way to "codify" this "well-defined criteria" then in > the sourceref view the relevant material would be displayed. The > sourceref would act as a "filter" that would only show the relevant > information present in the source. > (bla bla bla) While I still like this idea - and could be changed to something like highlighting the image/notes in the source, etc. - I want to make it clear that I'm not saying that sourcerefs should become "shared". Having read the GEDCOM standard one particular note caught my eye: source text in sourcerefs should be something only relevant to the specific reference *and* event, something as short as one sentence. This would in my examples be different for every event, so my "let's just sync it all!" approach fails miserably, it is exactly like Benny said (I never doubt it, was just looking at it from a different perspective), everything related to the Source stays in the Source, everything specific to the event and reference goes into the reference. As an example that could be potentially hopeful to others, the complete source text of a specific page of a book goes into the Source. The sourceref can have a "Source Text" note with a quick citation like "... born here in the 12th of Mars 1843...", which is entirely specific to the Birth event. My mistake was of always thinking in terms of "reference", the term "citation" did change my perception of it. The above works wonderfully when using the Narrative Web report, which is generally a good indication... the Source contains all the (extensive) source text, and the individual pages just some small snippets. I still think that the following would be desirable though. 1) Quick way to find all identical sourcerefs 2) Some way to identify the images/notes in the source that are relevant to the sourceref Regards, Frederico |
From: Frederico M. <fs...@gm...> - 2009-04-13 02:55:13
|
Sorry all for the multiple messages, I've been experimenting, exporting, importing, making reports, reading GEDCOM and trying to see how it works. I read this in the very beginning but only now have I saw it as something that makes real sense (as opposed to something utilitarian yet only done because of lack of better alternative) Benny said: >I believe you say that sourceref needs a description. A running text to make clear what it references in the source. > Do a feature request. You can misuse the page field to achieve it somewhat: I've been reading about it and it makes sense to put that description in the source reference. Especially given that I found this: ---- EVENTS_RECORDED:= {Size=1:90} [<EVENT_ATTRIBUTE_TYPE> | <EVENTS_RECORDED>, <EVENT_ATTRIBUTE_TYPE>] An enumeration of the different kinds of events that were recorded in a particular source. Each enumeration is separated by a comma. Such as a parish register of births, deaths, and marriages would be BIRT, DEAT, MARR. ---- So, this answers a small discomfort I was having with my Source definitions since I have "Baptism Records of the Parish of St. John", even if the "baptism" part is something that has not direct relation with the actual source (as I learned upon finding mixed books in Parishes that contain baptism, wedding and death records intermingled). Changing the Source to "Parish of St. John" makes it perfectly natural do add in the source reference something like "Baptism record of John, 1856, Pg. 13". This eliminates the need for the "description" field I was talking about, and actually looks better in reports (I was having duplicate TEXT entries). Also, in terms of handling the different relevance of different sources, I found this interesting: -------- EVENT_TYPE_CITED_FROM:= {SIZE=1:15} [ <EVENT_ATTRIBUTE_TYPE> ] A code that indicates the type of event which was responsible for the source entry being recorded. For example, if the entry was created to record a birth of a child, then the type would be BIRT regardless of the assertions made from that record, such as the mother's name or mother's birth date. This will allow a prioritized best view choice and a determination of the certainty associated with the source used in asserting the cited fact. -------- This would also aid in arranging the more relevant sources in some kind of order. After better analysing all of this - and reading the GEDCOM standard closer was a big help in understanding why things are done the way they are and why some suggestions are difficult or make little sense - most of my doubts are answered, and much of what I said was due to a misunderstanding on my part of the role of source references. A way to find them and a way to identify images would still be good, but other than that I'm actually happy now. I've changed several records and the appearance of it is impecable and the more event-orientated sourcerefs (with small citations supporting the event, etc) are a big help. If nothing else this conversations are a good way to learn all of this, but my apologies if I wasted anybody's time! Regards, Frederico |
From: Theo T. <tj....@ph...> - 2009-04-13 11:26:55
|
Frederico Mun~oz wrote: > <snip> I still think that the following would be desirable though. > > 1) Quick way to find all identical sourcerefs > 2) Some way to identify the images/notes in the source that are > relevant to the sourceref 1) These can only apply to identical events, persons or families: can there be any? 2) Surely these will be specified in the sourceref? Admittedly, I scarcely use sourcerefs myself! Yours sincerely, Theo Tulley. tj....@ph... SFHG Member No: 11619 |
From: Frederico M. <fs...@gm...> - 2009-04-10 16:27:44
|
Hello, On Fri, Apr 10, 2009 at 4:00 PM, Theo Tulley <tj....@ph...> wrote: > Just joining in the conversation! > > I'm new to this having acquired a lot of information mainly from others' > work, and only recently moved it into Gramps. I have until recently > neglected to enter sources. Now I am gradually editing where appropriate, at > the same time entering new data (persons). Yes, you should tackle sources as soon as you can. After a while some information is hard to remember :) > To keep things simple, I use where possible single "sources" such as Birth > certificate, or Parish record transcriptions. The former is entered if I > have the certificate; the Place information supplements the latter. My > Places do distinguish between a location and the parish church of that > location. For some sources, the on-line collection specifies "xx's > transcriptions" in addition to the parish; this is inserted as an additional > source - but once quoted is an "existing source" in Gramps. I've written in the wiki the way I deal with Sources and Repositories (http://www.gramps-project.org/wiki/index.php?title=Repositories_in_GRAMPS), maybe this could help. I generally use stuff like "Baptism Records of the Chuch of Argh" as Source, then specifiy page/date in the source reference. Somce sources are present in more than one media, and that is were I use Repositories. > I am using "Personal knowledge" for my immediate family. I use "Oral Knowledge" as a Repository and then "Memory of John Doe" as Source, using the date were I've collected the information as the source reference. Just one of many different ways to handle it. > Another generic > source might be "Fiche" when I get better at reading them! Ummm. a fiche in itself isn't a source, merely a medium... I mean, you could have the exact same information on a fiche, a paper copy and a scanned imaged. Furthermore many different and unrelated sources are in fiches. A scanned image of page 2 of a Baptism Book and a fiche that contains page 1 of said book have more in common that two fiches from totally unrelated books. > I guess that the details which can be entered in the upper half of the > Source entry form in Gramps provide for detailed references to, e.g., IGI, > or perhaps Court rolls, which will be valid only for a particular entry. If you are referring to the Source Reference upper half they can generally be applied to most "original" media (or even immaterial things like oral knowledge), especially once one separates the physical representation of a source with the source itself. Regards, Frederico |
From: Theo T. <tj....@ph...> - 2009-04-10 18:05:42
|
Hi Frederico: It seems to me that your method involves much unnecessary recording: If a source is specified as Parish Transcripts then just which parish, and the date, are already known from the data for which it is quoted a source. On the other hand, if the source is only a transcript, possible errors are implied and if questions arise the more accurate source - the actual Parish record - may become accessible - it is a different source. Similarly the identity of a fiche is already implicit in the data for which it is quoted as source. It can be re-examined if questions arise, or access to the original record may be sought. Yours sincerely, Theo Tulley. tj....@ph... SFHG Member No: 11619 <snip> |
From: Frederico M. <fs...@gm...> - 2009-04-10 18:31:13
|
On Fri, Apr 10, 2009 at 7:05 PM, Theo Tulley <tj....@ph...> wrote: > Hi Frederico: > > It seems to me that your method involves much unnecessary recording: If a > source is specified as Parish Transcripts then just which parish, and the > date, are already known from the data for which it is quoted a source. I'm not sure I'm fully understanding your reasoning (my fault). I'm thinking that the data you deal with must be very different from mine, and if it works fine for you then that's another point in favour of GRAMPS. If I had a "Parish Transcripts" as a source this would make it nearly impossible to properly identify sources. I have all sorts of different Parishes, with different collections of books. I wouldn't know how to use a "Parish Transcripts" source to correctly address all this. One way to look at how useful is the source reference method is to make a Narrative Website report and look at the bibliography footnotes. If from what appears there one can easily see exactly where the information is located then it is a good sign. > On the other hand, if the source is only a transcript, possible errors are > implied and if questions arise the more accurate source - the actual Parish > record - may become accessible - it is a different source. I guess this depends on the nature of the transcript, but it is a good point. I sometimes have birth records which are transcripts of Catholic church records. For them I use a different source since they actually say "this is a transcript of the entry 12 of the Baptism Books of year 1901, etc". They have their own entry number and year and refer to a different set of sources (generally this is done by the Civil Registry), so I add them just as I would do any other Civil Record source. I also add the Baptism Books they refer to as a source - even if I don't have the data myself - and add a source reference that points to this "original" books in the event, with a note explaining the situation (at least until I get my hands on this actual source). > Similarly the identity of a fiche is already implicit in the data for which > it is quoted as source. It can be re-examined if questions arise, or access > to the original record may be sought. I do the same as the above: a fiche is almost never the actual source but a specific representation of a source, most commonly a Parish Book in my case. I add the Parish Books as the source and the fiche as a Repository Reference - I do this for the LDS microfilms. This makes it possible to separate the physical representation from the source and if some other way of accessing it appears (a website that contains scanned images for example) I can simply add it as a Repository. Regards, Frederico |
From: Michael L. <mic...@pc...> - 2009-04-10 23:03:40
|
On Saturday 11 April 2009 04:31:09 Frederico Muñoz wrote: > On Fri, Apr 10, 2009 at 7:05 PM, Theo Tulley <tj....@ph...> wrote: > > Hi Frederico: > > > > It seems to me that your method involves much unnecessary recording: If a > > source is specified as Parish Transcripts then just which parish, and the > > date, are already known from the data for which it is quoted a source. > > I'm not sure I'm fully understanding your reasoning (my fault). I'm > thinking that the data you deal with must be very different from mine, > and if it works fine for you then that's another point in favour of > GRAMPS. > > If I had a "Parish Transcripts" as a source this would make it nearly > impossible to properly identify sources. I have all sorts of different > Parishes, with different collections of books. I wouldn't know how to > use a "Parish Transcripts" source to correctly address all this. > Let me put my oar in here... I use four levels of information: Source Type Source Source Reference Source Text (and sometimes other notes) For example: Source Type: 1851 British Census Source: 1851 Census, Cornwall (FreeCEN but on it's own website) Source Reference: A particular Piece, Enumeration District, Civil parish, Folio and page in the Volume/page field with the census date in the date field Source Text Note: The actual text of the transcription of the particular schedule My reasoning is this. I could use the standard "census" source type, but for the U.K. I find that not specific enough. There are several transcriptions of these censuses (FreeCEN, Ancestry, FindMyPast, etc.) so I naturally treat these as separate sources. Where I am entering that data for a family, I create a source reference with appropriate notes for the first family member and then using the clipboard, copy that to the other family members or other events as appropriate. This replicates the source reference but shares the notes (and of course the sources.) Parish records are somewhat similar if there are several transcriptions available (for example my own marriage transcriptions and the secondary source of Phillimore's transcription). Each of these transcriptions are separate sources and each test of those sources is then attached to a source reference. This appears to work well in reports that I have produced so far. -- ==== Michael Lightfoot Canberra, Australia OPC Merther & St Breock, Cornwall see http://www.cornwall-opc.org mic...@pc... ==== |
From: Frederico M. <fs...@gm...> - 2009-04-11 01:51:09
|
Hi, On Sat, Apr 11, 2009 at 12:03 AM, Michael Lightfoot <mic...@pc...> wrote: > I use four levels of information: > > Source Type > Source > Source Reference > Source Text (and sometimes other notes) > > For example: > > Source Type: 1851 British Census > Source: 1851 Census, Cornwall (FreeCEN but on it's own website) > Source Reference: A particular Piece, Enumeration District, Civil parish, > Folio and page in the Volume/page field with the census date in the date > field > Source Text Note: The actual text of the transcription of the particular > schedule I do the same, with an added Repository Entry: Repository: Parish of Belém Source: Baptism Records of the Parish of Belém (linked to the above repository, ideally with the book number in the "call" field) Source Reference: Log date and page of the specific baptism record > My reasoning is this. I could use the standard "census" source type, but for > the U.K. I find that not specific enough. There are several transcriptions > of these censuses (FreeCEN, Ancestry, FindMyPast, etc.) so I naturally treat > these as separate sources. Where I am entering that data for a family, I > create a source reference with appropriate notes for the first family member > and then using the clipboard, copy that to the other family members or other > events as appropriate. This replicates the source reference but shares the > notes (and of course the sources.) I do the same (use the clipboard and then add the source reference). My main complaint was that if I somehow need to correct or add something to the source reference I will have to manually track where does the source reference appear to make the change. An example: I decided to add a General Comment in Source References saying "Baptism of John Doe" since source references can look rather cryptic when looking at them. This means that I have to find out every source reference. This could be made easier by filtering like Jerôme said. > Parish records are somewhat similar if there are several transcriptions > available (for example my own marriage transcriptions and the secondary > source of Phillimore's transcription). Each of these transcriptions are > separate sources and each test of those sources is then attached to a source > reference. I agree and to the same if by transcriptions one means a document that contains the information and refers to the original source. Say, like a marriage paper. But I do not do that if we are talking about different physical representations of the exact same thing: I do not add a different source for a scanned image of Page 12 of a Book, microfilm that shows page 12 of same book, and a paper copy of page 12 of the same book. To this end I use Repositories. Regards, Frederico |
From: Benny M. <ben...@gm...> - 2009-04-11 07:51:56
|
2009/4/11, Michael Lightfoot <mic...@pc...>: > On Saturday 11 April 2009 04:31:09 Frederico Muñoz wrote: >> On Fri, Apr 10, 2009 at 7:05 PM, Theo Tulley <tj....@ph...> >> wrote: >> > Hi Frederico: >> > >> > It seems to me that your method involves much unnecessary recording: If >> > a >> > source is specified as Parish Transcripts then just which parish, and >> > the >> > date, are already known from the data for which it is quoted a source. >> >> I'm not sure I'm fully understanding your reasoning (my fault). I'm >> thinking that the data you deal with must be very different from mine, >> and if it works fine for you then that's another point in favour of >> GRAMPS. >> >> If I had a "Parish Transcripts" as a source this would make it nearly >> impossible to properly identify sources. I have all sorts of different >> Parishes, with different collections of books. I wouldn't know how to >> use a "Parish Transcripts" source to correctly address all this. >> > Let me put my oar in here... > > I use four levels of information: > > Source Type > Source > Source Reference > Source Text (and sometimes other notes) > > For example: > > Source Type: 1851 British Census > Source: 1851 Census, Cornwall (FreeCEN but on it's own website) > Source Reference: A particular Piece, Enumeration District, Civil parish, > Folio and page in the Volume/page field with the census date in the date > field > Source Text Note: The actual text of the transcription of the particular > schedule > > My reasoning is this. I could use the standard "census" source type, but > for > the U.K. I find that not specific enough. There are several transcriptions > of these censuses (FreeCEN, Ancestry, FindMyPast, etc.) so I naturally > treat > these as separate sources. Where I am entering that data for a family, I > create a source reference with appropriate notes for the first family member > and then using the clipboard, copy that to the other family members or other > events as appropriate. This replicates the source reference but shares the > notes (and of course the sources.) Yes, this is the approach I advocated by working with notes. The fact that since version 3.0 the source text note is a shared object takes away most problems, as the note only needs to be changed once on finding a typo. I believe this was one of the main reasons to make notes shared objects in GRAMPS, and before 3.0 I strongly discouraged this workflow (You might remember source text was a specific tab of the source reference editor causing many people to make the mistake of adding all info to source reference). The problem is still there when you discover a mistake in the source ref (page number), or want to add a note to these source references (eg transcript of the original latin text). Then you need to track down all source ref that where copies, and change them all. This is why in my workflow I keep most notes in the source object, where the media files (scans) also live. This is why I think about a subsource implementation, it does not hurt the people who want to keep working as they do now. The main thing holding such a thing back is GEDCOM though. The more we deviate of that in GRAMPS, the more difficult to map our internal data to something that can eg be uploaded to websites logically as you want it. It is really hard to keep having to drag an old and dead standard with us :-( Benny |
From: Theo T. <tj....@ph...> - 2009-04-11 12:46:51
|
What is the source reference for? My presumption is that it is to enable the user to check in a source the information for which it is quoted. To do this, the user already knows the information provided; this will generally identify within a type of source, the detailed specification of the actual source. Thus, if my source reference is Census81 (which for me generally means a printed page image) then recorded individual or family details identify the location to which the census applies. Similarly, recorded data identify a parish, and the source reference specifies the church and the type of source - transcript, fiche or direct view (i.e.personal traanscription). I have yet to enter most of my sources: to minimize keystrokes, as far as possible I want them to be shared "existing sources" in Gramps terms! If I need to create new source references, they are designed to be easily shared. For what purposes is this policy inadequate? Yours sincerely, Theo Tulley. tj....@ph... SFHG Member No: 11619 |