Keywording photos for family history purposes

(20 posts)
  • NJHeart2Heart
    Member

    I want to begin keywording or otherwise identifying my digital photos by the people in them as one criteria among others.   Are there other family historians that manage a large collection of family photos?  My specific question is- is it better to use a keyword for each name (first/last) or do I make one keyword for each person's full name?  What other tips might fellow historians have for managing heritage photos?

    Thanks,

    Dawn D. - NJ

    Posted On April 25, 2009 - 03:25 AM (7 months ago) (Permalink to this post)
  • RobotDude
    Member

    Hi Dawn

    I have faced a similar problem and my solution is to tag using their usual known name.  That is, if my name is Fredrick John Smith but was commonly known as Fred Smith, "Fred Smith" is the tag I would use.  For parents, I use their usual known name i.e. "Roger Smith" rather than "Dad Smith" as "Dad" is only relevant to his children plus there would be a lot of "Dad Smith"s in my database.  If, for example, my brother's name is George David Smith but was always known as David Smith then I would use "David Smith".  For pets I have used their name plus the animal type i.e. "Sammy Dog".  By naming in such a way, using ACDSee and other software, I can search on "Smith" to get the Smith family's images and with "Sammy Dog" just the dog Sammy rather than using "Sammy" and getting the human "Sammy Smith" also! 

    Using someone's full name is fine for the present generation but the next generation would be unlikely to know your sister's three middle names, plus do you want a database with the following entries for the same person: "Sally Alice Mary June Smith", "Sally Alice Smith", Sally Mary Smith" and so on.  Much better to use "Sally Smith" together with a date range when the image was taken if there is more than one "Sally Smith".

    The major question that needs to be addressed is could your grandchildren in 50 years time, with their collected images/videos from previous generations, easily find you?

    This is still "work in progress" for me and includes deciding what other keywords to include ie. the location ("Eiffel Tower", "Paris", "France"), name of the event ("Birthday Party", "60th") and identifying objects ("Cake", "Presents", "Car"). 

    I know I will not get it right first time but over time hopefully it will evolve into something other generations can reference.  This is what I'm trying to address.

    It does worry me that everyone is making up their own keywording "standards" and when images databases are combined, as they will be by future generations, it could be a bit of a mess!

    Hope this helps.

    Posted On April 29, 2009 - 10:20 AM (6 months ago) (Permalink to this post)
  • NJHeart2Heart
    Member

    Robotdude,

    Thanks so much for the information. I had all but given up on using the ACDSEE community since I got no responses at all.  Your comments are definitely helpful. I agree that I will not use full names, but my question was more about using tags which have just one name "Phil" "Smith", vs. tags with a common name, ie "Phil Smith", "Tom Thumb".  Then again I haven't really *tried* to use a tag consisting of two words.  I guess I assumed based on my experience on Flickr (my photo site of choice) that I could put a multi-word tag in quotes and it be recognized as one tag.  Correct or not, I have since decided to separate out each individual name, "Phil" and "Smith" separately.  Makes more sense when I am searching for all people with a particular family name.  Now as I write it it seems like a rather silly question, although you answered an even better question - "How much of a person's full given name to you tag"?  My only issue with your reply to that is that I have a brother and a father who both go by "Bob", w/ same last name.  That means I have no way of searching for JUST ONE of those family members.  Same with my husband and his father.  I guess I could go with "Bob I" and "Bob II"- I just have to remember that I did it that way! (My husband would be a "Ted III"!)

    Thanks for your input.  So, are you a family historian?  Involved in any other online groups?

    NJHeart2Heart

    Posted On April 29, 2009 - 01:15 PM (6 months ago) (Permalink to this post)
  • NJHeart2Heart
    Member

    Robotdude,

    BTW, I do understand your dilemma with "the rest of the tags".  I have a pretty detailed system of file naming, so for me some of that information is inherent in the file itself.  Each file is named with "YYYYMMDD_Location_(Event, if any)_camera_####(the number given by the camera)".  My formatting has changed over the years (I used to use spaces between), but the basic info is the same.  As for tagging I think one of the advantages is that you can be flexible.  Add tags that make sense to you in your searches.  It seems to me that it would be worse to leave tags out then to have too many, so why not add whatever you think is relevant?

    Posted On April 29, 2009 - 02:03 PM (6 months ago) (Permalink to this post)
  • tibu
    Focus Group

    Just my thouhgts based on tagging for quite some time. I personally tage "First Last" instead of "First" and "Last". The reason is so I can quickly find the exact person I want using one tag. Using ACDSee search funtions you can search for more than one tag at a time, but that' sjust how I do it because it seems clearer to me when using smart tags to have the first and last name as one tage you can click on to get what you want (the keyword tags that are auto generated that you can click on to bring up all images with that tage). You could also do both, a tag for "First Last" and a tag for "Last".

    For family I use "First Last". For friends I just use "Last" as I see no need to tag everyone in my friends family separately. I search on their last name and on occassion burn a disc for them as I am usually the one taking the photos.

    For family members that have gotten married I inlucde the maiden name in the tag "First Last (Maiden)", ie "Terri Johnson (Smith)".

    One thing that comforts me in my tagging scheme is you can always do a search, select as many photos you need to and change the tag globally. 

    I "batch set information" to the IPTC field so it is permanently in the file for other external systems to read.

    The rest of my tagging (keywording) scheme is evolving too. More and more I remember to be more specific about what is in the photo or what the photo is about (Cake, Candle, Gracies Birthday, Pony Tails, etc.)

    Since you can multi tag (keyword) many photos at once, then batch set that to the IPTC field, you can accomplish a lot wasily. I actually went back into all my archive images back to 2000 (when I got my first digital) and added keywords. It took a bit of time, but it wasn't too bad and now I have good records. Once I was writing a blog article about taking pics of my children and it was fun to write "I have 8,834 picture of Benjamin to date....".

     

    Posted On April 29, 2009 - 04:31 PM (6 months ago) (Permalink to this post)
  • NJHeart2Heart
    Member

    Tibu,

    Thanks so much for the info.  That was very helpful and it seems at least two of you agree that using one field for "First Last" makes sense.  I do however also like the idea of having a last name as well, so that one can easily search for everyone with a particular famiily name.  Oh, this is going to be fun!!
    One question.  I am new to many of the features on ACDsee and just recently read about Software-specific keywording vs. IPTC field tagging. You mentioned multi tagging many photos at once (understood) and then batch setting that set to the IPTC field- That's where I am unsure- are you saying that there is a batch process by which each tag of each photo in a particular set you are working with is automatically embedded in the relevant photo's IPTC field?  That would be way cool, but is it really as easy as that?  So, you use both the software tagging solution as well as the IPTC field, or do you add software based tags soley because it's the quickest way to embed IPTC info to a group of photos?

     

    Posted On April 29, 2009 - 09:15 PM (6 months ago) (Permalink to this post)
  • RobotDude
    Member

    Hi HJHeart2Heart

    On the issue of one tag or two for keywording an individual, I always use one tag, i.e. "Phil Smith",  is I can search on "Phil", "Smith" and "Phil Smith".  If I was to use two tags, then the software I was using needs to be able to allow me to search on more than one tag with the relation defined between them, i.e. "Phil + Smith" in ACDSee.  Either method works but the one keyword approach I find is easier to enter and to understand.  For example, I would use the keywords for an image: [Sally Smith, Tom Smith, Bob Jones, Ann Bloggs], you would use: [Sally, Tom, Bob, Smith, Jones, Ann, Bloggs].  If searched for "Tom + Jones", my method would not find the image which is correct but your method would because the image has a "Tom" and a "Jones" but no one is "Tom Jones".

    As a test I checked for images of my sister having used the keyword "Sally Smith".  Searching for "Smith" brings up all the images with "Smith" anywhere in their keywords - 36 images found.  Searching for "Sally Smith" finds 21 images and searching for "Smith - Sally" finds 15 images.  "Smith - Sally" in ACDSee means find all the images with "Smith" but not those with "Sally".  Searching for "Sally" finds 286 images as I know other Sallys!

    For married women, I use their married name in the keyword.  So "Sally Smith" becomes "Sally Jones".  If she then remarries she becomes "Sally Bloggs".  To find her, if I was unsure of which surname she had, I would search on "Sally" and on investigating the images returned and their keywords, would repeat the search with the relevant surname i.e. "Sally Jones".  Keywording is not an exact science, although I wish it was!

    Thankfully we don't have the same problem with Bob I, Bob II etc in the UK.  My approach would be to use "Bob I Smith", "Bob II Smith" etc as everyone who knows them would know who "Bob II Smith" is.

    My filenaming is very simple: YYYY-MM-DD NNN.jpg and the images are stored in the folder structure YYYY/YYYY-MM-DD <Event name> (MMM).  NNN is a simple counter with 001 being the first image in the folder and 234 being the 234th.  By numbering, I can define the order they appear (even though this may not be in cronological order)  and importantly see if one had been accidently removed.  <Event name> is a few words that tell me what the event was i.e. "Sally Smith 34th Birthday Party".  These words would also be added to the images as keywords.  I don't have the location, event or camera in the filename as this adds to the time to process the image and would be added as keywords anyway.  The camera information is automatically added to the image when the image is taken.  (MMM) is the number of images in the folder to check that none have been accidently removed - my biggest fear.

    Like tibu I store my keywords in the IPTC field as this can be read by other software and is the standard location for this information.

    And almost finally!  In addition to ACDSee, I also use the free software iTag to keyword my images as it offers greater flexibiliy in the keywording process.

    I'm not a family historian but a keen photographer with 1000s of pictures of my family which I wish my own children and their children to be able to access in the future and not just to sit on a forgotten hard disk uncatalogued.  Plus I find the challenges this subject poses interesting.

    I'm not actively involved in any other online groups but pass by the ACDSee forums from time to time to see what is being discussed.

    Best regards

    Posted On April 30, 2009 - 08:23 AM (6 months ago) (Permalink to this post)
  • NJHeart2Heart
    Member

    robotdude,

    So you're saying that if I choose to use just one "tag" for "First Last" ("Bob Smith", "Tim Jones") ACDsee still allows me to search just for Smith and will find (if specified) all photos with tags which INCLUDE the name "Smith" (meaning that it doesn't have to be in a separate tag at all).  That's terrific!  A new realization which makes that decision SOO much easier to make!  If ACDSee is that flexible than there certainly seems to be no need to separate names in that way.

    Funny you mentioned the camera type- after reading my own post and pondering it for a bit, I realized just what you said- if the EXIF is going to include that information anyway, it seems redundant to add that to the file.  I took some time and using a nifty batch renaming program (Renamer, basic (free) edition) went back through my photos for this year and took out the camera info.  The rest of my file naming format I'm really happy with and it has always allowed me to find batches of photos easily.

    Posted On April 30, 2009 - 05:41 PM (6 months ago) (Permalink to this post)
  • Sam Dring
    Moderator

    So you're saying that if I choose to use just one "tag" for "First Last" ("Bob Smith", "Tim Jones") ACDsee still allows me to search just for Smith and will find (if specified) all photos with tags which INCLUDE the name "Smith" (meaning that it doesn't have to be in a separate tag at all).

    Just chipping in on this one - I don't use ACD for fam history but robotdude is correct.

    1.  If your keyword (either database keyword or iptc) is 'bob smith', QuickSearch will find all bobs and the rest of the smith family when searching for smith.

    2.  WARNING:  Currently, QS only allows 3 exclusions from your search as shown in the following so if, for example, your uncle was named Bob Smith and you or any other photographer in your dbase had the same surname (author), you would return all the Smith family and all Smith Authors.  That, of course, remains valid if keyword is 'Bob Smith' or keywords are 'Bob', 'Smith'

    Just needs thinking through.

    Attached Image:

    Attached Files

    1. 2009-05-01_083531.jpg
    Posted On May 1, 2009 - 07:46 AM (6 months ago) (Permalink to this post)
  • NJHeart2Heart
    Member

    Sam Dring,

    I gotcha. Thanks for the additional information!

    NJH2H

    Posted On May 1, 2009 - 02:43 PM (6 months ago) (Permalink to this post)
  • atnbirdie
    Member

    WIth respect to the NJHeart2Heart's question about IPTC data entries, I enter data in both the database and the IPTC fields using a Batch Set Information template I made.  I got the idea from Marc Sabatellas great 5 part DAM workflow piece from the Pro Blog (just go to Pro Blog and search on dam).  See Export Metadata in part 3/5 of Marcs series to find out how to make a template. 

    The trick is that you have to run your Batch Set Information template twice because entering the data in the database fields only registers them  when you Set the information.  So you have to enter the data, Set the information, go back to Batch Set Information and just Set the information again without entering any more data.  It sounds more complex than it is in practice.  The first Set enters the database metadata you entered; the second copies the metadata you entered the first time to the appropriate fields in the IPTC based on your template instructions.

    Hope this helps,

    Steve

    Posted On May 28, 2009 - 03:16 PM (5 months ago) (Permalink to this post)
  • NJHeart2Heart
    Member

    Steve,

    Thanks for the suggestion.  It's funny you mentioned that multipart series.  I've been hanging out at the DAM forum, and made my way to that very same series just today.  Going to read through it during my lunch break!  I'm not quite focused on your answer but I believe once I'm in front of ACDSee I'll better understand what you're talking about.

    Thanks,

    Dawn in NJ

    Posted On May 28, 2009 - 06:19 PM (5 months ago) (Permalink to this post)
  • Marc Sabatella
    Moderator

    atnbirdie said:

    The trick is that you have to run your Batch Set Information template twice because entering the data in the database fields only registers them  when you Set the information.  So you have to enter the data, Set the information, go back to Batch Set Information and just Set the information again without entering any more data.  It sounds more complex than it is in practice.  The first Set enters the database metadata you entered; the second copies the metadata you entered the first time to the appropriate fields in the IPTC based on your template instructions.

    I'm glad you're finding my workflow articles useful!  But I'm not sure I'm following what you are saying here.  Are you using Batch Set Information to enter the information into the database in the first place, and then to copy it from the database to IPTC?  I suppose you could do that, but it seems much more awkward to me than simply entering the database information directly onto the Database tab of the Properties window.  Unless I'm misunderstanding something...

    Posted On May 28, 2009 - 07:09 PM (5 months ago) (Permalink to this post)
  • atnbirdie
    Member

    Marc,

    Your misunderstanding must be due to my explaining myself poorly because I am basically following the process you laid out in the Export Metadata section of your part 3 DAM workflow :-)

    Yes, I am using the BSI function to first enter the info into the ACDSee DB using a custom workflow template.  Then I set the info to save it.  The template uses the insert metadata capability to copy fields from the DB to the IPTC.  I'm usually processing multiple files with most of the info the same (e.g., a photoshoot of garden flowers) so the batch capability is nice.  I can then go back and tweak any outlier files that need an extra keyword or two.

    I then rerun my BSI workflow template to do the copy of the DB metadata into the corresponding IPTC fields.

    It is redundant, but that is the method in my madness.  If I just enter data in DB field, I can only access it via ACDsee.  If I do the quick copy into the IPTC, it will be available wherever and by whomever the photo files are used.  I suppose in theory if I somehow lost all my ACDSee DB files I could reconstitute most of it by building a custom BSI template that copies IPTC into the ACDSee DB.

    I'm sure it's overkill, but that's the logic behind my workflow.

    Steve

    Posted On May 28, 2009 - 07:53 PM (5 months ago) (Permalink to this post)
  • Marc Sabatella
    Moderator

    Right, I'm just not understanding why you are using BSI to enter stuff into the database.  It seems a far clumsier tool to use for that purpose than the Database tab of the Properties window - so clumsy that I'm having having difficulty imagining how it could even be done at all, except to enter identical info for all files, as you say.  But even then, it would still seems clumsier to me than the Properties/Database window.

    Anyhow, once you have stuff in the db, then indeed you need BSI to copy it to IPTC, but I don't see the point of using BSI rather than the much more straightforward Properties/Database window to get info into the db in the first place.

    Posted On May 28, 2009 - 09:59 PM (5 months ago) (Permalink to this post)
  • atnbirdie
    Member

    Well, it turns out I was working from a bad assumption.  I incorrectly assumed that if I selected multiple files because I was going to enter similar data for them all, that I had to use the batch function of BSI.  Your questioning my routine made me realize (I just tried it) that if I select multiple files and enter data in the properties window it is quicker given the auto-complete nature of some fields.  So now all I have to do after entering the DB data is run my BSI template to copy it to the IPTC fields.

    Thanks (he said sheepishly)!

    Posted On May 28, 2009 - 11:09 PM (5 months ago) (Permalink to this post)
  • Marc Sabatella
    Moderator

    No problem!  Of course, the next question is, why does ACDSee only allow you to use the Database tab when multiple files are selected - why can't you enter info into the IPTC tab directly when working with multiple files at once?  That would eliminate the need for BSI at all, and I suspect many people wouldn't not even bother using the Database fields.  That's been a common request, but performance issues are generally cited as a reason for not doing that (writing a value directly to the IPTC area of 1000 image files is much slower than writing it to 1000 database entries!).  I don't think the performance issue is unsolvable, but for now at least, this is the workflow we get.

    BTW, another non-obvious advantage of the Properties/Database window over BSI - select an image, put your cursor in the Caption field, type a caption, then hit Page Down...

    Posted On May 29, 2009 - 10:54 PM (5 months ago) (Permalink to this post)
  • atnbirdie
    Member

    I'm not sure what the PgnDn is supposed to do.  The first time I tried it, it seemed to pop me into the IPTC entry field.  However, in subsequent attempts, even after restarting the program, it merely enters the caption in selected photos, deselects them, and moves to the next photo in the folder.

    An advantage of BSI I've noticed is that whereas in the DB entry pane you for example see all the keywords rolled into one chain, in BSI view, you see which keywords are associated with what files.  As I am in the process of moving from a subject folder based system to one similar to the approach you outline in your DAM workflow, I'm finding the BSI approach helpful in identifying which files I missed keywording or those that could use some more detail.  I don't want to dump files into my date oriented system if I won't be able to find them with keywords.  Once I'm finished my file management conversion, and as I am becoming more efficient at keywording, captioning, etc., that won't be so critical as I'll have pretty much got it right the first time.  Then the DB pane entry and BSI insertion of those metadata into the IPTC fields will definitely be easier for me.

    Thanks!

    Posted On May 30, 2009 - 01:57 PM (5 months ago) (Permalink to this post)
  • Marc Sabatella
    Moderator

    atnbirdie said:

    in subsequent attempts, even after restarting the program, it merely enters the caption in selected photos, deselects them, and moves to the next photo in the folder.

    That's the idea. A very quick way of entering captions for photos one by one, no clicking required.  This can be used in Keywords and Notes too, but the way I do things, those are less likely to need to be entered individually.  Occasionally keywords.

    An advantage of BSI I've noticed is that whereas in the DB entry pane you for example see all the keywords rolled into one chain, in BSI view, you see which keywords are associated with what files.

    Yeah, I can imagine there be situations where that would come in handy.  I'm glad we have both methods available!

     

    Posted On June 1, 2009 - 07:11 PM (5 months ago) (Permalink to this post)
  • atnbirdie
    Member

    Got it. That certainly is a nice function if you are inserting different captions in a list of photos.   Any time you can reduce clicks it's a good thing for data management processes.

    The flexibility of ACDSee to meet our various approaches to photo management is one of the things that brought me to the fold :-)

     

    Posted On June 1, 2009 - 07:21 PM (5 months ago) (Permalink to this post)

Subscribe to this topic via RSS

Reply

You must log in to post.