• Please use real names.

    Greetings to all who have registered to OPF and those guests taking a look around. Please use real names. Registrations with fictitious names will not be processed. REAL NAMES ONLY will be processed

    Firstname Lastname

    Register

    We are a courteous and supportive community. No need to hide behind an alia. If you have a genuine need for privacy/secrecy then let me know!
  • Welcome to the new site. Here's a thread about the update where you can post your feedback, ask questions or spot those nasty bugs!

Keywording Strategies

Andrew Stannard

pro member
Hi,

As a Rawshooter user I am shortly going to be getting free copy of Adobe Lightroom (when it is released). At this point I am intending to migrate my picture library over and go through the process of adding metadata to all my images - including keywords.

Does anyone have any tips/resources on good keywording strategies?

I am keen to avoid a complete overload of keywords, and would like to come up with a heirarchial system that enables me to effectively search my images.

Thanks for any assistance,
 
Last edited by a moderator:

Ray West

New member
Hi Andrew,

I have no idea what you require, but I do know whatever you choose first, will be the wrong one. So, run through a few temporary setups, and select the most likely. I believe that the M$ new operating system has a 'virtual filing system' or some such name, which may well be a partial solution to this sort of indexing problem.

Best wishes,

Ray
 

Andrew Stannard

pro member
....but I do know whatever you choose first, will be the wrong one.

Ha! I figured that might end up being the case :)

I think I need to spend some time writing down possible sets of heirarchies and see how they might work out.

Ideally I would like to come up with a system that includes keywods related to places, content, style, season etc. Just worried about ending up with thousands of different keywords!

Thanks for the advice,
 
Does anyone have any tips/resources on good keywording strategies?

I think the various strategies are somewhat dependent on the final use of the images. For example, for some it may be important to search on the overall/dominant color of an image. For others it may be the name/address of the copyright holder of an original work of art (e.g. paintings or some architecture), or a reference to a release paper. Geo-referencing is becoming increasingly common.

In general you may find some useful data for fields like sports and news events at: http://www.iptc.org/pages/index.php , e.g. at http://www.iptc.org/NewsCodes/#descrncd

You should definitely check out David Riecks' Controlled Vocabulary at: http://www.controlledvocabulary.com/metalogging/ck_guidelines.html

Bart
 
Forget about not wanting "a complete overload of keywords". That's a poor starting point - inevitably the list will grow, so let it. But keywording does have to be done in a way that places value on your time - time inputting and time saved in finding. That balance is individual.

As for strategies, ideally you describe pictures with as many keywords as are needed to describe the picture to someone who can't see it. For me that means looking at the image and simply listing what's there as though I was looking at it through someone else's eyes. If you don't do this, you simply don't add enough keywords and rely on your own assumptions about the image's contents. For example, I was approached by someone who wanted to licence a picture which I had keyworded with "Matsushima Bay" and "Japan" - she had wanted it because of a type of tree in that area, and wanted more pictures of specific types of tree. I then had to remember other places where I'd shot pictures including trees. As a result, I now actively list what's in the picture - so trees, islands, ocean, sea, mist.

Next I go abstract. Can I use words like anger, love, free speech, Japanese, minimalist, triumph? Words that sum up what the picture symbolizes.

Also work downwards from bulk input (Japan, Matsushima) and then slice and dice (Matsushima Bay, Islands, Bridges, Statues).

Obviously these principles mean lots of keywords and perhaps more work than it seems worth. The trick is to make your Lightroom input more efficient and the burden tolerable / worthwhile, because lots of keywords make the later output (ie finding and selecting) easier and reduce the time cost.

With Lightroom, use hierarchies to group those keywords you often apply and where the subject can be broken down into hierarchies. That will make it easier to apply lower level keywords and have all the parents applied at the same time or "implied". With each keyword, use lots of synonyms - again boosting the numbers of keywords entered at least effort.

I would also look at using Bridge and metadata templates prior to LR - I could have a Matsushima Bay template that populates the country - location fields and adds a load of keywords at the same time (LR presets don't do keywords - doh).

John
 

Ray West

New member
Personally, I would keep any indexing system completely seperate from the image. Provided the image has a unique name, of course, and you keep track of it, if you move it, etc. If you are starting off with a few hundred/thousand images to index, you are going to need something easy and quick to apply. It is different if you are starting from scratch, indexing as you go along. If you have the space, then keep it as ascii text, then you should be able to get at the info easily, in a number of programs.

For example, use comma seperated key words, abbr. if necessary, and any spreadsheet can sort/count/find the records.

Best wishes,

Ray
 
I thought the question was about the use of Lightroom for keywording.

Spreadsheet-based or other external methods are always slower than dedicated cataloguing programs. Use them for data migration, not for finding pictures.

John
 

Cem_Usakligil

Well-known member
....Spreadsheet-based or other external methods are always slower than dedicated cataloguing programs. Use them for data migration, not for finding pictures.
John
Hi John,

I do not necessarily agree with you on this, if you don't mind me doing so :).
An external MS Access, MySQL or SQL Server db application will be as fast (or as slow) as any dedicated cataloguing program, considering the fact that CPs also use the same databases as their back-end data storage facility. Excel would be somewhat cumbersome, indeed, but slow? I don't think so.

Having said that, I agree with you that finding images using external programs introduces an extra step that causes an overall slowing down of things.

The best of two worlds can be achieved as well. LR makes use of SQLite database (see here towards the end of the page). One can use a front end program to access the SQLite database to mass update the keywords. One can even use a so-called ODBC driver to read the LR database from within Excel or Access. The possibilities are endless :).

Cheers,
 
Cem

Yes, fast in the sense of running queries and returning data, probably faster, but as you acknowledge that's a pretty limited slice of the actual human task.

I've connected both Lightroom and Aperture to external (Access) applications - both use SQLite databases. Maybe one day they can talk together?

John
 

Andrew Stannard

pro member
Hi,

Thanks for all the advice and ideas in this thread, it is proving really useful in helping me decide what to do.

For the moment I am looking at doing my keywording in Lightroom, rather than an external system. I do have some concerns about migration of the data should I switch apps in the future, but as Cem as pointed out it is SQLite based. In fact this could prove really useful for me as my website uses a mySQL database, so I am hoping to be able to program up a utility to migrate meta-data from Lightroom straight into my site's database.

John - I see what you saying about accepting the fact that keyword lists will grow long. It is the synonyms that I am concerned about.

For example using 'river' as a keyword against one shoot, and then inadvertently using 'stream' for another shoot. A subsequent search against 'river' then not finding all the shots that I would like it to. I'll have to investigate the best way to mitigate this when LR V1.0 is released.

Cheers,
 
Andrew

Not everyone can cope with passing data between databases, but that's certainly something that will be possible with Lightroom - assuming they don't decide to password protect the database....

Remember that Lightroom can export metadata in XMP files, which can be read in Bridge and in non-Adobe programs such as iView, IdImager, Photo Mechanic (I think) and is slowly becoming more common. I have a simple "prenup" test - if you can't get your keywords out of a program when divorce time comes, should you ever put it in?

Accept you will get those rivers and streams. In LR you define each keyword with its close synonyms, so when you set up "Stream" you add "Brook, Burn etc". One of these could be "River". But no program can really cope with human inconsistency - you just live with it.

John
 
Last edited by a moderator:

Asher Kelman

OPF Owner/Editor-in-Chief
Andrew,

This is perhaps one of the most important topics since it is the key to actually exploiting all your work.

John, I personally appreciate you being here to help us get our teeth in this.

Asher
 

Andrew Stannard

pro member
...perhaps one of the most important topics since it is the key to actually exploiting all your work.

Indeed - Feel like I am opening up a whole can of worms by deciding to keyword my images, albeit a very useful can of worms!

My current system revolves around filenames that contain the date of the shot and a directory structure that is based upon geographical location. It works really well when I am looking for a single image - but to find every image that features water? Not a fun thing to do!

Once Lightroom is released I shall try out a few different systems, and find a workflow and system that works best for me. It's great to have a forum like this to come to for advice.

Cheers,
 

Asher Kelman

OPF Owner/Editor-in-Chief
Andrew,

The "key" thing about all this is image mining. There's hopefully gold in your work; somewhere!

Asher
 

Ray West

New member
Hi John,

The question was about using lightroom for keywording, as you mentioned. I was trying to point out that I personally do not think any proprietory system is a good idea, based on my past experiences with similar systems which make me concerned about the future. I am quite happy for others to disagree. Again, it all boils down to how Andrew wants to slice the pie, what is the purpose of his keywords, etc.

Best wishes,

Ray
 

Asher Kelman

OPF Owner/Editor-in-Chief
Hi Ray,

Your point is one of the key statutes of Digital Asset Management.

We accept that current solutions are all transient. So a major qualifification for use of a filing system is not just how efficent it might be but how robust it will be when it comes the time to migrate.

Asher
 

Andrew Stannard

pro member
....do not think any proprietory system is a good idea, based on my past experiences with similar systems....

Ray, I agree with you on this point - but only if the user has no way of extracting the data into a non-prorietory system. For me personally, I am lucky that I have the programming skills to pull the data from the Lightroom Database/XMP Files if and when I need to.

If I didn't feel that I could do this, then I would have strong reservations about investing time and energy into an application specific database - be it Lightroom or something else.

Cheers,
 

Ray West

New member
Hi Andrew,

I'm glad to hear you can escape;-). Always remember, the software manufacturer is merely involved in your data, whereas you are committed - if you don't know the difference, then think of the standard english breakfast - the hen is involved, the pig is committed.

I hope you've not got 5000 sunset shots to catalogue....

Best wishes,

Ray
 

Cem_Usakligil

Well-known member
... the hen is involved, the pig is committed...
LOL. That's a very effective example I use for almost 20 years explaining to my customers the difference between involvement and commitment. I have first heard it in 1987 from a chap called Mike Salmon, who was an Oliver Wight Associates MRP-II trainer in the 80s. I wonder what he might be doing right now?

Re. the importance of keywords, that is indeed an extremely important area within the context of DAM as Asher and others have pointed out before. It is also one with which I've recently started experimenting, I am learning the rules while I'm already playing. I'm sure that I'll dislike what I do right now in a year's time or so, but I have to get going right now. The golden rule is to not moving into a corner from which there will be no escape later.

Cheers,
 

Ray West

New member
Be your own art gallery

This, and a whole load of other problems disappear, if, for example, you only keep 50 images. At the end of the year, or more frequently, you go through your fifty good images, and replace the poorer ones with the best of the new. Consign the poor ones to the 'digital sock drawer', or maybe some stock photo library.

The fundamental problem is that it is easy to keep images, difficult to delete, because it's no extra cost in taking one or fifty shots at a location, but look at the overheads you get into......
 

Asher Kelman

OPF Owner/Editor-in-Chief
Ray, there's little difficulty in finding ones brilliant images. For me at least, that's true.

I do not need a DAM system for that. A 1Gig disk would be overkill!

asher
 

KrisCarnmarker

New member
A lot of interesting thoughts going on in this thread.

First of all, I think the approach taken depends on the type of photography you do as well as the amount of images you catalog. A wedding photographer will probably have quite different needs than say an entomologist or a forensic photographer. That being said, I think it is better to have too many keywords than too few.

Now, I don't use LR for cataloging so you will have to take what follows further down and adapt it, if possible. Frankly, I think the cataloging abilities in LR are severely lacking. I am also an RSP owner and will use LR for its RC abilities, which I like a lot, but I will (for the time being anyway) be cataloging with IMatch and printing with QImage.

Regarding keywording, I use a hierarchical system of categories with many "roots". Categories are converted to IPTC keywords (via scripts), but not all categories in a categories-path become IPTC keywords. Other categories get converted to other IPTC fields than keywords, such as location tags. However, for the purpose of a keywording strategy, lets just assume that keywords and categories are the same.

The root categories I have set up are: Events, Locations, Subject Type, What, Who, Workflow States and some others. Some of these roots only have one level below them, an example being Workflow States, which simple has the child categories New, Reviewed, Edited and Final. Subject Types is the same, with e.g. Abstract, Action, Still Life, Landscapes, etc. Others are quite deep, such as the Locations root which goes from continent all the way down to local area, for example:

-Locations
----Europe
------Spain
--------Andalucia
----------Granada
------------Alhambra
...

The What root is the Universal IPTC category structure provided by IMatch, to which I add missing entries.

-What
----Animals
------Birds
------Insects and Spiders
--------Bees
--------Wasps
------Mammals
----Nature & Environment
------Beaches
------Deserts
------Water
--------Canals
...

One of the most complex ones to work out is the Who root, which includes people by name. I chose to create the hierarchy starting with the people's relationship to me, and then in order from oldest to youngest unless they are part of the immediate family. Just below the root, I have Unknown, Colleagues, Family, Friends. In the following example, each person's name has been replaced by their relationship to makes the example clear.

-Who
--Family
----Me
------Wife
------Daughter
--------Husband
--------Son
----------Wife
----Brother
------Wife
------Daughter
------Pet
--------Dog
--------Cat
----Mother
------Father
------Mother
------Sister
--------Husband
--------Daughter
------Brother
--------Wife
--------Daughter
----Father
...

Something to keep in mind is that IMatch allows me to copy categories "by reference", so I can have the same category in several paths, but assigning one of them assigns them all. For example, under What, I have "Cars" under both the path "Transportation/Automotive/Cars" and "Sports/Auto Racing/Cars".

Now, when it comes time to start assigning categories to the images, here's a method that's worked well for me, especially when there are a lot of images to categorize (and this is where the "Workflow States" category comes in handy).

-During import, all images automatically get assigned the "Workflow States/New" category.
-Do a search to show only images with the category "Workflow States/New"
-Start at the top of the list of images returned by the search and select the first image.
-Choose a category/keyword this image belongs to but do not assign it yet.
-Go down the list of "new" images and select all other images that also fit this category/keyword (e.g. using Ctrl-Click).
-Now assign the category/keyword to all the selected images in one go.
-Deselect all images except the first one.
-Choose the next category/keyword that the first image belongs to.
-Go down the list of "new" images and select all other images that also fit this category/keyword (e.g. using Ctrl-Click).
-Now assign the category/keyword to all the selected images in one go.
-Deselect all images except the first one.
-Repeat until there are no more categories/keywords left to assign to the first image.
-Select the first image and all other images that also have no more keywords to be assigned to them.
-Assign them the "Workflow States/Reviewed" (or Final).
-Refresh the search (for "new" images)
-Select the first image and repeat the process.

You will notice that as you get down the list (which gets shorter as you work), there are less and less keywords to be added and the process becomes quicker and quicker. Psychologically, I find this method easy to use because, while the process is slow in the beginning, it speeds up rather quickly and you feel like progress is being made. Usually, by the time I've tagged 10 or so images, it is real fast from there.

Another thing I do during this process is that I do not look at the image and think of a keyword, but instead start at the top of my keywords list and see it they fit the image. Not until I'm at the end of the list do I look at the image and assign any missing keywords.

/Kris
 

Jerome Marot

Well-known member
A lot of interesting thoughts going on in this thread.


I did not notice that thread in 2007, but it is still interesting.


The What root is the Universal IPTC category structure provided by IMatch, to which I add missing entries.

-What
----Animals
------Birds
------Insects and Spiders
--------Bees
--------Wasps
------Mammals
----Nature & Environment
------Beaches
------Deserts
------Water
--------Canals
...

I did not know that the IPCT had a whole predefined hierarchy of subjects. It can be found here:
http://cv.iptc.org/newscodes/subjectcode

Contrary to the list above, that hierarchy does not include categories for animal species. But there is already a hierarchical classification systems for animals (and plants, etc...) and we could probably simply use that one.

Locations root which goes from continent all the way down to local area, for example:

-Locations
----Europe
------Spain
--------Andalucia
----------Granada
------------Alhambra
...

With modern cameras being able to tag pictures with GPS coordinates, we don't need anything else. OTOH, we may be interested in degrading the accuracy of location data for privacy reasons, but there are location systems designed for that purpose, for example the radio locator system.

One of the most complex ones to work out is the Who root, which includes people by name. I chose to create the hierarchy starting with the people's relationship to me, and then in order from oldest to youngest unless they are part of the immediate family. Just below the root, I have Unknown, Colleagues, Family, Friends. In the following example, each person's name has been replaced by their relationship to makes the example clear.

-Who
--Family
----Me
------Wife
------Daughter
--------Husband
--------Son
----------Wife
----Brother
------Wife
------Daughter
------Pet
--------Dog
--------Cat
----Mother
------Father
------Mother
------Sister
--------Husband
--------Daughter
------Brother
--------Wife
--------Daughter
----Father
...

Here again, there have been some changes since 2007, with most social networks and even photographs database applications being able to recognise faces and classifying people automatically. This is an even bigger privacy problem as the location, obviously.
 
Top