Keywording - To or not to

[surenj]

Who uses a image catalog software? If so which one?

Do you use keywords to organise your images? How do you choose keywords? Do you use keywords to find your images?

What directory structure do you use?

I use Lightroom and only recently started cataloguing my pictures. I usually put keywords for each batch that I download.

[biggerry]

None at this point, I have dabbled in NX2 with keywords, but it has not yet taken off for me, I would rather spend more time photgraphing than worrying about this. However I will probably regret this in a few years ...

Folder structure, I keep it simple:

Year (eg 2008)
|
---> 'Date' (eg 07-27-2010) 'Description'

then in each folder I will have raw files then (when I feel like it) folders for low resolution jpegs and his res jpegs
The great thing about .nef files is that you only ever need a single .nef file, all edits contained in teh one file.

[ATJ]

Suren,

I use Lightroom and I keyword most of my images. As almost all of my images are of organisms, the bulk of my keywords are the scientific names of the organism (or as close as I can get). This may or may not be relevant for you.

I have my keywords set up hierarchy, basically, Kingdom, Phylum, Class, Order, Family, Genus, Species. If I know what species something is, I add that keyword to the image (and the whole hierarchy is associated with the image). Later, I can easily select every image I have of that species, or every image of species in that genus, or every image of species in that family, etc.

It also works for some genera or even families where the photo is such that I can't be sure of the species. For example, I have some images of the eyes of a stingaree. I don't know which species or genus it is, but I do know the family so I tag it with the family.

While you may not benefit from this, you may benefit from other keywords like "sunrise", "beach", "portrait", "boat", etc. And you can also mix and match. e.g. a boat at a beach at sunrise would have all 3.

I also keyword with events, e.g. DSLRUsers 2x2 April 2010. That way I can easily filter on those image.

[DaveB]

I primarily use Lightroom.
My folder structure is used to split the files up so they're not clumped with too many thousands per folder. It's a hierarchical system, with Year/YearMonth/Week, and when I'm in a busy period there'll be day sub-folders. But this structure isn't critical to the way I work.
If a disk fills up and I start putting new images into a new disk, I may end up with some folders repeated on different disks (but containing different images). That's fine: the folder structure will at least let me navigate by folders if I have to, but I rarely do.

With the files indexed in Lightroom (or in Expression Media, the other tool I use) I can select files by any combination of folder location (just as if I was in a directory browser) and/or by any other metadata (across the entire database, no matter which folder they're in): capture time, geographic location, camera model, geographic location, ratings/labels, even keywords. So I'm not too worried about where the files reside (as long as they're safely backed up of course!).
When I import files into the system I do give them unique filenames so I'll never have filename collisions when rearranging folders later, with the date/time at the start of the name so that if I do search through folders they'll be in a sensible order. But thereafter I don't change the names (other than to give exported copies appropriate names).

All the information about what's IN the photos goes into keywords. So I don't rename a file to have "Duck" in the filename (or put it into a "Ducks" folder, etc). But I may assign the keyword "Pacific Black Duck" to it, which when I export copies will automatically assign "duck", "anseriformes", "bird", "waterfowl", and "Anas superciliosa". Similarly, if I assign "petroglyph" it inherit the keyword "rock art". All of that happens because I have the keywords assigned in a hierarchy. I'm continually refining my keyword list, and this year I'm making a big effort to at least finish a first pass of keywording my main 76,000-image database (lots of old stuff never got keyworded). As of today my keyword list is 3216 words.
If you want to learn more about such keyword lists, I suggest you do a web search for "controlled vocabulary".

Having all these inherited keywords facilitates your own searching through your database, and it also help when you export the images to external stock libraries, Flickr, etc.
Some of my keywords are set to not be exported (and thus are only useful within my database). For example, I have a private hierarchy with lists of names. Pets, friends, family, clients. I want to be able to find all the photos of my nephew, but if I post a JPEG somewhere that happens to have him in it I don't want to automatically be publishing his name...

Keyword hierarchies are also supported in Aperture, Bridge, and Expression Media, but I like that in Lightroom it doesn't actually assign the extra keywords until you export. You can rearrange your keyword hierarchy without having to manually go and reassign/deassign keywords from thousands of photos. You can of course search based on the inherited keywords without having to export images.

It's a big topic!

[Mr Darcy]

None at this point, I have dabbled in NX2 with keywords, but it has not yet taken off for me

Yep I was using NX2, but found it useless for keywording. I could assign the names, but not search on them. So there was really no point.

I have recently bought Aperture3, and am finding keywords a much greater help as I can simply look for all the photos with the keyword "hut" in it for example. Trouble is for this to be really useful, I need to keyword my entire back catalogue of photos. Not a job I look forward to.

What I need to do now is work out how to structure keywords. I didn't know they could be listed in an hierarchy. Will have to look into that. But even then what terms do I put into keywords.
If I photo Pam at Sunset in the Botanic Gardens, I guess those words go into it, but if I am later looking for a photo of an Illawarra Flame Tree, I might miss the fact that there was one in the background of the shot. Or Kikyuyu grass or whatever. Where does the list end? How specific do you get? How generic?

Andrew's case is much simpler. Technical photos are listed by their subject matter. His taxonomy makes it easy.

SIGH

[surenj]

Thanks for all your detailed answers.

Looks like I have opened a can of keywords!

\I have quite a few questions but will ask the most pressing first.

which when I export copies will automatically assign "duck", "anseriformes", "bird", "waterfowl", and "Anas superciliosa"

Does LR do this automatically???

[ATJ]

Does LR do this automatically???

If you tell it to.

e.g. look at the IPTC data on this:


I just had to add "Batrachomoeus dubius" from my Keyword list, and when I export, I also get:
Bactrachomoeus (genus - with a typo )
Batrachoididae (family)
Batrachoidiformes (order)
Fish ("class" - yeah, should be Osteichthyes but having "fish" is easier)
Animals ("Kingdom")

Now, in theory, the common name should be there as I set it up as a synonym. I'll have to check that out.

The good thing is (as Dave points out), I can correct the typo without having to retag the library. Obviously, anything that has been exported will have the typo, but any exports from now on will be fine.

[DaveB]

which when I export copies will automatically assign "duck", "anseriformes", "bird", "waterfowl", and "Anas superciliosa"

Does LR do this automatically???

When you add keywords they are by default "flat" (at the top of the hierarchy). But you can nest them. In LR's Keyword List panel, simply drag one keyword on top of another. When an image has a keyword, it will automatically inherit the parent keywords.

In my case I have

• ANIMALIA (which is set to not be exported: it's just a placeholder to organise the tree), containing
• vertebrate (currently set to not be exported), containing
• bird, containing
• anseriformes (which has waterfowl as a synonym: it's also attached to some other orders such as ciconiiformes), containing
• duck, containing
• Pacific Black Duck (which has Anas superciliosa as a synonym).

Unlike Trevor I've got the "scientific names" as synonyms, as when I'm assigning keywords I'm used to navigating via the common names. ANIMALIA is just one of the hierarchies in my list, and it's continually growing. The birds section has all the Australian and Antarctic species, plus some others I've come across in my travels. I haven't yet populated comprehensive lists of mammals, other than the pinnipeds and cetaceans I like to photograph.

Some of my top-level keywords are just containers for lots of other keywords (such as ACTION which has lots of "ing" words in it) while some (like ANIMALIA) contain complex hierarchies.
What set of keywords YOU need depends on you. You can either build up your own set of keywords over time, or you can purchase a dictionary that's been built already (e.g. CVKC, D-65, MarineLife).

[rflower]

It is funny that you raise this question. I was thinking a similar thing, only in the last week or so ...

All my photos are stored in folders on my harddrive with format of
YYYY.MM
-- YYYY.MM.DD.DESCRIPTION OF PHOTOS
(sorting like this - year first makes it easier to sort -IMHO, than trying to sort by day ... Sorting by day 21/08 will be before 25/01, whereas arrange year first, then month, latest photos will always sort to the end)

I use Bibble as my raw editor of choice. I purchased V4, and recently they have released V5 (with a free upgrade from V4.8 ). Bibble 5 comes with "Asset Management", including the ability to tag images with key words. The key word manager, also includes the ability to have a keyword hierarchy. After having a couple of attempts at this, I think I have almost got it sorted out (almost ).

I have a hierarchy for Family, with initial children levels for the main people in my immediate family. Also at this children level, sub folders for mine and my wife's family

Code: Select all

-- ME
-- Wife
-- child1 ...
-- |FLOWER
   |-- Mum and dad
   |-- Brother etc

I also have a hierarchy for
- Location, split into levels for major Regions (suitable to me), and then broken down into specifics as sublevels
- Events (weddings, baptisms, children concerts, DSLR Users etc),
- Wildlife. This has 2 main sublevels. Since most wildlife that I see is at a zoo, there is a sub folder for ZOOs (and then each zoo as a sublevel), and Animals, with a sublevel for each sort.
- Landscape (sunrise, waterfalls, beach, Urban etc)

In the keyword manager, keywords (or hierarchies) can be added to "themes" for quick selection in the main Bibble window. When a keyword is attached to the image, then any words in the hierarchy above it are included. Multiple keywords (or hierarchies) can be added to 1 image.

So a photo of my daughter at a waterfall in the Otways, would be keyworded as
Subject;Location;Country Victoria;Otways;Waterfall Name, Subject;Landscape;Waterfall, Subject;Family;Lucy

I suppose my biggest challenge (apart from going through and cataloging my photos ) is working out extra hierarchies / key word sets for things I photograph outside these types of topics, and how to keep these all relevant / synchronised / concise etc. EG when I take photos of THINGS; like macro of Waterdrops, or flowers in my garden.

[ATJ]

When you add keywords they are by default "flat" (at the top of the hierarchy). But you can nest them. In LR's Keyword List panel, simply drag one keyword on top of another. When an image has a keyword, it will automatically inherit the parent keywords.

You can actually add them directly as a hierarchy - you don't have to drag them.

When you create a keyword tag by right clicking on an existing keyword you have a choice of:
* Create Keyword Tag...
* Create Keyword Tag inside "the one you have selected"...

Choose the second one to create a child keyword.

[DaveB]

You can actually add them directly as a hierarchy - you don't have to drag them.

True. I wasn't going to go into the full list of ways you can manipulate them, but I figured drag-n-dropping would get people started.

[stubbsy]

Like Dave I keyword in lightroom wth a fairly detailed hierarchy (I'm not as organised as him, but I will be by the end of the year with luck). Before Lightroom I used a program called iMatch which was also pretty good.

[aim54x]

I guess some of you really do keyword...Im with Gerry....I dont but will probably regret it later

[biggerry]

(sorting like this - year first makes it easier to sort -IMHO, than trying to sort by day ... Sorting by day 21/08 will be before 25/01, whereas arrange year first, then month, latest photos will always sort to the end)

thats where the US style date is good, month/day/year - means they all end up in order through the year.

[DaveB]

The US date format is flawed also.
Why not use something like the ISO standard date/time format?

[Mr Darcy]

Why not use something like the ISO standard date/time format?

I've been using this for years. Though when I use it for myself, I tend to dispense with the punctuation. It saves bytes. (I started with computers when 8K of RAM was a big deal, and for that matter a 173Kb Hard disk was huge. - and yes I do mean KILO bytes)

so July 6th 2010 becomes
2010-07-06 in ISO but I use 20100706 unless someone else will be reading it regularly.

There is not only no confusion, but computers ALWAYS sort it correctly.

[ATJ]

There is not only no confusion, but computers ALWAYS sort it correctly.

Now, Greg, I want an honest answer to this question.... when you first started using this date format (back in the days of 8KB of memory, etc.) did you use yyyymmdd or just yymmdd?

[DaveB]

Now, Greg, I want an honest answer to this question.... when you first started using this date format (back in the days of 8KB of memory, etc.) did you use yyyymmdd or just yymmdd?

Hehe. When I started filing my photos digitally I must admit I used yymmdd, but went back and renamed everything to YYYYMMDD in around 1998. But my collection then was tiny compared to now! I have had some students ask why I use YYYY instead of YY, and the answer is in 2 parts: to avoid problems in 90 years' time, and because I still have photos in my database from last millenium!


Actually, saving bytes in filenames is still important today. To allow you to easily transfer your files and folders between OS platforms and between application software (e.g. do you really want to have to rename 20% of your files as part of a disaster-recovery procedure just because you've got access to a different system?) you should make sure they fit some lowest-common-denominator rules:

• Only use
  • letters (mixed case OK, but don't expect "Upper" to be treated differently to "UPPER" or "UpPeR"),
  • numbers,
  • underscores, and
  • hyphens (minus).
  No apostrophes, no slashes, no colons, even no spaces! If you want something that "looks like" a space, use an underscore or a hyphen.
• That also means there should only be one (and always one!) "." in the filename: separating the name and the extension that indicates the file type. Some braindead software gets upset with multiple dots, so don't tempt fate. Some systems can cope with no extension, but that invites a world of hurt for several reasons.
• Filenames should have a maximum length of 32 characters (including the extension).
  This is becoming less important over time, but I think there is still some old software around that truncates names at 32 characters, making it impossible to access the images in that software.
• Files should be allocated names ASAP that will be unique within your collection (e.g. the camera-generated names are only guaranteed to be unique within a single flash card).
• Keep in mind that if you have multiple files with the same name but different extensions, some software will pick one as the "master" and treat all the others as "sidecars".
  e.g. FILENAME1.NEF with sidecars FILENAME1.XMP and FILENAME1.JPG
  To avoid problems now and in the future, you should avoid creating a FILENAME1.TIF or FILENAME1.PSD. Some software will pick the .NEF file and bundle the extras in as sidecars.
  The common way of dealing with this is to add a suffix (such as FILENAME1-ver1.TIF). The name is obviously derived from the name of the original RAW, making it easy to correlate the files later. Note that Lightroom does this automatically, with the default suffix of "-Edit" being something you can change in the preferences.

So the punctuation in dates is a hassle because it chews up space. If you use a filename system such as YYYYMMDDhhmm_NUMB.ext (where NUMB is some identifier to separate photos taken in the same minute: either a sequence number generated by your software, or the number off the end of the camera name is sometimes good enough) you already have a name that's 21 characters long (assuming a 4-character NUMB). That means that any suffix you add to the filename (e.g. for derivative TIFFs) should be limited to 10 characters (plus the '-' separator) to avoid overflowing the 32-character limit. 32 may seem like a lot, but it gets chewed up quickly.
32 characters is not a hard-n-fast limit, but the more often your filenames go over this the more likely it is they will introduce complications when you have to transfer your images to different software.


When working out your photo file naming standard, keep the above issues in mind and you should have a fairly safe and stable system. There can be exceptions, such as:

• Files you export to send to someone else should match their rules (e.g. some labs are picky about filename lengths and characters, you might generate filenames for web JPEGs that include things like title and keywords to optimise SEO, etc). But by using these rules in your master/archive image collection you will avoid many problems.
• Software configuration files and presets (e.g. as used by Lightroom and Photoshop) often use the verbose title as the filename so it's really easy to end up with spaces and apostrophes in them. While I don't apply all the above rules to these names, I do try to limit the extra characters to those examples (spaces and quotes) especially on systems where I have the Lightroom settings stored with the catalog on an external drive and transfer the lot between Windows and OS X systems.

This is an extract of some of the stuff I teach my 1st-year students. Hopefully you find it useful too!

[biggerry]

Why not use something like the ISO standard date/time format?

make little difference when year folders are used, the only concern is to get the folders within teh year to be correctly sequential.

No apostrophes, no slashes, no colons, even no spaces! If you want something that "looks like" a space, use an underscore or a hyphen.

if you have used unix or linux systems and trawled around directory structures with teh command line you quickly realise the benefits of this.

[Mr Darcy]

There is not only no confusion, but computers ALWAYS sort it correctly.

Now, Greg, I want an honest answer to this question.... when you first started using this date format (back in the days of 8KB of memory, etc.) did you use yyyymmdd or just yymmdd?

May I cite Tony Abbott's views on honest?
I said I started using computers way back then. I didn't say I started using yyyymmdd back then. It has been a gradual thing. I guess I started using it about 1996 when I started writing payroll software for LendLease. And yes I was using YYYY. I remember having arguments about it with the management. Even then, only 4 years to Y2K, they saw it as a waste of valuable computer resources.

[ATJ]

I remember having arguments about it with the management. Even then, only 4 years to Y2K, they saw it as a waste of valuable computer resources.

Interesting. My first Y2K project was in 1983/4 when I worked for Custom Credit. We had to convert all our applications to YYYY and had to get it all working before 1985. Custom Credit had loans that went for up to 15 years. This was fully sanctioned, supported and funded by management. The project went by two names: "Date 1999" and "2001: A Date Odyssey".

We will now return you to your regularly scheduled programme.

[DaveB]

Why not use something like the ISO standard date/time format?

make little difference when year folders are used, the only concern is to get the folders within teh year to be correctly sequential.

True, but only up to a point.

If you have YYYY folders containing YYYYMM subfolders, at one level that seems like a waste.
However there are drawbacks to using YYYY/MM/ instead of YYYY/YYYYMM/ !
Take the example of:

Code: Select all

2009/
     01/
     02/
     ...
2010/
     01/
     02/


When you're browsing a folder often you only get to see the name of the subfolder, not the full path. So if you're browsing the folder "01" you're not being given any clues as to which "01" it's referring to. Also, occasionally I end up moving folders around (e.g. when consolidating files from smaller disks to newer [bigger] disks). So I much prefer the example of:

Code: Select all

2009/
     200901/
     200902/
     ...
2010/
     201001/
     201002/


It's "redundant", but can help avoid confusion. If at any point I move the sub-folders together I won't get collisions.
This is the same reason we put the full YYYYMMDD information into the filenames within those folders as well. Every file on the system should have a unique name, and ideally every folder would also have a unique "leaf name"! It's not so critical with the folders, but the extra information in the folder name definitely helps!

Note that the top-level YYYY folders are optional at this point: they're just there to tidy things up and aren't a critical part of the name. One of the reasons to have unique filenames is to let you experiment with different directory structures to find the one that is the right mix for you of convenience and structure.

[biggerry]

True, but only up to a point.

If you have YYYY folders containing YYYYMM subfolders, at one level that seems like a waste.
However there are drawbacks to using YYYY/MM/ instead of YYYY/YYYYMM/ !
Take the example of:

To clarity, I am not referring to having folders in teh format YYYYMM.

The folders within each year should always be unique I would never use a folder '01' or the like, as mentioned it would be '07-28-2010_description' with the month as the leader to purely organise them in the year folder.

One good reason to have year folders is to reduce scrollage and RSI

On a side note: how do people deal with actual image file names, particularly once the you clock over the sequential numbers from teh camera?

[ATJ]

On a side note: how do people deal with actual image file names, particularly once the you clock over the sequential numbers from teh camera?

I don't worry about it and leave the files the same name as what the camera gave them.

All my files are stored in a month folder (YYYYMM - although I don't bother with YYYY parents) and I doubt I will even even come close to taking 10,000 photos in a single month. I rarely do anything with the files on the HDD, anyway, other than back them up, so all my interaction is via Lightroom where the files names don't matter that much (to me as the user).

[DaveB]

True, but only up to a point.

If you have YYYY folders containing YYYYMM subfolders, at one level that seems like a waste.
However there are drawbacks to using YYYY/MM/ instead of YYYY/YYYYMM/ !
Take the example of:

To clarity, I am not referring to having folders in teh format YYYYMM.

The folders within each year should always be unique I would never use a folder '01' or the like, as mentioned it would be '07-28-2010_description' with the month as the leader to purely organise them in the year folder.

One good reason to have year folders is to reduce scrollage and RSI

True, and one reason why I don't have day-level folders under year, and have month-level folders in-between. Having to scroll down hundreds of lines to get to December would be a pain. Within each month I have week and day folders (sometimes with descriptive suffixes as you've mentioned, but usually not) but for my own system I almost always at least have "month buckets".
I wouldn't choose the MM-DD-YYYY format, but if it's working for you then that's the main thing.

On a side note: how do people deal with actual image file names, particularly once the you clock over the sequential numbers from teh camera?

Note that I pointed out above that the camera numbers are only to let the camera avoid generating duplicate filenames on a single card. The number will roll over eventually as you've mentioned, and if you end up swapping cards between cameras you can also get two cameras producing the same numbers on the same day!

As ATJ mentions, as long as the filenames are unique within the folder that they're stored in on disk, you might think that leaving the files with the camera name is OK. But I strongly recommend against this. By ensuring that your filenames will be unique across your entire database you will have the flexibility later to shuffle files around between folders. If you copy/move your "best" photos from 2009 into a single folder you don't want to lose some of the files because they used the same name (OK, Lightroom will add -1/etc suffixes to avoid losing things, but that's another mess).

I do the renaming through PteroFile as it's being copied from the card to the disk, but there are lots of software options you can use. Some of them (e.g. Bridge) have very rudimentary name formats.
My PteroFile config is currently:

Code: Select all

File format: <YYYY><MM><DD><hh><mm>_<ID>

where YYYY, MM, DD, hh, and mm are obvious tokens extracted from each file's metadata, and ID is a unique number (with per-machine prefix) that's stored in a database so I'll never get conflicts. PteroFile also inserts the ID into the file's IPTC Title field for me.
For people using Lightroom a convenient Filename Template to use when importing (or renaming within Library) can be

Code: Select all

<Date (YYYYMMDD)><Hour><Minute>_<Filename number suffix>

By including time information as well as date you're extremely unlikely to have multiple cameras producing the same sequence number within the same minute. By including the number suffix you can relate them back to what's on the flash card if that makes you feel better.

Also note that Lightroom finds the number suffix by searching back from the right until it gets a non-numeric character, and by separating it from the rest of the generated filename by a '_' we ensure that we can use <Filename number suffix> in later operations and get the same number.
If you left out the '_' then DSCN1234.NEF might become 2010072613011234.NEF and (quite apart from that being hard to parse visually) if you reapplied that rule the new name would become 2010072613012010072613011234.NEF!
201007261301_1234.NEF is easier to deal with, and will generate the same (or similar) name after reapplying the rule. Sometimes after correcting the Capture Time (e.g. because you forgot about daylight savings) you want to be able to clean up the filenames so the date/time info matches. Lightroom does have "Original Filename number suffix" but that doesn't do what you want in all cases.
For this reason if you wanted to put a descriptive string in the name I would put it between the date/time and the sequence number, not at the end. Note that I don't recommend doing this at all: descriptions can go into the metadata, and filenames only need to be unique - nothing more!

[surenj]

Wow this is complex stuff. I hadn't thought about this issue that much at all!!!

I usually name my files descriptively at the time of import ie Coogee sunrise Jul 2010 - unique image number. This allows me to remember vagule what it's all about even without a RAW photo or metadata viewer

I know the unique image number can be repeated but for my low volume use, I didn't think this would cause issues.

I let LR organise the folders according to date. It's relatively rare for me to do multiple shoots on the same day. If I did, I would import (and name ) them seperately but they will live physically in the same folder.

FYI: I used the search function in the LR catalgue yesterday and it is extremely effective!!

[surenj]

e.g. look at the IPTC data on this:

Andrew, the EXIF/IPTC on this image is amazing! You can even tell it was taken 12m below sealevel!

Thanks to everyone who has answered. I will need several weekends to read and re-read this info and learn something.

[ATJ]

By ensuring that your filenames will be unique across your entire database you will have the flexibility later to shuffle files around between folders. If you copy/move your "best" photos from 2009 into a single folder you don't want to lose some of the files because they used the same name

I will never shuffle my files around, so this will not be a problem for me. As I said, I never do any manipulation at the file level, other than backup. If I was to make a "collection" of my "best" photos from 2009, this would all be done within Lightroom itself - e.g. with tags or even ratings.

[DaveB]

Then as long as you know you can avoid duplicates within a folder (caused by the use of multiple cameras) you're golden!

I'm a little more paranoid than that.

[Remorhaz]

Dave - you mentioned how in Lightroom you set certain keywords to be non exportable - I'd love to know how you achieve this since this is exactly what I want to do (at the moment I strip all metadata from exported photos - the sledgehammer solution - to avoid having my childrens names in them)?

As above I also use Lightroom and have a pretty simple setup - the default:

YYYY/
YYYY-MM-DD/

folders and then on import I just rename the files simplistically to just:

YYYYMMDDXXX.NEF

where XXX is the sequence number allowing 1000 photos in a day (which I havn't hit yet )

I never name the files or folders after events or shoots or whatever - I just need the filenames to be unique. I tag the photos with keywords if I want to add anything like that. Interesting comments re adding the _ before the sequence number however - I might have to consider that.

I do keyword the photos but only fairly simplistically - I guess I'm getting better at it over time but my entire keyword heirachy probably only has one or two hundred keywords in it?

[ATJ]

Dave - you mentioned how in Lightroom you set certain keywords to be non exportable - I'd love to know how you achieve this since this is exactly what I want to do (at the moment I strip all metadata from exported photos - the sledgehammer solution - to avoid having my childrens names in them)?

This is how you do it with keywords in your keyword list. Right click on the keyword and choose "Edit Keyword Tag...". Under the Keyword Tag and Synonyms field you have:
* Include on Export (which is on by default)
* Export Containing Keywords (on by default)
* Export Synonyms (on by default but doesn't seem to work for me).

Deselect them to stop them exporting.

[ATJ]

Then as long as you know you can avoid duplicates within a folder (caused by the use of multiple cameras) you're golden!

I have 2 cameras, a D70 and a D300, but they have 2 separate parent folders so never the twain...

If I was to get a second D300, I'd either have a new parent folder D300-2, or set the second one to use the alternate naming convention - in fact, I'd probably do both.

[Mr Darcy]

Thank you all from me too. I am using Aperture3 though, so I need to work out what will work there.

Search by keyword, hierarchies seem OK. The file naming options don't appear to be as comprehensive & I will have to look into synonyms and private keywords.

It does have excellent geotagging options though. A big plus for me. It looks like I'll be dropping Geosetter from my workflow. One more nail in the coffin for the M$ O/S

At the moment I am working out what will make a good keyword hierarchy for me. I can see it will be a long slow process to get it right.

BG & Cameron, I am just a little further along the path than you. I have hit that "pain in the future" you allude to. I recommend you take steps now to avoid it. I am also moving away from NX2 because of its limitations in this area.

[biggerry]

BG & Cameron, I am just a little further along the path than you. I have hit that "pain in the future" you allude to. I recommend you take steps now to avoid it. I am also moving away from NX2 because of its limitations in this area.

yeah I am hearing ya loud and clear Greg, its a bit like taking medicine, always leave it the last minute.

Its a real shame about NX2 and its limitations in this area.. cause I reckon its an OK program...

[aim54x]

BG & Cameron, I am just a little further along the path than you. I have hit that "pain in the future" you allude to. I recommend you take steps now to avoid it. I am also moving away from NX2 because of its limitations in this area.

yeah I am hearing ya loud and clear Greg, its a bit like taking medicine, always leave it the last minute.

Its a real shame about NX2 and its limitations in this area.. cause I reckon its an OK program...

I really like NX2, but I do concede that it sucks for cataloging. I am not sure if there really is any other program that I will take a liking to....

[Mr Darcy]

cause I reckon its an OK program...

I agree. It is an excellent basic editing suite. I really like the way the control points work. for about 95% of my editing, it was more than enough for most of my needs.

What broke it for me was actually TransferNX. For Starters, I like to make my backups at transfer time. TNX seemed great for this, but one day I didn't have my external drive plugged in &TNX simply reconfigured itself to use the primary drive for backups too. No warning & no reverting to the external drive the next time it was online. It was months before I realised.
The real deal breaker is its recent behaviour. It has started transferring MOST of the files. No matter how many or few it needs to transfer, it will nearly always leave a few untransferred. I need to click "Stop" then "Start". Then another will transfer. Then repeat for every file left.

Aperture handles transfers in its stride, allowing renaming in the process, and using a backup destination, so that box is ticked. Its keyword handling appears exemplary, as its Geotagging (Once I worked out how to do it) I am still coming to grips with its editing capabilities, but it seems to do most of what I need, and I can always revert to CS3 (or NX2!) for the rest.

[DaveB]

Aperture handles transfers in its stride, allowing renaming in the process, and using a backup destination, so that box is ticked. Its keyword handling appears exemplary, as its Geotagging

I'm not that impressed with its keyword handling. Here's an extract from something I wrote recently on a different forum:

For this example, consider a hierarchy like this (my hierarchy contains a bit more detail than this, but it will illustrate the issues):

Code: Select all

ANIMALS
   mammal
      whale
         Humpback Whale


The keywords are nested, and in Lightroom if we apply "Humpback Whale", when the image is exported it will automatically be tagged with "Humpback Whale", "whale", and "mammal" (but not ANIMALS: that's marked as not exported cos it's just a bucket to organise the keywords into).
In LR if we start typing "hump" in the "Enter keywords" field it offers us "Humpback Whale" in the auto-complete list, and this is one way of finding keywords.
However if you know it's a whale but the spelling of Humpback is eluding you (sorry, bad example) if you start typing "whal" it doesn't come up: this auto-completion is only matching against the start of a keyword. Fair enough: there is another way.
In LR's Keyword List panel we can type "whale" into the Filter and it will reduce the list to only show us all the keywords that have "whale" (case in-sensitive) in them. Perfect: we can then just tick the box next to "Humpback Whale" (or drag the keyword from the hierarchy to the image, or ...).

In Aperture, the keywording HUD is similar to LR's Keyword List panel, but when you put something in the Filter field it only matches against the start of keywords. This makes it a pain to find keywords. The accepted and recommended workaround seems to be to have a copy of your keyword list in a text editor window so you can search there, and copy-n-paste the full keyword to the keywording HUD. Ouch! I'm amazed they didn't fix this in Aperture3.

The point of putting the keyword into the Filter/search field rather than just applying it to the images is so that the panel will then display the hierarchy containing that keyword. Then by selecting the hierarchy and dragging it all onto the images you can apply the parent keywords. Aperture doesn't automatically apply parent keywords (and thus obviously has no controls over things like hiding "ANIMALS")! Double ouch!

In LR if I reorganise the hierarchy this dynamically affects the inheritance of parent keywords. In Aperture I might have to go through my entire database of images and add/remove keywords to match the new hierarchy. Argh!

Also, using keywords this way doesn't support the concept of keyword synonyms. In my example of an image of a Humpback Whale, in LR just by applying that keyword any exported images (e.g. JPEGs) will automatically be tagged with "Humpback Whale", "Megaptera novaeangliae", "baleen whale", "cetacean", "mammal", "placental", and "whale" (and probably more by the time I finish). Some of those are parent keywords, some are synonyms.

Providing an easy and logical way to apply lots of relevant keywords is important. Not only to let you find things when you search your own databases, but also to let other people find your images: in stock libraries, Flickr, etc.

For a list of the other ways in which Aperture3 seems to screw up metadata in general, have a look at:
http://www.controlledvocabulary.com/imagedatabases/aperture3.html
Not exporting IPTC Creator contact fields to PSD files!?!?! OMG...

Bottom line: I'm glad I'm not trying to use Aperture.

Cheers

[Mr Darcy]

OK, I guess "exemplary" was in reference to what I was using before. I am still new to this whole keyword thing.
however, I am not sure that what you are saying is entirely accurate. THe references you cite refer to A3.01. I checked I am on A3.03. Metadata has been worked on since 3.01

Metadata

* Addresses compatibility issues affecting XMP sidecar files exported with masters.
* The Duplicate Preset command now correctly duplicates the data in all fields when duplicating a metadata preset.
* Fixes an issue that could cause a spinning wait cursor to appear when creating or duplicating metadata presets.

Metadata

* Query HUD now correctly handles dates entered into the Date search field.
* Search field on Keywords HUD now performs a “contains” search to find matching keywords.
* Fixes an issue with auto-complete in the Keyword Controls search field.

I will look into to it further in the next day or two.
I am NOT a professional. Nor do I aspire to being one, so many of the issues are not relevant to me. However, I do use CS3 now and again & would be bummed if metadata is lost in the round trip. I also like the idea of private keywords. I need to explore this further. I have noticed that recognised faces do not show names in keywords where I would expect this info to appear.

[DaveB]

* Search field on Keywords HUD now performs a “contains” search to find matching keywords.

Big improvement!
Still a few other concerning issues, but you might be prepared to live with those.

[Murray Foote]

On a side note: how do people deal with actual image file names, particularly once the you clock over the sequential numbers from teh camera?

I don't worry about it and leave the files the same name as what the camera gave them.

All my files are stored in a month folder (YYYYMM - although I don't bother with YYYY parents) and I doubt I will even even come close to taking 10,000 photos in a single month. I rarely do anything with the files on the HDD, anyway, other than back them up, so all my interaction is via Lightroom where the files names don't matter that much (to me as the user).

I change the prefix which I currently need to do on an annual basis so for 2010 I'm using 10M. If I start shooting more than 10K images a year I could use 101, 102 etc or 10A, 10B etc.