Options
file naming, keywords
3PCo
Registered Users Posts: 65 Big grins
Does anyone have a good system for file naming and keyword assignment in the context of websites and systems like Smugmug? Presently, I assign filenames based on date and a sequential numeral. Then I tack a descriptor on top of that at post. These filenames are useful to me, but they add a lot of distracting drivel to those little yellow popups that appear when you hover over something, and if I use SM's auto-keywording it makes a complete hash of things. I handle a fairly small workload, but I want traceability back to the original raw file as well as version control.
Thanks for any suggestions!
af
Thanks for any suggestions!
af
0
Comments
I'll start with what not to do. Don't use just dates and locations as identifiers.
Don't put off keywording or maintaining your files. Check this out -- http://www.kickstarter.com/projects/1285547290/captive-a-zoos-book [note to mods: please leave the link as it is the example that makes this example so meaningful. I ask for forgiveness if it is somehow against the forum rules] -- anyway, this project has me sorting thru about 1/2 million images digitally taken or scanned that are filed by date on external HDDs and have no other descriptions/keywords in their names or metadata and looking for approximately 25 +/- individual photos that I cannot remember when they were taken. And that is with the photos that are already existing and not for the ones that we've yet to take when the project is complete. I updated my workflow but figured it was too late to go back thru the previous photos. I'm constantly crashing my computer because it doesn't handle such large amounts of searched data when using generic searches -- such as *.jpg. My biggest tip is to start now and move forward with a system and stick with it.
Here are my other tips:
1) Using the method that best works for you, as soon as you are done shooting (never put it off!) rename the files with some sort of descriptor of what it is. Even general, broad categories works. Anything that allows easier searches in the OS and without having to use a specific program such as Bridge. Set up your camera before each shoot to incorporate the name/location in the file name so it is already there on image transfer.
2) With digital photos, fill in the metadata. There are programs such as http://www.snapfiles.com/downloads/msprophoto/dlmsprophoto.html (or similar) that can help with keywords. Do them in camera or with a program designed to keyword.
3) As for SmugMug, I've got that part of my workflow figured out. I keep a spreadsheet with my most used keywords listed in each cell. I then select each one I want to use in my keywords within SmugMug, copy them into notepad, replace the space with a , and then cut and paste it into the keywords slot in SmugMug. Takes only a few minutes to do a couple of hundred photos.
4) In my updated workflow (I learned thru trial and error (-: ) , my filenames are my keywords, title and revision all-in-one. Since we now have 256 characters available, I've incorporated them into my naming system. For example a file name might be like this: 11222013_Lions_SA.NEF would be the raw file and the next revision would be 11222013_Lions_SA__8x10_01.JPG would be the first revision saved as JPG. My naming system tells me the date it was taken (date), the subject(s) in the photo (lions), the revision it is (01), a location of where I took the photo (South Africa), whether it has been post processed in several manners (8x10) and whether it has or has not yet been uploaded to a SmugMug gallery (no it hasn't).
Good luck with whatever system you choose.
I like your system, but it seems that, since you use only the date and not a sequential numerical identifier, you must end up with multiple files that are similar: let's say several shots of the lion you took in the same place on the same date. I will probably keep a number in my system.
I'm also wondering how you deal with projects in which multiple initial captures are combined into a single final: for example panoramas or HDRs.
Thanks again!
af
http://www.pillarpointposter.com
1. All my cameras have EXIF data in the camera. My name, copyright info, etc. That way EVERY file I ever shoot is carrying something traceable back to me.
2. After my shoots, when I get home, I use a program called Photomechanic to bring my images into my computer. On ingest, I have the program automatically rename my images with the following format: PTFPHoto_data_originalfilename. The subject of the shoot is not in my filename but I don't find this a problem. This naming convention also ensures 100% unique file names, and allows for easy sorting.
3. Photomechanic (and Lightroom/Bridge/others) allow you to insert metadata. Photomechanic has a template for this. I fill out captions, headlines, subject fields, keywords, etc. The software also includes numerous variables so I make use of those to include exposure information, GPS data, etc. into the metadata of every file I shoot. This may seem like a lot of work, but it's just a template. And it can be applied during ingest, or after the fact. So it all happens automatically. I can apply a template to 1000 images in under 20 seconds.
4. Once all this work is done, everything stays with the files through the remainder of my workflow. So when I upload to Smugmug, or submit to my clents, everything travels with the files. My webpage, email and other contact info, copyright, keywords, etc. All of it becomes searchable on the web.
I spent about a year refining this workflow and it works extremely well for me. Name your files to something that works well for you, and add as much metadata as you're comfortable with. But be consistent, do it as early in the workflow as possible, and enjoy the fruits of your labor.
perroneford@ptfphoto.com
Does the metadata follow the files into composites like panoramas or hdr's, do you know?
Thanks again!
af
http://www.pillarpointposter.com
Beyond that, it would be hard for metadata to follow in composites because you are combining more than one images and more than one set of metadata. However, if you are using templates, you can apply the template AFTER you do your composite and HDR if you like.
perroneford@ptfphoto.com
http://www.pillarpointposter.com
I have no idea what this means. Lightroom is more integrated with Photoshop than Bridge is. In fact, the full name for Lightroom is Adobe Photoshop Lightroom...
perroneford@ptfphoto.com
Really? I thought it was more of a Photoshop alternative, with Bridge-y capabilities. Maybe I don't really understand what it does.
http://www.pillarpointposter.com
2) The most important thing about a filename is that its guaranteed to be unique so that no two images have the same name.
3) What you need to really do is harness the metadata to help you locate images later. Captions, keywords, location data, keywords, contact info, have I mentioned keywords yet?
Check out an "old" book that is still extremely useful for questions like yours: http://www.amazon.com/DAM-Book-Digital-Management-Photographers/dp/0596523572/ref=sr_1_1?s=books&ie=UTF8&qid=1385603227&sr=1-1&keywords=the+dam+book
A former sports shooter
Follow me at: https://www.flickr.com/photos/bjurasz/
My Etsy store: https://www.etsy.com/shop/mercphoto?ref=hdr_shop_menu
af
http://www.pillarpointposter.com
Thanks
af
http://www.pillarpointposter.com
perroneford@ptfphoto.com
That's exactly what it does, and since my filenames are similar to yours, with a date and some text, it looks awful. So I find a Headline field in my IPTC: I take it that's what you are using. I guess you must put text in all of these?
Thanks for all your help!
af
http://www.pillarpointposter.com
This is also where something like Photomechanic shines. Since variables can be used in these text fields, it just pulls data as needed into the fields when the template is applied.
perroneford@ptfphoto.com
I personally now stay away from metadata and templates because there is 1) no guarantee that it will be supported in the future, 2) won't change even if it is supported, 3) there have been times when metadata has been lost during a transition to newer OS or machine (Google "lost image metadata" if you want some examples) and 4) metadata is not a standardized format. Not using metadata is what works for me even though I personally know of many people who are happy using it. To each his/her own...
While I have different camera kits so each one is slightly different, I've listed below a generic example of what file names look like at beginning middle and end of my workflow. I've used an extreme example to answer your questions.
IMG201309010001.RAW = out of camera
IMG201309010001.TIF = saved to Tiff
IMG20130901_0001a_SA_LIONS_HDR_Finof12_8x10_01a_SIG_BRDR_Reduced_DA.jpg = the final name before upload to aliceannys.com
Then I'll rename a copy of the last one to 20130901_001a That is the final file uploaded to aliceannys.com Here is what the long file name tells me:
The image was taken on Sept. 1/2013
It is the first one taken that day.
It was color corrected in Raw before taking it into Photoshop
It was taken in South Africa.
It is of Lions
It is an HDR file
It is the final revision of the HDR
I used 12 photos in the HDR using photos numbered 1-12
It has been cropped to an 8x10 format
It has been revised in Photoshop from the original two times and has one other "sibling"
It has a printmark
It has a border
It has been reduced for upload to aliceannys.com
It has been uploaded to aliceannys.com
It is located in the Digital Art gallery on aliceannys.com
I can then find every revision of that file just by a Windows search on the 20130901_001 identifier. For me, the name is short enough to solve the little yellow tags issue. If I searched 10 years from now in my file archives, I'll most likely find that I've got raw, jpg, png, tif, psd and other extensions all associated with that name and all of them will be in chronological order of when it was post processed and in which program I used. My methods work for me because it gives me flexibility but doesn't make me have to use templates, open another program on my machine to read the metadata, use keywords in multiple software, or allow loss of future metadata. It does allow me to work across my 6 different image editing programs I use with many revisions in each native format. It also allows me consistent processing and tracking of images from the many various camera manufacturers I have used and/or currently do use. I can go back and repostprocess an image out of the archive with a simple one time windows search as well.
I'm sure your mileage will vary so what works best for you will be a workflow that you ultimately end up designing for yourself. My system works for me because it hits the important-to-me parts of my workflow as mentioned above.
Happy shooting!
To a degree this is true. Much like floppy disks are no longer supported. The question is, how much time into the future are you projecting? I can't play my old 8-track tapes any more. But I'd hardly turn down playing a DVD or CD these days. When done in a planned manner, moving forward need not be painful.
IPTC fields (aka metadata fields) are updated and deprecated every year. This does not "break" old files and need never break them. It is a natural evolution in how metadata is handled.
Not entirely surprising. However, it seems like this is an advertisement more for having good backups than throwing out an efficient system of tagging images and media.
Sorry, but this is patently false. If you note any program that deals with metadata, you'll note that it generally has two flavors. EXIF information which is written into the RAW, JPG, or other file recorded in the camera. And second the IPTC fields written by software once the file is moved from the camera onto the computer. While EXIF data is somewhat proprietary, the specs are published on how to read them which is why hundreds of programs can read and process the EXIF data from all major camera manufacturers. As for the IPTC fields, they are in fact standardized. And came about precisely FOR the purpose of creating a world wide standard. IPTC stands for International Press Telecommunications Council. They devised the standard for metadata for media, and keep it up to date worldwide. It allows a photograph taken in New York, to be properly cataloged and published in Tokyo, or Madrid, or Paris. Those interested can read more about this here: http://www.iptc.org/site/Photo_Metadata/. Once my work started being published on a regular basis, I had to take time to learn how to properly utilize the fields so that my work would be properly read by newspapers, magazines, and online sources like ESPN.
One other thing of note. Using exceptionally long filenames can cause problems on some computer systems because they are not always supported. In those cases, the names are truncated, and the naming system goes out the window. As long as you are the only person who ever needs to handle your files, this need not be a concern. But for shooters who might need to hand off work to newspapers, magazines, a client who needs a CD/DVD (long filenames can be a problem there), or upload to a printing service, etc., then care must be taken in the naming conventions.
Thanks for reading.
perroneford@ptfphoto.com
I'm currently using files that are 20 - 25 years old. Some of my data is coming off of my QIC 80 tapes and, yes, I can still access them at home. My disaster plan gives me access to my data 50+ years into the future. It is all about individual needs.
Actually it isn't false whatsoever. You have even quantified your own comment by stating that "If you note any program that deals with metadata" and again with "the EXIF data is somewhat proprietary". While I heartedly agree that IPTC is a form of standardized metadata, metadata as a whole component is not. I currently have a three year old image processing software program that uses metadata to store all sorts of information about an image taken from digital cameras EXIF but it has no way of handling keywords, it can't track internally the date of modification (it shows the date of creation instead), and it has no idea what to do with the metadata that comes across as a description, author, title, etc, etc, inserted when opening a psd file (which it will do but without layer support). The program only uses 5 EXIF to IPTC database fields and then stores proprietary info in the rest of the fields. So if one views a file and looks at the city/state one sees lat/long info instead. The help file is also in an Asian language that I cannot read.
If all image metadata (again as a whole unit) was standard, there would be no need in Photoshop to have both the IPTC tab and Origin tab too. Even the CD/DVD you mention isn't standardized completely (hence the problem with long file names on some systems and not others.) Someone always thinks they can build a better mousetrap
With this, we can agree. Depending on the needs of the OP, he/she needs to know what and for whom images are being taken. I am curious about one part of this, why would one need to hand off the interim files (in my example the ones with super long names) to those entities that you named? My clients only get the end result (I'd be so embarrassed for them to see my working files) and the final name on the deliverable/upload file is as about as short of file name as one can have and still be descriptive. I seriously would like your opinion on this?
Because when you shoot for a wire service, or other news source, immediacy is the name of the game. My photos are due less than 15 minutes post game. I do not have "iterations" in those instances. When shooting for schools, I provide 5-10 photos less than 15 minutes post-game, and a full album sometime later that evening (in which a more normal Lightroom/Capture One edit session occurs). When I am shooting models then your workflow more closely mirrors mine, but I prefer to use metadata for numerous reasons. I can store GPS data so that I can return to exactly the same location years later. I can store my info on model and property releases including needed contact info, I can populate fields I need for stock photo services and the like etc. None of this is possible with long file names alone.
Again, if you are handling your files end-to-end, and do not need to hand off your files to anyone else, your methodology is likely as viable as anything else. But for those who wish to collaborate, sell to stock companies, work with media or news outlets, etc., images without proper metadata will be rejected, or if it is used, no credit will likely appear because that information is also pulled from metadata. Copyright, licensing, usage terms, photographer, contact info, etc. All comes from metadata. If it's blank, then you're just out of luck.
perroneford@ptfphoto.com