DAM Workflow Questions

[cabrackett]

Peter:

I have a few questions. I have been following the guidelines for creating the DAM architecture that you suggest as it fits into my software and hardware constraints. For reference, I am using a laptop with a small HD (60GB) and two USB external drives. The first drive is for my archived originals and archived derivative files, structured in two separate folders, and also a lot of old archived legacy files. The second drive (300GB) is for backing up the archives and the laptop hard drive. The files I am working on now (about 1200 of them) are Jpegs of about 3 MB. So far, so good. Now for the questions:

1. After the files are archived as originals and we go to iView to sort into catalog fields, polish the ratings, and tweak things, we are working  on the archived originals still. We have imported the ratings from the metadata into the catalog fields. The thumbnails still have the early ratings attached under ratings and labels and I am puzzled about whether to do anything about those displayed ratings or not. This is not a how to question, but a what makes sense question. Remember, I am still working on the original files, none of them have become working files yet (at least that is the way I am doing it now.) So, after additional work, the drivative file will be catalogued and searching the derivative files and the original  files may yield different results. Comments?

2. It is my plan to import files on a file-by-file basis into the laptop for editing in Photoshop and hold them there as working files until I am finished, then they go to the archive for the derivative files in the derivative file folder on the archive hard drive. Further on down the line, further revisions to the ratings, captions, and other metadata may take place and it is my presumption that those changes are not to be reflected back into the archived originals. Once a derivative file is formed, do we essentially "lock" the original form of that image file?

3. A similar question arrises regarding the derivative file. Some will not be edited a lot, some more, and  I can envision filing these drivative files away in their archives, only to go back later because of some need or using some new  technique or tool, and bring some  of those derivative files back into the workflow in the working file directory. What is the best way to manage such a situation?

4. Catalogs. You mention that you use several, but caution against using too many. What are the legitimate uses of multiple catalogs and when should they be used?

5. In the book you show an example of Catalog Sets that include a set for keywords, separate from the metadata linked catalog field for keywords. What is that all about?

Thanks,

Chuck Brackett

[peterkrogh]

Chuck,
This is a bit more than I can chew at the moment.  I'll try to be helpful.  (My numbers do not correspond to yours).

First, start your system by working with new files.  Sorting out old stuff at the same time makes the whole deal more complicated.

1. Bring in JPEGS.  Add metadata template and Ratings, put in buckets and put away.  Catalog files, dividing into useful sets.  The more work you do on your original files, the less synchronizing between original and derivative files later.

2. As you want to work on these files, bring them into your working files folder in your hard drive.  As they get finished (for the time being, at least), move them into an archive structure for derivative files.

3. If I readjust derivative files, I save them into new buckets (and sometimes will delete the older file, if the newer one is clearly better).  This keeps you from having to go and re-backup the old material.

4. With the number of files you have, I would only be using one catalog for everything.  It is much simpler.

5. You can add and alter metadata (including ratings and keywords) all you want to the files.  Your catalog holds this information. Right now, there are not great automatic tools for synchronizing metadata updates between different versions of the same file (like an extra ratings star, for example), so you have to do this manually (or use one of the existing iView scripts).

Let me know which of your question(s) the above does not address.
Peter

[cabrackett]

Peter
Sorry to be so long winded and to take so much of your time.
You did answer a lot and I will get back to you if I need more help.
The 1200 image file is just the tip of the iceberg though, so I  am trying to use these to establish a good protocol.
Thanks again.
Chuck

[peterkrogh]

Expect the capacity of the software to grow as well.  My catalogs are still pretty happy at 30,000 images.  Capacity is theoretically 12X,000, so there is a lot of room to grow.

Only split your catalogs when you have to.
Peter

[cabrackett]

Peter
One more thing. Are there other legitimate uses for more than one catalog, or do you use them just for sheer volume control? I was thinkking of a  catalog devoted to Portfolios, for example.

And, what was the catalog set for "keywords" all about?

Chuck

[peterkrogh]

BEst reason I can think of for more than one catalog is job delivery.

Portfolio should be a designation within your main catalog.  Why would it be a different catalog?

As to the Catalog set for keywords, do you have a page reference?
Peter

[cabrackett]

Figure 7-10, page 206
chuck

[peterkrogh]

Chuck,
That was created by a script that created catalog sets automatically from keywords.  I have not looked at how all the iView 2.6 scripts interact with 3.0, so I can't tell you right now if it can currently be used.

If a script can be used to create catalog sets from keywords when you first import the file into the catalog, it could streamline the process of getting images into catalog sets quickly.

Peter

[ape-fish]

Hi Peter,
I have been using the Cumulus DAM tool in a large corporate environment for sometime as a profesional photographer. Your book prompted me adapt my current workflow to using NEFF/DNG exclusivly instead of JPG (always adequate but not a great archiving format).

I appreciate all your arguments for both RAW and DNG however I seem to be missing a point in the workflow system you recommend. What point do you do the the detailed correction edits such as retouching etc? Everytime I want to edit something not supported in the camera raw I cannot seem to save these as DNG CS2 I just get offered to save as a derivative.

All my databased images are available to staff on an intranet - so as soon as I publish the image it is available for download - Cumulus automatically prepares the derivative (jpg screen rez) from the originating file if a client requests a download. In effect the databased file is the end version and should be a DNG.

I recognise my application is bending significantly your workflow recomendations from the book - however I guess I'm not the only person using DAM to dynamically publish content.

I would be grateful for thoughts - and thanks for writing the book I thought I was on my own with these problems.

[Patrick Cox]

...What point do you do the the detailed correction edits such as retouching etc? Everytime I want to edit something not supported in the camera raw I cannot seem to save these as DNG CS2 I just get offered to save as a derivative.

I am new to all of this as well but I believe the primary workflow here is to rate, catalog & archive your DNG files which will contain a "pretty good print" in the form of the ACR adjusted full size jpg preview image that is contained in the DNG file.  For Peter (I believe) and certainly for me, the Raw corrections only will do for most images.  Whenever you need to make localized corrections, you would create a derivative file, such as a Tiff or jpg and then save this as a master to your "Derivatives" folder structure.

BTW, if most of your shots need further adjustments in photoshop, you will be losing a lot of the benefit of this workflow I believe.

I hope this helps.

Pat

[peterkrogh]

Ape-fish,
Pat's got it right.
To add to that, I would say that the requirements of an enterprise solution are beyond the scope of my book.  I do some consulting on the matter I was at Oracle earlier this week), but there are not any solutions I really like.  Hopefully soon.
Peter