With gigabytes of pictures accumulating over time, I'm wondering what systems (software, strategies, etc.) people are using to keep track of their collections? Let's say you remember a picture and you want to find the file on disk, how do you ensure that this will be successful?
I'm a fan of Adobe Lightroom - its way of working seems to fit my mental model so I find it fairly natural to use. That does depend on you tagging the photos with appropriate metadata of course and not everyone thinks the same as me, so you may not get on with it.
Picasa has matured nicely over the years and has a nice "face finder" that does a pretty reasonable job of auto-finding people in photos and matching them to people you've previously named.
Supporting this I have a fairly simple folder structure of
\photos \ shoot description
Because I rely on tags in the software I don't really need too much hierarchical organisation and IMO hierarchies are the wrong tool for the job anyway because a photo will naturally fit into multiple categories but a hierarchy only lets you express one.
I think the short version of the answer is "Tag everything" - doing it on import isn't that painful and means you can finds it later, but going back through gigabytes of history to add the tags is a soul destroying task.
For managing larger collections, 25,000 - 250,000 images, you just can't beat DBGallery (www.DBGallery.com). It's tagging and search support is excellent, and has numerous other tools for organizing a larger collection.
More specifically, here are the main ways this software is used to make larger collection manageable:
1) Logical Views. Explore your collections by year and month, by camera, keywords, photographer, and dozens of others. Basically it slices any data it has on images and lets you explore based on the results.
2) Virtual Sets. Over time most want to have a collection of "greatest skiing photos", or "photos that coulxd go in a calender". Of course copying images around folders is no way to do this. Virtual Collections simply point back to the original photos (they can be called anything, Light Boxes, Buckets, etc.). Doing this has extra goodness because if tags are added to the original photo or the photo is modified the Virtual Set is instantly updated as well.
3) Tagging tools. Examples include Data Templates, editing any number of images at once, and smart options such as a retrieving the data from the previously saved photo via a keystroke.
4) Great seach capabilities. Search all data for a photo, fine tune a search by searching jsut specific data fields such keywords/title/author. Also get all photos which were imported, viewed, or last updated recently, e.g. show all photos added in the past week; show all updated today; etc.
5) Will find duplicate images across the entire collection.
6) Has been tested with hundreds of thousands of images.
Also very important: this is an open system, where you can move photos to other tools or stop using DBGallery altogether...your data is still there because it's written directly to the file, not just the database.
Another vote for Adobe Lightroom here. It combines the Adobe Camera Raw processing engine with a powerful backend database to allow you to store your images and more importantly find them again easily. A side benefit is even the processing is stored in the database, you only need to store one copy of each image (the raw) and the processing is applied on the fly.
For Mac users Apple Aperture does much the same thing, but Apple have chosen to price this at about 1/4 of the price of Lightroom, so that is worth a look.
For a more consumer level option Google Picassa also offers cataloguing as well as image processing, it works well, especially considering it is free software.
I created software, the SeaRisen Image Browser to organize and quickly find your images. I too have gigabytes of images and find it a real time saver when looking for particular images, or series of images.
This management is exactly what Lightroom is designed to do. I used to be responsible for document control for a CAD department, and came up with lots of tracking and revision systems. I can say that using Lightroom is the easiest and best way to manage lots of images. I have over 40,000 images organized using Lightroom.
I have mine setup to place the images in the following structure: Year Month Day
Then I add lots of keywords. Using Lightroom I can find pictures quickly and easily. Including ones that are keepers and ones that the electrons can be recycled without missing anything.
Lightroom has a 30 day trial for both Mac and Windows, so you can try it out and see what you think. Adding in the ability to do editing and management just adds to the power of the program.
Well, I think Lightroom is your power answer, but I'm similar in the old fogey category in that I structure on disk. Basically, my layout is:
So, for example, one would be:
I do the numeric/textual month thing for quick glance and sorting. It's been working for me, though it may start to fail if my memory starts to slide. ;)
Before all this digital hype, I stored my slides in boxes after having numbered each of them (from 1 for the very first picture to more than 7000). In a simple text file I carefully noted the index of each slide, where and when it was taken and a little description of the subject.
Finding a picture was (and still is) just a matter of remembering some of these infos (where, when and what) and searching through the file.
In my long process of scanning my slides, I continue to use this text file for feeding infos in the cataloguing software I use (this is iView Media Pro).
For digital photos I use both iView Media Pro and iPhoto.