To understand this concept, it can help to think of the Snagit Library like a traditional photo album.
In a traditional photo album, each of your photos has two parts. There is the actual image and also the details you've written on the back of the photo-like the date or people in the photo.
In the Snagit Library, the .snag image files have two parts too. There is the digital image and also the information (called metadata) associated with the image-things like the time of the capture, what website or program was captured, etc.
When you take a capture with Snagit, the image is auto-saved to the datastore, a special hidden folder on your computer. This makes it very easy to capture a series of images, without stopping to save each, and may serve as a bit of a safety net for those of us who are a little less organized. The image remains in the datastore until you've consciously made the choice to save the image elsewhere.
To conclude the photo album analogy, the datastore is like the shoebox that holds all your photos until you have the time to sort and order them properly.
When you choose to save an image out of Snagit, the program assumes that you've accomplished what you needed to do and the original capture is removed from the actual datastore folder. You can still see the image in the Library though-and it shows you the type of file extension on the thumbnail. This is a good way to know the image has been intentionally saved.
When you save an image out of Snagit, the metadata (such as the time of capture, keywords you may have entered, tags, the application or website captured, etc.) is still intact. This can be really handy because even if you forget where you put something, Snagit's Tag tab may be able to help you track it down. Note that this only works if you save the file from Snagit. If you manually move images files around using Windows Explorer for example, Snagit will not track the file.