Background: I have a few hundred census images stored as PDFs each named for the head of household. Most are single page, but in some cases they span two pages. (This happens when the household is on the bottom of one page and continues on to the next page.) Currently, if I have two households on the same page, I have duplicated the PDF. DT will also list them as duplicates even if one document has only one page and the other has two (which is handy).
Question:I’m beginning to be concerned that I am making my database unnecessarily large. These are images, I’m not editing them, and there is no real reason to duplicate them, except for the one page/two page duplicates. Anyone has advise on the pros/cons of Duplicates vs Replicants?
You can’t rename a replicant without renaming both/all instances, so I’d say that pretty much rules out replicants for you.
After thinking about it a couple minutes, I’m not sure I could provide a better solution based on the small amount I know of the situation. Are these images with wavy lines and squiggles ‘n’ stuff, or are they images because they’re text but not OCRed and don’t have a text layer or whatever? In other words, is there a better format they could be stored in? (PNG or JPG if they’re images, PDF->OCR->RTF or something if they’re text)
Other than that, and knowing that any better solution is going to involve work, I’d advise you not to worry too much about it. Wasting space? Who cares? Not being “efficient”? Who cares?
If it works and it’s not causing any problems, I’d say just don’t worry about it. If it’s making the machine slow or reducing the number of backup copies you can keep around, well, maybe something should be done and I’ll try to help figure out the best solution.
I forgot that renaming one replicant renames all of them.
I’m running into some space problems because I carry a copy of my database (created using the Backup Archive script) on my iPod, too. I have other files on there that don’t change as often that I will just burn to a DVD and store elsewhere.
The census files are handwritten and it’s necessary to preserve the original image file. I have transcribed them to sheets, which is nice, but old handwriting is open to interpretation and my transcriptions may have errors.
