A mini-taxonomy of digitization.
Not too long ago there was a #critlib chat on the “ethics of digitization” (backstory, update on backstory). Which I think means the ethics of access, of proliferation, of sharing online, of open content. When I brought up that people seemed to be using “digitization” to mean lots of different things, one of the chat organizers said “Hrm, never really thought about it. Doesn’t it mean putting stuff online, with metadata?” And no. No it doesn’t.
Digitization: Taking an analogue item and making a digital copy of it.
Preservation copying: Making copies, analogue or digital, that preserve content or information about an item.
Access copying: Making copies, analogue or digital, that allow the original item to be preserved better, or kept out of harm’s way.
So, please note, preservation and access copying can be done to analogue or already-digital items. You can therefore do preservation digitization or access digitization. (Sometimes both, but not necessarily.)
Here’s a person digitizing one item. (A one-page item, not a book.)
Here’s someone else’s blow-by-blow account of their digitization of a book.
Now, once you’ve digitized something strictly for preservation, you’re going to be ingesting it into a digital preservation system that helps ensure the digital object’s authenticity. That’s going to be a LOCKSS model that makes redundant copies and compares them regularly to see whether any have suffered authenticity loss (corruption, bit rot, whatever). There’s also going to be a fixity check upon ingest that generates a “hash,” which is essentially a code for future testing-against. If the digital file in the future generates the same hash, the file is genuine; if something in the digital file has changed (corruption, bit rot, whatever), the hash generated won’t match and we’ll know something has gone wrong. Both systems (LOCKSS and fixity) are important, and we use both.
What if you’re digitized something not just for preservation, but for access? Then you have lots of things to worry about.
Perhaps you’re digitizing for access because one specific researcher or media outlet or somesuch requested the image or document for some public use. In that case, you’re digitizing it, tucking a copy into your own digipres system, and emailing them an attachment or popping a burnt CD in the mail. Simple, done.
You keep the digital copy you made for unknown future use, but you don’t necessarily do a lot of metadata work on it. You’re not doing a release. But you are setting clear guidelines for rights and reproductions, assuming that this one specific request is for one specific use, and you’re not letting them just have a copy for some unspecified perpetual usage. (Generally speaking, we’re not allowed to make money off reproductions, only recoup the cost of providing one - so, digitizing and sending.)
What if you’re digitizing for open content? Or a virtual exhibit? Or some other en-masse release of images where there weren’t no images before? Then you’re talking about a big project with a lot of considerations. You’re talking about bulk digitization, bulk metadata formatting and improving, platform implementation and UX design, copyright assessments and license decisions, and first and foremost, surveying your collection to see what can feasibly be put online without a bunch of complications from all of legal, technical, and aesthetic sides.
Again: you do that last thing first. You identify what can actually be shared ethically, with an eye to the privacy and copyright status of all the items, and to the donation agreement that governs their use. You put a lot of work into getting into contact with people who are represented in the collections. You enlist community members and organizations to help you with this, if you haven’t simply chosen material that’s 200+ years old and 100% inoffensive.
There are lots of alternatives to putting your digital objects on your website with complete metadata, if you genuinely want to find a happy middle-ground between collection discoverability and ethical/legal restrictions. You can catalogue things without sharing the digital surrogates. You can share the metadata in a variety of hard-to-search/index ways (hosted PDFs, anyone? I’m looking at you, archives), or share minimal metadata. You can password-protect stuff on your website and hand the password out upon request (not a great solution, but it would work in 60%+ of cases). Lots of options like these were shared in the #critlib chat.
And you can host all your digital objects in an awesome platform that’s installed only on an un-networked computer in your reading room. If takes away all the access that people generally conflate with digitization, basically making your digitization work for preservation only - giving people access to the digital versions and disallowing access to the analogue items.
The best part about this is: you can do the digitization NOW, and then put it online LATER, in XX years, when the stuff is in the public domain and all the involved folks will be dead, and after you’ve done the requisite research about opening up access! Digitize now, share later - a real thing!
(You can also install a really heinous TOU on your website that pops up in one of those little bars at the bottom of the screen that most people X out of without reading, but which nevertheless functions as contract law, allowing you to sue anyone who uses your materials in ways your TOU doesn’t expressly approve of. Which is, like, every use on the internet. You will be a troll, but you will be a RICH troll.)
If there are collections out there that have been digitized without looking at all these angles, those people have made grievous errors. Their leadership team, their advisory bodies, and their funders need to seriously reconsider their accountability models and decision-making structures.
The good news is, it wouldn’t take too much effort to start writing some emails to these libraries, advisors, and staff to tell them you think they’ve done a bad job of protecting the records creators and owners in their collections.
2087
ellievsbear
andulkaphoto
turnnoffyourmind
theartofmadeline
izzy806
