Is digitisation on topic?

There have already been a couple of questions focused on digitisation rather than digital preservation. I would assume that these are actually off topic and should be closed? The boundary is of course a little fuzzy.

I would propose that this community should be involved in decisions related to the results of a digitisation initiative. For example, the file formats and metadata schemas used, where/how the results are stored, and so on. However, questions focused on how to engage in digitisation, what types of scanners should be used or resolution to digitise at, will be off topic.

This question is about OCR, although is phrased in the title as a more general digitisation question. Is this in scope? Its a tough call.

As many of us in the digitisation community are well aware, confusing digital preservation for digitisation is a common mistake. I'd suggest adding some clear scoping detail on this to the "What kind of questions should I not ask here?" section of the FAQ. Examples may well be necessary to keep things clear.

Paul Wheatley

• discussion
• scope

Comments

• Andy Jackson: Late to the party, but just wanted to say that I think we should not be surprised at the confusion. That Digital Preservation means Preserving Things Digital rather than Preserving Things Digitally is pretty confusing, and we should expect to help newcomers disambiguate the two.
• Paul Wheatley: We could definitely use a migration path to Libraries and Information Science stack exchange, so any questions closed due to scope can be pushed in that direction.

Answer by Donald.McLean

(updated to reflect a more careful consideration of the subject)

It occurred to me that the answer should be a conditional yes, rather than an unqualified yes.

If someone is just interested in digitizing audio, video, or photographs, they should probably be asking in Audio and Video Production or Photography.

If someone is asking about digitizing as part of a digital preservation project or with the intention of digitally preserving their material, then I think that it should definitely be on-topic.

It seems to me that if someone wants to digitally preserve something, and they have a legitimate question about the digitization process, especially if it relates to the preservation aspect, then we shouldn't be turning them away. They're a legitimate part of the constituency of this site and we should be taking their questions seriously.

Comments

• Paul Wheatley: I would argue that digitisation is not digital preservation, and both wikipedia and the experts agree with this assertion. Hence digitisation is out of scope. Or the title of this site is wrong.
• Donald.McLean: I didn't argue that digitization was digital preservation. I argue that it is a closely related topic. Fanatical adherence to a very narrow topic set isn't healthy for the site.
• Paul Wheatley: I don't think this is fanatical adherance. We're talking about two related but very different topics. Confusing the two would not be a positive move for a best practice site. If it was called "Digital Preservation and Digitisation" I would not have an issue.
• Donald.McLean: Without digitization, there wouldn't be any digital content to preserve, so I hardly think that ruling it as off topic is either wise or productive.
• Paul Wheatley: I strongly disagree with that statement. Much of the digital information created today is *born* digital. If digitisation is done correctly, its digital preservation should be trivial. The main digital preservation challenge lies with born digital data. This is why its very important to avoid confusion with digitisation. Its a different process.
• Donald.McLean: Born digital content may be the norm for some companies and some industries, but there's a big wide world out there, full of folks with different issues and different situations. This site is for all of those people, not just those wrestling with problems inherent to born digital data. I still haven't seen a single argument that would convince me why such a closely related topic should be excluded.
• Paul Wheatley: *:-) I haven't seen an argument that would convince me why every digitisation question should be included in a digital preservation Q+A site. I'll bow out of this one for now and see what other people think, as this should be a community decision.
• Donald.McLean: I agree - it is a community decision. I'm just trying to point out what I feel are important points about how questions like this will affect the long-term health of the site. A certain amount of tolerance and inclusionism will help build a strong community where blind adherence to some arbitrary guidelines won't.
• John Lovejoy: I am a firm believer that digital preservation DOES NOT involved digitisation. Digitisation is a process to create digital artifacts, sure, but the long term preservation of those artifacts is a totally different area, with a totally different skillset required.
• Paul Wheatley: A further thought I had on this: Digitisation remains in scope on the Libraries and Information Science SE, where many from the digital preservation community have also been asking questions about digital preservation in lieu of this SE going live.

Answer by Courtney C. Mumma

I agree that the mechanics of digitization are not in scope of this site, but think that the preservation of the results are well within scope, and that includes the formats that one creates in the course of digitization projects since those formats may determine the course of preservation activities.

Comments

Answer by warren

If you're not digitizing, how can you be doing digital preservation?

Seems like it has to be on-topic.

Comments

• Donald.McLean: My thoughts as well, but said more succinctly.
• walker: To be clear, I believe by 'digitisation' Paul means the process of creating a digital representation of an analog object. Therefore one could easily have 'digital preservation' - __ensuring access to a digital object over time__ - without digitization, just by working with **born-digital objects**: tweets, images from digital cameras, Word documents, web pages, Pixar movie components, etc. Nevertheless I personally think digitization would be a candidate topic here.
• Adam-E: I agree, though one might argue that the content would have been born digital and how to preserve this rather than preservation by digitising.
• Adam-E: Ah, Looks like @walker beat me to the punch on my last comment :)
• warren: @Adam-E - given that not all artifacts are "born digital" ... it stills seems that digitization must be on-topic
• gmcgath: If digitizing is automatically on topic solely because it's a prerequisite to digital preservation (although it isn't, as has already been pointed out a couple of times), then wouldn't everything from booting up a computer onward have to be considered "on topic"? I agree that digitization is on topic when it comes up in a preservation context, but not unconditionally.

Answer by wizzard0

I think digitization of analog objects with the goal of digitally preserving them, or keeping them easily-preservable in the future is on topic.

Digitization by itself, however, may better be discussed elsewhere.

Comments

Answer by Nicholas Webb

As Donald and Courtney have pointed out, there's a continuum of relevance here. "What's the best scanner to use" and "what DPI should I scan at" seem obviously out of scope, "should I save as TIFF or JPG" less so, "how should I organize and describe the files once I've scanned them" clearly on-topic.

My experience on archives listservs and discussion groups is that basic scanning guidelines are a perennial topic of inquiry from small LAM institutions trying to bootstrap a digitization program. Perhaps the answer is to link to some good introductory digital imaging resources from the FAQ, with a newbie-friendly explanation of why these types of questions are out of scope.

A distinction that might be useful: is the question about creating an authentic representation of an analog object ("will this file format capture all the details of my book/manuscript/photograph?"), or is it about ensuring the sustainability of the resulting digital object ("will my institution be able to read this file format in ten years?"). This might be a workable guideline for determining relevance.

Comments

• Paul Wheatley: I like this approach. I don't want to avoid helping people out with "which scanner, what DPI" type questions, but I do want to avoid having the site swamped with low tech digitisation issues, which is likely to put off a lot of digital preservation experts, and dilute the quality of this forum. Having a bit of detail and some good references in the FAQ, could, as you suggest, be a win-win.
• Donald.McLean: It has been my experience that many first time posters do not read the FAQ before asking questions. The important thing to remember about this is that a snarky, elitist comment about how the question is clearly off topic and they should have read the FAQ first WILL NOT HELP. As long as everyone is clear on the fact that even people whose only shortcoming is failing to read the FAQ first should still be treated in a friendly and welcoming manner, I think this is a policy I can get behind.
• Nicholas Webb: Cosign Donald's comment. In my experience consulting with local historical societies, many longtime archivists/curators of analog material have a vague idea that digitization and digital preservation are things they need to be doing, but if they aren't technologists or at least power users themselves, getting them up to speed can require a lot of handholding on basic technical topics. This stack isn't the place for those questions, but given that it's likely to get them, we should encourage a culture of politeness/friendliness in referring them elsewhere.

Answer by Ross Spencer

Yes.

The phrasing is key however and it should be focused on the results of digitization - the formats, the metadata associated with it, the storage etc. and beyond that, errors that appear in the data stream.

I would be interested in seeing questions about the digitization process, Sean Martin's work at the BL for example comes to mind re: the difference lenses make between scans of the same document in the same position etc. Also DPI questions. These are questions that affect the value of the digital data being created and impact whether or not this data becomes an asset that we do want to preserve.

It might not start as a digital preservation question but it will become one. Best to start with preservation in mind. Some records are not kept after digitization - that makes these new digital records quasi-born digital.

The compromise would be to discuss just the results of the digitization process.

Incidentally in my mind it is clear the OCR question referenced in the OP is NOT a digital preservation question. It belongs on the libraries Stack Exchange. My rationale is simple - OCR can be done over and over again. The OCR output is data associated with the digital object we're interested in, almost a second-layer of metadata. The error is an artifact of the current OCR process that doesn't impact preservation.

A few naive OCR DP questions (not having handled OCR much in the past) might be:

1. What standard formats are there for storing OCR data?\
2. What long term metrics should I maintain about my OCR that will be useful to future users of the data?\
3. How do I tie my OCR data to my digital images for long term access?

Those may not be perfect so happy to discuss in the comments below, I'm eager to find the boundary here too.

I think the site has to be carefully moderated. It is a Q&A site so ensuring the same digitization question isn't repeated over and over will be important and moderators, maybe via the discussion forum should have a clear FAQ about what constitutes an acceptable digitization question before closing anything.

Hope that helps.

Comments

Answer by Bill Lefurgy

Let's think about it in narrow terms as creation of digital content from a preservation perspective. Questions about optimal design of a prospective website in connection with formats, metadata and architecture presumably would be in scope.

Someone undertaking digitization could ask the same kinds of questions, as long as they were focused on how best to create preservable output. The same could be true of any digital creation activity--photographs, videos, data sets and so on. Anything that doesn't relate to creating content for clear-cut preservation purposes (equipment, throughput, QC) would be out of scope.

Comments