Request to develop IIIF viewer embed in archival descriptions

[Alexandra Healey]

Hi there,

I know IIIF has been discussed a few times but from conversations outside the forum us here at Newcastle University have become aware of a few other organisations that are interested in sponsoring some specific development.

All of us already have the infrastructure to generate manifests elsewhere, and we would like a way to embed these manifests within AtoM, so that adding manifests to archival descriptions would render a viewer within the record in which the manifest appears. So for example, if a manifest is added to an item or series level record within a larger hierarchy, a viewer appears within that item or series level description.

We are aware of some institutions who've created proof of concept using the Universal Viewer, and we've been exploring Mirador here, but we'd be interested in sponsoring Artefactual to develop something within AtoM, rather than duplicating effort to develop this just for local instances.

We'd like to get a quote for this work, and also whether anyone else is looking for this development?

Already on board are:

Newcastle University (UK)

University of Toronto (Canada)

Queen's University (Canada)

Thanks

-Alexandra

[Dean Seeman]

Thanks Alexandra. University of Victoria Libraries (B.C., Canada) is also interested in this development.

Dean Seeman

Head, Collection Management Services

University of Victoria Libraries

[Vicky Phillips]

Thanks Alexandra. The National Library of Wales would also be interested in this development.  We  managed to do a proof of concept with the Universal Viewer a few years back now but have not been able to finalise this and get into production yet.  It would be great as Alexandra says to have this as functionality within AtoM itself rather than several of us developing something in-house.  I'll see if the programmer who worked on this is able to share what he did where to get the Universal Viewer to embed within AtoM. 

Thanks

Vicky

[Jonathan]

Thanks Alexandra. The University of Brussels (Université Libre de Bruxelles) would also be interested in this development.

Jonathan

[Brittny Lapierre]

Thanks Alexandra. The Canadian Research Knowledge Network would be interested in this development. IIIF backs our entire infrastructure development, so having AtoM compatibility would make the need of true archival-based access much easier for us to reach. Getting more CRKN names into the forum is great!

Brittny Lapierre

Application Developer

CRKN-RCDR

[Alex Healey]

Hi everyone,

I've chased this up with Artefactual via email, who have confirmed they're happy for me to share their response here.

They will not provide a quote for this work:

we've been moving away from the 'bounty' model of software development and ever-closer to a place where enhancements to AtoM are borne out of requests by our customers - those institutions who pay us annually to host AtoM or support their on-premise installation of AtoM. We are focusing on undertaking development work that our customers are asking for AND, in our view, makes good sense to offer as part of our product. 

...

IIIF functionality in AtoM is a great idea, by the way, and I’m not saying we won’t develop that functionality in the future. It’s not never, it’s more like just not now and not in this way.

Thanks everyone,

-Alex

[Dan Field]

Hi All.

We did some very preliminary tests of this back in Atom 2.2 I think. Quite simply we followed the now-removed UV v2  documentation at https://github.com/UniversalViewer/universalviewer/wiki/Embedding (possible newer working link for v4 here https://codesandbox.io/s/uv-embed-example-wyus9 ) and embedded this within the relevant view. Thie code was something like this to pull our PID (Fedora ID used as IIIF identifier) from the location field. In this example theres a little extra code to strip it out of the handle.net URL format and just get the integer value as that was all we needed to build the IIIF manifest URL.

<!-- Begin Universal Viewer integration -->

<?php $location = $resource->getLocationOfCopies();

preg_match('/hdl.handle.net\/10107\/[0-9]+/' ,$location, $matches);

if (0 < count($matches)):

  $urlparts = parse_url($matches[0]);

  $pid = basename($urlparts['path']);

  if ($sf_user->getCulture() == 'cy') {

    $viewer = 'syllwr.llyfrgell.cymru';

  }

  if ( $sf_user->getCulture() == 'en') {

    $viewer = 'viewer.library.wales';

  } 

?>

And then inject into the JS

<?php $pid = $resource->digitalObjects[0]->getName(); ?>

  <div class="uv" data-locale="en-GB:English (GB),cy-GB:Cymraeg"

   data-uri="https://damsssl.llgc.org.uk/iiif/2.0/<?php echo $pid; ?>/manifest.json" data-collectionindex="0"

   data-manifestindex="0" data-sequenceindex="0" data-canvasindex="0"

   data-rotation="0" style="width:670px; height:777px; padding: 0px; background-color: #000">

</div>

<script type="text/javascript" id="embedUV" src="https://viewer.library.wales/build/lib/embed.js"></script>

<script type="text/javascript">/* wordpress fix */</script>

This work was purely a proof of concept and was never put into production. It does not attempt to follow any Atom coding standards or best practices but if it helps in any way, then hopefully it's been useful. Do not just inject the code into your own install. Its very NLW specific and based on a very old version of Atom. It will not work for you without major adaptation.

[Kelli Babcock]

Hello,

Is it possible to get more clarity from Artefactual on this point: "we've been moving away from the 'bounty' model of software development and those institutions who pay us annually to host AtoM or support their on-premise installation of AtoM. We are focusing on undertaking development work that our customers are asking for AND, in our view, makes good sense to offer as part of our product."

I'm wondering if there were AtoM customers who host their instance with Artefactual and a large group of those folks desired IIIF in AtoM v. 2, could a sponsored IIIF functionality project become a possibility for Artefactual to consider from both its customers and institutions who self-host in the next 1-3 years?

Thanks to Alexandra for bringing this up for discussion and Dan for sharing the NLW proof of concept. 

The ideal would be for some type of IIIF functionality to be in a public release that could benefit all AtoM institutions. For example, so many archives in Canada, and internationally, use AtoM but also use the Internet Archive, which is IIIF compliant (helpful docs on this: https://archive.readme.io/reference/ia-iiif-faqs + https://training.iiif.io/iiif-online-workshop/day-two/image-servers/iiif-hosting-ia.html). Between these two platforms, there is a really big opportunity to enhance AtoM v. 2 so archives can begin adding standardized image data to AtoM descriptions through IIIF manifests & improve pathways from descriptions to IIIF-compliant digital images for end users in AtoM v. 2.

Take care,

Kelli

Kelli Babcock | she / her

Digital Initiatives Librarian 
Information Technology Services

University of Toronto Libraries
130 St. George Street | Toronto, Ontario 

Collections U of T
Discover Archives

[Jennifer Roberts]

Hi Kelli and all,

I'll take a stab at providing more context from Artefactual's perspective since I have said versions of this to various folks (clients of ours and not) over the last year or two. This point: "we've been moving away from the 'bounty' model of software development"

Means: we are not developers-for-hire. We have lots of developers who work for us and we do lots of developing and we sometimes get paid for specifically requested development work but we rarely do so anymore using the bounty-model. I realise that we have done so in the past (this is the last example I can think of - with SFU in, planned in 2020) but it isn't a service we offer.

You might ask why - and this is harder to answer. Or it requires a longer answer, at least. I'll try to sum up some of why we are doing things the way we are doing them. Bounty-model development is simpler when something is new and it's harder when software is mature. We are trying to make Artefactual a longterm, sustainable company that can continue to put free and open source software out into the world. And I mean sustainable for the software, for the company and also for the folks who work for us! This means we we can't take on development-on-demand. It creates a large maintenance burden and technical debt, among other problems. This burden is one we are dealing with super-proactively right now by having hired a team of developers whose focus is the stability, usability and security of AtoM and Archivematica for the longterm. I think of software maintenance as a lot like other archival work: highly skilled yet often undervalued and invisible. Maintaining something helps knowledge persist across versions or organisations. It's caring work. (Stay tuned to our News page for some features about the work of this team in the coming weeks).

No one at Artefactual thinks IIIF functionality is a bad idea. We aren't against it. We just aren't able to commit to it because we have other things we are trying to take care of right now. I love what my colleague Kelly said about IIIF functionality: not now and not in this way. We are choosing to focus on customer needs first and that will ultimately benefit the larger AtoM community too. Trust me when I say IIIF is on our radar and we know there are people out there who want it. But it's not a matter of a customer of ours saying "I want IIIF" and then we will build it. It's more complex than that. And yes we are trying to listen to our customers first. We can't be everything to everybody. We can't DOALLTHETHINGS (*insert Hyperbole and a Half meme here)

Lastly, I would return to the AtoM Foundation as the ideal place from which community needs can be expressed to Artefactual.

I wanted to keep this short and I haven't really succeeded. We do strive to be transparent about what we do and how we do it and I am sorry if some of what I am explaining isn't giving the clarity you desire. Conversations in the user forum are great for getting info out to folks but they aren't the greatest at nuance or back and forth.  We take our roles as stewards of AtoM seriously - our responsibility to the project and the people who use it.  Let us know if I've missed anything.

--

Kind regards,

Jenn Roberts (she/her)

Systems Archivist, Business Development Team

Artefactual Systems Inc.

@archivematica & @accesstomemory

[Kelli Babcock]

Hi Jenn,

Thanks as always for the thoughtful response. I can relate to "[m]aintaining something helps knowledge persist across versions or organisations. It's caring work" and know the work involved in maintenance is not insignificant.

I would push back on a discussion about multi-institutional sponsorship of IIIF, and past discussions through the AtoM Foundation on WCAG AA compliance, as bounty model requests. IIIF and WCAG AA compliance are not development-on-demand asks - they are major initiatives that would enhance AtoM's ability to support basic compliance that all archives, internationally, are now confronting. 

As a design principle for self-hosted institutions, it is not strategic (or financially feasible for many) to do this compliance work individually in our own siloed institutional AtoM instances. We really rely on standards compliance relevant to the archives community to be baked into the AtoM v. 2 public release. The AtoM Foundation is a very valuable initiative but there is no timeline for AtoM v. 3. We only have AtoM v. 2 to work with right now and it is difficult to see a pathway towards compliance in v. 2 at this point. I hear the "not right now" but I hope we can still all continue openly discussing possibilities and ways forward as a community.

Take care,
Kelli

[Jennifer Roberts]

Hi Kelli,

Just a quick reply to say that I totally agree and absolutely think we should continue to openly discuss possibilities and ideas for ways forward - both here and through other channels.  I hope it didn't sound like I wanted to shut that down.  In my reply I was trying to explain how Artefactual is approaching things right now and to be honest about our capacities and current focus.  And part of that was to be clear that we can't provide a quote for that bounty work right now.  We are stronger as a community with many people contributing. I appreciate your work and support of AtoM, too, in case that wasn't clear.

Best,

Jenn