Present: Lee Berry, Ron Brashear, Michelle DiMeo, Anna Headley, Hillary Kativa, Stephanie Lampkin, Cat Lu, Erin McLeary, Sarah Newhouse, Patrick Shea, Amanda Shields,
Absent: David Caruso, Jim Voelkel
Updates on Hydra development from the Digital Collections team
- Building technical server architecture; not going to be in house – in the cloud with Amazon Web Services (AWS); learning curve with AWS
- Site visits at Jenkins Law Library, where they use AWS; figuring out size and pricing of server is clunky and difficult on AWS
- Also figuring out on-site development environments: sandbox for development, staging server, production server (which will be public facing); also needed a place to store code and worked with lawyer on licensing for public facing code; needed to automate deployment process
- Basics now working, some ingest testing, trying to crash and figure out file limits, scaling up
- Imaging equip installed, can give demos anytime - ask Michelle
- Wrote imaging guidelines, MD, HK, CL, SN and JV have all tried it; goal to ingest real images during testing, will later start with Beckman project for testing batch uploads
- New scanner slightly intimidating at first, but imaging quality is great and support staff at Digital Transitions is good
User Testing of Alpha Sites and Open Source Software
- Idea of Alpha testing (vs beta testing), usually beta with vendors, alpha before, may crash and we want you to crash it now - let us know!
- Software development with open source, get things in several iterations, used to be waterfall type development but users would find it’s not actually what they wanted, or miscommunication, or changed mind
- In this testing, it’s okay to break things or change your mind, but tell us workflow issues or error messages
- In-house open source development means no other conflicts or different priorities, but we might have to prioritize requests for Anna since we have only one developer
- Report bugs to Michelle for now: please give details on exactly what was happening, time and date, and also screenshot if can get one
- Metadata template: get in touch with Cat; Metadata guidelines have been circulated – please give feedback
- If you prefer something specific for collections, we’d like to incorporate it
Live Demo of new metadata ingest template and discussion of Alpha testing
- http://staging.hydra.chemheritage.org/ is landing page for CHF's Hydra test site, all temporary styling that will change and can be edited later on, originally made for Penn State University (Sufia aka ScholarSphere)
- IMPORTANT to remember: nothing is saving and no passwords are encrypted! Don’t do real data for first 50 objects: find a few objects to test out and give us feedback
- One-page quick guide on uploads and cataloging was circulated – check email!
- Site is locked down to CHF IPs for now, so can't try it at home
- 1 gig limit about 15-25 high-res TIFF images; uploads 6 at a time and moves down the file queue
- Dashboard page -> upload of files > selected folder > ingest
- Right now can only do single ingest, do single images or files for now right
- You will see two templates, first is batch input template, then individual item template. We want to know how much is batch metadata and how much is individual, so let us know
- AH: ‘upload set’ is new name for what was previously known as ‘batch load’; more batch functionality in the works
- The ‘upload set’ form is a one-time thing during ingest, right now only title applies on a file-by-file basis; everything else is bulk (button to show all fields)
- On public display, will be working to eliminate empty spaces for incomplete metadata when item is shown
- Can only edit files that are yours; "All of Sufia" searches everything that was uploaded to Hydra
- Don’t use Collection icons - not populated yet
- External ID: Important later for ingest to match records in other systems (Need OPAC staff access if you want the Bib Number.) CL will request access.
- Maker: what we’ve decided on for now, can easily add more; uses VIAF auto-complete: names are here by “relation” (no relevance ranking in API)
- Date: open so you can add lots of info; add single year to Start field; NOTE field for additional info (ex. Roman Numerals)
- Place: uses FAST; if entry with "USE" is selected it will fill in the preferred authority
- Genre: add items in here so we can see what you want to use (will add config file later)
- Medium: most important to museum
- Extent: Free field, broad to capture everything
- Language: (will later have controlled vocab)
- Subject: uses FAST; AS: alchemy in art or just alchemy or both? CL: Choose two or three, leaving it to curator’s discretion
- Series arrangement and physical container for Archives
- Related URL: hoping to generate links down the road, but haven’t worked on that yet
- Archival: Can I add an individual item to a collection? Yes – select items and “Add to Collection” (ex. Beckman)
- Workflow: In the future how much info will pull from other CMSs (PastPerfect and Sierra)? Not positive about API for Sierra, but PastPerfect might be a manual process. Will explore Sierra API before PastPerfect. Could be that a new version of PP or a different museum CMS would make automatic data pulls easier, but to be discussed. Easier to clean-up data via batch functionality in Hydra than in PP. In future, maybe don't catalog all information in Museum CMS if not necessary (ex. maybe subject info done in Hydra instead. TBD.)
- PS: Concern that partial collections are online and gives users wrong impression. Will people ignore the OPAC? Can we link back? MD: We can add links to OPAC and/or Finding Aid, add disclaimer at beginning re: amount of collection digitized, but this is a common problem for digital libraries. Should also bring in different audiences via different access point. HK: Could possibly add link to “View this collection in the online catalog”
- General Discussion regarding linking archival and museum collections with the same accession number; how we define collection (difference between collection or set?); CHF defines a collection as mixed media (archives and museum). Ex.: Perkins Elmer: same accession number for museum and archives, mixed collection.
- Discussion regarding “Relators” functionality in future? Can it have a Related Items field? Curated list of related items? Currently only able to do Related URL
Vote on a branding name for our digital collection
- Digital Collections Team met twice with Communications & Development to brainstorm a short-list of naming possibilities. ChemVault was our original first choice, but it was under trademark.
- Finalists: SciVault; Digital Arcana in Science and History (DASH), or Hydrogen/Hydragen. Also possible to not have a branding name and just go with CHF Digital Collections.
- Most people could get behind SciVault or Hydrogen/Hydragen. SciVault is good for branding, but might get misunderstood when spoken and vault might sound too severe. Hydrogen would get too many hits and be tough to find, but Hydragen might be too much of an "inside joke" since users don't know the collection is built on Hydra. For either, we could add a tag-line regarding the history of science.
- DASH was liked because it had both Science and History in the title, but DASH as an acronym wasn't liked. Searches for DASH Collection yield the Kardashians' Collection (also called Dash!) We agreed that we'd be open to another acronym if anyone has one!
- While some preferred its simplicity, CHF Digital Collections might be problematic if the name CHF changes in the future. Might also be less inviting to outsiders who might deposit with us in the IFR.
- Agreed to revisit conversation later. MD will also report back on voting for the name within the Hydra community at Hydra Connect conference next month.
Rights statements as related to the metadata template in Hydra
- HK circulated current R&R procedure and agreement prior to meeting. Should be discussed in more detail later.
- Creative Commons http://creativecommons.org/ : Standard way to license works so that rights are human-readable and machine-readable, with the legalese behind to back it. This is built into Hydra, along with All Rights Reserved and Public Domain.
- New task-force between Europeana and DPLA, in consultation with Creative Commons, recommending new guidelines. White paper still in draft: https://docs.google.com/document/d/1x10JsIfi8Y74pgJJEAqMtyO5iYp0p6DO5DrOZK-5umY/edit#heading=h.c2yvhrd7pznp Highlights: Aggregation tool found over 86,000 rights statements in use across collections and they tried to distill it down to 11 options (ranging from in copyright, public domain, to rights unkown or orphan works). Another discovery is that faithful reproductions of digital surrogates should have the same rights as physical items (can no longer claim copyright on a rare book image you scanned – not enough creativity involved.) May be slightly different for 3D objects in museum.
- Group agreed they'd like to use Europeana/DPLA model in the future. MD to give updates on progress. These will be machine and human readable.
- Group would to add a free-text "Credit" line field to the metadata template. Future discussion must include an approved tag-line for all collections. RB and EM agreed to something like "Courtesy of CHF Collections" as blanket for all collections. Also, "Credit line" versus "citation" being a bit different.
- Website should also include an argument for Fair Use to protect ourselves and language that suggests we only "believe" the work is in copyright if we're not absolutely certain. Can have a note inviting users to contact us to take it down if they believe they are the copyright holder.
- CHF needs documentation on risk assessment for copyright and what constitutes "due diligence" here. Staff would be more willing to priorities access if they knew they had support from upper administration and documented policy.
- Future discussions need to involve Communications staff and Wikipedia since they may be applying a Creative Commons license to items for which we don't hold the copyright. Also need standardization across online access points and need to revisit agreements with places like Science Photo Library as we move to more open access.