universalviewer / user-stories Goto Github PK
View Code? Open in Web Editor NEWCommunity repository for documenting stories and use cases related to uses of the Universal Viewer
Community repository for documenting stories and use cases related to uses of the Universal Viewer
In their implementation of IIIF for the Polonsky pre-1200 project, the BnF have grouped Links under one heading, which might be a neat way of grouping the IIIF manifest address, catalogue link and links to Learning pages (e.g. https://demogallica.bnf.fr/pl/ark:/12148/btv1b105395234/f8?lang=en)
Suggested items include:
User searching extracted text in a multi-page digital object
To be taken to the first page that a search term exists on
When a term first occurs somewhere within the multi-page object (but not on the current page), the viewing pane won't leave the user stuck on the current page and require them to navigate to the first page that has a search result on it.
(Having trouble expressing this cleanly in “As a…” syntax, but anyway)
The UV loads a manifest that contains many distinct recordings, modelled as Ranges. There is a web page for each recording, so that the same manifest provides the content for multiple pages. Each page initialises the UV with a range to start on. One manifest provides the content for many web pages, because each web page is about the Range.
That part is fairly clear; the UV starts playing the Range provided by the initialisation parameter, and provides navigation over that range - the manifest might have hours of recordings, but the navigation offered to the user is just the initialisation range:
The use can pop put the UV to full mode, where they will see the rest of the manifest. This is the UX issue - what extent of time does the scrub bar now represent? In the full view, you have control over the clock for the whole Canvas/manifest, which means the time extent offered as the whole of the scrub bar in minimode is a small part of the scrub bar when popped out.
BL proposal is use of visual distinction (here green colour, but needs to be accessible) to make it clear that the mini scrub bar is the same time extent as the small part of the maxi scrub bar.
Does the UV need to support (via config) the difference between minimode “start at this range and only offer navigation within this range” and minimode “start at this range but allow the user to use next and previous to navigate the range structure (for example, when playing an album). Compare UIs that are track focussed like iTunes and Spotify.
The issue here is quality - print works, but doesn't behave as users might reasonably expect.
Printing from the browser currently prints the page, rather than the image, so the actual image size is much smaller than a user would expect.
Image controls (zoom, overview box) leave blank regions on the page.
In the tab "share" of the UV, you see a box with the url to the current viewer, and below an icon that links to the manifest itself.
But unexperienced users click on that icon, and are surprised to see a popup for download, either because it is Internet Explorer, or the content-type "application/ld+json" is not recognised as json.
Wouldn't it be more clear to put a textbox with the manifest url, that you can copy? Not everyone has the reflex to do right-click-copy ..
we're trying to learn our librarians about IIIF. Things like "get the manifest url, and paste it in another iiif viewer"
See issue:
Can controls be grouped more compactly so a) it's easier for the user to see all the functions available, b) more room is available for displaying item content
Current best example: https://sok.riksarkivet.se/bildvisning/R0000004_00001?beta=true#?c=0&m=0&s=0&cv=0&z=-331.382%2C-766.4352%2C7684.764%2C6533.8704
... developer of applications that use the UV to display dynamically generated manifests
... to be able to pass a JSON blob for the manifest to the UV and have the UV render the manifest as if it had loaded it itself
... I can build applications that don't rely on a static hosted manifest
Repository manager hosting PDF content in my repository
To be able to create manifest files for PDF content that support 2 pages side by side in UV
my patrons can have a book-view experience browsing the content of my PDF's
This may require updates to both the search box and the search service (regardless of other issue re indexing multilingual text).
With many thanks to Stu Snydman and Jennifer Vine for sharing.
These are for discussion at UVCON, and in the UX channel on UV Slack, I will leave further elaboration of their content to Stu and Jennifer.
See issue: UniversalViewer/universalviewer#594
scrollToZoomEnabled
parameter to SeadragonCenterPanel
Downloaded image file names (from the BL) currently contain part of the ARK but no other identifying information is available re institution or item.
Previous usability work found that the settings cog is ignored by most people, the chevrons which indicate open/close panels aren't noticed by users, the meanings of the embed and share icons aren't recognised, the 'More information' panel is missed by some and might benefit from a standard icon ('i'?)
Download options should be consistent across single, double page views
User should be able to: download entire item; download selection of pages; download current view (e.g. zoomed in view); download whole image.
e.g. if rights statements says an item can't be printed or embedded, does the viewer respect that?
Currently 'more info' goes to a github page, which is ok but not brilliant.
Builder of manifests that feature rotated images
To be able to see my manifests in the UV
Source: Jenna Schoen:
... I'm working on making a manifest from scratch, and having trouble rotating my images. I've tried rotating them directly through the URL (which worked on the URL but didn't load on Universal Viewer), with a selector (which didn't change the image on Universal Viewer), and via the CSS route (which didn't load on Universal Viewer). Do you have any suggestions? Thanks a bunch in advance!!
Researcher viewing a digitized PDF document in the Universal Viewer
to search the full text of the document (if available in the pdf)
I can find a paragraph/article of interest.
At Georgetown, we have implemented a PDF document viewer using a javascript based package called FlexPaper. See the "bookview" link at https://repository.library.georgetown.edu/handle/10822/1047689 . It will take a minute for the PDF to download. Once it has downloaded, you will see that full text searching is enabled.
Eventually, we would like to replace our FlexPaper viewer with the Universal Viewer. In order to accomplish this, it would be particularly useful to have a within-document search available in a PDF.
I suspect that this solution would require us to convert the document full text into an annotation.
Researcher viewing a digitized image of a record
to be able to manipulate the colours of the image I am viewing
it is easier to view handwriting and text
This introduces a new display layout/mode for the UV, which is significant:
(source: BL)
Questions:
what UI toggles between the two, and how does that relate to full screen, and/or toggling to the regular UV mode, which may or may not be F11-style full screen?
From Fluent tests, April 2019:
The viewer is not currently keyboard accessible.
Keyboard arrows can only be used as a way of navigating the item when the item has been clicked on. The visually impaired user in particular who defaulted to using the keyboard arrows was confused when this did not work due to the item needing to be clicked on first. Also, the thumbnails or the actual item cannot be selected. As a results, a keyboard-only user cannot navigate to different screenshots and cannot pan around the image with unless they select the image with the mouse.
Consider the keyboard accessibility of the viewer and make it keyboard accessible (e.g. by clearly showing where the focus is and by making the thumbnails and item accessible by tabbing)
Does it work as expected in target environments? Is the link to the BL version easy to access? Intersection with responsive view?
e.g. BL UV - you can set the language to Welsh but it doesn't change (as the translation files are missing?)
collection.json?#m=6
collection.json?m=aaaa-bbbb-123
which would look up the index of the manifestComment from a user: Will there be a plain text window? Like there is in the JISC databases like Historical Texts? These can be useful for people not used to reading early printed letter-press.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.