E-mail List Archives
Thread: How do you properly code an image viewer?
Number of posts in this thread: 2 (In chronological order)
From: Jeremy Echols
Date: Fri, Aug 30 2019 11:52AM
Subject: How do you properly code an image viewer?
No previous message | Next message →
I'm thinking of a site like Historic Oregon Newspapers where content is presented in a way that is fundamentally not usable by a blind person (e.g., https://oregonnews.uoregon.edu/lccn/sn84022644/1878-08-30/ed-1/seq-1/)
We provide an OCR capture, which is unfortunately not great, but the best we can do without hiring thousands of people to write transcripts of each page. But the viewer for those pages is something that cannot be used by a blind person, period. How do we handle that? Open Seadragon uses a canvas element, and that viewer can't be described textually (imagine trying to describe one part of a zoomed-in page). I'm just at a loss for how best to handle a site like this.
From: Andrews, David B (DEED)
Date: Fri, Aug 30 2019 1:13PM
Subject: Re: How do you properly code an image viewer?
← Previous message | No next message
I only looked at one page -- but the OCR was terrible. There were only random words recognized.
While it doesn't necessarily meet WCAG, I would say, don't worry about the viewer, as far as blind people go if it only works with graphical images. It doesn't matter!
Dave
David Andrews | Chief Technology Officer
Minnesota Department of Employment and Economic Development
State Services for the Blind
2200 University Ave West, Suite 240, St. Paul MN 55114
Direct: 651-539-2294
Web | Twitter | Facebook
-----Original Message-----
From: WebAIM-Forum [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Jeremy Echols
Sent: Friday, August 30, 2019 12:53 PM
To: WebAIM Discussion List
Subject: [WebAIM] How do you properly code an image viewer?
I'm thinking of a site like Historic Oregon Newspapers where content is presented in a way that is fundamentally not usable by a blind person (e.g., https://gcc01.safelinks.protection.outlook.com/?url=https%3A%2F%2Foregonnews.uoregon.edu%2Flccn%2Fsn84022644%2F1878-08-30%2Fed-1%2Fseq-1%2F&data=02%7C01%7Cdavid.b.andrews%40state.mn.us%7C2562f90b04104f74e5c608d72d72e8ca%7Ceb14b04624c445198f26b89c2159828c%7C0%7C0%7C637027843860613303&sdata=jFwgdd2riZdTnw%2BZwNrOVSjx5MetgwhKY0tH97ii1CM%3D&reserved=0)
We provide an OCR capture, which is unfortunately not great, but the best we can do without hiring thousands of people to write transcripts of each page. But the viewer for those pages is something that cannot be used by a blind person, period. How do we handle that? Open Seadragon uses a canvas element, and that viewer can't be described textually (imagine trying to describe one part of a zoomed-in page). I'm just at a loss for how best to handle a site like this.