W3C

Site Navigation

Nearby

Results of Questionnaire ACT TF - Rule Review: Image filename is accessible name for image

The results of this questionnaire are available to anybody. In addition, answers are sent to the following email addresses: team-wcag-act-surveys@w3.org,maryjom@us.ibm.com,wilco.fiers@deque.com

This questionnaire was open from 2020-03-25 to 2020-04-09.

6 answers have been received.

Jump to results for question:

  1. Instructions
  2. Consistency with ACT Rules Format
  3. Rule assumptions
  4. Implementation data
  5. Consistent with WCAG
  6. Remaining open issues
  7. Other questions or concerns
  8. Readiness for publishing

1. Instructions

Review the rule Image filename is accessible name for image and answer the questions in this survey.

If there are issues with the rule, you may either open an issue in GitHub or provide details in the entry fields for the applicable question.

Summary

ChoiceAll responders
Results

Details

Responder Instructions
Wilco Fiers
Jey Nandakumar
Charu Pandhi
Kathy Eng
Mary Jo Mueller
Trevor Bostic

2. Consistency with ACT Rules Format

Does the rule follow the ACT Rules Format 1.0?

Summary

ChoiceAll responders
Results
Yes 6
No. I have opened an issue in GitHub or have documented my comments below.
I don't know. My questions are documented below.

Details

Responder Consistency with ACT Rules FormatComments
Wilco Fiers Yes
Jey Nandakumar Yes
Charu Pandhi Yes
Kathy Eng Yes
Mary Jo Mueller Yes
Trevor Bostic Yes

3. Rule assumptions

Are the assumptions acceptable?

Summary

ChoiceAll responders
Results
Yes
No. I have opened an issue in GitHub or have documented my comments below. 2
I don't know. My questions are documented below. 4

Details

Responder Rule assumptionsComments
Wilco Fiers I don't know. My questions are documented below. The assumption is a little confusing to me. Would be good to make it a little clearer.
Jey Nandakumar No. I have opened an issue in GitHub or have documented my comments below. https://github.com/act-rules/act-rules.github.io/issues?utf8=%E2%9C%93&q=is%3Aissue+is%3Aopen+9eb3f6+
Charu Pandhi I don't know. My questions are documented below. Not sure i understand why the language of the page is important to this rule
Kathy Eng I don't know. My questions are documented below. Should the assumption include that the language of the filename matches the language of the page, not just that it can be determined?

"Sufficiently understood" meaning it is a real word? Understood by whom?
Mary Jo Mueller No. I have opened an issue in GitHub or have documented my comments below. Agree with others. Why is the language important for this test? The test cases don't seem to indicate that language is important to the test. Also...when speaking of language - programmatic language? Human language?
Trevor Bostic I don't know. My questions are documented below. I agree with Kathy that the 'sufficiently understood' leaves the assumption open ended. I would prefer to remove it and have the assumption state only "The rule assumes that the language of each test target can be correctly determined (either programatically or by analyzing the content)." This definition feels more concrete to me.Otherwise, we should elaborate on the meaning of "sufficiently understood".

4. Implementation data

Is the implementation data correct?

Summary

ChoiceAll responders
Results
Yes 3
No. I have opened an issue in GitHub or have documented my comments below.
I don't know. My questions or comments are documented below. 3

Details

Responder Implementation dataComments
Wilco Fiers Yes
Jey Nandakumar Yes
Charu Pandhi I don't know. My questions or comments are documented below. How does the rule determine which file name is OK and which is not? Some filename makes sense, some definitely not. What about extensions in filenames? or any other nontext characters or numbers in filename.
Kathy Eng I don't know. My questions or comments are documented below. I think another failure example may be needed where "graph" is the image's filename and the accessible name, but it does not provide the image's equivalent purpose.

Example 1 - "https://www.w3.org/WAI/demos/bad/img/w3c" is the link, but isn't the image's filename is w3c.png?
Mary Jo Mueller I don't know. My questions or comments are documented below. I have same concerns as Charu on this. "Pass" of test cases should actually need more test. I think it's ok for a test to check that the file name used as the alt matches without loss of info, but should result in further test needed to check it is sufficient. I have often seen images that use their file name as alt but the name is completely meaningless - numbers and such. I don't think we should pass such files as meeting non-text alternatives requirements.
Trevor Bostic Yes

5. Consistent with WCAG

Is the rule consistent with existing WCAG documents?

Summary

ChoiceAll responders
Results
Yes 4
No. I have opened an issue in GitHub or have documented my comments below. 1
I don't know. My questions or comments are documented below. 1

Details

Responder Consistent with WCAGComments
Wilco Fiers Yes
Jey Nandakumar No. I have opened an issue in GitHub or have documented my comments below.
Charu Pandhi Yes
Kathy Eng Yes
Mary Jo Mueller I don't know. My questions or comments are documented below. On quick glance through the techniques, I haven't seen where a file name is considered suffficient. However, I do understand that there are some cases where it could be - like a UI that is showing a list of files and has an image to represent each of them (like in Box). Or an app that shows thumbnails of pictures someone uploads.
Trevor Bostic Yes

6. Remaining open issues

Are there any remaining open issues for this rule that were opened prior to this review?

Summary

ChoiceAll responders
Results
Yes there are open issues that need to be resolved. I have listed them below. 3
Yes, there are open issues but they don't need to be resolved for the rule to be published. 1
No, there are no open issues. 2

Details

Responder Remaining open issuesComments
Wilco Fiers Yes, there are open issues but they don't need to be resolved for the rule to be published. https://github.com/act-rules/act-rules.github.io/issues/1244 Proposal to extend the rule to include srcset. Not a problem with the rule though.
Jey Nandakumar Yes there are open issues that need to be resolved. I have listed them below.
Charu Pandhi No, there are no open issues.
Kathy Eng Yes there are open issues that need to be resolved. I have listed them below. "Filename is valid accessible name" (9eb3f6) should limit its applicability #465" (https://github.com/act-rules/act-rules.github.io/issues/465).
Mary Jo Mueller Yes there are open issues that need to be resolved. I have listed them below. David Macdonald's issue #1246, #1244 from Wilco, and #465 from Kasper should be resolved. These all reflect my concerns for this rule.
Trevor Bostic No, there are no open issues.

7. Other questions or concerns

Do you have any further questions or concerns about this rule?

Summary

ChoiceAll responders
Results
Yes, I have questions or concerns, described below. 3
No, I have no further questions or concerns. 3

Details

Responder Other questions or concernsComments
Wilco Fiers No, I have no further questions or concerns.
Jey Nandakumar Yes, I have questions or concerns, described below.
Charu Pandhi No, I have no further questions or concerns.
Kathy Eng Yes, I have questions or concerns, described below. I'm not clear this rule is necessary. Wouldn't a rule for "Image has an accessible name that describes its purpose" broadly cover what's tested here?

A filename with extension for accessible name would pass only when the image is used to represent the file and not the image displayed. Perhaps the rule should be "File extensions are included in accessible name for files available for download."
Mary Jo Mueller Yes, I have questions or concerns, described below. I think the expectation is unfounded for this rule. It requires manual test to really know if the name of the file serves an equivalent purpose. A file name rarely is sufficient in my experience. I like Kathy's suggestion here to define the applicability of the scope of the rule so it isn't misconstrued to be more broadly applicable.
Trevor Bostic No, I have no further questions or concerns.

8. Readiness for publishing

Do you think this rule is ready to be published?

Summary

ChoiceAll responders
Results
Yes, it is ready to publish as-is. 2
Yes, it is ready to publish with the following changes. 1
No, it is not ready to publish and the reason is documented below. 3

Details

Responder Readiness for publishingComments
Wilco Fiers Yes, it is ready to publish with the following changes. We should clean up the assumption.
Jey Nandakumar No, it is not ready to publish and the reason is documented below.
Charu Pandhi Yes, it is ready to publish as-is.
Kathy Eng No, it is not ready to publish and the reason is documented below.
Mary Jo Mueller No, it is not ready to publish and the reason is documented below. I think some work needs to be done to scope the rule's applicability and when a use of a filename is appropriate and when it usually isn't.
Trevor Bostic Yes, it is ready to publish as-is.

More details on responses

Non-responders

The following persons have not answered the questionnaire:

  1. Katie Haritos-Shea
  2. David MacDonald
  3. Romain Deltour
  4. Alastair Campbell
  5. Detlev Fischer
  6. Chris Loiselle
  7. Jonathan Avila
  8. Rachael Bradley Montgomery
  9. Charles Adams
  10. Daniel Montalvo
  11. Helen Burge
  12. Todd Libby
  13. Thomas Brunet
  14. Catherine Droege
  15. Suji Sreerama
  16. Shane Dittmar
  17. Nayan Padrai
  18. Sage Keriazes

Send an email to all the non-responders.

Compact view of the results / list of email addresses of the responders

WBS home / Questionnaires / WG questionnaires / Answer this questionnaire

Report issues on GitHub project w3c/wbs-design or by mail to sysreq.