ietf-tools / author-tools

Author Tools
https://author-tools.ietf.org
BSD 3-Clause "New" or "Revised" License
37 stars 30 forks source link

Problems with footer of online iddiff tool #487

Open klensin opened 5 months ago

klensin commented 5 months ago

Describe the issue

Hi, when I run the online iddiff tool [1], at least with txt input and in "side by side" mode, I see a two line footer at the bottom of the last page. The first line reads

This html diff was produced by rfcdiff 1.45. The latest version is available from [http://tools.ietf.org/tools/rfcdiff/]

In no particular order, (1) I thought rfcdiff had been retired and that iddiff was the replacement. Claiming that I used rfcdiff when I invoked iddiff as noted above is, at best, confusing. (2) Clicking on that link gets me, not to the "latest version" as promised but, because tools.ietf.org is gone, to the general "authors" page [2] which invites me to go on a tour of the tools collection. (3) If I go to the tools catalog page from there, "iddiff" does not appear but "rfcdiff" does. However, when I click on the "rfcdiff" link it takes me to the iddiff page [1]. (4) Or, if I go to the author tools page instead, I find "Compare drafts. Replacement for rfcdiff is available under iddiff."

Looks like something needs a bit of updating.

[1] https://author-tools.ietf.org/iddiff [2] https://authors.ietf.org/

Code of Conduct

kesara commented 5 months ago

(1) No rfcdiff is still being used here by default, due to the community favouring subtle changes of rfcdiff over iddiff. You can still construct a URL to use iddiff with by appending iddiff=1. Example: https://author-tools.ietf.org/diff?doc_1=rfc9001&iddiff=1. This is not exposed via UI yet. See #438.

(2) This is from the rfcdiff script. I've created https://github.com/ietf-tools/rfcdiff/issues/5 for this.

(3) This is because the author tools did not change the URL when it changed the default comparison tools from iddiff to rfcdiff. Maybe a generic path like /diff and adding the ability to select the tool might be better?

(4) That's an incorrect statement which should be changed.

Thanks for bringing up those issues.

klensin commented 5 months ago

--On Tuesday, May 21, 2024 17:03 -0700 Kesara Rathnayake @.***> wrote:

(1) No rfcdiff is still being used here by default, due to the community favouring subtle changes of rfcdiff over iddiff. You can still construct a URL to use iddiff with by appending iddiff=1. Example: https://author-tools.ietf.org/diff?doc_1=rfc9001&iddiff=1. This is not exposed via UI yet. See #438.

Ok. But some comment in the footer to that effect would be appropriate. Otherwise we are creating a situation which is very confusing to anyone not following both the main tools list and the github repository. That explanation might also reasonably be placed on the page referenced for obtaining the tool, but that would need to work first.

(2) This is from the rfcdiff script. I've created https://github.com/ietf-tools/rfcdiff/issues/5 for this.

Thanks.

(3) This is because the author tools did not change the URL when it changed the default comparison tools from iddiff to rfcdiff. Maybe a generic path like /diff and adding the ability to select the tool might be better?

Perhaps. On the other hand, a navigation logic that runs from a tool catalogue, not to the tools themselves but to a page that explains, then requiring that one make choices and then go through another link, and then requires making further choices, etc., is often an unfortunate user experience for those who basically know what they are going and are just trying to get it done.

(4) That's an incorrect statement which should be changed.

Thanks.

Thanks for bringing up those issues.

What makes all of this fragile is that, on a different day or with a different pair of documents to be compared, I might not have noticed that footer at all. And, on a day when either I didn't have as much time or was less generally annoyed/irritable for other reasons, I might not have bothered. To the extent to which others have similar reactions/ behaviors, we should probably be thinking about ways to better highlight important information than putting it in a light-type footer and/or to make reporting as painless as possible.

best, john