[Tools-discuss] Re: Collection of glitches on the nonwg mailing lists list

[Toerless Eckert <[e-mail]>]

This ?

On Tue, May 28, 2024 at 11:55:27AM -0500, Robert Sparks wrote:
> We'll address it with the next datatracker release.
>
> https://github.com/ietf-tools/datatracker/issues/7477
>
> RjS
>
> On 5/28/24 11:26 AM, Toerless Eckert wrote:
> > [ Sorry if this is duplicate, i didn't track all recent messages to tools-discuss ]
> >
> > https://datatracker.ietf.org/list/nonwg uses links to the mailing lists without
> > the full email address, e.g.:
> >
> > https://www.ietf.org/mailman/listinfo/Iot-onboarding
> > https://www.ietf.org/mailman/listinfo/enterprise
> >
> > When the name of the mailing list is capitalized, as in the first example (Iot-onboarding),
> > the behavior is weird:
> >
> > - The browser shows a "Page not found" page rendered by mailman3.ietf.org
> > - The URL says  https://www.ietf.org/mailman/listinfo/[e-mail]
> >    - aka: still capitalized, but now with the added @ietf.org
> >
> > When you then try to enter/load that page, it does actually load the right page
> > but (of course) changes it to non-capitalized:
> >
> > - https://www.ietf.org/mailman/listinfo/[e-mail]
> >
> >
> > Not sure how you tools folks want to fix this. One option seems to be to de-capitalize the 204
> > occurances of capital letters in those URLs in https://datatracker.ietf.org/list/nonwg.
> > Or else make mailman3 do the decapitalization itself correctly..
> >
> > Cheers
> >      Toerless


On Tue, Jun 04, 2024 at 09:29:45PM -0400, John C Klensin wrote:
> Sorry.  While the message below was sent to tools-help as directed on
> the relevant page, I have just discovered it is a more general
> problem.   Especially given Greg's recent note on the IETF list about
> other Postorium/ Mailmain 3 issues, it may be worth warning others
> and perhaps more general discussion.
> 
> Apparently, if one goes to https://datatracker.ietf.org/list/nonwg (a
> place some, I suspect many) of us go to track down information about
> particular mailings lists, what they are called, and the location of
> subscription information and archives, many of the entries have
> broken links. In a non-trivial number of cases, the problem is that
> the entry for list "foo" points to
> https://mailman3.ietf.org/mailman3/lists/[e-mail]/ where it
> should point to
> https://mailman3.ietf.org/mailman3/lists/[e-mail]/, where the
> actual information is available.
> 
> While "i18ndir", as discussed below, may be obscure, I note, as
> particularly interesting, that
>    alldispatch
>    architecture-discuss
>    art
>    eligibility-discuss 
>    gen-art
>    hotrfc
>    iab
>    i-d-announce
>    ietf-announce
>    ima
>    int-area
>    mib-doctors
>    mtgvenue
>    ops-area
>    rfp-announce
>    rsab
>    terminology
>    tools-help
>    tools-discuss
>    training-wgchairs
>    travel-grants-committee
>    tsv-art
>    wgguide
> that is by no means a complete list but illustrates that the problem
> spans a broad range of IETF Areas and IETF-related activities and
> groups including the LLC, IAB, IRTF, and RFC Editor policies
> (although RSWG does not appear on the list at all, something I think
> has been noted previously).  Many other lists work.  I don't know if
> they have been individually reported and patched or if there is some
> pattern I cannot discern.
> 
> But it is probably time, before we conclude that the transition to
> Mailman3/ Postorius is complete, for someone to go through that page,
> test all the links, and fix the ones that produce "no found"
> errors... and do that before IETF 120 for which a subset of those
> listed above (and maybe others) might be particularly important. 
> 
> Suggestion if that cannot be done Real Soon Now: It appears as if
> there is a dedicated "page not found" catcher in Postorius because,
> all of the bad links produce the same "Page not found": "This page
> either doesn't exist, or it moved somewhere else." page.   Can that
> be modified to suggest that people inspect the URL that produced that
> page and, if it contains upper-case characters, they experiment with
> changing them to lower case and trying the URL again.  I've tried
> several, but not all, of the cases listed above and those that I
> tried worked.  Obviously not ideal, but better than leaving people
> wondering and frustrated.
> 
> More below about digging into a specific case and a few other
> problems.
> 
> best,
>    john
> 
> 
> 
> --On Tuesday, June 4, 2024 20:11 -0400 John C Klensin
> <[e-mail]> wrote:
> 
> > Hi.  
> > On <https://datatracker.ietf.org/list/nonwg>, the entry for
> > "i18ndir" points to
> > <https://www.ietf.org/mailman/listinfo/I18ndir>, which produces a
> > "page not found" message.   
> > 
> > This appears to be another instance of a case-sensitivity problem
> > because https://www.ietf.org/mailman/listinfo/i18ndir produces a
> > relevant page.  
> > 
> > I'm uncertain about the status of that list.  On the one hand, the
> > directorate was closed some time ago.    On the other, it might be
> > the last useful centralized list for i18n discussions or, if there
> > is another one, it should probably be reflected on the i18ndir page
> > once it can be found in the obvious way.
> > 
> > Also, IIR, while the directorate was active, directorate (and hence
> > mailing list) membership was by invitation.  This page seems to
> > imply that anyone can now join, which might or might not be what is
> > wanted.
> > 
> > The ADs should probably make up their minds and then the link should
> > be corrected and the mailing list page tweaked and/or the entry
> > moved from the "nonWG" page to the "concluded" one.
> > 
> > General suggestion to the tools team: Unless we have decided that
> > the "owners" of a mailing list should be secret, it would be a good
> > idea to list them, and maybe their addresses, on the mailing list
> > pages. If we have decided their names are secret, I question that
> > decision given that, e.g., we make WG Chair names and the like
> > public.  At worst, the names should be omitted only if list owners
> > request that and get approval some someone like a relevant AD.   
> > 
> > Probably even more important, there have been enough problems with
> > reaching people via ietf.org address aliases that, unless you are
> > confident that those problems have been solved --now and forever--
> > identifying mailing list "owners" by name (and probably address) on
> > the mailing list pages is particularly relevant.
> > 
> > best,
> >    john
> 
> 
> -----------------------------------------------
> Tools-discuss mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]
> https://mailarchive.ietf.org/arch/browse/tools-discuss/

-- 
---
[e-mail]