Re: [Tools-discuss] de-authenticate?

[Henrik Levkowetz <henrik at levkowetz.com>]

Hi Tony,

On 2009-10-21 16:38 Tony Hansen said the following:
> Looks nice. Thanks!
> 
> You might consider adding Elwyn's information on using "clear recent 
> history".

Done.  Thanks!


Best,

	Henrik


> 	Tony Hansen
> 	tony at att.com
> 
> Henrik Levkowetz wrote:
>> Hi Tony,
>>
>> On 2009-10-20 23:38 Tony Hansen said the following:
>>> Henrik, given that there are issues with the various Logout mechanisms, 
>>> could there be a Logout link that goes to a page explaining the various 
>>> techniques and their issues/shortcomings? For example, the page could 
>>> include a link to 
>>> https://logout at wiki.tools.ietf.org/group/nomcom/09/input ** with an 
>>> explanation of the consequences of using it.
>> Done, try http://wiki.tools.ietf.org/group/nomcom/09/logout-info
>>
>>> 	Tony Hansen
>>> 	tony at att.com
>>>
>>> ** yes, I intentionally changed "dummy" to "logout".
>> I did exactly the same change when I implemented this earlier for test
>> purposes :-)
>>
>>
>> Best,
>>
>> 	Henrik
>>
>>> Elwyn Davies wrote:
>>>> Samuel Weiler wrote:
>>>>> Aside from messing with cookies on my end, is there a trivial way to 
>>>>> deauthenticate from the tools site(s)?  I suggest adding such a link 
>>>>> to most pages, PARTICULARLY the NomCom pages.  (I was asked for NomCom 
>>>>> feedback at one address, but I'm typically authenticated to the tools 
>>>>> site from a different address, and I'm having challenges swithcing 
>>>>> between my identities.
>>>>> -- Sam
>>>> I had more or less the same problem... Henrik sent me this 'hack' as a 
>>>> workaround.
>>>>> BTW it would also be useful to have a means to logout from a session - 
>>>>>> once logged in with the 'wrong' email address, there is no way to > 
>>>>> relogin with the 'right' email address apart from clearing the session 
>>>>>> with browser capabilities.
>>>>>  
>>>>> Yes.  One of the woes of using http auth.  Why couldn't whoever specified
>>>>> that have provided a logout mechanism?  (Oh, wait, that was maybe us...)
>>>>> And why haven't the browser manufacturers provide the ability to drop an
>>>>> ID?
>>>>>
>>>>> You can always achieve the effect of logging out by appending a dummy
>>>>> user ID to the URL, to make the browser drop the current ID:
>>>>>
>>>>>   https://dummy at wiki.tools.ietf.org/group/nomcom/09/input
>>>>>
>>>>> but that's a hack.
>>>> On Mozilla Firefox you can also de-authenticate by using the 'clear 
>>>> recent history'/details/active logins option from tools menu (in the 
>>>> recent 3.5.3 release .. the menu option is different for older versions 
>>>> - it is buried in the 'options' tabs).
>>>>
>>>> Regards,
>>>> Elwyn
>>>>
>>>>> _______________________________________________
>>>>> Tools-discuss mailing list
>>>>> Tools-discuss at ietf.org
>>>>> https://www.ietf.org/mailman/listinfo/tools-discuss
>>>> _______________________________________________
>>>> Tools-discuss mailing list
>>>> Tools-discuss at ietf.org
>>>> https://www.ietf.org/mailman/listinfo/tools-discuss
>> _______________________________________________
>> Tools-discuss mailing list
>> Tools-discuss at ietf.org
>> https://www.ietf.org/mailman/listinfo/tools-discuss
>