On Wed, Aug 24, 2011 at 11:25 AM, Oleg Kalnichevski <[email protected]> wrote:

>> Still: forcing the users to implement that code snippet every time
>> they need to use preemptive auth is just this: forcing the user to do
>> unnecessary stuff because you want to.
>> You can and have put an explicit statement about why preemptive auth
>> should be used carefully or best not used in the documentation
>> everywhere (with javadoc its even available "during programming"), and
>> this for sure will raise awareness at the users side, but the code to
>> be implemented does not show at all that preemptive auth is not
>> advised, it is just a PITA and this indicates a certain mindset of
>> you.
>
> Please RTFM.

trust me, I did. I guess you cannot see this, but all of the code is
just self-serving boilerplate. You know, all I want to make httpclient
do is sending this:

GET /redmine/users/4.xml HTTP/1.1
Host: example.com
Connection: Keep-Alive
Authorization: Basic dTQxMzQ4MTp1NDEzNDgx

instead of omitting the Authorization line in the first request and
sending it only after a 401 was received.
To me, libraries should be as useful as possible, making things easy,
which is clearly not the case with httpclient and this particular
issue - and all of this without a real need, but just because the lib
dev thinks he is smarter and needs to force-educate his users.

regards,
christian



> You are entitled to your opinion on the matter, I am entitled to mine. I
> contend that preemptive authentication is conceptually flawed and poses
> major security risks in the overwhelming majority of cases. In rare
> special cases where preemptive authentication may have certain benefits,
> the users are expected to command a certain understanding the concept of
> credentials caching and should be capable of writing a few lines of code
> to pre-populate the cache instead of just dumbly flipping a boolean
> flag.
>
> Oleg
>
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [email protected]
> For additional commands, e-mail: [email protected]
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to