On 6/11/2013 8:54 PM, Doug Miller wrote:> richard <
nor...@example.com>
wrote in news:db4a3lruunc0.g24bw2k2tws0.dlg@
>
40tude.net:
>
>> On Tue, 11 Jun 2013 09:41:22 -0400, Jerry Stuckle wrote:
>>
>>> On 6/11/2013 9:03 AM, j wrote:
>>>> I've been seeing URLs rewritten with 303's. With the somewhat
>>>> inscrutable "see other". I never see a manual link with these.
>>>>
>>>> As near as I can tell these are probably being used for location or
>>>> device redirects.
>>>>
>>>> So whither the 303? What is it really for?
>>>>
>>>> Jeff
>>>
>>>
http://www.lmgtfy.com/?q=http+303
>>
>> I give people who post this shit a good hefty bitch slap.
>
> Why? Jeff had that coming. So do you, whenever you get the same.
Why don't you read what Google yields. HTTP 303 is quicksand.
#1 hit:
303 See Other has been proposed as one way of responding to a request
for a URI that identifies a real-world object according to Semantic Web
theory (the other being the use of hash URIs).[1] For example, if
http://www.example.com/id/alice identifies a person, Alice, then it
would be inappropriate for a server to respond to a GET request with 200
OK, as the server could not deliver Alice herself. Instead the server
would issue a 303 See Other response which redirected to a separate URI
providing a description of the person Alice.
303 See Other can be used for other purposes. For example, when building
a RESTful web API that needs to return to the caller immediately but
continue executing asynchronously (such as a long-lived image
conversion), the web API can provide a status check URI that allows the
original client who requested the conversion to check on the
conversion's status. This status check web API should return 303 See
Other to the caller when the task is complete, along with a URI from
which to retrieve the result in the Location HTTP header field.[2]
#2
The response to the request can be found under a different URI and
SHOULD be retrieved using a GET method on that resource. This method
exists primarily to allow the output of a POST-activated script to
redirect the user agent to a selected resource. The new URI is not a
substitute reference for the originally requested resource. The 303
response MUST NOT be cached, but the response to the second (redirected)
request might be cacheable.
The different URI SHOULD be given by the Location field in the response.
Unless the request method was HEAD, the entity of the response SHOULD
contain a short hypertext note with a hyperlink to the new URI(s).
And going down the Google list gets further in the weeds. What I see for
303's is nether from a POST, nor have a hyperlink nor does it fit
whatever I can make out from hit #1.
So, I had consulted the all knowing Google and 303's are still
swampland. It seems to be used for a catchall when 302 doesn't quite fit.
What I had hoped is that someone who actually uses 303s would weigh in.
Just forget it.
Jeff