wilt:
Hi,
If you really want to delve into the history and technicalities of cookies then pls
see
http://www.ietf.org/rfc/rfc2965.txt
http://www.ietf.org/rfc/rfc2109.txt
There are a couple of reasons why the setting was added.
If a cookie is set with out a leading period the cookie is only valid for that domain, and not for any subdomains.
So if my cookie domain is zen-cart.com then the cookie is not valid for https://www.zen-cart.com. This is obviously important given that a lot of sites these days eschew the use of the www prefix.
It's also complicated by the fact that different browsers report the host differently, eg some will add the leading . others won't (although this was mainly Microsoft browsers, and they are a lot better these days) and by the way different servers (apache vs IIS) report the domain to the browser.
and even further complicated by how shared SSL hosts might report the domain. eg where the ssl domain is something like https://somedomain.sharedssl.com
Thanks Wilt for the clarification and the links to read more (my head hurts now) :)
I know each situation is unique, in our site's case we use https://www.clevershoppers.com and actually have a redirect that makes sure anything that doesn't point to https://www.clevershoppers.com gets redirected to the www version. So if someone types clevershoppers.com it goes to the www version. It seems odd then in that situation to have the cookie with a period in front of the www.
Again, the logic makes sense from what you've described. I have dropped the leading period from ours and so far so good, it's not affecting site accessibility. At least not that I am aware of so hopefully, I didn't break anything by setting it to false.
Thanks!