I'm going to respond to a bunch of e-mails from different people, so
in order to avoid sending a bunch of replies I've combined them into
this one. Also please read it all, including my previous lengthy post
on other software that uses the "be lenient in what you receive..."
policy. Also, I'm posting this to TBOT
(http://groups.yahoo.com/group/tbot/) as well because this is getting
off topic so please post replies there.
First some more general questions/ideas in all e-mails:
No, this is not a bug in TB, it's a error on the part of
person who wrote the malformed mailto that I think TB should be
interpret as the original errored author intended. I think originally
I was just lazy and preferred to say bug instead of "error made by
web authors which should be interpreted as intended by TB," sorry :)
Also if there is confusion I'm referring to a malformed link in this form:
<a href="mailto:[EMAIL PROTECTED]?subject=send me some cheese">mail
me</a>
[note the enclosing quotes]
Regarding this being specific to Outlook Express, I've also tested it
with the follow programs which interpreted the malformed mailtos as
intended by the author:
Eudora 5.2
Mozilla 1.2b mail client
Netscape 7.0 mail client
You can see http://www.gummibears.nu/test.html for an example of one
correct mailto, and one incorrect if you'd like to test for yourself.
The rest of the stuff is addressed below, tell me if I missed anything.
----------------
Thursday, November 28, 2002, 10:59:57 PM, Thomas Fernandez wrote:
>> Yes, but it's beside the point. There's that internet saying,
>> (paraphrasing) "be lenient in what you receive and strict in what
>> you send."
TF> Who said this?
" 1.2.2 Robustness Principle
At every layer of the protocols, there is a general rule whose
application can lead to enormous benefits in robustness and
interoperability [IP:1]:
"Be liberal in what you accept, and
conservative in what you send"
"
-- Jon Postel, 1989, RFC 1122
TF> knowing about them, let alone following them. If you don't agree with
TF> the rules, do what any democrat does, become a member of IETF and
TF> change the rules. Many are being changed over time as necessity
TF> dictates. Just ignoring the rules leads to anarchy and helps nobody.
This is already a rule the IETF follows (see
http://www.apacheweek.com/issues/01-03-02 ).
TF> I did not wish to use an email client that follows this philosophy.
TF> The philosophy is wrong.
Please read my earlier post, this is a philosophy used by very many
major internet software clients and servers.
----------------
Friday, November 29, 2002, 4:41:03 AM, Dierk Haasis wrote:
DH> So, TB should "send" mailto links correctly coded and "accept" (in
DH> other mails) them wrong?
DH> First, I am not sure what that means in this context.
In this context it only applies to receiving since The Bat! isn't
actually generating mailto: links. Keep in mind this is a general
rule, so the entire thing may not apply in a given context (in this
case just the first half). So basically it means The Bat! should be
able to understand malformed mailto: links. Lenient in what you
accept means accepting data that is possible to get the intended
meaning of (in this specific case you can), but is nonetheless
malformed.
DH> Second, how the hell should TB know what part after a mailto: belongs
DH> to the link and what is the next normal text?
Good question, for clarification I was referring strictly to mailto:s
embedded in an <A> tag. I've noticed while certain websites don't use
%20 instead of spaces they do enclose the link in quotes so you can
still extract the original meaning even though it has spaces.
Ex: <a href="mailto:[EMAIL PROTECTED]?subject=hi how are you"> With the
quotes you can still extract the meaning of it as it clearly marks
where the text ends. Of course without the quotes you're screwed, but
no software can interpret that the way it was intended anyway.
________________________________________________
Current version is 1.61 | "Using TBUDL" information:
http://www.silverstones.com/thebat/TBUDLInfo.html
