Google Blogoscoped

Forum

Gmail abuse of "Cc"-only message?

Ianf [PersonRank 10]

Saturday, August 16, 2008
15 years ago3,663 views

I've been using email for more than 20 years, and this is the first time I encounter a client that dispatches a message with ONLY the "Cc" adresses, neither "To" nor "Bcc" -line content.

Either the RFC822 rules have been superseeded, which I doubt, or Gmail in Basic HTML mode does something really *nasty* to clients' privacy (I usually put the future recipients in the Cc-field, while keeping the To/Bcc clear, all in order to PREVENT the msg from being dispatched prematurely/ or by accident; and, later, to be explicitly told to supply some addresses to send it to – ie. move the Cc ones to Bcc). Only not this time, Cc-adresses with null content in To/Bcc were enough.

I wonder therefore if you could test it in other setups, preferably both Gmail 2.0, Basic and Mobile views with/without CSS, and tell me

IS GMAIL MISBEHAVING FOR EVERYONE OR JUST ME ???

(and where, apart from mailto:mobile-support[put at-character here]google.com I need to report that to).

Grega M [PersonRank 2]

15 years ago #

Hm, I have been using Thunderbird for a few years now and it has no problems with sending e-mails with CC fields only...
I thought it is quite common to put only BCC addresses and the e-mail is sent to "undisclosed recipients"

Tony Ruscoe [PersonRank 10]

15 years ago #

I thought that Outlook and Outlook Express both allow this too.

Ianf [PersonRank 10]

15 years ago #

Hold on, this is getting confusing. For most of the 15 or so years before I migrated exclusively to Gmail, I used but a few different mail clients, chiefly Eudora on a Mac, ELM and now-and-then PINE on Unix/ Linux. As far as I remember none of these would accept a sole "Cc'd" address – either a "To'd" or "Bcc'd" one had to be there as well, or the client would complain and NOT dispatch anything. Cc was wholly OPTIONAL, To and/or Bcc were REQUIRED. I assumed either there was some explicit, IETF RFC-stipulated rule to that effect (which made sense to me), or an informal/implicit-but-universally adhered-to M(ail)U(ser)A(gent) convention/ standard.

Having checked the RFC822 now, I see that this isn't addressed there, only the correct syntax for various header fields. So where could that rule, if it indeed IS A *RULE*, originated, and now be found? (I know I didn't make it up).

Could it have been one of SENDMAIL's requirements rather than Eudora's; a rule governging a M(ail)T(transfer)A(gent) rather than a MUA?

In any event, I *AM* sure that that's how Gmail acted not so long ago – ie. didn't accept a sole Cc'd address for dispatch. And now you're telling me both Thunderbird and OE always done that by default?

Tony Ruscoe [PersonRank 10]

15 years ago #

I can confirm that I've been able to send solely to CCd addresses in Outlook at least since 2005.

I'm sure you're right about what you remember, but I find it hard to understand why any mail client would enforce the rule that CC is optional but To and / or BCC is required – because if you received an email with a blank To field but the CC field was completed, you would immediately know that it must have been BCCd to at least one person. While you wouldn't know *who* it had been BCCd to, you would know that it's not for your eyes only.

Ianf [PersonRank 10]

15 years ago #

Tony, the "Cc" implies it's but a copy of something intended primarily for other eyes. Philosophically speaking therefore, a Cc'd letter with blank "To" line is a logickal aberration – you might as well move the Cc'd recipient to the "To."

On the other hand a letter with no visible adressee (which the receiving client sometimes fills in/ enhances with "Recipient-List;" or similar label) is, obviously, a Bcc-dispatched one, with one's address in the RFC822 MAILTO-"envelope" that preceedes delivery of body content (allso applicable when the "To" bears some to us unknown, but remotely-valid alias....).

I believe that, apart from disclosure, one of original intentions for distinction between "To," "Cc," and "Bcc" were how to proceed when messages replied-to/ bounced/ forwarded onward. After all, early mail delivery networks were very much patchy affairs, where the goal was to prevent occurence of endless loops etc.

In any event – I am *pretty sure* Gmail observed that no-unaccompanied-Cc distinction from the beginning. Only now no longer, at least at my end. What made them change it – and was it done explicitly, by design, or was it merely an oversight when some key nth-level library code accidentially stripped out previously resident "check for sole Cc" rule?

Tony Ruscoe [PersonRank 10]

15 years ago #

> Tony, the "Cc" implies it's but a copy of something
> intended primarily for other eyes.

I completely agree that's what CC implies but I didn't write the mail clients... ;-)

Ianf [PersonRank 10]

15 years ago #

I've done my best to bring this discovered sudden "Gmanomaly" to Gmail-tainers attention, with, so far, no confirmation whether it reached the --or anywhere near-- correct party. I've mailed both the mobile-support[put at-character here]google.com and mobileblog[put at-character here]google.com – both of which responded with no-reply form letters informing me that they cannot respond individually, and directing me instead to Gmail Help pages (which, as it happens, don't deal with that issue at all), and to Googlegroups, as therapy and/or panaceum. I dunno, I almost get nostalgic for the times when one could mail anything, any silly query, and bill.gates[put at-character here]microsoft.com would respond practically immediately with witty, informed letters.

Forum home

Advertisement

 
Blog  |  Forum     more >> Archive | Feed | Google's blogs | About
Advertisement

 

This site unofficially covers Google™ and more with some rights reserved. Join our forum!