IMO, users can't be trusted to make this kind of decision. You will end up with an unusable spam database.

That's simple: A company can add a simple condition in their NDA for that kind of possible "nasty ones"...

What you call "nasty ones" is users adding sh*t to the spam database, which makes legitimate mail bounce.

Another case as an example:
I wonder how Companies use (according to your objection) embedded Firewalls with integrated Spammail filtering function in countless companies. All these rules are controlled by the service deliverer!?

Companies trust on this service (Sonicwall etc.) and not on their own colleagues? That makes no sense. That would be embarrasing btw..

Thanks god Domino's integrated mbl filtering function does a significant good job (if you poll the right MBL servers) at least.

The concept in itself is (IMHO) the correct way, nothing wrong in here: centralized serverside Spamrule Management is better than muddled personal junk folders.

Keep the mail.box clean, avoid wasted internal traffic & wasted Server CPU cycles for junk mail transportation to the clients...

The keyword here is CENTRALIZED... If You want a centralized serverside spamrule management - then you CAN NOT de-centralize the administration of the same! The serverside spamrule management must be handled by central administrators and not by end-users. What if one user thinks that mail from abc@xyz.com is spam, but the next think it's perfectly valid sender?? Then You have a rule that goes back and forth depending on each users own opinion. That is nothing we want to see...

It is perfectly ok for an end-user to flag a message as spam, and the rule is saved in the users own mailbox and all mails from that sender ends up in the Junk-folder. If the user thinks that this is a bigger problem and everyone should have that junk-rule - then he can ask the administrators to add such rule at the server. The admins can then ask the organisation if this is perfectly ok and then You have a working rule that will not be changed back and forth...

And, as You point out: If You configure the Domino server to poll the correct MBL servers, then spam is a minor problem - and therefore there is even less need to implement Your suggested functionallity.

Kenneth: My opinion exactly! Thanks for expressing it!

You're completely right, Kenneth. I didn't phrased my idea correctly, so I added this administrative revisioning to the Idea.

Thank you very much!

But neither you nor Peter gave me a reasoned answer, why millions of business users/admins trust on Spamrules delivered,controlled & updated by Hardware Firewall vendors...'External Spamrule databases' can be accepted, but (_revised_) internal Spamrules wouldn't? Come on..;)

I didn't say that you couldn't trust an internal spam rules database. I said that an internal spam rules database that's being fed by your users can't be trusted. Normal users can't make the distinction between spam and real mail.

" why millions of business users/admins trust on Spamrules delivered,controlled & updated by Hardware Firewall vendors...'External Spamrule databases' can be accepted, but (_revised_) internal Spamrules wouldn't? Come on..;)"

We trust spamrules delivered, controlled and updated by H/w firewall vendors, just because the fact that it is administrated by people with EXCELLENT knowledge of determine spam from legitimate mail.

Internal spamrule databases fed by end-users can't be trusted, and as I stated: Even when You change Your suggestion to have some administration approval process for entries to and from the database, the whole suggestion is unnecessary since external MBL does EXACTLY this and with excellent accuracy. So there is very little need to implement Your suggested feature...

This is a good idea, it is suggested implementation is flawed. You just need the user flagged mail to go to a "suggested spam" db, where a knowledgeable admin can control which rules are deemed usable and which rules are not.

Wayne = Oh, great, something else for the Admin to check. That database would have to be manually acted upon at least once a day. And the volume in it could be unmanageable for an Admin at a large organization. I voted no because this has already been resolved by many industrial strength anti-spam appliances. If you want the user to have some sort of control, most of them (if not all) allow the use of an End User Quarantine. If you want to use a Lotus branded product, look at Protector.

Seems to be a bad idea I threw in:) There seem to be no need for this...

Still there are questions..
- why personal spamrules when serverside MBL's are working so great?
- trusting _EXCELLENT_ 3rd party Anti-Spammers and not _EXCELLENT_ own corporate Domino Admins?

Won't get this. Anyway.

@Gregg Eldred: On BIG Organziations, you're totally right. Think about all the same Anti-Spam rules applied by 50.000 mail users :)))

And as you say, Lotus Protector seems to be the correct answer in here. Don't know if something like SpamAssassin-like works on Lotus Protector....

Approach as you want - of course there is a need for centralized spam prevention in a 'open-to-the-world' Domino Maildomain.

Let all the corporate end-users fiddling around the same spam from same senders parallelly with similar rules & deleting it every morning, througout the day and especially each monday doesn't make their work cost-effective, believe me.

I saw a similar concept working on a well known webmail service here in Turkey. You just sign the "D'oh - viagra-me!" mail as spam. The mail is in the trashbox, the service backend puts the spammy sender to the general blacklist of the mailservice for all users, and after a given period, the spam address is going to be deleted (FIFO). That keeps the own corporate spamlist rule moderate...

I agree with @11

Next, users that are in-charge of their spam, have over a 100 mail rules, that get checked each time a new message arrives. They have duplicate rules, rules that don't run, rules which create conflicts, and race conditions--causing inconsistent results based upon the search criteria and placement order in the rule list. Then, admins have to go and visit these machines to clean-up the rules so that Notes works again, as most help desk team members have knowledge outside of this arena and escalate the tickets.

Protector is a better way. But for the few messages that might squeak by with a Protector implementation, sure, then use the "Add to Spam folder" action, which works well enough in this case.

Atilla is asking for a domino domain-based rule. Normally those rules are set by Administrators which would mean that if you don't want it you just don't enable it.
There are Linux gateway servers that filter 99%+ of all Spam without license cost or third party software. What Domino offers out of the box is still very limited and everything that would help improving it is welcomed.
Unfortunately without proper support for a Bayesian-like Spam or better content analysis system domain-based rules indeed do not make a lot of sense.
At least it would make more sense than the block sender button in the Notes client.
I still voted yes because choice for me is better than not being an option at all.

Nobody has explained the real reason why users shouldn't be trusted to add domains to a domain-wide spam blacklist, so I will: If you have 10 users who have subscribed to weekly mailings from acme.com, the one user who is too lazy to figure out how to unsubscribe properly will end up blocking the messages which the other 9 users still want to receive.