Fog Creek Software
Discussion Board




End of the road for SMTP?

Just saw an interesting article at news.com (http://news.com.com/2100-1038_3-5058610.html?tag=fd_lede1_hed), with one quote standing out as interesting, especially to Joel and many of his readers:

"I would suggest they just write a new protocol from the beginning," Suzanne Sluizer, a co-author of SMTP's immediate predecessor and a visiting lecturer at the University of New Mexico, said in an interview.

"In my experience in computers--which at this point, is quite extensive--trying to fix problems in the existing thing is almost always more difficult than just sitting down and thinking about what you want and coming up with something new," she added.

Thoughts?

Dennis Forbes
Friday, August 01, 2003

SMTP Replacement...

This one's already come up.  I quote from the beginning of a thread started by GiorgioG on Tuesday, April 22, 2003:

Joel wrote "Let's find a protocol with decent authentication and with micropayments to make spam uneconomical, and let's set a deadline, maybe two years in the future, when SMTP will simply be turned off."

Giorgio went on to pan.  Myself... I liked Joel's plan.

John Aitken
Friday, August 01, 2003

The thing I found most intriguing about her comment was the belief that it's better to just toss it all out and start anew, and this is a practice that Joel has come out fervently against many times in the past. Personally I believe that it's easier to _start_ the process of throwing it out and starting anew, but it's much harder to _finish_ the process.

Dennis Forbes
Friday, August 01, 2003

I think Joel is talking about re-writing an existing software for the purpose of making it faster or easier to maintain.  This can be done by refactoring and should not be done from scratch.

However the problem with SMTP is a fundamental design flaw which cannot be fixed by refactoring. (Or maybe you can refactor the whole email system in this world??)

Jack
Friday, August 01, 2003

I see no great difference betwen choosing to continuing to enhance and fix the world's SMTP system, or throwing it all out and starting from scratch, or doing exactly the same thing : I'm not seeing the great philisophical gap that you are.

Regarding whether one will "refactor the whole world's email system", what do you think each enhancement of the SMTP protocol is? Do you think it is _easier_ to break the whole world's email system?

Dennis Forbes
Friday, August 01, 2003

"with micropayments to make spam uneconomical"

So - no more free mailing lists? You'll need a credit card to sign up to anything (since distribution of a mailing list otherwise would be cost prohibitive)

Philo

Philo
Friday, August 01, 2003

Lets suppose we implement a new mail system with micropayments.  Who determines the cost of the payment, and whats to prevent ISPs and governments from adding piggyback fees and taxes? 

I can already hear the next politician, "Whats an additional $.009 per email if it will help establish internet connectivity for the poor"

Would this also be the end of anonymous email? 

notAlGore
Friday, August 01, 2003

The best solution that I've heard is that the payment amount for email is set BY THE USER. The actual proceeds of the payment go TO THE USER. And any user can set up a whitelist of email addresses that don't have to pay anything.

I think it's a nice idea.

Benji Smith
Friday, August 01, 2003

I don't think micropayments are the only answer:  Decent authentication -- and the accountability that would entail -- would also help curb spam.  Personally, I'd favor a new protocol with some kind of barriers to setting up accounts (a waiting period, a deposit to be forfeited if you violate ToS, physical verification like in the old days of BBSing, or others), thus eliminating the 'throwaway account' phenom.

Of course, this does preclude a certain level of anonymity, so strong encryption would be desirable as well.  But there's no reason we'd have to get rid of SMTP right away.  I hate to hear myself saying this (because I think markets are really bad at some things), but you'd just need to get the new protocol out there, get a few major players to sign on (to avoid the IM fragmentation effect), and see if people think not getting spam is worth the tradeoffs.

Sam Livingston-Gray
Friday, August 01, 2003

The micropayments would only have to accrue to a sender of unsolicited emails and could be billed through the sender's ISP... and they'd only need to be 1/10 of a cent or something... just enough that freeloading bastards couldn't hijack the infrastructure at everyone else's expense.

If you were subscribing to a newsletter you could append a publisher specific 'he's okay' validation key to your email address when you sign up.  At least that's how I imagined people were envisioning such a system.  I don't know enough about cryptography and IP routing to know how much of a headache it all would be (ie. publishers and ISPs might have to trade keys offline or something if addresses were spoofable), but I can see that some such approach is theoretically workable. 

John Aitken
Friday, August 01, 2003

I don't see how you can have email be both free and anonymous without spam abuse... either one, yes, but not both.

True anonymity is pretty tough anyhow I guess...

John Aitken
Friday, August 01, 2003

Changing a widespread protocol is like trying to change the engine in your car while you are still driving it.

Could everyone on the internet really agree on a replacement for SMTP and arrange a date to switch STMP protocols.

The slow uptake IPV6, XHTML and others are testomony to the difficulty of changing a protocol.

Matthew Lock
Saturday, August 02, 2003

...which is why we should all design our Web pages for 2.0-level browsers, right?  Because someone's grandma is still using that 66MHz Pentium with the floating-point bug...

With the proper amount of publicity, it could be done.  Get AOL and Microsoft to agree on a protocol and spend a piddly (for them) amount sending out a few press releases about a new email system that was "spam-proof," and they'd get enough attention to attract a self-sustaining population.  There'd still be a place for SMTP, but just because it's the "dumbest common denominator" doesn't mean we shouldn't think about better systems.  (=

Sam Livingston-Gray
Saturday, August 02, 2003

AOL will fight it - it would cost them a TON of money and they wouldn't get anything out of it. (You can't use "secure spam-proof email infrastructure" to attract subscribers, since everyone else has to do it for it to work, so you're not unique)

I *suspect* MS can't do it on their own because the /. crowd would call it an attempt to control email and sell more Exchange servers.

Who's left? Linus?

Philo

Philo
Saturday, August 02, 2003

Philo, nobody takes what the Slashdot mob says seriously; why on earth would Microsoft of all people?

Adam Fitzpatrick
Monday, August 04, 2003

*  Recent Topics

*  Fog Creek Home