IP: Re: why DSL providers are terrible email providers

[David Farber <dave () farber net>]

> Date: Mon, 9 Apr 2001 21:19:31 -0400
> From: Meng Weng Wong <mengwong () dumbo pobox com>
> To: David Farber <dave () farber net>
>
>
> | >I've suspected that some ISP's are better at delivering mail than others.
> | >Check out:  https://www.icgroup.com/login/announcements  for some
> | >interesting comments by a mail forwarding company that seems to really
> | >understand the problems.
>
> Now that I've been accused of "really understanding the problems" I
> suppose I'd better explain myself.  :)
>
> I have here an argument that vilifying a broadband provider for being
> bad at email is like criticizing an airline for serving bad peanuts.
> Broadband ISPs need to either reduce users' expectations of the email
> aspect of the service, or (for cultural reasons much harder to do)
> meet them.
>
> As DSL and cable replace dialup as the connection method of choice,
> more and more people are complaining about the shoddy standard of
> email services provided by broadband companies.  I have reason to
> believe these problems are more than just growing pains: they reflect
> a shift in what it means to be an ISP.
>
> I observe these problems firsthand from behind the consoles of
> pobox.com, the email forwarding service that redelivers mail to paying
> users.  Since 1995, we've been in the business of offering people a
> lifetime address, so that when they switch ISPs, the change is
> transparent.  When an ISP has trouble with email, we usually know
> about it before they do.
>
> Recently we observed a clump of problems with some major broadband
> providers.  Big problems, like losing DNS, bouncing mail, taking the
> mailservers down for two weeks at a time, or just dropping mail on the
> floor altogether.  Problems that make it look like those ISPs don't
> care much about mail.  Guess what?  They don't.  I'll go farther: they
> shouldn't.
>
> I learned something in MBA school: as an industry matures, competition
> moves along five frontiers:
>
>   functionality (can we get the damn thing to work at all?)
>   reliability   (will the damn thing please stop crashing?)
>   convenience   (let's shrink it so i can take it with me.)
>   price         (if it's a commodity, give me the cheapest)
>   fashion       (indigo or graphite?  hey, maybe key lime.)
>
> Only after one frontier is crossed does a market focus on dimensions
> relevant to the next.  I made the mistake of signing up with the
> cheapest DSL provider, and it was months before I got connected.
> We've all heard the horror stories.  Shouldn't have based my decision
> on price, when they still hadn't figured out functionality.
>
> Five years ago, an ISP provided a tight bundle of services: dialup,
> email, web access, even usenet.  Everything was roughly equally
> difficult: you'd spend about the same amount of time on SLIP as on
> POP.  All the technologies were crossing the same frontiers in synch.
> Tech support would walk you through each aspect of the Net.
>
> Today, a broadband ISP is a pipe provider first, an an email provider
> maybe fifth or sixth.  Tech support only cares enough to get you to
> "OK, sir, can you surf the web now?  Is it working?  Good.  Bye bye
> now.  Gotta take the next call."
>
> It's 2001.  Congratulations everyone: email works.  It's become a
> matter of convenience.  When it doesn't work, people get annoyed.
> It's so far along the curve that you can get free email anywhere:
> people sign up if the domain name looks cool.
>
> But DSL is still back where dialup was in 1993: busy signal?  Ah well,
> that's why I have two providers.  We're still on that frontier of
> functionality: if they drop your connection three or four times a day,
> you just sigh and try to connect again.  Shouldn't expect any better.
> Expectations start low and rise as the frontiers are crossed.
>
> Broadband ISPs know the market is going easy on them; they're doing
> their best, and nobody else is getting it right, either.
>
> Then they make their big mistake: they take those standards and use
> them for other aspects that carry higher expectations.  You say the
> mail servers were down for five days straight?  So what?  You're lucky
> it works at all.
>
> It's an understandable position: DSL providers with sales to make and
> COs to provision are all too worried about being the next Northpoint
> to pay any attention to a trifling thing like email.
>
> But customers expect much more of email than of DSL.
>
> Due to the modularity principle that guides computing, the Internet
> access industry will fractionate into highest common denominators:
> break down the products and services into as fine a level of
> granularity as possible, and those are the natural domains of
> competition.  Each technology, maturing at its own rate, might fall
> out of sync with its complements.  Each technology needs an
> appropriate culture to support it.  If you can't separate the cultures
> by division, the market will be happy to separate the cultures by
> company.
>
> When I proved to @home that they were losing mail, they stuck their
> heads in the sand, refused to admit the problem, and said, tell the
> users to contact their local support representatives.
>
> It's the old adage about core competence: a broadband company should
> be a broadband company exclusively and outsource everything else.  Or
> maybe not even outsource: just trust the users to find a solution for
> themselves, and pass the savings on accordingly.  After all, there's
> always some fool out there who'll provide a service for free --- who
> wants to compete with that?  (Actually, I could answer that, but
> that's a different argument for another time.)
>
> DSL providers should just say to their customers, we'll just drop your
> price by $X a month if you decline POP --- that way we save on
> machines, sysadmins, and software licensing fees, and we get to say
> we're 20% cheaper than the competition ... and you'll just go off and
> use Hotmail, which is what you were going to do anyway!
>
> After market dynamics have had their way, businesses that try to do
> too many things at once will end up doing none of them at all.  Pick a
> core competence: email or dialup or DSL, but not all three.
>
> mengwong () pobox com 20010409



For archives see: http://www.interesting-people.org/