Blog: 
Feb
Jan 2008
Email Newsletters are Hard
Dec
Nov
Occasionally, I receive Microsoft TechNet's newsletter by email, which is incidentally delivered
to my Microsoft email program, Outlook 2007. I suspect hundreds of thousands of other people
also subscribe to this mailing. You might think that a $300 billion dollar company sending
emails from its own servers to people using its own software could do so fairly effectively.
But they can't, because email newsletters are hard.
Email is even better! Or is it?
An Officious Beginning
The very top of the newsletter is a wide splash of empty space plus extra instructions, an ominous
warning, and a red X error icon. This is no way to make your introduction. You expect the email to
begin with something welcoming, but instead you are required to think. Do I want to right click?
Will my privacy be at risk if I do so? Is this message really from Microsoft? Do the errors
indicate it might be fraudulent?
An official newsletter from Microsoft blocked by a Microsoft application to protect my privacy, presumably
from potential identity thieves at Microsoft.
The message is legitimate, so the problems are evidence of bureaucracy, complexity and incompetence.
The business people in TechNet probably focus on the juiciest potential market—new
customers—and probably any user testing they do at all cleverly employs non-Microsoft
email programs. The developers for Microsoft Outlook, however, are battling the
spam epidemic, and thus designed their application to
shield you from any potentially vicious emails. The two teams don't work together because although
both deal with email, they do so in entirely different ways. Nevertheless every biweekly TechNet
newsletter I have ever received starts like image pictured above.
Security and Convenience
Although it should not even appear, the error message, too, is confusing. The software
proclaims magnanimously, "To help protect your privacy, Outlook prevented automatic download
of this picture from the Internet." It's great that Microsoft is playing information bodyguard,
but how will showing me some picture leak my personal secrets? Could this really be a risk?
The answer lies in some gritty technical details. These images don't work the same way as text
in email or file attachments, which are sent along like regular mail. Instead, the underlying
mechanism contains references to the pictures stored elsewhere on the Internet, and when
you open the email, Outlook jumps online and downloads them for you. This act means your computer
establishes an active connection with some remote system to get the image, proving you saw the
email. Who cares if people know you saw their message? You do, if the sender is a spammer.
Once Outlook displays the image, the spammer knows they have the email address of a live human,
and can begin bombarding you with unsolicited junk mail. So Outlook errs on the side of security,
refusing to show you pictures in email lest you become trapped in an inescapably deep inbox.
This is thoughtful, but still frustrating. Users just want things to work, not be forced to
understand how they are trading current inconvenience for future comfort.
No Good Solutions
There are two readily apparent ways for Microsoft to address this problem, but neither one
is desirable. One route is to establish an entirely new back-end authentication system for email
content, so that your own copy of Outlook can negotiate with particular trusted authorities,
determine the sender, text and images are legitimate, and confidently display the message in its
intended form. This is probably what Microsoft will end up doing, since it's the best way to
ensure their own messages look good with their own software.
The other option, which will never happen, is to return email to what it was: plain text
messages. No more complicated formatting and images, no more risks of privacy loss or increased exposure
to those nasty spammers. No one can imagine technology moving backward, not even to correct problems
created by moving forward too quickly. Rich, beautiful emails were hastily implemented by leveraging
available web technologies instead of assessing the unique nature of electronic messaging. Future
engineers will spend much of their time addressing these past mistakes instead of praising and
extending on robust successes.
Further Reading:
Would you like to leave a comment?
Read this.
###
