Given the growing interest in "dynamic" (a.k.a. "active") email, I thought I'd write up a

Very Brief History of Dynamic Email

and my role in it.

Dynamic, or active, email is email with content that invokes computation at the recipient's system, possibly having results that vary over time. These days, the best known examples are Google's AMP for Email and Microsoft's Actionable Messages. A specialized long-standing form of it is HTML's <IMG>, which can produce a different image each time it is loaded, depending on what is currently on the responding server.

I was heavily involved in the history of dynamic email, being best known for my role in the first attempt to define a common language for dynamic email, Safe-Tcl. Way back in 1991, I used it to sell my daughters' Girl Scout Cookies. However, the story of dynamic email actually goes back at least a decade before that.

The earliest citation I've found is a 1981 paper by John Vittal that, remarkably, seems to be unavailable online. I think I have a paper copy, I should scan it in...

John Hogg published a 1984 paper about an experimental active mail technology, which was apparently only briefly used.

From 1985-1989 I worked on the Andrew Message System project. This system was notable for making a number of features available to a substantial community of users for the first time, including multimedia email and five primitive forms of dynamic email, such as messages that allowed the readers to vote on multiple choice topics, to subscribe to new information sources, and to authorize information redistribution. The idea of generalizing this was discussed in section 6.5 of Architectural Issues In the Andrew Message System.

That led me to develop the ATOMICMAIL language, which I believe to be the first language to begin to address the security issues in dynamic email. This led in turn to the aforementioned Safe-Tcl language, which was also the first openly proposed standardized language for dynamic email.

Roughly concurrently with the latter work, a company called General Magic developed a language called Telescript as part of an even more ambitious business plan. Had they only developed Telescript, and had they deployed it using ordinary email instead of entirely reinventing electronic communication, there is a chance that dynamic mail might have caught on in the 1990's.

Of course that didn't happen, but that was when the Internet really began to boom. At that point, the email community was so consumed with security issues and entrepreneurial ventures that there was no real will to pursue the complexities of dynamic email for decades. But Google and Microsoft have brought it to center stage at last. I am excited because I still believe in the potential of dynamic email. However, I'm also a bit alarmed at the lack of standardization and community input regarding the design of the current technologies. I hope that the industry will soon see fit to pursue open standards for dynamic email.

However, the truth is that dynamic email won't become a fundamental everyone-must-have-it technology until the advent of the Interplanetary Internet. When TCP/IP packets face significant speed of light limitations, the web as we know it will be unusable, but dynamic email will let you send interactive mini-web-sites by email, which is barely affected by interplanetary distance delays. (I have no idea what, if anything, will be useful over interstellar distances, at least for short-lived creatures like us.)