Former MSN Messenger Service developer recounts battle to reverse engineer AIM's chat protocol
Chat Wars — In the summer of 1998 I graduated from college and went to work as a programmer at Microsoft in Redmond, Washington. I was put on the group that was building MSN Messenger Service, Microsoft's instant messaging app.
Context & Ripple Effects
This n+1 memoir drops into a long-running argument about whether consumer messaging should interoperate. In the summer of 1998 Microsoft's MSN Messenger team fought AOL by reverse engineering AIM's chat protocol rather than licensing it — an admission that the first big messenger war was won through unauthorized cloning, not cooperation.
The arc since then has bent toward negotiated openness and then back toward walls: Google Talk eventually reached an interop deal with AIM, as covered in TechCrunch's look at AIM-in-GTalk, and the protocol designer himself resurfaced in 2011 pitching a personal-app platform. Meanwhile Microsoft's own messaging-security record came under fire with the Guardian's report that Microsoft had handed the NSA access to encrypted messages, giving the old protocol-war stories a darker aftertaste.
First-order effects
- For readers tracking messaging history, the essay reframes Microsoft's role: MSN Messenger's early growth depended on breaking AIM's protocol, making Microsoft both the era's most aggressive interop hacker and, per its later NSA disclosure, a company later criticized for message-access practices.
- AOL's position looks different in retrospect — its protocol was cloned repeatedly, so its control over AIM's network was enforced by constant patching rather than legal or technical closure.
Second-order effects
- The clone-then-negotiate pattern set the template rivals followed: once Microsoft proved reverse engineering worked, interop became leverage, culminating in deals like the GTalk-AIM arrangement rather than open protocols by default.
- Vendors building messaging products learned the opposite lesson too — proprietary, frequently changing protocols are a moat, which pushed the industry away from the shared-protocol model AIM's cloners assumed would win.
Third-order effects
- If the pattern holds, consumer messaging consolidates around closed platforms whose owners alone decide who connects — with interop granted commercially, never technically — leaving regulators and users without the protocol-level escape hatch early messengers had.
- The surveillance dimension raised by the 2013 Guardian reporting suggests the next structural question for messaging is not interoperability but verifiable end-to-end encryption, a concern absent from the 1998 protocol wars entirely.
The trend: Consumer messaging has moved from adversarial protocol cloning in the 1990s to commercially gated interoperability inside closed platforms, with security scrutiny replacing openness as the battleground.