Why didn't AOL open-source its IM client?

AOL is getting a lot of credit for "opening" its ubiquitous AIM instant-messaging software "to open source." However, like Microsoft did recently by revealing documentation to its APIs and protocols, all AOL has done here is open access to OSCAR protocols necessary to create open-source implementations.

This is great, but consider just how much more AOL could have done--and for its benefit--such as open-sourcing its instant-messaging server and client software.

Think about it. What revenue does AOL protect by keeping its IM software closed? Sure, there's advertising revenue from the obnoxious ads it sprays around the client, but that is thinking far too small.

The real money is in abundance. Or in "adoption-led markets," to borrow Sun Microsystems' nomenclature.

Let's imagine that AIM were the default messaging platform for enterprises and consumers. Open source makes that much more likely because it would likely lead to a more widely distributed and used collaboration platform. It would get embedded into all sorts of applications.

A platform used by everyone is much more valuable than a platform used by many people. AOL would almost certainly continue to guide the platform because it would attract more developers to grok the code. It would therefore still have the opportunity to provide advertising, support (in the enterprise), and other services around the core platform.

An open-source AIM could be the instant-messaging collaboration platform. Not just first among equals or one of many.

But AOL didn't do this. It simply made it easier to read up on how AIM works. That's great, but it's a poor shadow of what AOL should have done.

[FellowConspirator]

AOL probably didn't open source their client principally because there was no incentive for them to. On their part, open-sourcing the client is work with minimal benefit and the onus of having to maintain a code base suitable for public consumption. It's not what they know, or what they do, and there are already a number of open-source AIM clients that demonstrate the only interesting part (messaging).<br /><br />On the part of the community, nobody cares. Of all the AIM clients available, the AOL one is the least compelling -- limited to a single protocol, feature poor, with a crude GUI. The other AIM clients exist in part because AIM itself (which was free) was not a well regarded piece of software. I would hazard to guess that a majority of the code base is devoted to hooking into the OS, carrying ads, etc. The value for the consumer is in the protocol -- which has been largely reverse-engineered already. There are a number of open-source implementations of OSCAR clients.<br /><br />The "opening" of the OSCAR protocol is also a non-event. While AOL promises to publish the protocol specs -- which have largely been reverse-engineered by programmers already -- they do so at a price that makes it effectively worthless. If you use their specification, you: cannot use it in GPL'd open-source software, you are required to implement several unrelated features (like "AIM toolbar" or ads), and you can't use it in software that interoperates with any other IM system (like Trillian, Adium, etc.).<br /><br />Essentially, the AOL move explicitly tries to thwart the development of the types of instant messaging software people want, in favor of instant messaging software that might use AIM as a conduit for advertisement delivery.

[lmasanti]

The usual suspects... on non-opening code are patents and copyright infringements... Sun has survived to them a long time after initial release.

[daftkey]

I'm not sure you've connected all the dots here, Matt. <br /><br />You are saying AOL would benefit by making AIM open-source because it would become the dominant player (a pretty bold claim already given the "success" rate of other open-source IMs and MSN's ubiquity and dominance currently; but for the sake of argument, I'll just humor you and say yes, AOL will become the powerhouse IM player if they open-source AIM.) <br /><br />Assuming that this is true and developers are enticed to "grok the code", what makes you think that any open source developer is going to leave intact AOL's advertising routines in their own IM software? I'm pretty sure, given the implementations already in place with Trillian and iChat, that this simply wouldn't happen. So to get your daily dose of ads as a user, you would have to download AOL's own IM software (which, as it turns out, is exactly the code used right now - completely closed and proprietary). Of course, as a user, I don't much care about ads, and most others don't either, so I use Trillian. I think the majority of users out there share my viewpoint. I think given a real alternative, AOL's ad revenue would pretty much disappear. <br /><br />As far as enterprise consulting for IM services goes - AOL could provide this now. So could MS (ahem, dominant market player, developer of an enterprise server exactly for this purpose, completely closed technology). They don't. There's probably a good reason for this.<br /><br />Of course, making AIM the dominant IM client WOULD make traffic to AOL's IM servers go up. But, you know, bandwidth and server hardware and maintenance are all negligible costs when compared to the value of being the leader in market share, right?<br /><br />I mean, at least that's what they thought in the late 90's when all those really really successful online businesses like Pets.com and kozmo.com were running, right?