While I am flattered that AOL recognizes what many of my users have been telling me: that my server is the most reliable and highest-quality Jabber server on the Internet; much of it's success has come from it's very reliable AIM and ICQ Transports (server-side plug-ins to the Jabber server to automatically translate between different protocols). With the loss of AIM and ICQ, I'm wondering if my users will flee in droves to other (less reliable) servers.
It has been suggested to me that I bring back AIM and ICQ support to my server by using a different IP and trying to block IPs coming from AOL itself in an attempt to prevent them from seeing where my transports are. After some thought I think this would be a very poor solution since IPs are a very unreliable way to track users, and I would certainly do more harm for my user-base than good. Besides, it would only be a matter of time, likely very short time, before they found the new IP and blocked that, too, erasing all that work in an instant.
Also, I don't think any solution that would force alteration of the jabber clients would be a viable solution at all. This would in effect create what's called "multi-protocol clients" like GAIM and Everybuddy, which handle AIM and others alongside Jabber. Anything which requires a client-side modification is doomed to failure, since users would be slow to pick it up and it would only complicate their user experience.
You may next think a server-side solution is the only way, then. And that is partially correct, but messing with the Jabber servers on such a low level could easily introduce new problems just when the code has stabilized.
The only technical solution feasible is a transport-side one, where the IM transport itself is modified to somehow evade AOL detection. This is slightly different than the server-side, since a transport is a server-side component independent from the Jabber server itself. I had been thinking a possible solution would be to have the transports form an "OpenAIM" network, with each knowing about the other AIM Transports on the Internet and intelligently "shuffling" user connections around between themselves in a random fashion so to avoid AOL from knowing what connections were coming from where. But then someone pointed out that AOL engineers connect to an AIM account through a downloaded Jabber client, and watch to see what IP that account comes in from, so they would be able to get the end IPs anyway. So this is also not a viable solution.
So it seems I must accept the fact there is no way I'm going to be offering AIM and ICQ again any time soon. I will simply have to reserve Theoretic to handling non-IM Transports such as my increasingly popular SMTP Transport and hopefully IRC and POP/IMAP in the near future.
But, that doesn't adress the Bigger Issue, I realize. That is, AIM and ICQ connectivity for Jabber in general. With Theoretic's blacklisting setting a precedent for other non-official Jabber servers to be blocked, it is only a matter of time before more and more Jabber IPs are blocked as well. And I don't think chasing AOL's tail while avoiding the bite is a long term solution. I have been working with Jabber for almost 9 months now, and would hate to see it's superior technology fail because a Big Dog crushed interoperability for it. Jabber has alot of potential, even with it's current feature holes and wantings.
Should we then reason with this Big Dog? Try to negotiate for them to open up their system at least to Jabber servers, to try and do for IM what was done for email in the Early Days? I have set up an unofficial petition to request AOL open AIM and ICQ to Jabber, but can it do any good? The key to answering this is to ask "What's in it for them?". email became standardized because a critical mass of the server admins at the time were academic types who had interoperability in their mind to begin with. SMTP was the result of this de facto standardization, and all of the proprietary email protocols were left in it's dust. However, as has been pointed out, the IM-scape today is very different. It is dominated not by university computer labs, but by corporate titans using it for profit. They would gain nothing to let users freely flow from network to network, and everything to gain by trying (even with mild success) to chain users into their network. To these titans, controlling 25% of the market with an iron fist is many times better than merely influencing 60% of it. In this environment, negotiation without having to hand over cash is about as likely as I am to stumble upon a bag of unclaimed cash (slim to none, for your information, and you know what just happened to slim... yep, it left town).
Perhaps good technology is the way to win after all, then? Perhaps Jabber needs to be so damn good that few users can deny it. Could Jabber and it's clients be so appealing that hackers and newbies alike will rush in droves to download and use? Sometimes, even I am skeptical. I mean, Jabber has great potential, but it has taken a long time (3 years) even to get to this purgatory stage.
Jabber still does not have a standard file transfer mechanism. No video conferencing support, not even voice over IP (VoIP) abilities. Why is it taking so long? Is it the difficulties of implimenting these resource-intensive features in low-range client/server architectures? Or is it lack of time, the age-old nemesis of open source projects? Or possibly even as strange as it may sound, lack of interest? After all, Jabber is still largely the realm of it's own developers, who are typically at home in plain text and sometimes buggy interfaces. Can a developer "think user" when it comes to IM?
Can the Jabber protocol and network survive the corporate investments? Can it survive being tweaked and ripped apart by various companies each wanting to be the Big Dog of Jabber IM? There are a couple now, and many more on the horizon fresh with investment monies. Few have any ties to the open source project that began the whole thing. Few have any open source advocates among their ranks or any open source plans. I know of a couple that might end up being downright hostile to the open source project. How can Jabber as an open platform for everyone hope to retain that coherence?
There does exist the Jabber Software Foundation, modeled after the newly formed Python Software Foundation and inspired by the Apache Software Foundation. But the JSF is very young, and has not really had time to build it's muscles, let along flex them already. Besides, some would argue it's too easily controlled by those corporate interests anyway. Not enough focus towards the individual member (although my personal opinion it's leaps and bounds above many other such software foundations). Should the member base organize under a "Union" of sorts to make sure it's always in the driver's seat? Or perhaps a radical approach of joining or merging with a more experienced and respected organization should happen? I do believe the JSF is the answer to keeping Jabber intact and making it successful in face of AOL and Microsoft, I just hope it has the strength to do it.
My mind is swimming with all of this. All of these options, all of these possabilities, all of these disasters waiting to happen. After a quiet sit on the couch and one cup of orange tea later, I have realized I cannot retaliate against AOL like I was about to. I cannot make them see the Internet and Jabber through my eyes. No amount of flames, polite petitions, or reasonable negotiation will sway them. There is only one thing they will listen to: money. And if I can't give them money to make them give in, then I will have to take it away to make them give in. I now see the only, the *only*, way out of this mess for me and everyone else is to fight to make Jabber so damned irresistible to everyone that the power of superior technology and open standards can't be refuted.
But as I said, there is an incredible amount of work to do. Lack of features, a need to clean up the current infrastructure, too few developers, too little investment, too much investment... Where to begin??? So, is my reasoning flawed? What can this open project do to survive let alone thrive?