I'm sorry to have tricked you, but this is actually going to be an auto-biography that leads up to the primary topic.

Back in 2004, I joined BYOND with the sole intention of playing cool games, little aware of the technical world. My programming knowledge up to that point consisted of limited VB6 that my mother had taught me, and another BASIC variant entitled 'Learn to Program: BASIC' that my friend who introduced me to BYOND had also introduced me to. Oh, also failing to understand C entirely while modifying another program, but that's largely irrelevant.

After playing MLaaS for quite a while, my friend introduced me to the developer side of BYOND, pitching it as incredibly cool. I spent a year or two making dumb mistakes and not quite 'getting it'. Around 2007, everything just sort of 'clicked'. Programming made a lot more sense, and I found myself a lot better at it. I found myself interested in things like networking and taught myself C++. It took a few months and I finally started getting pretty good at it, to the point where I could write simple applications like DreamConfig, which was used to write or amend a hostban.txt for DreamDaemon on Linux.

After I started actually getting incredibly proficient in C++, March of 2008, life happened and I had to take an extended leave of absence as I moved into a completely new life with completely new people, studying in my free time for when I actually got a computer hooked back up. It wasn't until I graduated and went to college in 2010 that I actually got a stable computer running and was able to get back into the BYOND scheme of things.

At the time of writing it's December of 2011. A fully year and a few months have passed since I first managed to re-acquaint myself with the virtual world. Since then, I've been involved in research projects, some private and some public, with other members of BYOND, tearing apart the internals of BYOND slowly. When I say slowly, I mean really slowly.

== End autobiography

The projects I've been directly involved in have either completed due to the full extent of the subject matter being researched or halted due to lack of motivation. The latter part of this is what the title of this post is about. At one point, I was having a discussion with someone about Linux. The picture we were painting, so to speak, was of a full set of BYOND on Linux- the Pager, Dream Seeker, Dream Daemon, and Dream Maker, aptly code-named Dream Link, Dream Wanderer, Dream Daemon (already exists, just supplement it with DreamConfig), and Dream Builder.

After tossing the idea back and forth a little bit, I set forth doing some mock-ups. I switched back and forth for a few months between re-drawing the various controls that interfaces offer that didn't already exist in Qt and doing some research on how the Pager communicates with the hub and, to a much lesser extent, how DS communicates with DD/DS. The Pager <-> Hub research wasn't going too bad until I hit a brick-wall with the authentication. I tried numerous things, including writing a proxy to analyze my traffic with BYOND.com's hub and writing a hub emulator that sat on the other end and forged the intro packet to the Pager over and over and over again with the same data, only to reach the conclusion every time I looked at it. After talking to Tom about it a bit, he expresses interest, to some extent, in our project, but is unable to offer any help.

Fast forward a little bit and Tom publically announces a Flash client. Suddenly, the general consensus is that multi-platform isn't an issue anymore because of the Flash client. This is "Removes Enthusiasm #1".

Don't get me wrong, I can appreciate the effort that Tom and Lummox are putting in with the Flash client. It will be interesting to see a working server/client model with Flash that allows for larger-scale multiplayer games than have been previously done with Flash (that I'm aware of). The problem is that while it is a somewhat-ok solution to multiplatform issues, a Flash client will never be able to completely beat out a stand-alone client. With the stand-alone client, you have the freedom to actually play single-player games on Linux (using DreamDaemon to host and connecting locally) that you aren't allowed with the Flash client. You can actually play games with full interfaces as the developers may or may not have intended, or at least exactly as the equivalent Windows users would play it.

I said the last one was "Removes Enthusiasm #1", so that must mean there's at least one more, right? I mean, I did number it at after all. After Tom announced the Flash client, it seemed like people made a lot of unrealistic expectations about the Flash client. We'll call this #2. Will it allow BYOND's developers to develop for a wider audience? Yes. Will it bring in more players and more developers? Yes. Will that really help BYOND? No.

So now I seem like I'm contradicting myself. I made it sound two paragraphs ago like it would be absolutely amazing because it will provide a greater server/client functionality than seen in Flash to-date by myself, then I go off and say I don't think it'll help BYOND. The problem is that I don't see it really helping BYOND all that much. You're going to bring in new users, that's alright. Most of them will probably be better than the current user-base. You're also bringing in new developers. That's alright as well, but what type of developers are they?

Most of the new developers brought in by the Flash client are going to be the type that grew largely when people started realizing that Flash games could generate a lot of money. They're the ones that are great at making small games that will catch your attention for limited amount of time probably 4, 5, or even 6 days a week, when you're bored listening to some professor lecture about thermodynamics or what-have-you. They'll be good at making these games, and they'll make quite a few of them. Or at least, we can hope they will. We all love those games. This will improve the cash-flow to BYOND and actually generate some funding to the project, that's true, but how will it actually benefit BYOND?

Looking at the bug tracker and the feature tracker, I can see stacks of feature requests and bug reports lined up just waiting to be hammered. Will they get done? Most of them. The feature requests will slowly dwindle, as will the bug reports. What's left? A large player-base that absolutely loves these short Flash games, and a large developer-base that absolutely loves these short Flash games. Ultimately, you don't really have people pushing the engine to its absolute limits all-around trying to figure out what it can or can't do, what it should be able to or shouldn't be able to do, they're mainly in this one, now incredibly large, corner of BYOND. It'll get optimized to do what they want to do, and then it'll just sort of hit a brick wall.

Is that the only possible way for it to end? No. I've played out countless scenarios in my head, the majority of them leading to no real extension to the software outside of the Flash client, some even leading to Dream Seeker being mostly phased out or unmaintained. It could always go the other way, though, but we can't accurately predict that.

It's clear that my points of "Removes Enthusiasm" aren't in order of importance, because #2 isn't THAT big of a deal, where-as #3 sort of is. The Flash client comes out. It succeeds, it fails, who knows. I hope for the former. I, along with many others that I know here on BYOND, am the type of guy that likes to take something, study it, analyze it, try to figure out how it works, and re-create it. If there's three things that I like they'd have to be continued maintenance, improving compatibility, and improving efficiency, no matter how terrible I am at them. Writing a new DS parallel in Linux with C++ and Qt would hit not only the first and the second, but it would also hit the third as we could make recommendations to Tom and Lummox ways that they might be able to improve their protocol or other things.

The Flash client is already coming out, so that removes motivation (See: #1) to work on a native Linux, et al. client to parallel Dream Seeker because people are expecting the Flash client to alleviate multi-platform issues. What's next? Reversing the protocol the Flash client uses to communicate with DD. That's fun, for a little bit. Once it's completely documented, there's no where else to go. It's a new protocol, new technology. It's already probably going to be pretty dang optimized as it is. There's not going to be any room for improvement because the feature-set is set in stone until it's updated by Tom or Lummox, then that's another few hours spent figuring out what actually changed. Sure, it hits point 1, but there's not much of point 2 or point 3.

The point of it is, the flash client's going to come out and it's not going to be all that entertaining for very long (for me, at least). There's not going to be much more that I can actually do around here for entertainment. This is #3.

== End rant

As a summary, for those of you who decided to skip to the very bottom because it's a decent-sized book, the points of demotivation are:

#1. People feel that the Flash client is going to alleviate multi-platform concerns.
#2. The announcement of the Flash client generated unrealistic expectations from a lot of people.
#3. After the Flash client, there's not much left to do.

And to summarize my position on the Flash client itself- I support it despite the fact that I don't feel like it will actually help the actual development of the software that much and it'll essentially drive off anything left keeping me here. I appreciate the work you do, Tom, and the attempts that you are making to keep the project a-float. As such, I will continue to support BYOND by renewing my membership when it comes to an end (as long as I can afford it), but my general activity (that was already pretty low) will start dwindling. Thank you for providing me with an awesome starting point and experience.

P.S. I'm sorry to any that endured reading the entirety of this post. If you exist, you deserve a medal. I started writing this at 6:15 AM CST while having a hard time sleeping, yet being tired. As such, the structure and general efficiency in conveying coherent thoughts is probably non-existent and incredibly low respectively.