What follows is my analysis of what happened. All TSP members may add their own or comment
as they will.
In Dec. 2000, the idea to make a System Shock 3 mod with Shock2's engine, ShockEd, was launched at Through The Looking Glass' forums. The project was to create a mod with a SS3 setting, in order to make what could be considered an unofficial System Shock 3, as the chances to see a "real" SS3 released one day were (and still are) very slim.
At that time, most of the ttlg members who would soon participate were involved in a RPG recreation of System Shock 2 based on logs, posted on the forums. Called the "System Shock 2 revival", it was also an occasion to assess the talents of the wannabe writers in terms of style and skill. A little group was formed, and soon the title "the Shock Project" was used to designate the mod.
I didn't know ShockEd at that time, I could only write texts, that is plot and logs. I also planned to do some voicing, as I assessed by the time of the release I would have a microphone. The project's first months were pretty shaky, as there were disagreements and discussions about where and when the story would take place. Would it be on a planet, on a ship, why, would Becca and Tommy survive or die, etc. The basic plot, which would eventually be used, had already been elaborated by JediK and Zach Jepsen. Finally, it was adopted almost as is. The Shock Project would take place on a remote TriOp secret base, on an icy planet, and the player would be a TriOp or UNN member with special augs, part of a project conducted in this base. The rest was just coating.
We lost a lot of time discussing the plot. Not to say that discussion should be avoided, but even long after some things seemed certain, there were still some (me included sometimes) who believed another version was still being used. Although in the first months of 2001 there was a lot of discussion over AIM, the plot wasn't really defined, that is, REALLY complete, until mid-2002, almost one year and a half later. This lead to several discrepancies in logs and level design, which, although they were technically not that hard to correct, slowed us down a lot.
For example, the chronological chain of events that made "the Shock Project" was always under discussion to know where the NPCs were, where the player should go, and what path he would go throughout the game. The most severe misunderstanding happened when discussing the enemies. For more than a year, several people, me included, believed the secret project the player was a member of, "Osiris", would provide the main stock of enemies after Shodan's attack, after their AI would go berserk and trigger their defence mode, whereas the other "faction" always thought they were going to be the victims, the main enemies being mostly the Fort's robots. We had to change the formulation of several logs in order to correct this, and we lost precious time trying to re-imagine the project's plot from another point of view.
During the first few months, everything was cool. We had AIM (instant messaging) meetings up to thrice a week, sometimes even more, and the project was going very well, in spite of the usual little disagreements that occur anytime a team is formed for the first time. Up until mid-2002, the project went on really well. Strangely, it was after the creation of proper forums, which we though would increase our chances of success, that the project started declining. Several members never bothered to post or even register, and stuck to the project's "old" Y! Group. (This is not a critic per se, merely an observation, but I still can't understand why.)
Emails replaced IM as the main communication channel, with all the backdraws it involves: delay, random redirection (from the Y! Group), several reactions to one single email scattered among a long list of other issues... We lost sync, and started to do our thing without really bothering to check what the others were doing.
Communication I think is what we lacked most, and I'd say it's the main reason of the project's failure. The fact we didn't really know about the others made us do it without really caring, I'd say. Some peoples, namely the music guys, did a wonderful job because they didn't really need to talk, as once they were told the basic mood wanted for a given track, they could do it without much further discussion. But with levels and logs, communication was the key. Level designers could do a level, but they had to keep in mind where the NPCs, where the keys, where the plot-critical objects should be. Log-writers had to know the levels in order to modulate and to adapt, so that they were coherent with the environment. And that didn't happen.
We could say time and dedication were also missing, but my opinion is that it's this lack of communication that killed the project.
3- Time & Dedication
These two go together. The more you're dedicated, the more you'll spend time on a project. In the beginning, everything was almost too perfect. There was too much enthusiasm, and that lead to the few clashes I mentionned. But it had a positive effect. During the golden age, we kept discussing about each and every aspect the game would have. Understand it, we were youngs, mostly teens, embarked in a great adventure to recreate System Shock! Pardon my French, but it was fucking great!
But then came the time when we didn't discuss that much. I remember reading my emails and only nodding "Huhum, good idea" without bothering to reply. I personnally had to work on other things than TSP, Uni projects, internships, and RealLife affected the others too... Little by little the emails became rare. By the end of 2002 it was down to three or four per month, when we wrote about 20, 30 per day in January 2001.
In the course of the three years the project lasted, there also were several changes in the team. Some members didn't go on and disappeared or left, most of them giving perfectly good reasons (RealLife's a bitch) but leaving a gap that would have to be filled. The new members, when we found them, had to be briefed on what was done, what was left to do, and what we expected them to do, and this also delayed the project.
Then, even emails disappeared.
The less you get emails, the less you'll feel motivated. The less you'll feel motivated, the less you'll write emails. The same thing applies to all. A few of us tried to rejuvenate the project, several times. New forums, new site, new screenshots... Now I think of it, and I've already said that in the other thread, if we had wanted to, we could have released the game for Christmas in 2002, or perhaps Easter 2003. I mean it. Instead, we lost time trying to remotivate each other. We didn't want to believe the project was dying. Sadly it already was.
The side effect is that we showed nothing of this to the community. For everyone, the project was going on, slowly but steady. The lack of updates didn't mean we were dead, it meant we were busy, which we repeatedly told them and ourselves.
We managed to complete a lot of things, but another mistake we made was to leave all the details for the end. In spite of them being the final stage of design, we didn't set up objects and logs placement, TriggerTraps, Quest&Objective marks as we designed the levels as we could have done. It would have been a better solution though. We did the easy part first, and what was left was the part that'd require a lot of dedication and time, both things we hadn't anymore. The fact nobody posted anything at the forums or elsewhere didn't motivate those who were "left".
4- Time & RealLife
Another two that were kind of forgotten. As I mentionned in the previous paragraphs, RealLife didn't help. First, the organisation of IM meetings is not really easy when the project's members are scattered around the globe. Australia, Europe, and the United States: a meeting for the three of them at the same time is a pain to set up, and it's not even guaranteed that all will attend. None of the project's members were professionals, they were, basically, a bunch of kids doing that on their spare time. Add Uni, school or college works, exams, internships to find and arrange, or just the everyday tasks, and you'll break the schedule to pieces beyond repair. After all, technically, the project wasn't a priority compared to these. Objectively, it's normal that everyone should first do their own business before doing their part of the project.
From mid 2003 (IIRC) on, there started to be open talks about the end of the project, but not the successful release we dreamed of... The other end, the one we didn't want. In December 2003, JediK took his decision, and left the project. He had been the leader all along, so it was a major blow for the remaining crew. I tried to take over, without realizing how much was left to do. I tried to add plot-critical objects to some levels, but either I couldn't make it work, or it took ages to make something functional. I became aware that it wasn't a question of a couple of months, but more of about a year or more, before I could get something done that could be released. And I couldn't work on TSP that much anymore.
This is the reason why, after discussing the matter with JediK, we decided to release TSP "as is" last week. I wrote this to explain our decision, to try to explain why we failed, and to help those who may set up another project like this.
We lacked motivation, because we lacked coordination. We lacked coordination, because we lacked communication.
Communication is the key. On Internet projects, people can't meet at the office to discuss. They can't call each other on the phone unless they live in the same area, which is a matter of coincidence. As long as we could IM each other, there was a flow of ideas and suggestions and this interaction was the heart of TSP. When it stopped, the fuel was gone. We ran on the reserve, then it died. I may be oversimplifying, but more or less, it's what happened.
On an Internet project such as this one, there should be a minimum amount of IM meetings so that the staff don't lose track. Keeping in touch allows to share ideas, to assess the development and to re-distribute the tasks if needed. The creation process is maintained according to the needs, and the projects stays alive. Take this away, and you'll have a bunch of people working without knowing if what they do is still what is needed. A great idea, a major bug will be subjects to notifications, but the rest will go unnoticed.
Communicate, and the bond won't be broken. And your project may have a chance to survive.