First, let me say that I did not enter the competition because of the restrictive rules of built-in content only. If you allow content from the DLS which is not built-in, you have a much better chance of getting some real competition going.
We are thinking of allowing rules and driver orders from the DLS. We still have a focus on built-in content (and thus small downloads) - locos and rolling stock still need to be built-in items.
All the confusion expressed in this thread about what makes a session and how they are created drives home the request I made on the original Contest Anouncement thread - that it would be helpful to have a specific Forum category for Session Creation. One place where we could share ideas and discuss problems and solutions to creating award winning sessions. If you are serious about showing off this fine feature of Trainz, surely it deserves its own Forum Category.
Looking at what we have at the moment, we have Content Creation Support (and making sessions is content creation), and the scripting forum - which is rather more at the 'making rules and orders' end of the spectrum.
We could encourage discussion of sessions in Content Creation Support, extend the Scripting forum to also cover session creation, or create something new. We'll have a think about this
I have looked at very few session from the DLS because the ones I have tried are so lame. Go to industry A, pick up cars, go to industry B, drop off cars, etc. Surely you are looking for someting a bit more challenging. Can you give us some quality examples of what you are looking for?
I think the key thing is to set the player a task, and check up on how well they are doing it.
What the task is, and how you define 'well' is up to you.
Can you expand on the "black boxes"? What defines a black box?
You won't hit this unless you've done significant work outside of surveyor - either by writing your own rule(s), or creating your session using a 3rd party tool rather than the surveyor interface.
If you are writing your own rule:
* Make it perform a clean task - if you are writing one to detect badly formed consists for example, do only that task - and trigger subrules to take action when the condition is detected. Don't apply a fixed penalty to your own scoring system from within your rule's code, as someone else may have a different scoring system, and want different action, yet still want to use the same test.
* Don't hard-code anything inside it. Don't record the name of any object in the map or session within the code for the rule. Make sure they can all be specified in surveyor, via the rules interface.
If the rule is good, creators will want to use it in sessions for a number of maps. If something in it's coding means it can't do this, then you have a problem.
Is there an optimal time limit you are looking for?
I'd say between 30 minutes and 2 hours is good.
That's not to say that a session that takes only 15 minutes can't be fun, or that people won't play and get enjoyment from a session that takes an entire day, but most people will want something within roughly the above timeframe.
The longer the session, the more likely people are to want to save it and continue later, so if you've made a long session, you need to pay close attention to saving and continuing from a save.
Making save work properly is more the realm of rule creators than session creators, but you will want to test it anyway, because you might have used a rule (particularly a 3rd party rule) that doesn't work properly in this regard. If you can either avoid this rule, or persuade it's creator to fix it, it may be beneficial to your entry at judging time...
What if the session is open ended, permitting operations to go on for multiple days?
Open ended could be a problem. You need to set the player a task, and check on how well they do it.
If you have a session that sets a task, and watches (say) about an hour of player activity - giving them a message (and maybe granting an achievement if they complete it perfectly), and then allow them to play on afterwards, that's fine - but if you use "open ended" as an excuse to
not check up on the player to find out how well or even if they are performing the task, then that's not good, and will count against you. In the extreme, it could mean your session is regarded as "not substantially complete".
Are prototype operations important or can I create any silly way of running the railroad as long as it is entertaining?
The judges will be looking for an enjoyable experience from the session. So the question is "Does prototypical accuracy have an impact on enjoyment".
There's also the related question of just exactly what is fun.
Speaking personally, (and I'm a great fan of prototypical operations), the danger area for me is in trying to be prototypical and failing. Or starting out with a high level of prototypical behaviour and having that deteriorate throughout the session.
However, start out obviously unprototypical, and that's not a problem. As an example, I distinctly remember a long time ago a scenario about crashing as many 'krokodil' locomotives as possible in the shortest time. As I recall, you got extra points for speeding and passing red signals, as well. Utterly unprototypical - but also
hugely fun. And memorable too - it was for UTC, and if I can remember it a good 8 years or so later, it must have been good