Browsed by
Author: Paul Tschirgi

Efficient Systems and Immersive Experiences

Efficient Systems and Immersive Experiences

Not trying to make any big technical claims, I just want to acknowledge that there is a battle between games with user-tools that work quickly and are ‘labor saving’ and emotional experiences for users from cutscenes to series of steps or tasks that take multiple player actions.

I think it’s really important to see game design in the realm of all meaningful choices versus all choices in our awareness, AND all choices which are not meaningful.

It is not always clear in life what choices are meaningful in some gray area between what to choose to wear and how it affects people’s perception of us to something like how long to hold eye contact which may or may not have an impact. Infinite numbers of choices. A good game systematizes choices with meaningful impact so that there is win and lose and also learnable skills to apply.

Villain vs. Omniscient Player

Villain vs. Omniscient Player

I may have a unique opportunity in storytelling within the realm of time travel.

Many games and movies exist in which characters are aware of the possibility or mechanic of time travel. Some even make references breaking the 4th wall in awareness of the players non-understandable abilities.

But what if a villain not only gets it, but specifically wanted to craft a plan to try to force the player to use this time travel loading ability for his own gain.

This could work by the villain having in game world resources to detect and monitor the uniqueness of the player and devise series of plans to manipulate and control the player. The example being killing your companion or girlfriend which would incentivize the player to go back and use mechanics to prevent it.

Now we can give the villain a trick to use strange words and test to see whether the player reacts.

Now you might say that since the game world has no existing actors besides the player that everything would be the same every time and so the villain would always do the same thing. However the villain could reasonably deduce that, if put forth as an incredibly intelligent person.

By that discovery the villain could also simulate randomness outside of the game world by monitoring the players decision making across the game and driving their set of evil plans based on those choices because they represent randomness.

In fact if the villain was especially dedicated they could basically do what encryption is and with enough randomness simulated by player injected choices it would be impossible to figure out the villains formula.

Now the villains systems for writing out all his possible plans and wiring enough of them to be executed or decoded would be exposed. Especially to a time traveling player looking to Snoop.

Probably the clear defense against this is for the villain to realize his system planning and decoding would be pursued by the player and plant massive amounts of false leads and false information through other people crossing over this faulty information.

Via this system as long as the villain is aware of the player being aware of time travel/the villains plan, the player can never fully avoid or outdo the villain in terms of in game consequences maliciously inflicted.

However it may be then possible to pick up clues about the problem to solve not being the systems, but actually the villains awareness of the players awareness of the plans the villain has.

Theoretically the player would then play a fresh new game without the + and use that incredible knowledge gained to avoid the first loop evil plan.

And so the lesson is taught to the gamer by the game that they should not act by the arbitrary rules of the world of the game, but by the full extent of their power as a true agent.

This is represented in the difference between using the in game technology to rewind current memories to old and using the start new game button from the launcher manually.

And I would never do this from a self righteous finger wagging villain. Or at least through player dialog options I would make the counter argument that games are still incredibly valuable, but we should be aware of the temptation to apply constraints of video games to ourselves that don’t need to apply outside of games.

Part of me does worry someone could do a good job of stealing this idea, but hopefully I’m the only one with the interest and capacity to go this meta.

The Big Computer Tool Revolution

The Big Computer Tool Revolution

In an age of functionally infinite information, some useful and some harmful, it is critical to sift. Understanding how to pick knowledge out of the internet is more valuable than retaining great amounts of knowledge locally/in memory. The problem is that no computer tools really offer a fun, powerful and expandable system for sorting or organizing information. Search engines were a big advancement, but their closed system based on their own prioritization even supposedly on behalf of other people’s measured interests feels like an ancient approach to me.

Surely the best tool of conquering the internet in a productive positive journey must involve a very serious wrestling with the organization and ordering of that information. I’m not talking about simple search engine customization, but breaking search engines into components of every possible or measurable variable including new ones added and calculated by any person or group.

I guess this isn’t too mind bending of a declaration, but it is important. We need to focus on getting tools to the average user where they can interact graphically with the set of all possible search engine customizations and compare the set of all possible results from all possible combinations.

For anyone thinking this isn’t realistic- we already do this. Just in a manual way, we search a couple terms look at the results, maybe browse a couple sites then often times change search terms. But search terms are a very imperfect medium for communicating what we are looking for to the computer.

It’s impressive it works as well as it does, but we should be much more interested in reexamining this relatively ancient interface. Typing in keywords and sifting through matches ranking by variables outside our control and some outside our understanding is not the end all for information collection.

Games v. God : Discarding the Author’s Intentions.

Games v. God : Discarding the Author’s Intentions.

To myself and all future Electra writers,

Narrative is a long tradition, spanning thousands of years. One way we understand stories is as a mechanism to pass on knowledge to future generations in a compelling and fun way.

But what happens when the writer reliquishes total control over the narrative? When the writer is creating a system with a set of inputs and outputs?

Well at that point it definitely becomes more complex to interpret the author’s intentions. It definitely still can be done, with games for instance in almost all cases we see outcomes as good or bad endings.

Is this meta thought process of what the author wants, from the player’s point of view a good thing?

Well how about the parallel of reading the Bible?

Religious experts spend their entire lives trying to interpret the intentions of God through the text of the book.

So clearly at least a large number of people find incredible value and meaning in trying to understand the writer from the writing.

Now if we apply gamer critique psychology to the Bible, you might find some comical comments;

“God is overall clearly a gifted writer (including his writing through selected individuals), but man sometimes he really goes on and on about the lineage of some very minor characters. I’m all for the backstory of King David, but listing off a dozen generations of names really pulled me out of the immersion of who he was.”

Or more to my point about discerning intentions and reacting to them:

“God seems to be really fixated on driving home the point that idol worship and drifting from full commitment to him is bad and you’ll get invaded if you get too distracted or into sin. I mean first off God seems to be correlating kingdom strength to their devotion to him at the time. Classic example of correlation is not causation. In fact the pattern of attack, kingdom falls, kingdom returns seems a lot more connected to the strength of neighboring enemies and the overall strength of the believers forces, not the per capita worship rating.”

Let’s reframe using concepts.

Understanding a story is one level of interpretation. Trying to understand the author’s intentions based on the meta of the story is a meta level of interpretation.

Emergent Properties of Language

Emergent Properties of Language

We generally understand that life is an emergent property. A single atom is not alive, and not all combinations of atoms in a large enough group form something that is alive. But with plants, animals, and humans and our specific collection of many many atoms, the emergent property of life applies.

Now within interactive storytelling I think there is a serious misconception along the same structure. A single word by itself has a use or a definition, it is like an atom. It definitely exists, but it does not do very much on its own. I would argue that 1 word by itself does not contain meaning. By meaning I mean a concept or idea within a specified context. 1 word by itself is more like a topic for a brainstorm than representative of a consistent part of a whole sentence which has meaning. To make things simpler, assume it took 100 atoms to reach the critical mass at which the emergent property of life was part of the whole. 1 atom does not contain 1% life. The atom is in no way by itself alive. In the same way a sentence built of many words (assumingly constructed just like atoms to life, correctly to achieve meaning (life))-

A sentence built of many words at that point has meaning. It has an idea within a context. But no single word has 10% of the meaning. The meaning is only a property of the whole.

This is the error that the architects of many previous interactive writing tools have made. They think that because most sentences can be altered by changing one word or many different one word segments within a sentence, that this is the correct level of interactivity for the user. I think if we reflect more critically on the formulation of language as the vehicle for communicating ideas then we may realize this is a poor mechanism to systematize human language use.

Swapping out individual words of a sentence is meaningless because the individual words do not have inherent or consistent meaning. In fact sentences are not even reliable conveyors of meaning. It is a great tool for getting across meaning, but in practice has to be restructured multiple times both from the speaker and reiterated by the listener to achieve a reliable shared interpretation of meaning.

It is not lost on me, that I’m trying to communicate a consistent idea across many sentences, but with enough attempts at the same idea structured in different ways- if you do a good enough job the idea can be seen as the pattern across all your attempts to express it in language form.

Through this theory, exchanging words within a comparatively fixed sentence structure is futile. Because the words themselves do not have a consistent meaning across all sentences, and therefore the words themselves are not useful tools to the player. I do not mean to say swapping out words leads to the same meaning in a sentence. In fact very obviously if we take the sentence ‘I love her’ and made it ‘I hate her’, we clearly have a completely different meaning.

But the question becomes what the pros and cons are of extending to the player the option between ‘I love her’ and ‘I hate her’ and the pros and cons of extending the action words of love and hate which can be swapped into sentence structures with the blank of ‘I ___ her’.

At first having the swap mechanics of ‘I ___ her’ open to list of verbs (love, hate, admire, envy, ignore, etc) is really appealing. Almost immediately we see the parallel to normal game mechanics where simple rules are exposed to the player to use in any and all situations.

But! If the mechanical power we are trying to extend to the player which we also want the game to understand is the ability to express a set of meanings, we have a serious issue of interpretation. I love her is actually pretty complex and means a lot of different things in different contexts as far as the actions you are hoping to evoke on behalf of your own character players as well as others. Is it romantic, is it familial, friendship, mocking, hesitant, angry? There are contexts where it could be all of those things.

And if we operated on words alone, as if they have consistent meanings across all sentences with the format to swap in that word, then there’s actually a really high probability that in the most important moments of a game based conversation, you completely disconnect from what the computer understands about the word love you’ve been swapping in all game.

What seems like a sincere confession, could actually be a despairing release of the feelings, hopeless for the prospect of reciprocation. It could be used as emotional manipulation to reinforce virtuous feelings of care and demand action or change on behalf of the other person. It could an uncomfortable forced and panicked response that upon deeper thinking turns out to be empty.

And normal conversation is a lot more complicated than even this in my experience. So what’s the solution?

Stop using swappable words into dialog choice. The meaning is at a bigger picture level just as life starts beyond the atomic level. It may be impossible to communicate through language without a consistent risk of misinterpretation. I think most of us have experienced how surprising someone’s interpretation of our words can be and it’s a separate topic the degree to which language can be improved or made worse in search of maximum or optimal clarity, but when the specific meaning of the language is important in strongly divergent player options, players have a right to know what they’re saying.

One of the best solutions I can offer to this problem is to have multiple sentences that all carry the same meaning, but with different expressions. So I’ve basically done the opposite of the traditional word swap approach, by creating multiple different sentence versions that all attempt to communicate the same idea, rather than different ones.

And then I also allow multiple options that have these multiple expressions. Through this process the player actually gets a very clear idea of what they are contemplating saying within the game. Don’t swap words, swap meanings and use expression type to further communicate the meaning.

Interested how this works in practice, check out my recent rundown of my writing tool Electra:

Feature Rundown

Feature Rundown

I just put out the latest version of Electra and decided to do a crash course of using the tool and talking about some of the exciting features my writing program contains!
For those of you who have not yet seen it, you can watch it below, don’t forget to subscribe!

Feature List

  • Auto updating patcher/launcher hub program
  • Multiple game modes to load writing files into
  • 7 Personality defining traits
  • Character list organization
  • Nested Flow Chart- Every step is a pairing of player and ai with depth calculation
  • Tone dictionary 132 combinations based on 12 parameters top 2 absolute value
  • Easy Enter to Enter typing for fast lengthy dialog creation
  • Text messaging output window for easy simulation of possible paths and understanding long conversations
  • Calculation window for understanding characters reaction to player choices and selection of ai response
  • Multiple filter and chunk modes for variability in flow of conversation
  • Meaningful monologue where multiple choices in a row can be chained into a personalized chained response back
  • Premise variables of Location, Action, Tag and Time, Situation, and Relationship
  • Ai initiated conversations based on action events
  • Revolutionary meaning and tone grouping structure to organize deeply expansive dialog choices and be able to manage the outcomes
  • In game support for topic switching (tree branch jumping to ongoing conversation trees without direct connections)
  • Conditions for managing what topics are accessible when or based on certain variables like relationship
  • Multi character conversation support
Self aware replay game theory

Self aware replay game theory

Imagine a game with an usually broad series of outcomes with a huge number of in game characters.

Perhaps at the end of the first playthrough you reflect on the outcomes you experienced. What if you were to simulate the game being played by an AI that perfectly emulated the decision making process you yourself employed when you played the game.

Let’s call that emulation of you, your personal emergent behavior.

What if the game included an end game set of god-mode tools to understand and edit that personal emergent behavior.

Decisions like dialog options, who to fight, who to team up with, who to pursue romantically were all predictable and predestined according to the AI behavior that emulated your playthrough.

Now in editing the core structure of your personal emergent behavior it resimulates the entire game.

Again let’s imagine a game with an incredible amount of disparate outcomes such as some characters die or live, love or hate the player character, etc.

Imagine in editing yourself being simulated in the playing of the game to make decisions according to different rules or priorities.

Imagine if you could then preview the outcomes of that alternate version of yourself by any criteria you could think of.

For example you change your personal emergent behavior ai to prioritize saving lives and the outcome does indeed assure none of the characters die.

What if you could then replay the game, BUT the simulated decision making version of yourself that prioritized other characters surviving would tell you what to do in order to assure that result.

Now in your guided playthrough there is no certainty that every single suggestion is required to achieve that outcome, but any decision could be.

I think that would be a fascinating way to replay a game.

Comparing all decision making priorities with all desirable or undesireable outcomes and experiencing the game pursuing or avoiding those paths.

Designing design principles

Designing design principles

Everything can be analyzed from a set or group view. Iteratively both reflectively or predictively.

Good design! What is good design? Well that depends on what your goals are, and then how you rank the priorities of probably a set of goals.

For a lot of artistic mediums, aesthetic effect and conveying themes clearly is top priority, but within games the top of the wanted list is much more contentious.

Easy to learn? Well that can be a trade off with functionality or effectiveness. User error friendly?

I think it’s useful to focus on the defining characteristic of video games which is meaningful choice making in a specific context.

By that central mission, designing game mechanics should be first and foremost enhancing the key decision and not worsening it.

So how is this applied practically? Well things that get in the way are long repetitive animations that occur between user inputs because it creates an artificial window that the user cannot control, but will still contain real time consequences. However animations that give users information ahead of time leading to better or faster user decisions later are great.

But that’s something I needed to figure out because my last several attempts and multidimensional conversations in my dating game have been pretty excessive in animation and not enhancing the gameplay essence which is deciding between options and understanding their different properties and similarities and result effects.