McDonalds is the most successful restaurant chain ever. Does not mean it is good for you. They are just very good at selling crap to the masses. Same for Blizzard.

...they do cater to the masses...it's not a matter of McDonald's and Subway.

I think that, if one is going to use a food analogy, a much better comparison would be:
WOW = Starbucks
PS = local neighborhood coffee shop
It's still suspect but, as they say,all analogies are.

BTW, I've never played WOW; I looked at it, but it wasn't really my style (PS, on the other hand, is).

I looked over the page for "Join the Team". I noticed lots of openings, but I don't know that I fit into any of them. I am also a programmer, but am not free to program for the project because of the no compete contract I have with my current employer. I am, however, free to offer my Linux system administration, system maintenance automation, and security services to the project. Which link should I click?

It's actually a bit academic at this point: those links all lead to laanx, which is down, so you can't submit an application until it's working again, which could conceivably happen faster if you were able to help, which you can't do until you apply, which you can't do until laanx is back up, which could conceivably happen faster if you were able to help, which you can't do until you apply, which you can't do until laanx is back up... and there's a hole in the bucket, dear Eliza, dear Eliza.

Edit: you're at nine posts, so just reply to this and then pm me.

This makes nine, which apparently isn't enough for PM.  Here are my directions:

From Hydlaa to Gugrontid:

Exit through gate by Octarch's tower and go into the woods.  Go left at the fork in the road in the woods.  Shortly after exiting the woods through a short canyon, the road will curve to the right--instead of following the road, go straight ahead over the ridge.  You will see a large stalagmite in the distance: the entrance to Gugrontid is to the left of the base of that stalagmite.

From Gugrontid to Hydlaa:

After exiting the canyon out of Gugrontid, you will see a fallen stalagtite ahead and slightly to the right.  Go towards it until you see a large boulder ahead and to your left.  Go past the boulder and continue over the ridge, then turn right onto the road on the other side.  Once in the woods, turn right at the fork in the road and follow the road to Hydlaa.

It should be kinda difficult to get lost from place to place. There are clearly marked paths from place to place (note they are different colors), and there are signs all over the place.

That statement is, at best, a misleading half-truth.  Let's break down some of these "clearly marked paths" and "signs all over the place:"

Kran city to Hydlaa: if you ask any of the Kran how to get to Hydlaa, they will tell you "go out the gates and turn right at the fork in the road."  So, you go out through the stone "gates" into the next region and...road, what road?  There's just a great big empty green area with lots of hills, a few trees, and a lake.  If you just turn right and start going, the edge of the area will gradually force you to curve your path around to the left until you hit the stone "gate" to Hydlaa.  (Note to newbs: this is actually a very reliable, albeit excruciatingly slow way to get around: follow the edge of the area and you will always end up a at a "gate" to the next region).  If you try to go the other way, from Hydlaa to the Kran city, you emerge from the stone "gate" onto a very well defined road: it's a different color, and it has steep walls closing it in on either side.  Unfortunately, following this road will never get you to your destination.  You just have to know that at some point, you have to defy all common sense by climbing up that steep wall on your left and setting out cross-country.  As for signs, there are a few along this route, but none of them mention either city.  All of these signs just sit right next to some sort of geographical feature and provide it's name.  While knowing that I'm at "Cutthroats Bane" certainly adds some nice flavor to the game, it doesn't tell me how to get where I am going.

Hydlaa to Ojaveda: exit the gate behind Jayose's library and just follow the road.  This works pretty well at first because the road does seem to be reasonably well defined by its different color, but then the borders of the road become vague in a few places, and there are even some small branches here or there.  You think to yourself, "Should I take those branches?  No, I'll just stay on what looks like the main path."  And then, you get to a large, round area that's all the same beige "road" color, with several "roads" branching out from it.  Which one do you take?  Well, fortunately, there are some signposts.  Oops!  None of those signposts say anything about Hydlaa or Ojaveda.  You'll just have to guess!  Isn't getting lost just one great big fun adventure?

If the intent is to make this a world with some verisimilitude, then the roads and road signs are an abject failure.  We are talking about a society that, at this point, has just three major settlements plus the bronze doors.  Such a society would have very big, very well-defined roads between those settlements to facilitate travel and trade.  There would be clear signposts at every intersection to tell travelers which way to go.  If you want to take a page from the ancient Romans, you could even put mileage markers along the roads (10 leagues to Hydlaa...8 leagues to Hydlaa...6, etc.).  There would be inns along the roads, certainly at every major crossroads.  There would be a constant stream of NPC traffic going in both directions, especially trade caravans.  I realize that those last two (inns and NPC traffic) will most likely not come until a later stage of development, but the roads and signs really need to be improved as soon as possible.

This thread actually hits upon a point I made in another thread in the general discussion forum.  To paraphrase a much earlier post in this thread by Xillix, a lot of different people seem upset about different aspects of the economy, often for apparently contradictory reasons.  Some might say that the when some people say "too much" and others say "not enough," then the balance is just right and the people complaining are just the extremists.  I don't think that this is the case.  I think that the economy in Yliakum is so completely out of whack that a majority of people (or at least a large minority) are upset about it.  Why is this?  It's because the economy is centrally managed.  Historically, centrally managed economies always perform poorly, even ones that have relatively good central management.  The management of this economy is at the extremely poor end of the spectrum because the people running it are apparently oblivious to the "law of unintended consequences:" economies are very complex things, so when you change one thing, other things that seem completely unrelated will change as well, often in a negative way.

If I understand correctly, the designers are trying to set up the economy with certain goals in mind: primarily, they want to encourage player-player interaction and they want player-based market prices.  They have made certain tweaks to the economy in order to achieve these goals.  Unfortunately, from the number of unhappy people out there, these tweaks are having undesireable side effects, and are possibly even havng an adverse effect on RP.  I have a proposal for a system that will fix most of these problems while still encouraging players to buy and sell from each other (instead of NPCs) and also make the market truly free and player-based:

1.)  Have the server track all player-to-player trades and store them in a matrix, which would have a size roughly equal to [number of different types of items that can be traded]x[number of transactions that take place in a given period of time].  This is actually not all that big compared to some of the other data that is running around on a server like this.  The size of the matrix can also be controlled by setting the "lookback period."

2.)  At regular intervals (say, once a day), perform a least-squares fit reduction on the matrix to find average values for all goods in terms of tria.  This will work well as long as a sufficient percentage of player-to-player transactions include at least some tria.  This is actually very straightforward math: I can do a small matrix by hand in less than a minute.  A big matrix requires computers, but this whole thing is running on computers, so no problem.

3.) Set all NPC store prices according to the average player-to-player prices from the least-squares fit reduction.  Have NPCs sell at a slightly higher price than players (say, 120% or 130%) and buy at a slightly lower price (say, 80% or 90%).  Remove the stock controls on the number of items an NPC can sell, but restrict NPCs to only selling mundane items, so that the only place to buy high-quality items is from other players.  Players seeking to buy mundane items will have a choice: if they are in a hurry but don't care about price, they can buy from NPCs.  If they are low on cash but have plenty of time to find a seller, they can buy from other players.  Players selling their goods can do the same.  This will give Yliakum a true, player-based, free-market economy.  If you still feel that too many people are doing business with NPCs, increase the price margin between NPC prices and market prices, which will provide more incentive for players to do business with each other.

4.) In the tutorial, designate a location in each city as "the market" and encourage newbs to go there to buy and sell gear.  Spread the word to existing players, as well.  The central plaza in Hydlaa would be a great place, and you could even set up sign-posts designating areas for different types of goods.

4.) In the future, use supply and demand if you want to effect prices.  Want iron to be more expensive?  Make it harder to mine, thus decreasing supply.  Or, require more iron to go into the production of popular items, thus increasing demand.  Either will make iron more expensive.  Be warned, though:  either action will also make iron-based items more expensive.  Want to make iron cheaper without changing the price of iron-based items?  Make iron easier to mine but require more to go into the production of items.

5.) As long as most prices are set by the free market, you can control the prices of certain items directly.  For example, gold and platinum don't have a whole lot of utility: they are mostly only good for selling to make money.  This makes them, in essence, a form of currency themselves.  You can control their prices through the NPC prices: want gold to be worth 250 tria per lump?  Have NPCs buy unlimited quantities at 245 tria per lump and sell unlimited quantities at 255 tria per lump.  If you do this to commodities with utility, however, your price will cascade through the rest of the economy: if you set the price of iron hard and fast in this way, items made from iron will adjust their prices accordingly.

6.) Make different methods of money-earning relatively equal in output.  Right now, even players who aren't really interested in mining do it anyway because it is by far the best way to make money.  This certainly does not encourage the "individuality" that the developoers claim to desire.  If you have taken the first five steps above, you can use fixed-price commodities to help maintain this balance.  For example, adjust the difficulty of mining gold and platinum so that mining a lump of gold or platinum should take about the same amount of time as any other activity that would produce the same profit.  In other words, if a lump of gold is worth 250 tria and the hide of a particular creature is worth 55 tria, mining one lump of gold should take about the same amount of time as killing four or five of that particular creature.  In this way, each player can choose his own method of making money without putting himself at a disadvantage because all means of making money will produce roughly equal profit for a given amount of time invested.

Before I get into my counter-responses, let me re-iterate that I didn't really want this thread to be about my ideas, I wanted it to be about everyone's ideas.  I was hoping for more of a brainstorming session, where just throw out their own ideas instead of commenting on the idea of others.  I hope that people will read each other's ideas, and think about them, but commenting on each other's ideas will quickly tie this conversation down to a handful of topics and stifle new thoughts.

All that being said, I will now proceed to comment on previous posts.

Quote

so NPCs buy at 130 to 140 tria and sell at 90 to 100 tria.

Why would anyone buy high and sell low?

Oops.  That should have read: "...so NPCs buy at 90 to 100 tria and sell at 130 to 140 tria."  If you read the proceeding paragraph, you'll see that I described it properly the first time and only got it backwards when I summarized at the end.

Prices:
I think the idea of basing NPC prices off of the player market is a VERY good one (although the player market is somewhat inflated right now, when will the bubble burst? ).

I think that prices in the player market are not inflated at all: they are perfect reflections of the value that people place on objects in the game.  I will, however, agree that the relative value of many items is out of line with what most people expect.  Why is that?  It's because this economy is, for all intents and purposes, centrally managed.  Historically, centrally managed economies always experience huge price distortions for some goods and severe shortages for others.  You may have noticed that both of those problems exist in PS.  The first step in solving this problem is to track prices, so that you will know definitively what effect, if any, your tweaks make on prices in the player-market.  Take iron for example.  Iron is expensive because it is required in the manufacture of some very valuable items.  Two steps will fix this:

1.) Require more raw iron to go into the construction of more valuable items.  Right now none of the smiths out there will waste iron making cheap weapons, because they can use that same iron making more valuable weapons.  Scaling the iron rquirement for items according to their value will restore the incentive to create all kinds of weapons and armor.  Unfortunately, this step alone will not lower the price of iron, it will just drive up the price of high-value items even further, which is why step 2 is needed.

2.) Make iron easier to mine.  If it's easier to mine, there's more of it, and we all know how supply and demand interact to set price.  This will also solve part of the drudgery problem.  As with step 1, this step alone will not solve the whole problem: if you make iron easier to mine without taking step 1 as well, you will flood the market with cheap but powerful weapons.  Oops.

From this example of iron, you can see how a free-market economy can still be influenced by the developers through tweaks to the ease of acquisition (mining, etc) and utility (how much is needed to make something) of goods: ease of acquisition directly effects supply and utility directly effects demand, which in turn combine to set prices.

As far as the math goes though: Crystal Space (aka the 3D engine) contains some matrix math stuff already, but it only works for small matrices.  This leaves the programmer types with two options: pull in an existing (C++) numerics library (Boost::uBLAS, Blitz++, or POOMA) or roll our own linear-algebra code (probably based on the STL valarray<>).

The client runs on Crystal Space, but my proposed price-tracker would have to run on the server.  I don't know a whole lot about the nuts and bolts of the behind-the scenes stuff of PS, but I would have to assume that the server is running on some very different stuff, probably mostly home-brewed.  Depending on what they are running, there are some great math packages out there.  If they can somehow incorporate scripting language into their code, they could just steal directly from www.sagemath.org, which is opensource.

Quote from: svuun on February 07, 2008, 05:34:29 pm

Ideally the npc should understand meaning but that's probably more difficult to code.  I'm hoping in the future that the NPC's will understand multiple phrases at least and therefore make a step closer to understanding meaning as opposed to exact "input."  I remember playing king's quest back in the day (late 80's) and exact phrases weren't needed, but multiple variations of words and phrases were understood by those npc's.  There's no reason why that can't be today.  Then again that game was purely quest driven and probably had a decent engine to parse sentences so was sophisticated in that aspect.  That being aside allowing NPC's to understand multiple phrases at given points in the quest would bring them closer to understanding meaning and add reaslism.  A good example of it currently working is how we ask for quests-- there are multiple phrases.  I think this should be employed within various points of the quest too.  crap, this turned into a wishlist item. ah well.

There is no reason why that is not today, because it already is . Quest responses are parsed to allow at least some synonyms. ALso, many quests allow for different responses, or even a single word from a response. If you find a case where an answer makes sense but is not accepted, look here how to help solve that.

I agree that this problem is best solved by setting up the parsing system properly.  I disagree that quest responses are currently parsed in a manner that is even remotely acceptable.  I have seen instances where a single word from a response is enough to trigger the proper NPC response, but I have seen far more cases where EXACT verbage is required.  Many people on these forums say, "All you have to do is take careful notes and use the exact words of the NPCs."  That's one of the stupidest things I've ever heard.  I SHOULDN'T have to write down exact verbage (or check my logs) because it is NO FUN.  This is a game, so above all else it should be fun.  Moreover, there have been times when I DID write down the exact verbage from the NPC who issued the quest, but ended up having to use a different verbage to get the correct response out of the "target" NPC.

Here's a specific case: there is an early quest that involves taking a drink to someone and then reporting back to the bartender if that person liked the drink.  I said to the bartender, "She liked it."  No dice.  OK: "She liked the drink."  Still doesn't work.  How about: "[Insert person's name here] liked [insert name of drink here]."  Nope.  I finally got it by saying, "[Insert person's name here] said she liked the drink."  The difference between a successful sentence and an unsuccessful one?  The word, "said."  That's inexcusably bad design.  A good parser would have accepted any combination of: ("[person's name]" OR "she") AND ("liked" OR "enjoyed") AND ("drink" OR "[name of the drink]").  That would be very easy to code, especially if you ignore the order of the words (which you should ignore, because English grammar is WAY too complicated, with too many possible word orders for the same meaning).  If quest-related conversation were parsed in this SIMPLE way, most people would trigger the desired NPC response after one or two attempts, every time.

Another specific case: I'm currently trying to complete the prank quest for Taulim Wilaal.  It started with him telling me to go to a specific person and get a specific item.  I found that specific person, but can't seem to get him to understand what I want.  I spent twenty minutes typing in different phrases, all to no avail.  I know exactly what I want from him, but seem unable to communicate that fact to him.  The item I am asking for is simple enough that it should be very easy for a parser to pick out a single key word from my phrases and trigger the proper NPC response, but it just isn't happening.  One common way to deal with this (I am extremely surprised that I have not seen this implemented at all in PlaneShift) is to have an NPC respond to a key word in an otherwise suspect sentence by saying, "Were you asking about [insert quest item here]?"  Of course, you have to be careful with this and take measures to prevent "quest-fishing."  An easy way to control this is to set quest flags.  For example, if Bob goes to Jane and gets a quest assignment from her, a flag associated with that quest is set for Bob.  Let's say the quest is for Bob to fetch the magical feather from Adam.  Bob can walk up to Adam and say just about any sentence that contains "magical feather," and Adam will say, "You want the magical feather from me?"  Now, if Sharon hasn't gone to Jane to get the quest yet, and she walks up to Adam and says anything about the magical feather, even if she says, "Jane sent me to get the magical feather," Adam will say he doesn't know what she's talking about, because she doesn't have the proper flag set.  Simple!

Bottom line: players have to spend waaay too much time trying to guess the right way to say something, and not nearly enough time figuring out the substance of what to say.  This sucks the fun right out of the game.  I realize that many people are probably working very hard on this, but I think the biggest problem lies in their design choices: design of anything is always about trade-offs, and this game needs to trade away some strict control over quest execution in favor of smoother NPC interaction for players.