20 Mar, 2012, David Haley wrote in the 61st comment:
Votes: 0
Hades_Kane said:
(a bunch of stuff to customize)
OK… so… who said you had to customize absolutely everything to that extent? You can customize to the same granularity as your current color scheme. You and Lyanic are making an argument ad absurdum and frankly you might as well be arguing with somebody saying that rainbow barf is the best way to go.
Hades_Kane said:
I don't see how going in and adding values to each and everyone one of those along with the code to configure and load each instance of those would be anything but a headache for something I don't really think would be widely used (which is based on my suspicion, from personal experience, discussion with others, and the fact that color customization hasn't been brought up by anyone ever to me or any of the other staff members in the 6+ years of development).
You're missing the point. You just advocated for a "mid-point" approach where you offered several color schemes. If you have several color schemes, then you have a way to change colors based on which scheme is active. If you have a way of translating text into colors based on a scheme, then you have a map that says which colors get translated to what. If you have a map of color type to color code based on active scheme, you have a color map. If you have a color map, you can let its values be customized rather than fixed with no more effort than a save and load routine. So, the point is that your mid-point approach is basically the same development work as a fully customized color system.
20 Mar, 2012, Hades_Kane wrote in the 62nd comment:
Again with multiple people citing years of players "not asking for a feature" as some kind of indicator that said feature would somehow be ill-received or unused. I really don't understand this line of thinking at all.
I posit that the vast majority of features in your respective games weren't requested by players. I also posit that all three of you have introduced at least one such feature and had it turn out to be quite popular.
People don't always know what they want before you give it to them. Apple made a killing off that very premise.
Much of the tone of the conversation has been advocating customizable colors as though it's a in-demand feature that can help make or break recruiting swaths of potential players, or at least that's the tone I've been picking up on.
I'm basically taking the position that if that were the case, I would assume that over the course of 6 years, someone would have complained about the colors or suggested a way to customize them, and if it isn't a critical or highly in-demand feature, at this point I don't think it's worth bothering with.
Would it be popular? Perhaps, but I really don't see that making or breaking the playerbase, and I -suspect- it would only be sparingly used, and more than likely by those trying to recreate the look or feel of their first/previous/favorite MUD. We all eventually run out of major things that need coding or get to a point where the game world, area wise, is relatively complete and much of the building from that point turns toward filler and extra areas, and further customization on the display may be worth addressing at that point.
The combat spam (and to a lesser extent the combat messaging [decimates, disembowels and the like]) has come up (unlike colors), although not much, but that is something I do find worthwhile to address. This isn't something that a player can address themselves client side, and this is something that could actually interfere or hamper with their ability to play the game if they aren't a quick reader. Dark blue and dark purple can be lightened in the client (not that those colors are commonly used) and other than just actually being able to read a dark color, there's really nothing I can think of that is difficult to read or differentiate in the game, and a preference for a different color scheme is normally just that, a preference.
If players start requesting more control over the colors or anything along those lines or what has been discussed, then I would certainly work to address it.
As far as the comment that my list was going to an extreme, I don't really think so. If someone doesn't like our color scheme or would want to customize it, we are rather consistent throughout much of the game with the grayish/red color scheme, and there would be a lot to consider there. Just stripping color and allowing the assignment of a color on room titles, room descriptions, exits, and channels would be trivial to add, but does this address the overall concern that people are expressing in the thread of people not liking a MUD's color scheme or what others are calling "full" customization? Maybe I've misunderstood and just a few points of customization is what is being discussed, but I've interpreted it as being the ability to change the color scheme of most anything that is presented to a player, and even in non unicorn-puke colored MUDs, that can still be quite a bit.
OK… so… who said you had to customize absolutely everything to that extent? You can customize to the same granularity as your current color scheme. You and Lyanic are making an argument ad absurdum and frankly you might as well be arguing with somebody saying that rainbow barf is the best way to go.
Hades_Kane said:
I don't see how going in and adding values to each and everyone one of those along with the code to configure and load each instance of those would be anything but a headache for something I don't really think would be widely used (which is based on my suspicion, from personal experience, discussion with others, and the fact that color customization hasn't been brought up by anyone ever to me or any of the other staff members in the 6+ years of development).
You're missing the point… So, the point is that your mid-point approach is basically the same development work as a fully customized color system.
The point is that if you don't agree to do what DH thinks is a good idea just because he thinks it's the same amount of work as your idea, then you must be a silly pot-head.
Why do you keep missing his point, HK? :smile:
20 Mar, 2012, David Haley wrote in the 64th comment:
Votes: 0
You know, Plamzi, it's a little unbecoming to accuse people of throwing out crude strawmen if you're going to make a post like that.
HK gave his mid-point solution as an easier alternative that is more worth implementing than full customization. The fact of the matter is that it's the same amount of work. So, if you believe that full color customization is better, but not worth it, and believe that HK's mid-point solution is worth implementing, then in truth you might as well do full color customization because you will have already done the work.
Hades_Kane said:
Much of the tone of the conversation has been advocating customizable colors as though it's a in-demand feature that can help make or break recruiting swaths of potential players, or at least that's the tone I've been picking up on.
Maybe this explains why people like Plamzi are getting so uppity about this. For my part, I'm just saying it's a nice feature and not hard to implement. I never said that it would "make or break recruiting swaths of potential players", nor do I believe anybody else said such a thing. I think it says something about all y'all if you're taking the discussion this way. :sad: Nobody is attacking your MUD or calling you stupid for not wanting to implement this in your own game. I'm not sure why some people feel the need to call others stupid for thinking it's a good feature that's easy to implement.
Hades_Kane said:
Just stripping color and allowing the assignment of a color on room titles, room descriptions, exits, and channels would be trivial to add, but does this address the overall concern that people are expressing in the thread of people not liking a MUD's color scheme or what others are calling "full" customization? Maybe I've misunderstood and just a few points of customization is what is being discussed, but I've interpreted it as being the ability to change the color scheme of most anything that is presented to a player, and even in non unicorn-puke colored MUDs, that can still be quite a bit.
The most critical color elements to customize are those that give information that needs to be processed quickly. Room titles, exits and channels are great examples of this. Combat messages (damage to you, damage from you, and if you want to get fancy, damage to your allies) are another example of something that is valuable to customize.
The point is that you can get a lot of return on your investment with even simple customization, without having to go nuts allowing customization on absolutely everything. As you say, it is trivial to add this customization. I guess I'm not sure why there is such visceral, even mean-spirited, opposition to the idea of customizing colors. (This hasn't been from you, HK.) What's the big deal?
@Hades: I'll echo DH here. Nobody that I've seen is taking that stance. We're just disputing the idea that it's some huge project and that nobody will use or enjoy it simply because they haven't asked for it.
Also, players can customize your colors client-side, anyway, so it's really only an illusion of control to not offer it server-side.
20 Mar, 2012, Rarva.Riendf wrote in the 66th comment:
Votes: 0
Quote
Also, players can customize your colors client-side, anyway
It is VERY annoying to do it client side as you have absolutely no way to differentiate objects from people from room desc unless you add msdp or mxp. You end up with triggers all over the place. Client is really not the place to parse text and identify what it is supposed to be.
Also, players can customize your colors client-side, anyway
It is VERY annoying to do it client side as you have absolutely no way to differentiate objects from people from room desc unless you add msdp or mxp. You end up with triggers all over the place. Client is really not the place to parse text and identify what it is supposed to be.
Well, I think it's a plain text problem rather than a general architectural commentary. Ideally I'd want customization done on the client for interface. Coloring included.
@Rarva: If you know regular expression syntax and use a client that supports it, it's actually relatively trivial to do.
Regular expressions are of limited use. It's not uncommon for muds to colour things that couldn't be done easily (or even at all) through regular expressions.
Then I'll give you a simple example. In my mud, character names are highlighted white (by default), but if you have Assassin Training and the person has their back to you, their name is highlighted red instead.
How would you do that with just a regular expression?
So you're suggesting replacing every instance of bold white text with red text?
While that would certainly do a great job of messing up the ASCII maps, it wouldn't actually solve the problem I described. It would colour everyones name red, when all I want to do is colour the names of people who are facing away from me.
Furthermore, the only reason you can identify names at all is because you're relying on the server-side colouring. If I hadn't added colour to names, you wouldn't be able to identify them from sequences of text - unless you're proposing writing some sort of language parser to break down the text and work out the names from the sentence structures? But even then you'd have considerable trouble differentiating between the name of a person, an object and a place.
Rarva said that you can't "differentiate objects from people from room desc" just through client-side regular expressions, and he's correct.
It would be comparable with writing an application that reads in the text of a novel, and automatically generates a list of the characters, a list of objects they interact with, a list of geographical locations, etc.
You gave no literal example, so neither did I. But, no, that was not what I was suggesting. Output from a MUD is structured and consistent. Names, room names, room descriptions, etc. are all in the same place relative to surrounding patterns. Unless your MUD output looks like:
Here is a room name and here is the descriotion and here are the exits and here is a player.
…then I find it rather odd that you and Rarva think it difficult to pluck out the individual pieces. Give me a literal example of MUD output, and I'll give you the regex to match it.
WRT the colors, Mush exposes style runs for a line. You can first match said line (e.g. PlayerX is here.), and then look at what color "PlayerX" is using GetStyleInfo() (if I remember the name correctly - it's been a while). It's very easy, and no, it won't mess up your ASCII map.
The fact that no one's requested it is hardly a good indicator of how popular it would be if you implemented it. I'd wager you'd get quite a surprising result if you pollled your players about it. I mean, it's totally alright if you just don't want to, because that's all the reason you need, but don't make such baseless assumptions. :-p
I'd argue it's a really good indicator of how popular it'd be. You see, players have a tendency to ask for all sorts of things. I even have players logon and tell me that if I implement X, Y and Z features immediately, they will stay. I usually scoff and wave goodbye, but that's beside the point. Some features get requested quite frequently - those are the "popular" ones. Then, there are features like full color customization, that only come up in threads like this, mentioned by a small handful of people. Is it possible that other players may get some value from it, too? Sure. Do they really care, though? Probably not. I took the liberty of polling some of my players on the idea. I even tried my best to pitch it as something fantastically awesome, so as not to influence decisions against it. As expected, it was met with apathetic shrugs. You should be careful about throwing around terms like "baseless assumptions".
David Haley said:
OK… so… who said you had to customize absolutely everything to that extent? You can customize to the same granularity as your current color scheme. You and Lyanic are making an argument ad absurdum and frankly you might as well be arguing with somebody saying that rainbow barf is the best way to go.
It's not an argument ad absurdum. What was advocated as "full color customization", and is still being advocated as best I can tell, is to allow each player to remap any color X to any other color Y via server-side configurations. For instance, a player could have all grey text remap to white, and all white text remap to blue. Feel free to correct me if I'm wrong on that. I still say you, and others, are operating under a false assumption. The assumption is that each component of text is assigned a single, static color. If all room titles were magenta, it would be easy to allow a configuration that changes them all to green. If, on the other hand, the base color varies by the plane of existence and also includes words that may have other information encoded via color, the problem is quite different. Sure, I could easily allow the color remapping, but it would lead to a loss or confusion of the information encoded by the color, along with making the aesthetic presentation objectively worse in almost all cases. The non-trivial aspect I previously mentioned relates to implementing it in a way that A) doesn't decrease usability and B) doesn't detract from the aesthetics in more cases than not. I'm not arguing against the virtues of the feature in general. I'm just pointing out a case where it's not practical.
Hades Kane said:
We all eventually run out of major things that need coding or get to a point where the game world, area wise, is relatively complete and much of the building from that point turns toward filler and extra areas, and further customization on the display may be worth addressing at that point.
Speak for yourself. I've been working on my game for 10+ years now, and I typically keep To Do lists for 2-3 years out. I'm convinced I'll never run out of major things to work on.
@Lycanic: I used the term appropriately, as when I made it, you had no base for your assumptions (hence "baseless assumptions"). It wasn't meant in a derogatory way.
21 Mar, 2012, Rarva.Riendf wrote in the 77th comment:
Votes: 0
* *(*)* + Temple of All Immortals [Area City of Silverleaf] [Room 7600] [Sector inside] [Light 1002/999] { no_mobs indoors safe area_init} | | ` | + [Exits: north east south west down] *-*-*-*-* + A magnificent temple has been built of marble and gold. The halls are ` , | | + bustling with activity, yet peaceful. The grand city of Silverleaf is all *-*-@-* * + around you. On the wall there is a beautiful mosaic map of the city. , | + S-*-*-* S + The Silverleaf Bank lies to the east. , | | , | + The Silverleaf Bar lies to the west. *-*-*-*-* + + (Perfect) (Glowing) A bright ball of light shimmers in the air. + A misty door hangs in midair, leading to the Arcane School + A Large Demon grins evilly at you as he attacks! + A goblin slave lies here asleep. + (Charmed) An angel is here to keep you safe. + An Atlantis Guard stands here, guardian the Entrance of Atlantis. + (White Aura) Panthea, the +|divine priestess|+, raises her hand in benediction.
Give me a literal example of MUD output, and I'll give you the regex to match it.
Okay, here are a few example combat messages from my mud - there are literally thousands of them, even before taking into account the fact that they're dynamic (modified based on body shapes, equipment, etc). I've bolded the parts that the regex should colour - the names should be red if the creature is facing away from you, white otherwise:
Quote
Two hundred and eighty-one feet north of you, the ratling matriarch turns south and carries on running.
Nine feet east of you, Tulmyn the huge demon lashes at you with his left whip, driving the tip through your feral eye and into your brain!
Six feet south of you, dark chains burst from the ground and coil around Shin's legs.
A shadowbolt shoots from a blurred shape's fingertip towards you.
You spin your darkblade swiftly across the wrinkly old woman's neck, decapitating her! Lightning arcs from your darkblade, engulfing the wrinkly old woman's headless corpse.
Seven feet south of you, a hideous ice troll rakes the filthy talons of its left hand across Kyoujin's face, drawing blood.
You breathe a blast of fire at the mob of angry villagers. Dozens of villagers fry as your flaming breath passes through the mob.
The large promethean splits into two smaller prometheans.
You swing your katana in a curved arc, severing the black-sash martial artist's right hand!
You shoot an arc of black lightning at the watchtower, blowing a gaping hole through its wall!
A desert nomad stabs his right scimitar over his left scimitar and into a scruffy watchman's face, drawing blood.
Four feet east of you, a tiny feathered imp staggers from the force of the huge dire gorilla's thunderous roar.
You hurl a python at Draco, hitting him in the face.
You rip your stilettos from Grimm's stomach, disembowelling him!
As an aside, the messages themselves can also be coloured to indicate if the attack was a critical hit. So the regex would also need to identify which of the thousands of messages represent regular hits and which represent critical hits (note that sometimes the messages for regular and critical hits are identical).
Here's another example, this time with equipment:
You are equipped with the following: A black cotton bandanna on your head. A black leather belt of catlike grace around your waist. A bloodsteel katana of sharpness strapped to your belt. A pair of slightly torn black leather trousers on your groin and legs. A pair of blood-stained black leather boots on your feet. A pair of steel knives tucked into your boots.
I've bolded the parts that should be coloured. Note that the katana should be coloured cyan because it's a rare magical item, while the belt should be magenta because it's an uncommon item. The others should be red. In theory you can also select colours for magical type as well, but let's keep this example simple.
Yeah, even if you identify the correct thing to modify with a regular expression you'd have an impossible task of knowing the state of the thing being modified. Like aggressive vs nonaggressive npc, etc. Markup exposed to the client would make this trivial as well, which is actually my preference. I don't like backend controlling interface and presentation so heavily.
@KaVir: /^You hurl (.+?) at (.+?) hitting (him|her|it) in the (.+?)\.$/
You now know that $2 contains the player name. You can use GetStyleInfo() in a loop to run through the style runs. If the styled text == $2, read the color. If the color == white, the player's back is not turned. If it is red, their back is turned. If you want to check if the hit is critical, you can check the color and position of the first style run. If the color is red and position == 0, it's critical, otherwise, it isn't.
I will grant you that the very dynamic nature of your messages makes it a much more tedious task, but nothing about it is "hard", and certainly nothing is impossible.
Also, note that I'm on a cellphone, so I've only done 1 of your provided messages, since it's a PITA.
OK… so… who said you had to customize absolutely everything to that extent? You can customize to the same granularity as your current color scheme. You and Lyanic are making an argument ad absurdum and frankly you might as well be arguing with somebody saying that rainbow barf is the best way to go.
You're missing the point. You just advocated for a "mid-point" approach where you offered several color schemes.
If you have several color schemes, then you have a way to change colors based on which scheme is active.
If you have a way of translating text into colors based on a scheme, then you have a map that says which colors get translated to what.
If you have a map of color type to color code based on active scheme, you have a color map.
If you have a color map, you can let its values be customized rather than fixed with no more effort than a save and load routine.
So, the point is that your mid-point approach is basically the same development work as a fully customized color system.