Friday, January 30, 2009

Gaming Articles Suck!

It's been a while, so I figured I should post some observations about the web pages and traffic. Things work differently than I had expected, though honestly I didn't know what to expect anyway.

The biggest surprise so far is that all of the articles we've written and put up over at the Mudding and Gaming Articles page are worthless.

Ok, not completely worthless; in fact, there's a huge amount of good information in them, ranging from general gaming hints to specific strategies for advanced players. But in terms of web traffic, they've so far shown to be worth substantially less than I expected. The obvious impact of the articles on the overall site, from the standpoint of driving external traffic, seems to be minimal.

Take for example the heaviest hitting articles:

1) [48] The glossary of mudding terms;
2) [45] The muds and mudding faq;
3) [37] Alternative mudding clients;
4) [30] Equipment optimization;
5) [28] High level attack strategies.

The number out front is unique pageviews per month. The heaviest hitter has 1.5 views per day, while the main AA pages have over 100. Note that even fan fiction, artwork, and the historical archives beat several of these. Further, an eyeball's estimate places the -total- hits from the articles pages at approximately the same value as the new player mudding guide. In short, one of the more time consuming aspects of the web pages, article development, has virtually no obvious impact on the overall site quality.

[Note - 'no obvious impact' is not the same as 'no impact'. The pages have definitely improved the site in other ways. It's just not obvious.]

I can think of a couple of good reasons for this. One might be that the articles are buried, and having to click through two layers to get to them is a real hindrance. To unbury the articles, it would be simple enough to have an 'article of the day' show up on the main page, if I can find screen real estate for it.

The other, more obvious reason is that the articles don't have wide appeal. They are for the most part written for a very narrow audience of Alter Aeon players. The search engines seem to be reflecting this, in that the most general of the pages, the glossary and mudding FAQ, are the only pages with nonzero page rank.

Wider scope articles that still sell AA are more problematic. Adjusting some of the existing articles to have broader gaming scope may help. Even so, there's a limit to how broad an article on turning the undead can be.

On the plus side, each article contributes some non-zero amount of traffic to the site, and helps give it breadth and depth.

As a side note, this blog is equally boring as a source of information for people. I don't post anything overly controversial or interesting to the general public. Because the site is so narrow, and because my prose is so mechanical, the Alter Aeon Mob Factory drives virtually no traffic as well. It probably doesn't help that the layout is still the basic default.

Tuesday, January 27, 2009

Google Adwords - The Content Network

In preparation for the advertizing bump due in a little over two weeks, I've started looking at little harder at statistics, and at some of the other options. One of the easier things I can test out is to turn the Google 'content network' back on, as this is known to bring in a lot of web page hits for not a huge amount of cash. I've been fiddling with this for the last couple of days. But it's clear that the content network has some problems:

1) Even after these last couple of days, I still don't have things set up properly.

The content network has huge numbers of impressions, and above certain thresholds it's possible to blow your entire advertising budget for the day in a -very- short period of time. I've finally got it cranked back far enough that I don't run out for the day, but it's still about four times higher than the level I'm shooting for.

My advice if you're going to use the content network is to start out at $0.01 and raise it up by a penny a day until you figure out the appropriate level for what you're trying to do.

2) Content network click rates are stupidly low.

Expect at best one tenth the click through rate of regular adwords ads. Rates like 0.04% (four clicks per ten thousand impressions) so far seem typical.

3) Content network clicks are damn near worthless.

If you don't have some way of tracking or testing content clicks for click through, you're fooling yourself. Keep in mind that there's a fundamental difference between someone searching for information, and someone presented with information incidentally while they're looking for something else.

Regular adwords suffers much less from this problem than the content network, because of 1) the negative feedback ad system which kills bad ads, and 2) because people searching for something are still looking for something instead of being shoved things they don't care about.

My own experience and statistics for this show that the content network is 10 times (!) less likely to result in a saved character or a new player than regular adwords search traffic. Price your content ads to reflect the difference for your own site.

4) Content clicks can be gamed, sometimes badly.


I have so far had only one serious incident of content network click fraud, but it can and will happen if you run an extended campaign or one with wide coverage. Sometimes, Google will refund and clear the damage to your account; in my case, they did nothing, even though it was clearly and obviously fraudulent.

The only recommendation I can give for this is set your daily limit to the lowest possible value that still serves everything you want to serve. Then watch carefully for days when the budget is exhausted more quickly than usual. Assume there will be no recovery for click fraud; your only recourse is to limit the damage.

Saturday, January 24, 2009

Where do we go from here?

There's been a lot of development and other Alter Aeon activity over the last couple of weeks, but the bigger question is, to what end? While making ad-hoc changes can improve things, it's not necessarily the best way to get to a particular destination.

There is a method to the madness here, and a specific end goal. In February, I'll be approximately tripling the advertising budget, and if I'm going to be spending that kind of money we damn well better be ready for it. I have no intention of getting traffic to the site only to fail the new arrivals in other ways.

This still leaves a lot of things to be done:

- Getting proper ad and propaganda materials ready.

- Level 33 should be in and available. Happy high level players advancing to a new goal will help new players feel like there's something interesting going on.

- A proper new version of the dClient should be available, hopefully with a prettier automap and better blind support.

On the plus side, the statistics collection on new players is rapidly improving, and we have a much better idea of where people drop out and where problem spots are. The 'saved character' rate is actually quite high now that false negatives have been properly weeded out; of the new connections that begin the character creation process, approximately half of them get to level 1 and save. This is excluding multis - it's a pretty close measure of brand new people who have never seen the game before.

The 50% 'player save' value is also consistent against the dClient numbers, which I had not expected. I would have thought that the dClient numbers would be higher, and they are after a fashion - but not nearly as high as they should be. The dClient clearly needs more work and polish.

In addition to this, I'm in the process of collecting statistics for the islands to see where new players drop out. This is a longer term collection, which will hopefully tell us how to streamline and improve areas on the islands.

Friday, January 23, 2009

Status Update

Events so far this week on Alter Aeon:

- The first pass of microleveling is installed. It's required to get to level 31 and level 32. We're still thinking about the best way to handle level 33 and higher. There's also a new mode for showing the who list that can be toggled with the 'set who' command.

- We got a new article up on the web, courtesy Morpheus: The Black Art of Poison Brewing, which talks about brewing poisons using the 'brew poison' skill.

- Minor additions and updates to the Mud Terms Glossary.

- Various web page updates, which I've been slacking on recently. One of the more interesting ones is the Alter Aeon Quests List, which is now sorted by geographic area. You can see which quests are on which islands, or on the mainland.

- A 'resurrect' spell has been added, which is effectively a more powerful summon using the corpse of a dead teammate. Resurrect ignores distance and various nosummon flags, making it easier to put a group back together after deaths. You must be grouped and have a corpse to use the spell, so there's not much potential for abuse.

- A new 'faq' section has been added, detailing how the in-game text editor works.

- A new version of the TinyAeon custom Alter Aeon client has been released, with hopefully improved screen and scroll handling. Even better, a new version is on its way and should be uploaded in the next couple of days.


There's also been a boatload of minor fixes, to everything from specp 25 (the shopkeeper reload specp) to channel replay. This week has seen a lot of cleanup and improvement to various things, though the microleveling additions spiked a lot of controversy. It's looking fairly good now though, thanks to good comments from lots of players.

Saturday, January 17, 2009

High levels

As a result of the recent high level observations, I've been making minor changes and cleanup to the player experience. One of the things I looked at was int/wis requirements on spells.

The reason for looking at int/wis was mostly newbie based; new players have limited practices available, and have to decide where to spend them on which stats. If you train stats based on the minimum-cost method, you'll find that stats are naturally separated by two, for example int 18 and wis 16, or int 18 and wis 14.

Under the old int/wis system, the values were hand picked and mostly but not completely followed an algorithm for this that left the int/wis requirements separated by 3. This left new players with a subtle but weird choice to make: should my secondary casting stat be my second highest stat, or third? 3 is between 2 and 4, so keeping the costs even was inefficient.

With the new system, the separation is always 4, so that's never an issue. Also, all the values are calculated, with (currently) no exceptions, and all the spells are consistent. Plus, I got to delete a lot of lines of code.

Unfortunately, there are some complaints, and this seems to be where it always breaks down. No matter how small a change you make, someone gets upset, or pissed off, or threatens to leave. Most of the new players will never even know that there was a change, or the ones that do will adjust their characters within a few days at most. The real complaints come from the highbies.

One such player confronted me while I was in the middle of writing this post. In this particular instance, the change has moved the 'sanctuary' spell from 17/20 int/wis to 18/22, which honestly is a bigger change than I expected (the high mana cost of the spell is bumping up the requirements from what would otherwise be 17/21.) This player is a high level thief/warrior, and doesn't have the natural int/wis to cast sanc now, though they barely had it to begin with. After months of arduous effort, the total 105+ in question finally managed to get a 20 natural wisdom, just as the spell requirement changed from 20 to 22.

I'm not sure how I feel about this. Part of me wants to say 'WTF, a total 105+ with a 20 wisdom?', and other parts of me think other things. But an additional point was raised that brings up yet another design related question:

Why bother to have thief and warrior classes if it's effectively mandatory to have caster classes to do anything interesting?

I'm not sure how I feel about this one either. The game has always been multiclass. I expect people to make full use of whatever skills and spells are available in each class, and to use those spells and skills to compensate for weaknesses in other classes. To have a level 20 cleric incapable of casting sanctuary or other cleric spells simply isn't something I would have expected; after all, sanc is an EXTREMELY powerful spell used by virtually everyone virtually all the time. The game is practically built around it.

In short, I don't know that it's legitimate to have a warrior-only character and expect it to be able to stand on its own and compete with multiclass characters of similar level (or total levels.) The game just isn't designed that way. That type of operation is generally reserved for single class games.

Still, the player is pissed and will likely continue to be so. I have a shitty set of choices to decide from. Compromise a subtle change that I know in the long run is what should be done? Make an exception for particular spells? Make an exception for bitching players? Ignore it?

I'll let you, the readers decide. And then probably ignore the result.

Wednesday, January 14, 2009

The death (dearth?) of high levels

It has finally been made sufficiently clear to me that there is a problem with boredom at high levels. As near as I can tell, it's not that people actually run out of stuff to do; it's that they run out of -fun- stuff to do.

Consider the plight of a total 118 or higher. When it costs you a billion experience per level, you pretty much have three options:

1) Run experience for weeks straight to get one measly level, ignoring anything dangerous like EQ or quest runs

2) Do EQ and quest runs and forget about getting a level

3) End up halfway in the middle without incentive to get more exp and not wanting to do anything else because you'd lose the exp you already have

Of these, option 3 is the worst case, and unfortunately probably the most likely. Fortunately, all of the above have an easy solution: microlevels, also known as level staging.

In short, it's a way to bank off experience or whatever the level requirement is in the short term, and give people the ability to work on the level at their own pace. It's even better if you can display that somehow, so that others can see them making progress. Everyone likes feedback about how they're making progress.

This brings me to a whole slew of ideas:

- display of the microlevel on the who list somehow?
- more use of fame for microlevels?
- use of quests for microlevels?
- abandoning exp completely for microlevels?
- going to levels 33-40
- limit rate of acquiring microlevels and levels?

All of these things are something to think about, but either way we're looking at opening up level 33 this year. Whatever I set up, I'd like to do it such that we only open at most one level per year - it's far too easy to get massive inflation otherwise.

For the display thing, the current display model is something like:

[30 32 23 32] Jerk doesn't exist

Suppose Jerk has 16 microlevels on his way to level 33 warrior:

[30 32 23 32] 16 Jerk doesn't exist

I like the way this looks; perhaps other formats would be prettier. Microlevels for each class, with only the highest being displayed sounds like a reasonable way to handle it.

I also really want to do more with requirements other than experience. We did some of this with fame, in requiring fame to gain levels, but right now a quad 32 only has to use 700 fame to get there. The current level requirements for fame right now are:

30 - 25 fame
31 - 50 fame
32 - 100 fame
33 - 200 fame

Note that the level 33 setting has never been used, but 200 fame sounds like a good value for it. For a dual-32 having used 350 fame, another 200 to get to level 33 seems fair. Pro-rating this per microlevel would help people to know where exactly they stand.

Other possibilities for level requirements could be as simple as explorer tokens, or some periodic event. I'm definitely open to more ideas in this area.

For implementation of this, if I had my way I'd make microleveling required for any level higher than 30. This still might be doable, so long as the level and other costs associated with it don't go out of bounds. Microlevels seem kinda stupid and redundant if you can get 15 of them in an hour. 50 million exp spread across 30 microlevels to get to level 31 leaves only 1.6 mil per microlevel; perhaps the real cost for it should be more along the lines of 300 million, or 10 million per microlevel.

This would be a huge increase in the cost of level 31, but with the idea of giving some bonus for each microlevel it might be well tolerated. Consider a temporary bonus for each microlevel, perhaps hp for each of the 32 microlevels on the way to warrior 33. When the final level is actually attained, the microlevel bonus goes away and is replaced by whatever the real level gives you. Players can then start accumulating another set of microlevel bonuses.

Ok, enough ideas for today. Feel free to post back.

Google Page Rank

As I posted a while back, the AA website's page rank recently went up; since the site went from 2 to 3, there's been a factor of 1.5 increase in overall traffic to the site. Newbie logins, and Google Analytics stats all seem to reflect this general increase.

However, I'm not sure I buy it. When I look at the competitive analysis trend lines in Analytics, I see a pretty regular pattern leading up into December, then things start to look a little different. Going into January, it appears that the global amount of traffic in all categories has increased by the exact same factor of 1.5 I'm seeing on AA. I'm not entirely sure what to make of this. I've seen January increases before, but I never paid much attention to it. I always figured it was people waiting to go back to school, bored and without much to do.

One possibility is that the competetive analysis numbers I'm seeing are related to reclassification of the web site. When a page/site changes page rank, does Google reclassify the site to compare it to the new group? It's possible that the wierdness I'm seeing is simply the new competition in the new page rank category. It would be nice to have more information on this.

Whatever the reason, this influx of newbies is really refreshing. I can only hope that it's not a temporary blip, and that it continues going forward. For too many months did I work on the web site to see no gain whatsoever.

In other news, I found out the reason for the massive drop in players around the October time frame: competition. A rather large group of dedicated blind mudders for all intents and purposes got bored and left to go to another (very large) mud. I've learned several things from finding this out:

1) I need to pay more attention
2) My mortals are bored
3) It's possible to do what I want to do. The target mud has on the order of a thousand players on it. That's the short term goal for AA.
4) I'm slacking. This is my fault. See #1 above.

So I suppose I have a belated new-years resolution: I will pay more attention to the competition and what they are doing.

Monday, January 12, 2009

More xml

The ongoing XML saga continues, with updates to the code to properly escape the font color tags and the addition of the wholist at http://dentinmud.org:8080/xml/wholist.

Beyond that, the main web pages got another facelift, and I think I'm done propaganda blitzing for a while. That gets old real quick.

Time to try to get some bug fixes into the server and the client if time permits today.

Friday, January 9, 2009

Deathlogs

One more real quick post before I bail out for tonight: I got Alter Aeon an entry on deathlogs.com, which is a repository of PK kill logs for a handful of muds. While I don't anticipate that this will bring in a huge amount of traffic, it's yet another external link.

xml xml xml

I really hate that hitting enter while in the title bar automatically posts, even if the post body is empty.

Today, I added support for XML dumping of player data via the web interface. This was at the request of the player Cu, whose XML data can now be seen at http://dentinmud.org:3004/xml/players/cu. The idea was to export various player statistics so that he could dump it into a database and do databasey things with it; in short, a third party app using AA data.

I really dislike XML, but I like this plan. I also hope to export other data, hopefully without compromising the game itself, so that more things can be done with it.

This reboot also brings in a handful of minor changes to newbie sends and the email command, but more importantly it brings in a major change to area checking. We now have code to check linkages between areas and figure out how far off in terms of relative position they are. This should help us line up the world and get things to be a little more grid-like.

This gridlike preparation is in anticipation of graphical maps in the dclient.

In other news, there's a bunch of presentation/toolkit bugs in the client, pointed out by various people. Some of them do not look easy to fix.

Wednesday, January 7, 2009

Propaganda

I've been reading SEO stuff off and on for a few months and I've been trying to incorporate various principles into the AA pages as I run across them and understand them. Recently, I got a guide with a whole bunch of short tips in it, and I've been going through it (very slowly.) This week's tip was deep linking.

So tonight, I spent several hours submitting various AA pages for deep linking to other web sites. This was mostly the glossary and article index, but I learned something very unexpected from this adventure: trying to get sub-pages linked with indexes means that I have to make the sub-pages useful to said indexes.

In short, by doing this, I'm improving the web page content in the most meaningful possible way: I'm making it more useful to external parties. In retrospect it's obvious. But it's something I could have missed for a very long time if I hadn't bothered to try it out.

Xmas event log!

I finally got the 'will christmas be saved' event log posted on the web. To see who all participated and in what way, you can find the log here:

Alter Aeon 2008 Christmas System Event

It's also linked to off our Alter Aeon System Events page.

This looks like it was a pretty fun event, as there were a lot of participants. There were even several deaths during the final battle, all of which are recorded for posterity in the event log.

In other news, version 971 of the Alter Aeon Custom Client is available. This version has mostly bug fixes in it, but also moves some of the direction buttons around and in general has a slightly nicer looking interface.

Tuesday, January 6, 2009

When will we have a real client?

Client work continued hard and heavy today, with two major additions going in: first, the hp/mana/movement status bar is functional. This is a bunch of progress bars, red for hp, blue for mana, and yellow for movement, that goes between the command-entry window and the main window. As you get beat up or heal, your stats update in real time, and the bars get redrawn to give a graphical representation of your current status.

A lot of this code was portable from the old client codebase, but between two different sets of library bugs on two different platforms, I ended up spending several more hours than is reasonable getting it all working. And in the end, I ended up doing a full client release with a trivially obvious bug in one of the display settings: the mana bar displays movement numbers if the client window is made wider than a certain amount.

The second thing I got working was to use the autoexit data to highlight/darken the various direction buttons. So if you're in a room with 3 exits showing, those three direction buttons will be lit. You can use the others if you want anyway (eg. to find doors by bumping into them), but if you're running from something or exploring, this should help.

The surprising thing about the autoexit buttons is how ugly they are. The display is just plain ugly, and I'm not real sure why. I'll have to poke around with it a bit more and see what else I can do with it.

I looked a little at the scrolling issue listed in the previous post, but it's entirely unclear to me how you would hook such an event with this library. I'm also pretty well stuck on the accessibility functions. Unfortunately, noone makes howtos for something this specific.

Tomorrow, other things permitting, I'm going to try to get new style automaps working. That should be fun, and probably overly ambitious: I have enough bug fixes and miscellaneous cleanup to probably consume the day already, and web page work is starting to fall behind.

For web pages, I have a large number of directions for the blind maps project, and an event report from christmas which needs to be put up.

This would be easier if I could fork copies of myself.

Monday, January 5, 2009

Client updates

I'm in the process of getting another client build out. This version will be almost entirely bug fixes, with a number of wierdnesses with scrolling and the dialog boxes fixed. It should be available at http://dentinmud.org/TinyAeon.exe by the time anyone reads this post.

I'm still running into problems with the toolkit regularly, but I'm slightly more effective and figuring them out and getting it to work anyway.

Next up on the client todo list is going to be auto-scroll-lock on mouse scrolling, and autoexit highlighting of the direction buttons.

In other unexpected news, there's been some serious pk stupidity on the game, and I've made some code changes to try to improve things.

Sunday, January 4, 2009

Merging merging merging...

Keep them dawgies mergin!

Over the last couple of days, I've been trying to get stuff merged in from one of the other coders, as well as fixing up various issues and trying to get some things caught up. While doing this, I'm starting to get a better idea of what I really can and cannot do in a certain period of time.

One thing I should probably do is set aside just one day per week that is going to be dedicated to doing nothing but implementing one single feature. I can actually get quite a bit done in a 3 hour or 5 hour block, and with another 3 to 5 hours to clean things up, it can be booted into the game and functional. But in order to get this time, I need to ignore things like minor bug fixes and all the crap detail work that so easily can bog one down. Not that detail work isn't important, but sometimes big new features are important too.

In other news, I got a freaking pile of directions and 'blind maps' from one of my blind users. The only problem is that there's so much of it, it's going to need a database just to make it all accessible.

Thursday, January 1, 2009

Hooray!

After all this time and effort, something finally paid off on the main Alter Aeon web pages: at some point in the last week, we went from page rank 2 to page rank 3. Yay!

I'm actually really jazzed about this. Finally, something I did had a positive effect. That said, looking back at the historical analytics data for this year, I really don't see any change in overall traffic to the site. A long term average since may seems to be about flat. Granted, I didn't start really hitting the site hard until september or october, but not seeing any impact at all over the course of several months is discouraging to say the least.

The only real solution to this is simply to spend time linkbuilding and getting us listed everywhere I possibly can. I suspect my latest push for that here is what raised the page rank.

In other news, the quest to add useful content continues. I added a bunch of stuff to the Blind and Visually Impaired Player Support page, including more downloads and scripts for GMud and a raft of MushClient sound and data files.

I also spent a good chunk of the morning going through Scribe's old mudding site, and imported a couple of articles for the current AA pages. A number of dead links were killed in the process, which hopefully will cause the google webmaster tools to complain less. The two new articles are:

Player Types and Styles, and

So you want to be a Player Killer?

This brings the article total up to 21, not including other things scattered around the site but not in my high-tech document management system. This is pretty low as far as SEO goes, in that most SEO sites recommend at least a hundred pages of solid content, but it's a start.