It seems that quite a few people are confused by CAVCON, moreso following recent changes RAWA made to what was being reported, so I thought I should try to explain my interpretation of what the CAVCON Meter reports mean and at the same time address a few of the more recurring questions associated with it.
RAWA created CAVCON (CAVern CONdition) shortly after MO:ULagain opened in February 2010 in response to requests from the fans for information on how well it was performing with regard to collecting donations. Now, here’s where one of the important things about MO:ULa comes is and creates “difficulties” for giving a simple status: MO:ULa is provided as “free-to-play” and donation is entirely voluntary – no-one is obliged to make a donation, no matter how much or how little they play. Cyan simply asks that people donate what they can reasonably afford and there was never any suggestion that there was a “recommended donation”. So, at one end of the spectrum there might be a few people with a substantial monthly disposable income who feel that donating $100 per month is quite comfortable and at the other end some who really struggle to make the domestic budget cover the essentials of living and couldn’t really afford to donate at all. And a lot of people somewhere in between.
So, Cyan needed to find a way to report how healthy MO:ULa was without placing any undue pressure to donate on folks who simply couldn’t afford it, or who were otherwise reluctant to donate. Expressing thing in terms of the cash value in the fund would lead to people trying to work out what that meant per player and concluding “Hey, we need to donate $5 each per month to keep this going” and at that point it becomes more like being charged a fee than a voluntary donation.
The CAVCON concept has a precedent: Many of the “old guard” of Uru Live players will recall the “Cautious Optimism” (C.O.) meter that began being used when the GameTap deal was being worked out to indicate the likelihood of Uru Live returning. Both the C.O. Meter and CAVCON share a root in the DefCon (Defense Readiness Condition) indicator used by the Armed Forces (and much loved by movie makers, e.g. in “War Games”).
However, CAVCON’s scale runs the opposite way round from DefCon, in that a high CAVCON number is a good thing. It’s probably best to quote RAWA’s description of the levels, taken from the original post on the MOUL forums:
CAVCON Level Definitions
CAVCON 5: Unlike some movies, which get the DEFCON levels backwards, CAVCON 5 is the best state. It means that donations are exceeding MO:ULagain expenses, there are a couple of months of MO:ULagain expenses in reserve, and extra money can be put toward server upgrades, bug fixes, new content, incorporating fan created content, etc.
CAVCON 4: This level means that donations are exceeding MO:ULagain expenses, and we are building a small reserve fund for future slow periods.
CAVCON 3: Donations are roughly equal to MO:ULagain expenses.
CAVCON 2: Donations are less than MO:ULagain expenses, and we’re currently using the reserve fund to help pay for MO:ULagain expenses.
CAVCON 1: Donations are less than MO:ULagain expenses, and the reserve fund is empty. This would be A Bad Thing ™.
So, put simply CAVCON is an approximate measure of how donations compare to expenses. But it also implies some other things: To be meaningful, you need to know whether CAVCON is reflecting the position for the current month or over the longer term. I always interpreted it as being related to the current month and I think this was correct. However that left some other issues, for example CAVCON 5 implied that CAVCON 4 had existed for some time, and there was now way tell by how much CAVCON 4 was topping up the reserve or how badly CAVCON 2 was eating into the reserve. A number of people called for more information, seeing CAVCON as inadequate in various ways, with two points in particular recurring:
- A scale of five points, especially since only levels 2, 3 and 4 seemed to be relevant, didn’t give any indication of how far short of or in excess of expenses the month’s donations had been.
- There was nothing to show the overall “health” of the reserve fund.
For October 2012, RAWA introduced some changes, one of which attempted to address the first of those points by adding an additional decimal to the CAVCON value. This in itself created some further confusion, since for level 3 (donations near equal to expenses) a decimal part would be meaningless, so RAWA posted a clarification on October 14 and gave an example of the revised scale:
4.9 (Donations = expenses + 90%)
4.1 (Donations = expenses +10%)
3.0 <--- monthly minimum goal (Donations = 100% of expenses) 1.9 or 2.9 (in both cases, Donations = 90% of expenses, the whole number portion just tells who covered the difference) 1.8 or 2.8 . . . 1.1 or 2.1 1.0 or 2.0
For values of 1.x Cyan pays the difference while 2.x means that the difference is paid out of the reserve fund.
Speaking personally, I was actually quite content with the original five point scale despite its shortcomings, and feel that with this new scale RAWA has created something that is not very intuitive, but as JW Platt commented to me “He’s a puzzle man.” And I guess that as Uru players we all ought to be able to cope with a small puzzle in reading the MO:ULa status.
The other change that RAWA brought in was the Donation Meter. This is a measure of how donations are progressing for the current month. Reset to 2.0 at the beginning of each month, it is generally updated on a weekly basis and progresses towards 3.0 when the donations have met the expected costs for the month. If the donations exceed the target then the Donation Meter will move to 4.1 and beyond. Quite what happens if monthly donations exceed 190% of expenses and we need to go past 4.9 isn’t clear!
CAVCON and the Donation Meter are two distinct measures; CAVCON tells us what the final position was at the end of the preceding moth, while the Donation Meter tells us how we’re doing as we progress through the current month. However, as both get reported together, they are often both referred to as “CAVCON” leading to some folks wondering “Why did CAVCON take a sudden dive from 4.7 to 2.0? What happened?” I expect in time we’ll get used to differentiating between the two.
So what are the expenses anyway?
There are a number of questions that come up about the nature of the expenses for MO:ULa so I’ll try to address some of the more recurring ones, at least so far as I’m able.
Why should I donate to MO:ULa, especially since there’s been no new content for two years?
No-one is obliged to donate to MOULa – that’s the long and short of it. But if someone doesn’t like what MOULa is doing (or not doing) and sees “some other shards” as better alternatives then presumably they don’t use MOULa either. If they do still use MOULa, then there must be some “value” to it, in some form, so it must be worth donating “something” for that value. Provided, of course, that they feel able to donate in the first place. I think that Cyan created MO:ULa for “its fans” and not just for “the fans who are able to pay for it.” Encouraging or reminding the community at large to donate is no bad thing, but badgering individuals over it is something to be avoided.
What does the donation money get used for?
Fundamentally, two things: The leasing of the servers and “hiring” Cyan staff to support MO:ULa. Contrary to what some people assume, Cyan do not own the servers they use, and the MO:ULa servers are leased from Amazon and incur a fee for each server instance used and another fee for the volume of data passing through the servers. In addition, MO:ULa has no “staff” so it needs to pay for Cyan employees to take time out of the working day for support work, like server re-starts or updates – it is essential that MO:ULa is seen to be self-sufficient to maintain the separation from Cyan’s business.
How much does MO:ULa cost to run?
We simply don’t know, it’s not information that Cyan is prepared to share for the reasons given earlier in this post. There have been estimates posted elsewhere, derived from the published Amazon EC2 pricing and guesses at the labour hours to support MO:ULa, but they are only estimates and based on a lot of assumptions that may or may not be accurate.
Why does Cyan have these expenses when shards are free?
No shard is truly “free”, but in most cases the shard operator has decided that they’re able to bear the running costs without asking for money from the users, although I have seen at least one shard that invites donations. The shard’s server costs may well be a lot lower that Cyan’s and the labour to run and maintain the server is effectively free. Even a home PC acting as server will have some “cost” associated with it even if it’s just the usage of a portion of the internet bandwidth.
Can’t Cyan use cheaper servers?
Possibly. The choice of server setup is probably a legacy from the work done for Magiquest Online and its beta testing, since work on MO:ULa began right after the MQO beta closed. It’s possible they could move to an alternative, lower cost server setup, but the move would take time and effort (= money) so it starts to get questionable whether it’d be a good idea or not. There are probably a lot variables related to the server setup that we can still only guess at, since the MOULa server is not the same as either MOSS or DIRTsand. That said, Cyan have announced that the intend to move to an annual contract for the MO:ULa servers in order to reduce costs, saving maybe 30-40%. While this removes some of the flexibility of the current arrangement, I expect that the usage metrics are showing that flexibility is an unnecessary luxury.
As a final thought, I’ve seen a couple of comments to the effect that “Open Source means that Uru can’t really die now – the fans will keep it going.” Technically, I guess that’s true, but I’m of the opinion that even Open Source needs MO:ULa: At least for the present, if MO:ULa shut down then we’d lose all (legal) access to the Cyan ages as there’d be no legitimate location to load them from. Without MO:ULa to act as a “base” or “hub” it would seem likely that shards would become increasingly insular communities that would mostly wither and die in time, and although it’s not a driver for all potential content creators, I think that the absence of a Cyan shard would remove some of the incentive for people to work on new ages or even learn age writing. There’s a symbiosis here: MO:ULa needs fan support, both in terms of financial donations and the donation of effort from fan developers and writers, in order to survive, and Open Source Uru needs MO:ULa for it to have a focus and reason.