Elder Basilisk said:
With a computer, the only real reason to upgrade the OS is because
A. the new OS offers features and support that your current one doesn't
B. There are some great new applications/games on the new OS that your current one doesn't support
C. You're getting a new computer anyway and might as well get the new OS as the old one.
To carry that analogy forward, you've missed one other compelling reason:
D. Continued Support is only available for the new OS
Ask anyone who tried to get a fix for the Daylight Savings changes to Windows 2000 Server earlier in the last year how that worked out. Microsoft graduates support for older versions of their OS over time. Windows NT is unsupported. Windows 2000 is not actively supported. Windows 2003 has full-time active support.
When 3.5 came out, 3.0 material stopped being compelling in the marketplace. WotC stopped producing materials that were meant for 3.0 and any future material (including material in Dungeon and Dragon, which were in Paizo's hands) was in 3.5 format. 'Lame Duck' releases like Savage Species and Fiend Folio were half-a-loaf...not quite 3.0 and not quite 3.5.
And we should also be clear here that sales and actively playing a game are not the same thing. Many players may not adopt 4e for some time or at all...but if they purchase the 4e rulebook, it's no different to WotC. They cannot tell
how those books are being used, if at all...merely that they were purchased.
trollwad said:
Over time, they listen very closely to their customers (or more likely as in the case of D&D, their most verbal leading edge customers) but in doing so, over time they tend to "overshoot" the needs of their average customers or new customers (how many kids have you seen playing d&d vs. the 70s) -- ie they keep tacking on a bunch of "crap" functionality that no one wants which clutters up the game or change for change sake.
There's a lot of assumption in there that's hard to quantify, because we don't have any data about D&D players other than the limited numbers WotC has and has chosen to share. D&D, as a game, didn't truly catch fire until the early 1980s, when it was a cultural fad. Most of the people who picked up the game at that time didn't really stay with it for any real length of time. While we know that TSR was extremely profitable at that time, we don't really know (and apparently TSR didn't really know) exactly
how profitable. We don't even really know how many active D&D players have existed at any time in the game's history (though we have estimates for the total number who've played the game over it's entire life).
The issue with overshooting the needs of it's customers is the assumption that the customers are entirely aware of all of their own needs (and with the majority of OS consumers, this clearly isn't the case, vis a vis Security) and that an RPG can be guilty of this, which is hard to establish because of the nature of RPGs. Unlike an OS, an RPG can be readily and completely gutted or retooled to a purpose that it was not designed for by even the most amateur of enthusiasts and with far fewer repercussions.
As for playtesting a version for 10 years....when has this ever been done for ANY RPG, ever? The gap between OD&D and AD&D was what, 3 years (74-77)? Then 2e came 12 years later...but it certainly didn't feature 10 years of playtesting. 3e came 11 years after that, but while it, afaik, was one of the most formally playtested versions of the game ever done, it still had no more than a couple of years playtesting while the game was in development. 3.5 came 4 years later and now 4e is coming 5 years after that. I have my concerns that there won't be enough playtesting for 4e myself, but I think the ideal playtest period is much shorter than 10 years. What gives me hope is that so many of the mechanics and concepts have been culled from supplemental material released in the last two years, meaning that some of it has been playtested and real world tested in piecemeal.