State of the Industry Editorial

[HinterWelt]

Let me start out by saying that I am in favor of distributed development for small companies. This is based on my 15 years of working in the IT industry as a consultant, in house IT, and contract freelancer and the two cost analysis we did on outsourcing/distributed software development (one at Unisys and one at HTI). So...

There are differences between software and publishing but the development process is surprisingly similar for a cost point of view. In our studies we found in house development was 2-3 times as productive in the terms of actual lines of debugged code submitted to source control of an equal period of time. This was averaged over (at Unisys) 12 4-person teams working from home and the same number from the office. Cost, as evabluated over a year, including lost days (due to sickness, other commitments of kids and such) came out to be about office work being 2-3 times as expensive as the office. This cost actually decreased if the home worker was closer to the office since the home worker would draw on supplies from the office.

My personal experience is that I can manage about six distributed teams from home at the same time before quality is seriously affected. At work it is more like 10-12. The study at HTI was more oriented on this aspect and what happened there was that wfound you could almost double the work accomplished with in office teams ofver home workers. This means you could do the same work with 1/2 the people at the office.

The point I would make is not the RPG Industry is "broke" (whatever that means) but that small companies doing a few projects serially can function quite happily and efficently in a distributed environment. Larger companies that have the revenue to handle the capital investment of an office can benefit from the cooperative nature of the office space. Large Companies can also benefit from distributed development, the benefits increase the more you need to call on non-local specialists. The reason usually comes back to workflow and document management in the sense of one release per year vs thirty.  

This of course are only my observations and they do not represent the universe and all its glorious working. In other words, YMMV,

Bill

[Christopher Weeks]

you could do the same work with 1/2 the people at the office.

I'm not disagreeing with your observations.  There are a lot of factors that go into these kinds of analyses and our situations are certainly different.  And I'm looking at it from the bottom up instead of the top down.

But in my particular case, I am absolutely positive that I'm more productive working from home.  When I'm at home, I work on and off in spurts at all times of day and night and carefully track my time.  When I'm an in-office employee (like right now), I come and go when the office culture expects me to and I loaf.  I might only do four hours of development per eight hours in the office.  

Now, if I'm working on my own, I can't call on my team members when I'm stuck on a problem at 3AM, so sometimes I encounter a cost with that workstyle too.  But I don't think it evens out with my vastly increased productivity.  But I have always worked alone or on small teams on pretty small projects when doing development.

My $.02,

Chris

[HinterWelt]

Chris,
I want to emphasize that the two studies were carefully designed to take average. With individual interviews we conducted we almost always got the same answers as you goive along the lines of "I can be more productive because I work when I want". As I have said, I am a proponent of distributed development, simply because not every thing can be boiled down to the balance sheet. People who do that kind of thing often are very unhappy. That said, you will be equally unhappy if you ignore the balance sheet.

What we found was, if the team working on a project had separate caompartmentalized tasks then productivity was impacted less but still effected. Strangely enough, without the developer being aware of it. This happens because of the little word I snuck in "Bug-Free" code. We found that working at the office had a higher level, and more thurough code reviews. If all you are doing is spewing out HTML it is still effected but more from design point of view. You could mitigate this somewhat by pushing the envelope on video conferencing, SWS software and fcused and shared data dumps but you start to incur serious software costs at this point. Your benefit begins to dwindle. I should say, th comapny's benefit begins to dwindle.

Also, I want to stress what no study can effectively communicate, worker satisfaction. Some people enjoy working from home and having family/pets/comforts of home around them. I happen to be one of them. Some people found the structure/facilities/comraderie/break from family of the office benefical to focusing their efforts. Impossible to quantify but very real in terms of retaining employees.

Feel free to ignore my ramblings,

Bill

[ryand]

One of the defintions of insanity is "doing the same thing over and over again while expecting a different result".

The d20 RPG publishers, in masse, are currently exhibiting this behavior pattern.  Many of them are locked into a business model that requires them to publish a product each and every month, with a minimum assumed unit volume, in order to pay monthly expenses.  Without a monthly release schedule, these companies will stop paying salaries, royalties, and rent when the tail end of their distributor invoices dry up in 60 to 90 days.

This business model is predicated on the idea that consumers want to buy something every month (in fact, it is predicated on consumers wanting to buy something weekly, but I digress).  For several years, it was sufficient to just print anything reasonably centered on the fantasy / science fiction axis common to most hobby gamer's interest pattern, and that product would meet the minimum monthly unit volume required to sustain the business model.

Starting in the 4th quarter of 2002, this pattern of behavior broke.  Buyers in large measure have stopped purchasing products unless those products are exceptional.  Sales volumes of new product, on average, have been declining roughly 50% per quarter since then end of 2002.  At the end of 2003, several publishers are reporting that they can no longer count on hitting their monthly minimum unit volumes.  As a result, they are laying off staff, cutting future products (announced or not) and trying to retrench into a smaller cost basis so they can survive with a smaller unit volume.

In response to a discussion about this phenomenon in a private industry forum, I composed the gist of the message quoted by the original article as a way for publishers to reconsider the fundamental business model itself as opposed to bemoaning the "collapse" in RPG sales.  Because fundamentally, I believe there has been a "collapse" because publishers keep trying to sell substandard product with substandard marketing to an increasingly interconnected community of increasingly savvy buyers.  And the buyers, well, aren't buying.

Comes a product like Draconomicon, or AEG's Spycraft, or Fantasy Flight's Midnight, or Monte Cook's Arcana Unearthed, and those customers get into the stores, open their wallets, and buy.  Clearly, the consumers are still interested.  And the consumers still have money to spend on RPG purchases.  And the channel is still functioning well enough to get those products to the stores so those customers can make those purchases.

I am therefore compelled to conclude that the problem is not with the customers, the retail stores, the distributors, the licensing framework, the game systems, or competition from other entertainment options.  I am compelled to conclude that the problem is that, on average, the products that are being produced aren't good enough to trigger the "buy" response.

My original concept, loosely quoted in the original article, was to suggest to publishers that they get off the treadmill, and stop trying to publish a product every month.  That if they instead focused on producing just a few products each year, but crafted each product with care, attention to detail, and high production values, they would be able to sell substantially more units of that product then they would sell in aggregate of the "monthly treadmill" products they're currently making.  And because there are economies of scale in RPG publishing, and they are attainable at reasonable unit volumes, following that kind of limited development schedule creates sizable pockatable profits as well as being a sustainable business model.

There are three reasonable reasons to ignore this advice.  First, the publishers might think I'm a crackpot without any insight about how RPG publishing works, and thus conclude that I don't know what I'm talking about and my suggestions are meritless.  Second, the publisher may be targeting a very small niche auidence for personal, artistic, or asthetic reasons, and the fact that such markets produce small unit volumes is an acknowledged problem that cannot be fixed.  Third, the publisher may have spent themselves into a black hole of debt or continuing obligations, does not have access to free capital, and believes that without any realistic way to stop producing those monthly products and regroup for a higher quality, higher unit volume business model the company would bankrupt itself in the process.

In 2004, d20 companies will go out of business because their business models are broken.  A lot of them will go out of business.  Some will stop publishing paper products and go to PDF distribution.  Some will soldier on racking up debt to freelancers, artists, and printers, delaying the reckoning while they hope for a wholesale industry return to high volume sales of low quality products.  Some will decide to bid for a better business model, and my prediction is that they will be successful, and their success will provide a roadmap out of the current quagmire into a long term, sustainable design cycle.

Ryan