Monday, January 30, 2012

Pentaho won't launch on MacOS X

Download Pentaho, unpack it.  You double click "Data Integration 64-bit" and nothing happens.

When you try it on the commandline, it gives you:

LSOpenURLsWithRole() failed with error -10810 for the file /Users/trimbo/Downloads/data-integration/Data Integration 64-bit.app.

The solution is to give the JavaApplicationStub file execution permission.

chmod +x ~/Downloads/data-integration/Data Integration 64-bit.app/Contents/MacOS/JavaApplicationStub

Now you can double click the icon and it will launch.

Hopefully this helps someone.

Saturday, January 14, 2012

Why I didn't go work on Facebook games

Many people were surprised when I left the games industry last year and moved over to a completely different kind of ecommerce engineering position.

Some background.  I started out of college working on computer graphics (CG) for commercials.  This was in 1995, and a lot of money was made doing this.  Most commercials with any kind of CG in them would cost $500K, and commercials that were entirely CG would be $1MM plus.  These numbers are obscene today, right?   And I *loved* doing commercial work.  Commercials were fun, 4-8 week projects that you could just rip through and move on.

Dreamworks had an arm that did commercials, believe it or not -- and I asked Jeffrey Katzenberg if he would invest in the business more.  This was in 2000 or so and were doing very well with some spots for Intel and Visa, earning a good profit. But shortly after I left that company, he killed the whole division.  He made the right decision in retrospect though it was growing at the time.  The reality is that those margins began to drop, fast.  By 2002, when I left commercials for good, you started to see home PCs capable of doing real time editing of standard def video.  Within a few more years, HD could be edited on a home PC. The whole idea was commoditized.

Reflecting on that over many years it became apparent to me that I needed to always watch for industries becoming commoditized.  This is what Katzenberg saw happening in commercials and in film effects.  I did another gig in film effects, but the next opportunities were all overseas (ruh-roh, again).  So I moved into video games.  And now games are doing that.  Here's why.

Let's start with Facebook games.

First of all, Facebook games are almost never games.  They are designed to be obligations, like Tamagotchi or SeaMan or Animal Crossing.  One of the best essays ever written about Farmville is this one, which outlines theories on what defines a game, and why Farmville is not one.  But let's pretend they're actually fun, entertaining, escapist games at the moment, and they're just freemium in business model.

How do you make a freemium Facebook game successful?  You must reach as many people as possible.  Ok, how do you do that?  You make people enlist their friends into the game.  Ok, how do you do that?

To make people enlist their friends in the game, you must optimize on revenue by way of cross-promotion.  Right, that means optimized towards either spamming friends or paying instead.  This takes game design and turns it on its ear.  In my book, there is no such thing as game design once you are revenue driven in game mechanic itself.

Few people I know want to be a "game maker" in that environment, myself included.  It's no small wonder that so many people I've talked to who have worked on those titles didn't like it.  It's probably because they're not actually designing fun games there, they're making design decisions based on analytics.  Those two concepts are nearly orthogonal.   Starcraft wouldn't exist if it was designed to be freemium, and Farmville wouldn't be the money machine it is if it wasn't run that way.

Anyway, that's all fine and dandy because maybe some people like making obligation games armed with Tableau and Vertica, good for them.  But this leads into the next stage:  creating content for these platforms must be commoditized in order to survive long term.  Not every customer of one game will want to play the next title.  So you must offer more and more choices as time goes on and customers specialize to unique experiences in their mind.

Do the math.   There are 20 friends playing Farmville.  One pays for virtual junk, the others do not -- this is the rate (<5%) that Zynga has disclosed to the SEC.   Now let's assume there's a 50% churn of customers per month.  That is, a 50% chance that one user who pays will not be playing a title in a month.  That's the rate I've found reported on the interwebs for social games.

To stay revenue neutral, you will need to migrate your entire customer base to a new game every 2 months.

And the chances are low that you'll interest the same customers in the next game.  Maybe one Farmville player migrated to Cityville, and one migrated to Castleville.  Over time, you'll need more and more titles to keep the interest and bring in more people to the spam loop.  This is the only way they can achieve revenue optimization across a large audience of unpaid customers in freemium games.

As a result, making these games must become a commodity to the extreme.  They will become cheaper and of less quality because many more of them need to be made to satisfy growth in light of the facts that there is extreme churn and 95% of Zynga's audience pays nothing at all.

If you don't believe me, you can see this in action right now.  Sims Social dominated 2011 for the most part, but EA overall is losing customers while Zynga stays somewhat steady.  EA had nowhere to put customers and their friends once they lost interest in Sims Social.

The business plan of "I'm going to make a facebook game" just won't work for very long.  I worked for a short time at a company that dropped several million taking their top line game franchise and converting it into a Facebook game.  I realized the tough position of Facebook enterprises while I was at that company.. even if that had been successful, you can see what happened with EA.  Overall your revenues decline unless you have somewhere to put customers immediately when they lose interest.

Zynga has been able to escape this fate for years because they had the luck of growing as Facebook grew.  But now that Facebook itself is leveling out (and shrinking here in the US),  Zynga needs to expand production significantly.  Spending less on more titles and hoping that the customer churn goes into itself, rather than someone else's games or to another distraction.

In my next part, I'll take on "Why I didn't go work on mobile games (well, actually I did, but then quit in one month)"