March 2011 Archives

One of the external factors that I believe drained many yahoos during my time there was the constant bashing from analysts, press, bloggers etc. Internally I think it was reasonably obvious what problems existed but hearing it constantly from "the outside" lead to a lot internal navel gazing about what we could have done, should have done. Google was at the time the darling of the web and there was so much discussion about what Google was doing that a separate "what would google do" mailing list was setup.

So its interesting to read What Larry Page really needs to do to return Google to its startup roots. slacy's post is almost a perfect summary of the commonly raised issues at Yahoo! during my later years there (I was laid off in October 2008).

For example, languages like Python are shunned upon because they're "too slow for web frontends". Let teams use whatever tools and languages they want, and are most efficient in. Don't pass judgement on infrastructure, pass judgement on Products. If someone launches a great system based on Oracle and a bunch of Perl CGI scripts running on Sun Sparc 5′s, then you should praise them. If they're crushed under load, then praise them even more for their success.

pipes.yahoo.com was built in a way considered non-standard for Yahoo!. It went down almost immediately after launch and I remember Rasmus saying almost exactly the same as the quote above in response to the criticisms that the Yahoo! Pipes team received for launching on a non-standard platform.

In-house incubator

Yahoo! launched and then killed Brickhouse.

Finally, one of my all-time favourites, from a comment on slacy's post:

Looking from the outside in, I think what Google currently looks most is: a coherent strategy and focus. Its products are all over the map and new products tend to be neglected after a while.

True or not, if I had a penny for every time I heard the same about Yahoo! ...

Overall, slacy's post makes me wonder if such organisational problems are inevitable no matter how bright the minds are, and no matter how well intentioned the people are. It sure will be interesting to see what changes Google will make over the next few years, and also how Facebook evolves as an organisation, being the current fast growing behemoth that's still pumping out some amazing technology & product.

Anyone who knows me knows I play a lot of TF2. In fact I've been playing variants of Team Fortress since 1996, almost half my life time.

I think its a common belief that gaming, whilst good stress relief and general entertainment, is still largely a waste of time. I thought I might share how TF2 in particular parallels the "real" world.

One of the things I enjoy about TF2 is observing human behaviour, as much as one can within the confines of a virtual game. The game is not unlike soccer - individual skill matters greatly but so does team work. You can have the most amazing striker/shooter but without strong support and co-operation her team will still be beaten by a less skilled, more cohesive group of players.

In-game you get to observe the usual spectrum of players - the couldn't-care-less-what's-going-on-I'm-here-to-play-my-game players, usually recognisable because there are 5 snipers in a team of 12 which is a well known recipe for disaster, but all refuse to change class. On the other end you have the players who will switch class and play whatever the team needs, folks who will help other team mates for no individual reward, e.g. engineers who upgrade or repair their fellow engineers' gear are not given any points for doing so.

There are players who have to win no matter what. These players will join a server and upon realising that they have joined the weaker team will switch to the other side. They might even sit in Spectator mode until a slot opens up on the winning team rather than play with the losing team. And on the other end you have players who for whatever reason will switch from winning to losing team when there is a noticeable imbalance. Maybe through a sense of fairness, or maybe they just want a challenge? Who knows.

There are players who can turn losing teams into winning teams through natural leadership ability -- even when the team is composed of complete strangers.

All of these player types are easily recognisable and I think there are valuable lessons to be learnt that can be applied to any team situation in future. e.g. teams need a mix of personality types & skills, or that the individual's role in a team may not always be what they desire but if they do it well the team can still win; is winning as a group a sufficient trade-off?

This is all the same stuff you find in management books & courses. Does its presentation in a game format make the lessons any less valuable? Or does it make the material more accessible, even fun, to a wider variety of people?

Recently a whole new side of the game was added - the MannConomy. Basically there are in-game virtual items that can be bought, found, and traded. There is an official store where a limited set of items are available for sale 24x7 but more interesting is the "underground" economy that has sprung up.

Players are now setting up "shops" on forums such as SourceOp.com. This particular link is to a list of "trusted sellers", each of whom have their own forum thread listing items for sale, some with prices and stock levels! The buying process typically involves the buyer paying the seller via PayPal, and upon receipt of moneys the seller will initiate an in-game trade request with the buyer to transfer the virtual goods. This is currently the most efficient mechanism since the game itself does not support monetary transfers between players.

Although the mechanisms are haphazard all of the typical characteristics of a market economy are present. There are buyers & sellers, bids, asks, even escrow ("middlemen"), and price discovery happens as expected. Reputation of the participants is a huge part of the market's integrity, particularly because of the disjoint nature of payment vs goods transfer.

Browsing and speaking to a few traders its clear that there's at least two types of market makers.

One is the unusual hat trader who seeks to profit from a superior understanding of the value of each hat based on market demand & supply. S/he preys on the ill-informed participants who might let their unusual hats go for below market rate.

The other is typical of most games - the farmer who spawns multiple accounts in order to collect more items over time. The game here I guess is volume, as the price of the TF2 commodity "metal" is fast falling and their margins must be getting slim.

Whilst some of these traders are likely to be college age or older there must be more than a few who are in their teens, or possibly even younger.

I wonder how many traders are monitoring their P&L, cash flow and margins over time? They may not know the terminology but I'm pretty certain that the serious traders are running profitable ventures. How many will use this experience in future as they become "real" business owners, managers, and participants in "real" world markets?