Disqus - Friends of spencerfry

Re: Testing - Continuations

albert — Fri, 29 Feb 2008 21:33:27 -0000

Testing a comment

Re: Fred Wilson Dot VC

albert — Sat, 01 Mar 2008 08:43:47 -0000

This is a building in Vienna, Austria by an architect named Friedensreich Hundertwasser

Re: Fred Wilson Dot VC

albert — Sat, 01 Mar 2008 08:44:53 -0000

More pictures here http://www.phototravels.net...

Re: Welcome to Monte Carlo, Excel - Continuations

albert — Mon, 17 Mar 2008 12:25:13 -0000

Spot on -- am planning to post about how to introduce correlation into the model and how the use of historic correlations has been the major downfall in modeling mortgage backed securities.

Re: Location, Location, Location - Continuations

albert — Sun, 30 Mar 2008 11:41:33 -0000

Thanks! The Lijit implementation is neat -- I am planning to add lijit search to this blog. Quova looked like the gold standard based on my admittedly brief research, but in terms of getting something up and running quickly I did not like that they did not have a self service option with published pricing.

Re: Multiple Displays and (Developer) Productivity - Continuations

albert — Sun, 30 Mar 2008 11:47:00 -0000

I think a lot of developers do that too where they have IM or an IRC channel open in one corner of the screen. IRC is great for distributed teams, but even then I wonder what the tradeoff is between individual and team productivity. Maybe I am simply too old, but when I really want to concentrate on something (e.g. coding), I need to turn off all alerts and other interruptions first.

Re: I (May) Have a New Platform

albert — Tue, 08 Apr 2008 11:37:45 -0000

That's definitely an important concern -- also one of the reasons why google entering this arena with an interesting offering is just the beginning.

Re: I (May) Have a New Platform

albert — Tue, 08 Apr 2008 11:40:43 -0000

Always happy to learn more -- either post here or send me an email.

Re: I (May) Have a New Platform

albert — Tue, 08 Apr 2008 11:43:02 -0000

Completely agree on competition being good for cloud computing and hope to see others (with differentiated offerings) joining the fray. There are a number of interesting other offerings already out there (e.g. salesforce) and others just launched or about to be launched.

Re: Structural Change and Marketplaces

albert — Wed, 09 Apr 2008 08:06:05 -0000

That's a great observation - heavy handed monetization via ubiquitous and high transaction fees can be a huge hurdle. The trick seems to be finding the right mix between growing a marketplace and charging for only a small fraction of transactions (eg Craigslist) or keeping the transaction fees tiny compared to the value of the transaction (eg Etsy).

Re: Traction Revisited - Continuations

albert — Tue, 15 Apr 2008 13:31:46 -0000

There is no doubt that it would be better for the entrepreneurial ecosystem, but the question I am trying to address is whether it makes sense from the investors' perspective. That's where I believe the disconnect is because the risk reduction from adoption is so significant.

Re: Traction Revisited - Continuations

albert — Tue, 15 Apr 2008 13:34:52 -0000

I would say best of both worlds. There are a important issues with the approach that Paul proposes for the companies themselves. For instance, suppose you take money from a VC firm that follows Paul's approach and you do make progress, but not enough to make that VC's cut for follow on among all the companies that they have decided to spread money to. How do you explain this to new investors? That's one of the great benefits of angel investors from a company's perspective. Nobody coming in later expects the angels to continue funding.

Re: Team Versus Traction - Continuations

albert — Tue, 15 Apr 2008 15:07:11 -0000

I don't think you can infer that. Clearly the company has a team that has gotten something right, but that team could have lots of issues that would prevent it from achieving continued growth and realizing the company's full potential.

Re: Team Versus Traction - Continuations

albert — Tue, 15 Apr 2008 15:11:50 -0000

Agree -- the risk reduction tends to be larger though than the increase in valuation, which is the main point of my post on Paul Graham's exhortation to VCs to invest in companies before they have achieved traction. On reason for this is that valuations are bounded from below because you don't want to take away too much equity early on.

Re: Traction Revisited - Continuations

albert — Tue, 15 Apr 2008 22:32:23 -0000

I spent a few minutes tonight running some numbers on the type of investment strategy that Paul describes of making 5 investments of $400K instead of 1 investment of $2 million. I assumed a simple base case of $2 million resulting in a 20% ownership stake at exit of a company that is worth $100 million with a 20% likelihood and $0 with an 80% likelihood. With these numbers the investor will earn a 10x return 20% of the time and lose everything 80% of the time for a 4x return on expectation. Now compare this to 5 investments of $400K each. Even if you invested at a pre-money as low as $1.5 million, i.e. starting out with a 21% ownership it's unlikely that you would own more than 15% of the company at exit. Now at first I assumed a linear risk relationship where each of the 5 companies has a 4% chance of being worth $100 million. Turns out that almost 85% of the time you would lose all the money (so no improvement here), about 15% of the time you make 7.5x return and very rarely (about 1% of the time) you make a 15x return, but on expectation you are only making a 2.6x return. In fact I believe these assumptions are still too favorable for the 5 investment case for both the final ownership and the increase in risk.

Re: Traction Revisited - Continuations

albert — Wed, 16 Apr 2008 08:28:02 -0000

Two points really work against us here. The first is the one that Paul rightly emphasizes often -- it has become much cheaper to start. As a result many more things get started and that means many more things get started that are potentially / actually competitive with each other. Add to this the second point that why some services take off while others don't in any one category may seem obvious after that fact but is really hard to tell up front. If someone had shown you del.icio.us and any number of other social bookmarking services without any information about traction which one would you have picked to invest in based solely on seeing an early version of the product?

Re: Traction Revisited - Continuations

albert — Wed, 16 Apr 2008 08:30:08 -0000

Cem, I would modify this slightly. Traction is not a test of a products mass market viability because of the early adopter effect, but I do believe in a world of capital efficient startups it is a much more important test than it ever was.

Re: Clouds over Simple Paid Apps / Services - Continuations

albert — Wed, 16 Apr 2008 08:37:03 -0000

Thanks. Although I am lost as to the "Spime, much?" bit at the end.

Re: Large Hadron Collider is a Sucker's Bet - Continuations

albert — Wed, 16 Apr 2008 11:27:45 -0000

That's one of the claims that scientist have been using to promote the LHC research, but as I wrote if you look at "reasonable" time frames, by which I don't mean next year, but would happily go out 10-20 years, we could get a lot more bang for the buck in funding research on solving the few key problems (e.g. batteries) that keep us from a rapid shift to existing clean tech. I am also for funding more research on fusion and spreading that money a lot more widely than the Tokamak approach which has been sucking up all the funding (ITER is another European Union pet project).

Re: Clouds over Simple Paid Apps / Services - Continuations

albert — Wed, 16 Apr 2008 12:21:03 -0000

Haven't read the book - will check it out. Thanks!

Re: System and Organizational Scaling - Continuations

albert — Mon, 28 Apr 2008 09:35:02 -0000

I think there are very few generic ones. Etsy would have a pretty different set of internal services from say Wesabe or Twitter. One good way to identify internal services is to think about the overall API for the site or service and look for logical groups. A candidate that might apply to a lot of sites is a 'social graph' service.

Re: System and Organizational Scaling - Continuations

albert — Mon, 28 Apr 2008 20:22:48 -0000

Agree that APIs need a good approach to versioning for changes that are not backward compatible. Ideally though, most changes to an API make it more powerful without breaking existing apps.

Re: System and Organizational Scaling - Continuations

albert — Mon, 28 Apr 2008 20:24:51 -0000

Good point -- discipline is required to prevent the internal services from being overloaded with features.

Re: System and Organizational Scaling - Continuations

albert — Mon, 28 Apr 2008 20:29:15 -0000

You are right. I did not mean to imply that these layers disappear, simply that they are encapsulated. They beauty is that the implementation of layers might differ across various internal services, using the technology that's most appropriate for the particular service.

Re: Gasoline Pandering - Continuations

albert — Wed, 30 Apr 2008 22:07:21 -0000

So true about the lack of public transport -- even some of the big cities are accessible by car only. Was just in Phoenix/Scottsdale, which is a perfect example.