andy britcliffe

Simple and effective

select {
    case <-chanToReturn:
      // Do work
    case <- time.After(5 * time.Second):
      // out of time so carry on
    }

blakemasters:

Here is an essay version of my class notes from Class 6 of CS183: Startup. Errors and omissions are my own. Credit for good stuff is Peter’s entirely. This class was kind of a crash course in VC financing. I didn’t include all the examples since you can learn more about VC math elsewhere, e.g. here or here. As usual, though, I’ve tried to include all the key insights from the lecture.

CS183: Startup—Notes Essay—Thiel’s Law

I.  Origins, Rules, Culture

Every company is different. But there are certain rules that you simply must follow when you start a business. A corollary of this is what some friends have (somewhat grandiosely) called Thiel’s law: A startup messed up at its foundation cannot be fixed.  

Beginnings of things are very important. Beginnings are qualitatively different. Consider the origin of the universe. Different things happened then than what we experience in everyday life. Or think about the origin of a country; it necessarily involves a great many elements that you do not see in the normal course of business. Here in the U.S., the Founders generally got a lot of things right. Some things they got quite wrong. But most of the time they can’t really be fixed.  Alaska has 2 senators. So does California. So Alaska, despite having something like 1/50^th of California’s population, has equal power in the Senate. Some say that’s a feature, not a bug. Whatever it is, we’re likely to be stuck with it as long as this country exists.

The insight that foundings are crucial is what is behind the Founders Fund name. Founders and founding moments are very important in determining what comes next for a given business. If you focus on the founding and get it right, you have a chance. If you don’t, you’ll be lucky at best, and probably not even that.

The importance of foundings is embedded in companies. Where there’s a debate or controversial claim at Google, one says, “The Founders have scientifically determined that x is true,” where x is his preferred position. If you think that certain perks should be extended since happy people are the most productive, you say that Larry and Sergey have already settled the matter. The point is that all the science is done at the founding. No new data can interfere with the founding moment.

Foundings are obviously temporal. But how long they last can be a hard question. The typical narrative contemplates a founding, first hires, and a first capital raise. But there’s an argument that the founding lasts a lot longer than that. The idea of going from 0 to 1—the idea of technology—parallels founding moments. The 1 to n of globalization, by contrast, parallels post-founding execution. It may be that the founding lasts so long as a company’s technical innovation continues. Founders should arguably stay in charge as long as the paradigm remains 0 to 1. Once the paradigm shifts to 1 to n, the founding is over. At that point, executives should execute. 

There is, of course, a limit to how much you can do with rules. Things can and will break down even with perfect rules. There is no real chance of setting things up correctly such that the rest unfold easily. But you should still get the early stuff as right as possible.

Read More

I watched a very good interview with David Sacks of Yammer at Launch 2013 which covered a range of topics but the section that stood out for me was how David looks to do his Angel investments.

He listed 4 key things he looks for when investing in a startup:

1. Product hook. That simple interaction that keeps users coming back
2. Solving a market problem that exists. This will often require iterating towards a solution
3. Distribution model. How does it spread??? David is looking for virility.
4. Non-copyability. What sets you apart from others -e.g the network effect.

3 out of the 4 will usually get David on board as the 4th can be added.

David is a very impressive CEO and while watching a thought crossed my mind - possible future MSFT CEO - they could do a lot worse.

The full video is here: http://www.youtube.com/watch?v=ps61OjznUp8

I wonder if we will see one in the next 5 years! #notconvinced

Sequoia Capital (@Sequoia_Capital) tweeted at 6:12 PM on Mon, May 20, 2013: One of the fatest decisions we’ve ever made, looking back at when we first met @davidkarp http://t.co/mlq6HBE0VE (https://twitter.com/Sequoia_Capital/status/336529792453193728)

Little trick I’ve discovered so I can read the package docs for Go while offline.

If you have go installed execute:

godoc -http=:6060

and then head to http://127.0.0.1:6060/pkg/ in your browser.

As part of daily work life I’ve been doing a decent amount of thinking around platforms and the structure of technology teams. I just wanted to reference a couple of very useful posts I’ve found on these topics. At the very least it will be useful documenting them here for my future reference and as a stretch goal maybe others reading this will find them useful.

First one is Steve Yegge’s Platform rant - a post that was due to be internal to Google only but managed to sneak out, has great insight into how Amazon went about transforming themselves from a products company into a platform one and also some of the challenges a forward thinking company like Google still face.

The second one is a great post from Marc Andreessen giving his thought process on what he sees as a platform (and from my own discussions it matches the view that many people have many different interpretations on what a “platform” is) and also the 3 levels of platform on the internet with “level 3” being what we now know today as PaaS. Again another insightful read IMHO.

We’ve launched a series of short videos showing some of the key use cases of sharpcloud to help solve key business problems such as:

1. Visualize your business data - Go beyond Excel
2. Capture and develop your knowledge - Go beyond sticky notes and white boards
3. Communicate your plans to your business and customers - Go beyond Project
4. Make an impact through dynamic presentations - Go beyond PowerPoint
5. Connect people and knowledge - Give life and ‘context’ to your SharePoint content

Don’t forget you can sign up and try sharpcloud for yourself here

So node.js is hot and it’s been something I’ve been playing with for a while now in a purely kick the tyres type way. Now sharpcloud lives in a Windows Azure world and while we’ve been busy utilising amazing tools like Redis and MongoDB in our Azure production environment Node.js has been something that we’ve not been able to utilise as realistically it needed a native port to make use of IO completion ports to get the full node.js async goodness. 

So back in May Ryan Dahl (the creator of node.js) presented the future of node.js and in it was a view to bringing a native version of node.js onto windows. Obviously this was exciting news but it was unclear what the timeframe would be so I withheld too much joy :-) …then I saw this post indicating Microsoft were aiding in the effort which increased my excitement (geeky I know). 

Jump forward a few weeks and low and behold an unstable v.0.5 release is available as a native windows binary - fantastic stuff. Anyway my attention immediately turned to trying and getting a test deployment on Windows Azure and that’s exactly what I’ve done.

I’ve created a skeleton Azure VS2010 solution with a worker role that houses node.js and binds it to port 80, and I’ve succesfully run it on the Azure cloud which can be seen here (note: this may disappear at some point). I’ve also pushed the solution to my GitHub account here so feel free to use it to starting taking advantage of Node.js goodness on Windows Azure.