Just yesterday I blogged about why Mediassembly switched from Amazon AWS to Windows Azure. In that post, I mentioned that I hoped to have time to describe some of the technical pros/cons of each that influenced the decision. Well, had I done that, the details would have been outdated already. Amazon just announced enhancements to SimpleDB and EC2 that address two of the key challenges and frustrations we experienced. You can read the announcements, here:
Even though we switched to Windows Azure, I’m still a huge fan of Amazon AWS. They set the bar for cloud platform providers and few have been able to reach it. While these recent announcements would not have changed our decision, they certainly narrow the gap between the two platforms for developers building on a Microsoft stack.
I’m looking forward to seeing the competition progress between these and other cloud platform vendors. As each continues to push the bar, we customers ultimately win.
Back in December I made a decision to migrate Mediassembly from Amazon’s AWS cloud platform to Microsoft’s Windows Azure. This post explains why. I’ve been holding this post in draft, waiting for time to add more detail. But to keep this from getting stale, I’ll just start with the the bottom line, which is this:
Developers! Developers! Developers! Developers!
That famous mantra from Steve Balmer is clearly reflected in the combination of Windows Azure platform services and development tools. If you’re building on a Microsoft stack, Azure offers the most productive solution with a much richer set of cloud platform development tools and infrastructure services. If you are a business or development team leader, this translates to Productivity, Productivity, Productivity, Productivity! Since I’m running an Internet startup, productivity (time to market) is a critical consideration.
The key enablers of this productivity are:
- Good .Net Wrappers around the REST Services
The .Net Azure wrappers are much richer than the comparable Amazon AWS wrappers. - Tight integration/compatibility with advanced .Net features and frameworks, such as LINQ, Ado.Net Data Services, Entity Framework, etc.
Developers can easily transition from traditional web application development to Azure. - The Azure Development Fabric
In my experience, the biggest challenge with developing for the cloud is dependency on the cloud infrastructure. The Azure development fabric provides a robust simulation environment that runs on the developer’s local machine. No internet or cloud connectivity is required, so developers can continue to work wherever they are. The development fabric includes nearly all Azure services and APIs and allows you to configure and run multiple concurrent web and worker/background role instances – as you would in production. However, unlike production, you can attach the debugger and step through the code to discover those often extremely hard-t0-find concurrency-related bugs. - Robust Diagnostics
Good diagnostics are critical to identifying and correcting problems. Azure’s diagnostics allow you to use standard System.Diagnostics.Trace APIs and enables you to collect IIS Logs, Failed Request Logs, Performance Counters, Windows Event Logs, and Application Crash Dumps – and transfer the collected data to your storage account either on-demand or at scheduled intervals.
I hope to have time to discuss some of the advantages and disadvantages of Microsoft’s and Amazon’s equivalent compute, storage, messaging and data infrastructure services. They each have their strengths and weaknesses, but the those had far less influence on the decision than the points above.
No, not the heavenly kind. I was recently invited to attend a local angel investment group meeting. It was an opportunity to meet local investors, listen to a few pitches from other entrepreneurs, and learn how the group operates. I gained some good insight and would suggest that anyone planning to seek angel investment for their endeavor find a way to attend a meeting informally before pitching your opportunity officially.
While I didn’t present, I did get to talk to several investors about Mediassembly and social media. You might think that social media, being a very hot topic these days, would naturally be of interest to investors hoping to ride the next big wave. But many waves crash into rocks. The investors I’ve met tend to be conservative and skeptical of anything that is overly hyped. They’re interested in finding companies that provide innovative, yet practical solutions to tangible problems. The clearer the problem, the easier it is for them to understand the opportunity.
Social media is indeed a phenomenal wave that is changing the way people interact with each other and the way businesses interact with customers. And it is tempting to lead into discussions with how social media is changing the world and citing amazing Facebook and Twitter statistics. But doing so will likely trigger the skeptic’s spidey sense and disengage your audience. Statistics are important, but only after you demonstrate that you’re offering a valuable solution to a clear and significant problem. Sell your solution, without incorporating the hype. The good news is there are plenty of problems to be solved with social media. You don’t need the hype.
While the context of this post is speaking to angel investors, I believe the advise is broadly applicable. It seems there are many agencies and social media champions inside organizations who encounter skepticism while proposing a social media strategy to senior management. I have found that a healthy skepticism is a trait common to both executives and angels. Angels are often retired executives, after all.
There have been several great posts and presentations recently that address the ROI of Social Media. Whether you’re talking to investors or executives, you should be well versed on the bottom-line impact of your solution and how ROI will be measured. This is a good place to start:
