Four short links: 25 April 2011

E-Referral Evaluation Interim Findings — in general good, but note this: The outstanding system issues are an ongoing source of frustration and concern, including […] automated data uptake from the GP [General Practitioner=family doctor] PMS [Patient Management System], that sometimes has clearly inaccurate or contradictory information. When you connect systems, you realize the limitations of the data in them.
c64iphone (GitHub) — the source to an iPhone/iPad app from the store, released under GPLv3. It incorporates the Frodo emulator. Sweet Freedom.
mlpy — machine learning Python library, a high-performance Python package for predictive modeling. It makes extensive use of NumPy to provide fast N-dimensional array manipulation and easy integration of C code. (via Joshua Schachter)
What is The Truth Behind 9 Out of 10 Startups Fail? (Quora) — some very interesting pointers and statistics, such as Hall and Woodward (2007) analyze a dataset of all VC-backed firms and show the highly skewed distribution of outcomes. VC revenue averages $5 million per VC-backed company. Founding team averages $9 million per VC-backed company (most from small probability of great success). The economically rational founding team would sell at time of VC funding for $900,000 to avoid the undiversified risk. (via Hacker News)

Healthcare Data, C64 Emulator, Python Machine Learning, and Startup Success Stats