For example: using the phrase "butters no parsnips" in the USA butters no parsnips.
My usual helper, Wikipedia, failed me, so I have to fly solo.
This phrase denotes that the action of the sentence does not help in any useful way. Buttering parsnips (ahead of roasting, perhaps), is clearly thought of as a positive outcome. This nifty little English phrase pours scorn on an attempt to do something by saying that the action will have no good outcome, not even buttering any parsnips (which, the phrase suggests, ought to be easy).
Of course, explaining this to Americans butters no parsnips, either.
The Eight Fallacies of Distributed Computing
Essentially everyone, when they first build a distributed application, makes the following eight assumptions. All prove to be false in the long run and all cause big trouble and painful learning experiences.
- The network is reliable
- Latency is zero
- Bandwidth is infinite
- The network is secure
- Topology doesn't change
- There is one administrator
- Transport cost is zero
- The network is homogeneous
For more details, read the article by Arnon Rotem-Gal-Oz.
I am not a sophisticated software engineer, and I rarely create distributed applications (who am I kidding? I never do!). However, I know exactly what all these fallacies mean. I even know what #6 is about (which Andy says mystifies him). It means that you can never find the right person to fix your problem if the application is spread all over the internet.
This leads me to wonder what are the fallacies of being a Venture Cyclist ... hmm, a topic for a subsequent post.
First, I don't want to die.
Second, it's interesting.
Third, I'm a venture capital investor ... Healthcare is a major (and growing) area of our economy and as such presents investment opportunities. Finally, healthcare and IT are converging.
One area of healthcare that is of interest to an IT investor is known as "personalized medicine". As usual Wikipedia explains this well. My precis is that personalized medicine is old-fashioned medicine plus computers. Personal genetics is a case in point.
Take 23andMe, the genetics company founded by (Google founder) Sergey Brin’s wife Anne Wojcicki that launches today. You can send 23andMe a saliva sample, and they extract lots of information from your DNA about your ancestry and other information that may be health related such as genetic markers for predispositions to certain conditions.
My (lay) understanding is that 23andMe surveys your genome for SNPs (single nucleotide polymorphisms). Competitor Navigenics will offer a similar service, but will also offer partial or full sequencing of your DNA as well. Sequencing is more expensive for lots of technical reasons, but can provide more information because not all genetic markers are found only in SNPs (or so I'm told).
All this information about your genes is discovered by computerized lab systems, stored in large computers using sophisticated databases, protected by hi-tech security, accessed by complicated software designed to uncover interesting facts and generally spends its life inside IT infrastructure. In the same way Google thinks this is an interesting business (Google invested in 23andMe), I think so too.
The policy implications of all this personal data overlap, for all sorts of reasons, with policy issues related to personal health records, which I wrote about last week. The US Congress is currently considering the Genetic Information Nondiscrimination Act (GINA). The issues of privacy for this information are magnified in the US because of the perverse healthcare system. Insurers and employers are directly charged for the healthcare costs of their subscribers/employees and so may be tempted to use genetic (or other personal health) data to discriminate against applicants. Some physicians do not even order some tests because they know that an insurer or employer could use the fact of just the diagnostic event to exclude someone from employment or coverage (whether or not the test is positive - which also ought not to be a factor).
Between managing the data, creating secure appropriate access for family, care-givers and health providers, and using anonymized data in aggregate for research there are lots of opportunities for a venture capital investor to dabble in this field, as many already have.
However, as you can see, there are good reasons for everyone to be interested and informed.
By the way, one reason NOT to get yourself tested by 23andMe or Navigenics, at least for now, is the fact the medical community has no idea how to deal with the incidentalome.