I just published the source code for one of my side projects, isvalid.org.
IsValid started as a little tool for myself to automate data analysis. It will take the results of an A/B test, perform some basic statistical tests (significance and confidence intervals), and return some data and charts to help you understand what the results mean.
If you’re into that sort of thing, check out IsValid on Github.
If you save anything online, assume you will never be able to actually remove it from the web.
Alternatively, assume that once you delete something, you (and only you) will never be able to recover it.
Everything else will be simple.
Core revision #18742
wp_strip_all_tags() in wp_trim_words() to remove CSS and JS within tags. Props evansolomon. Fixes
#18726.
One month ago I bought a DNA kit from 23andme and I just got my results. There is far too much data to make any real sense of on first glace, but it’s certainly fun if nothing else.
I have lower than average risks of Alzheimer’s and Parkinson’s — about 30% and 15% lower, respectively. I have a 75% lower than average risk of type 1 diabetes, but a slightly higher (8%) than average risk of type 2.
Of all the diseases I am at high risk for, Venous Thromboembolism (causes blood clots and things) and prostate cancer are at the top. I have about a 1/3 chance of getting each of them. Additionally, I carry one known problematic trait, Alpha-1 Antitrypsin deficiency, which makes parts of my body (particularly fragile lung tissue) less able to protect itself from an enzyme released by white blood cells. This increases my risk for lung and liver diseases. I am more sensitive than average to warfarin, a commonly used blood thinner, which means that if I ever need it my optimal dose would be smaller than normal. On a bright note, I have decreased risk of male pattern baldness.
Overall I’m quite “typical” according to 23andme. Most of my appearance factors (hair curl, freckling, etc) fall into the typical range, as do most of my chemical predilections (food preference, alcoholism risk, etc). My ear wax stands out as “wet”, my non-verbal IQ (whatever that is) is likely to be slightly higher, my likelihood of complications if I contract malaria are higher than normal (something I’m glad I didn’t know before going to Vietnam) and my odds of menopause are unsurprisingly not applicable.
My genes corroborate the stories that my mom’s family was northern European and my dad’s was eastern European Jews. This makes me — and this will come as a surprise to literally no one that has met me — overwhelmingly white:
As has become the norm, a new web security meltdown erupted recently, starring KissMetrics, among others. And as has become the norm, I think it’s much ado about nothing.
I was actually thinking about this earlier today, and Nik Cubrilovic’s post on yet another way to secretively store data just reminded me of it. There are a few reasons why I think the concern over this is mostly misguided.
First, it’s not new. Store owners can write down your shirt color, your height, your race, the time you came in, what you bought, etc. They can track plenty of “personal” information, which I put in quotes because I think the whole concept of some special class of information leads to more confusion than awareness. The key difference, obviously, is that computers make tracking this information cheaper and faster. I tend to think that we shouldn’t legislate against actions based only on their relative efficiency, so I don’t see why this should make a difference. Worse yet, it’s far more likely that real world businesses can connect your actions to your real “personal” information, like your name, family and address.
Second, it’s a never-ending game of cat and mouse. There are limitless ways to store data and acting like anyone can stop it is foolish. The whole thing seems eerily similar to the war on drugs, a failure I think we’ll have a hard time contending with any time soon.
Third, for the overwhelming majority of people there is nothing to worry about. Awareness of risks is great, but confused fear based on misinformed media reports is awful, and that’s mostly what we’ve created.
I’d liken responsible web security education to something along the lines of wearing your seatbelt and not driving drunk. What we have today is much closer to fear-mongering along the lines of urban legends about exploding engines or murderers at drive in movies. Those things can certainly happen, but I don’t think it’s reasonable for anyone to treat them as likely consequences of driving a car.
To be clear, I think it’s great if websites and web services don’t do these things, but some of them will and you should probably assume all of them do, particularly if it’s something that concerns you.
At a basic level, metrics are incentives. By measuring quality, features, and schedule and discussing them every staff meeting, my people intensely focused on those metrics to the exclusion of other goals. The metrics did not describe the real goals and I distracted the team as a result.
I wrote a post for a new blog I helped start called Perfecting The Upset.
The 1997 Masters is my earliest golf memory, which probably doesn’t make me unique among mid-20′s Americans. What does make me unique is why I remember that tournament. When asked about anyone in the field catching Tiger Woods after Saturday’s third round, Colin Montgomerie said:
“There’s no chance humanly possible that Tiger is going to lose this tournament. No way.”
Read the rest of the post over on Perfecting The Upset.
You don’t want people using your product because it’s cool, because then you’re a fad. What you want is people using your product because it’s a part of their life and they can’t stop using it.
-Sean Parker