HNHacker News
TopNewBestAskShowJobs

citrablue

366 karma · joined May 1, 2018

submissionscomments
citrablue··on I'm a Heart Doctor and Wary of the New Apple Watch
The argument against more monitoring is one of false positives. Suppose the Apple Watch has a 5% false positive rate, and detects problems with 1% of 100 million users.* That's 50k extra patients going to their doctors, thinking there might be something wrong with them. The next test may have a 1% false positive rate. That would put 500 extra patients on anti-coagulants.

That may not materially change the public health calculation, or it may. But that's the calculus -- not that a GP would blindly prescribe, just that increasing the top of the funnel is not necessarily a good thing. (Plus, extra cost.)

Positives and negative impact of additional testing must be measured for optimal decisions.

* numbers to make the math easy

citrablue··on Inside Facebook’s Election ‘War Room’
Private interests and public have always interacted. Whether this fact matters is open to debate, but I find dealing with current reality more useful.

As someone unfamiliar with Australian politics, is The Guardian AU a reputable source for this story? I don't generally consider that domain, granted a .uk, very useful to provide information.

citrablue··on Inside Facebook’s Election ‘War Room’
TV and print media went through a period of regulation when they had a similar market share. We may be able to learn from that time.
citrablue··on Is “data scientist” the new “programmer”?
I think of engineering as the practical counterpart to science. Science is finding patterns/uncovering truths/building and testing models; engineering is the deployment of technology to fulfill an objective.
citrablue··on Monolist: working hard to preserve the spirit of Inbox's organizational features
I use (and pay for, ~$50/mo.) an awesome plugin called Gmelius. It really upgrades the gmail experience to a professional-grade system. Also lets you strip out a ton of the unnecessary gmail web chrome.
citrablue··on Stripe Terminal – Programmable point of sale
I would be impressed if a company had 100 employees today and planned for exponential headcount growth (say, 50% growth per annum). And had taken $440 million in funding. And had a product that wiped the floor with the competition.*

* YMMV, but for this dev, Stripe was a godsend.

citrablue··on Stripe Terminal – Programmable point of sale
The numbers quoted here[0] are soft and from a year ago, but at that time, Stripe claimed to handled handles tens of billions of dollars in internet transactions annually. If you were doing 10+% of that, kudos to you!

[0 ]https://www.bloomberg.com/news/features/2017-08-01/how-two-b...

citrablue··on Linus Torvalds apologizes for his behavior, takes time off
To take it one step further, I'd say that it's more of a probability function than a 2d chart. I think of it more as your chart, plus a 3rd dimension of "context", then a density cloud.

Put another way, people have probability distributions of alignments based on context.

citrablue··on Could sonic booms stop hurricanes? (2008)
I think it's likely, having just harnessed the power of the atom after growing up w/o electricity, Teller assumed the problem of fallout was a simpler one to solve than what he'd just worked on. We see the standstill, where radiation was unresolvable... but wouldn't he see it as a smaller hill after he just climbed a mountain?
citrablue··on Let’s bring back the Sabbath as an act against ‘total work’
Sacred usually means something that is above questioning. If that's the common definition, there's not a lot of value in that assignment.
citrablue··on Let’s bring back the Sabbath as an act against ‘total work’
I'll advance a more American/Individualist comment for perspective.

Government should be limited in what it can control. Let the people and social norms make the decisions. The problems with government are worse than those of the free market in the long run, because they create friction and power imbalance between what people want and what another entity tells them is OK.

If your core value is the Government is too strong a force to be allowed to do much anything, then this makes sense. (Personally, it makes some sense to me -- limit the power of the gov't then the worst thing a Trump administration can do is deadlock. Imagine if he had more power?)

citrablue··on What Ötzi the Iceman’s Tattoos Reveal About Copper Age Medical Practices
The hypothesis that hunter-gatherers spent more time getting food does not seem to be supported or accepted in academia. Lots of research out the past few years that indicate the HG lifestyle had much more leisure time than agrarian.

This may be confounded by different lifestyles -- without incentives to create and store surplus, no need to work more. No extra benefit.

I am not an expert, but it seems very unsettled whether HG had more or less time spent at work.

citrablue··on Ask HN: HIPAA Compliance?
Approach compliance like security -- it'll never be 100%. Except with HIPAA compliance, your goal is both to prevent data leaks as well as document to cover yourself legally.

Ignorance is no excuse, so it's important to take the recommended precautions. I would suggest hiring a 3rd party auditor with some sort of certifications -- that'll be an important CYA.

In general, you want to have written protocols and documented auditing for the following (plus some general concepts):

* Encrypt at rest, in transit

* Who has access? What is the policy for granting/revoking access? Is there an audit trail to see who accessed what, in the event of a breach?

* What data / systems are considered PHI/PII?

* How do you train employees about HIPAA?

Most of this is documenting processes and then showing that you review the logs every 3/6/whatever timeframe.

A more complete list: https://www.hipaajournal.com/what-is-a-hipaa-violation/

citrablue··on Google Optimize now free for everyone
The different between "traditional" A/B testing and Multi-Armed Bandit has nothing to do with the number of treatments. A/B is really shorthand for A/B/n.

In an A/B test, the probability of selecting an "arm" (a treatment to show to the visitor) is equally distributed. Google Optimizer differs by adjusting the traffic distribution, sending more traffic to better performing options, what's usually called MAB approach.

The difference is that MABs will more quickly converge on the "winning" variation, but are more likely to get stuck in a local minimum. EG, if an "worse" option performs better right off the bat, the MAB might send most traffic to it. This would take a while for the algorithm to "recover".

The major advantages of the MAB approach are minimizing opportunity costs and the ability to capture seasonality because you can continuously run tests. Traditional split testing runs for a while, gets a result, and moves on. With a MAB, you can assign an "explore" budget that keeps tests running in the background to capture the seasonal/periodic change in conversions.

That comes with a cost: every visitor who doesn't see your best page is lost revenue.

citrablue··on Google Optimize now free for everyone
Google Optimize offers MAB testing, which is awesome. Personally, I don't see it as a "when do results become reliable" problem -- it's a "what is the cost of a false positive, false negative, or 'do nothing' decision?"
citrablue··on Google Optimize now free for everyone
Some pretty easy rules of thumb, assuming you have a decent grasp on your economics. Look at it as a "low hanging fruit" optimization problem -- do you put resources into running a test (+ opportunity cost for lost sales), or into something else?

Suppose you have 10k monthly sessions with a 0.5% conversion rate (50 conversions). How many more customers would you need in order to prioritize running a test? If 55 conversions in a given month means you crush important KPIs, then that's probably worth testing -- you just need a 10% lift.*

Also keep in mind that running A/B tests (1 control, 1 treatment) is suboptimal. That tests, "does this beat what I have now?" The more important question is "what is my best option?".

OTOH, if other things like messaging and product are stable, you can test a smaller traffic site by leaving it running longer.

My rough estimate is 100 conversion events in the time the test runs. So if I have 100 conversion events in 1 month, it may make sense to run a 2-3 option + 1 control test for 1 month.

(You can also test much larger things than buttons. For startups, I like to suggest trying out positioning or value statements and seeing how visitors respond!)

* however, it'll take a long time for you to reach statistical confidence for a 10% lift in rate, with only 50 conversion events across all tests.

citrablue··on How the Brain Experiences Time
The super fascinating thing is the difference between how the brain experiences time and how the brain remembers itself experiencing time. It seems possible that intense experiences cement more details in your memory, which your brain interprets as having taken longer, without actually affecting the "in the moment" realtime experience.
citrablue··on Cognitive Biases and the Human Brain
I'd include Hindsight Bias or perhaps Survivor Bias as well. But excellent list!
citrablue··on Ask HN: What do you struggle with?
Have you tried putting yourself out there for work? As patio11 recently said, try to fail out of 20 interviews. It's un-possible -- the hiring process is so broken, you'll get an offer.

Then it's up to you to catch up to their expectations. If you fail in that, so what! You'll have your first gig on your resume, and I bet you'll have learned a lot about working as a dev in that time. You'll be miles ahead of grinding it on your own.

Just have several strategies for coping with failure or disappointing yourself, otherwise it can be extra rough.

citrablue··on Ask HN: What do you struggle with?
I can empathize with this. Add in the fact that the older I get, the more impressive my project needs to be in order to impress myself. It's a form of self-sabotage, but by realizing that I don't have to build something great in 1 day, it helps.

What are your strategies to cope with this?

citrablue··on Insured, but Still Owing $109K for a Heart Attack
I don't think this follows. Insurance profit margins are restricted, but they're a percentage of overall payments to hospitals. If the total charge goes up, their total profit also increases.
citrablue··on Using AWS CodePipeline, CodeBuild and Lambda for automated UI testing (2017)
In my experience, people shopping for solutions will latch onto one that claims to do what they need. They then build a vendor comparison matrix that results in selecting that technology, but makes it look like they considered alternatives. This is because people don't make rational decisions, but instead rationalize decisions.

The tack they took here is, IMO, the most efficient. It's an easier effort to publish this article than one exploring other options, and provides a linear path to success for naive readers (rather than introducing the "other options" branching paths).

citrablue··on Artificial Intelligence Is Now a Pentagon Priority. Will Silicon Valley Help?
I disagree, I think the military exists for the purpose of advancing American interests at home and abroad. Think of ships like the USS Mercy, or other humanitarian efforts.

What makes you think those innovations would have been created without the military? The US military does a lot of research that doesn't have immediate practical/commercial applications. They cost a lot of money, and don't have obvious payoffs, which make them a poor candidate to be funded by private investment.

citrablue··on Need More Self-Control? Try a Simple Ritual
Agreed, what you say reflects my own growth as I've gotten older. I'm hesitant to call it wisdom, but it's served me better than the outlook from my youth.
citrablue··on The Student Debt Problem Is Worse Than We Imagined
Similarly, people who survived the system should have just as much access to remedies as people currently in their own difficulties.
citrablue··on The Student Debt Problem Is Worse Than We Imagined
3 questions 1/ Are there solutions on the table where the cure isn't worse than the disease? That money has been spent, it came from somewhere. Can it just be vanquished without repercussions?

2/ Objectively, what is the likelihood that continuing the status quo crashes the system?

3/ What are the long term consequences from potential solutions? E.g., if current debts are forgiven/cancelled, would lenders refuse to lend for colleges? Would well off people with significant debt take advantage of the system we're protecting, leading to less trust in it, which could cause it's own collapse movement?

The defensible logic is that we don't really know that current student debt will "crash the system", and we definitely don't know what the 2nd and 3rd order consequences would be if any of the solutions were implemented.

citrablue··on Need More Self-Control? Try a Simple Ritual
As a counterpoint, biological systems grow stronger when stressed then allowed to recover.

Routines are great when interspersed with periods of non-breaking stress. Growth is dependent on stress.

I refer to a more general term of stress, not necessarily psychological. Although the latter can make you a stronger/more empathetic person; the danger is in not being able to let go of them for the restoration/growth period.

citrablue··on Show HN: A side project in medical devices
Of course not, it's our own biases that lead us to think we are the gods through the expression of a creative process. Because we are human, and cannot see past what's in front of us, we mistake our tools for creating.

The first steam engines were riveted by metal workers who may have felt similarly.

citrablue··on Power Worth Less Than Zero Spreads as Green Energy Floods the Grid
> utilities love to just install more generation because it's in their playbook and they don't want to innovate.

I'm curious, how did you reach this conclusion?

citrablue··on Power Worth Less Than Zero Spreads as Green Energy Floods the Grid
A desal plant is a useful thing and can be run around the clock off traditional energy sources. The "free energy" hours would help lower the costs.

I can even imagine a SETI-like application where people who over-generate power are able to donate it to causes of their preference...

← PreviousPage 4 of 7Next →