HNHacker News
TopNewBestAskShowJobs

aabajian

1,743 karma · joined December 29, 2010

aabajian aaaaaat gmail.com 697b79
submissionscomments
aabajian··on Removing SMS support from Signal Android (soon)
Alright, it's fine for a company to remove features, if they are honest about why they are doing so. It's obvious none of the reasons given are due to user complaints. The truth is, they are removing SMS because they don't own the SMS platform (e.g. it's not a walled-garden like WhatsApp). Would it kill them to just be honest? Yes, it's less secure, but no end-user is saying, "Please remove SMS as it's not secure."
aabajian··on SAT score distributions in Michigan
The SATs, particularly the mathematics score, is very much a reflection of your parents' willingness to push you academically. The stratification is more granular than Asian versus Black or White versus Native America. For example, Jewish students excel beyond the Caucasian average while Cambodian and other Southeast Asian Immigrants perform significantly below the national average for Asian Americans. The challenge for large and heterogeneously diverse nations like the U.S. is to offer an educational system that provides access to quality education regardless of parental involvement. There is also the need to provide quality education to adult learners, a wholly different, but equally important topic in the area of rapid technological change.
aabajian··on NYC Proposal Offers Cash for Spotting Parking Violations in Bike Lanes
I 100% agree. I see those placards on the most expensive vehicles (Lamborghini SUV, Porsche 911s, etc). Either the NYC police are very well paid, or they are selling their placards.
aabajian··on Show HN: Red Goose – Convert your website to mobile app
I put my website in and all it says is, " Is that site live yet?"
aabajian··on What would a “good” WebMD look like?
I'm an interventional radiology resident near the end of my training. I think Google needs to index the clinician-facing side of UpToDate. It can't right now because it's paywalled.

For those unfamiliar with UpToDate, it is considered the #1 online reference for evidence-based medicine. What they do is write articles that summarize research articles from each domain. They have much more content than this, but it's the expert-summaries that make it a "killer app" for clinicians. We simply don't have time to read all the latest research.

Also, UpToDate is the #1 most expensive subscription that most academic universities pay for (e.g. in terms of journal subscriptions).

aabajian··on Robotic motion in curved space defies standard laws of physics
I get what you're saying, but this is why people are brushing off the article because they think it's either obvious, or due to friction. It isn't either.

The phenomenon you're describing is a gyroscopic effect. The entire system (you, the chair and the weights) rotates. That is not what the authors have setup.

The author's system includes the weights and their tracks, but not the arm connecting them. The arm is the "space" (or surface) in which the weights and tracts reside. In your example, planet Earth would be the "space" (or surface).

In order for you to demonstrate their phenomenon, your chair would have to change position (translate) from where it started, not just rotate in-place.

The source of the confusion is that they are measuring translation with an angle. They can do this because their universe is very small. You could also measure your translation relative to Earth with an angle, but it would be a tiny angle as the Earth has a very large radius.

aabajian··on Robotic motion in curved space defies standard laws of physics
Here's the pre-print: https://arxiv.org/pdf/2112.09740.pdf

I think their figures answer your questions...both theoretically and empirically. Figure 3 shows the difference between curved/linear vertical motion.

aabajian··on Robotic motion in curved space defies standard laws of physics
Edit: Here's the pre-print: https://arxiv.org/pdf/2112.09740.pdf

I'm not sure others have read deeply into the article. While there's always some subtle ways friction can sneak its way, the authors account for the obvious.

The article has a lot of definitions, which makes the crux difficult to pinpoint. They use the word "shape" many times to describe the configuration of the four motors. I think this adds confusion. Let me try to summarize as I understand it without using the word "shape."

The authors are measuring the net position of the arm holding the two motor tracks. The arm can only rotate clockwise or counter-clockwise in the horizontal plane, so it suffices to measure the arm position using an angle - phi, with initial position being phi = 0.

The two motor tracks are oriented perpendicular to one another. A key point is that the motors do not travel up, down, left, or right. That's Cartesian thinking. Rather, they travel clockwise or counter-clockwise along their respective tracks. One track is vertical, the other is horizontal.

With this picture, the authors claim that the net horizontal position angle (phi) of the arm changes depending on how the motors move along their tracks.

None of this is particularly impressive (e.g. the motion of the motors may cause the arm to "wobble" one way due to issues with synchronization, weight differences, and numerous other causes).

The interesting part is this: If the vertically-oriented motors are made to move up-down on linear straight paths instead of curved paths, the net movement of the arm is not observed. Think about that for a moment...why should the path that the vertically-oriented motors take affect the horizontal rotation of the arm?

In Newtonian physics, acceleration/force in one direction cannot affect acceleration/force in a perpendicular direction. At any given moment, the vertically-oriented motors are perpendicular to the arm's rotation direction, so their motion should not affect its horizontal rotation in space. In particular, both the curved-vertical and linear-vertical paths are perpendicular to the horizontal at all times. Newton would predict that changing the path of the vertical motors from linear to curved would not affect the horizontal rotation...but it does.

aabajian··on Boston University Undergraduate Costs Reach $80k
Personal example, I almost did an MD/PhD at BU. Almost every MD/PhD is fully-funded due to the time commitment and presumed loss of 4-5 years of attending-level salary. BU's MD/PhD program is not fully funded...you have to pay for your living expenses all throughout the MD years, not an easy price to stomach in Boston (see: https://www.bumc.bu.edu/mdphd/admissions/funding/).
aabajian··on Never hose out a Honda Element
The FJ has drainage holes on the bottom so it all clears out.
aabajian··on Never hose out a Honda Element
It has enormous cargo space for its size. It has more interior space than an FJ Cruiser, despite being 80% of the size. The stock rear seats of an Element fold flat across the windows: https://s3files.core77.com/blog/images/585272_81_61976_raLZL...

Also, there was factory-optioned camper model called the Honda Element ECamper.

Like the FJ, the Element has suicide doors allowing you to load things like wheelchairs and surfboards.

I think a big reason it has the false misconception of being washable is that there is no carpeting (except on the SC) models. It was also rated as the #1 dog-owner vehicle, which means it tends to get muddy. Honda even had a dog-utility ramp for it: https://www.hondapartsnow.com/genuine/honda~pet~ramp~kit~08l...

aabajian··on Never hose out a Honda Element
If you want a hoseable vehicle, buy an FJ Cruiser. I take my dogs to the car wash (it has heated water), and hose them off without taking them out of the back.
aabajian··on The computers used to do 3D animation for Final Fantasy VII in 1996
I'm surprised nobody has stated the obvious. See the recent post on Dall-E 2. It's likely to change the face of creative design and was done using modern supercomputers (e.g. cloud-based deep learning). To make a direct analogy, I imagine some of the newest game AI-graphics and engines require training on such supercomputers.
aabajian··on Ask HN: Software with biggest potential for positive impact in 5 years?
I think adult education needs reform. We prioritize childhood education, but kids don't choose their parents. Affluent kids have access to better education, and there's no later-in-life educational safety net for adults without those early childhood benefits. If you went to a crap high school because of where you lived, you have limited options after graduating, and even less if you dropped-out.

Your current options as an adult are to take a course online, take a course at a community college, or buy books and watch YouTube videos all on your own. The problem with these approaches is that all motivation must come from within. There's no requirement that you learn. As soon as you get bored, you can quit. A large majority of kids, even stellar students, study because they are required to do so, out of a fear of the consequences imposed by their parents, by their teachers and even by their peers.

Imagine if you had to learn a subject outside your comfort zone. How attentive would you be to an online classes? Even basic courses like algebra are very uncomfortable to adults with minimal mathematics knowledge. It's not enough to hope adults who enroll will 'do their best', you have to motivate them to stay and push them to excel.

Now...I don't know what such a system would look like or even if it's possible to implement on a large scale.

aabajian··on Ask HN: If you used to be socially awkward and shy, how did you improve?
In high school I was very nervous speaking in front of people, but I enjoyed explaining the one-off calculus problem that was challenging to solve. In college, I tutored small group mathematics and chemistry classes (2-8 students per class, 1-3 classes per week). I then did Teach for America for 2 years (20-35 students per class, 5 classes per day). By the end of that, I could speak in front of thousands of people without breaking a sweat.

Pro-tip for public speaking: Never pause-out-loud. If you are going to say, "uh", "well", "like", etc. just don't say it. From an audience perspective, hearing, "Thank you all for coming, uh, we have a great special for you today..." is much worse than hearing, "Thank you all for coming (pause), we have a great special for you today."

aabajian··on What are the most important statistical ideas of the past 50 years?
Didn't read the document, but hopefully it mentions PageRank, the prime example of using probabilistic graphical models to rank nodes in a directed graph. More info: https://www.amazon.com/Probabilistic-Graphical-Models-Princi...

I've heard that Google and Baidu essentially started at the same time, with the same algorithm discovery (PageRank). Maybe someone can comment on if there was idea sharing or if both teams derived it independently.

aabajian··on Final Fantasy VI – Ted Woolsey Uncensored Edition
This game is fantastic, I was fortunate to go to the 25th Anniversary concert in Chicago. It's reportedly Nobuo Uematsu's favorite game as well.

Also, FFVI is the only game (to my knowledge) officially released on 8 different platforms across four console generations: SNES, Playstation, Wii Virtual Console, GameBoy Advance, iOS, Android, Steam, and SNES Classic Edition.

The story is so good, I've thought they should make a movie or TV series from it (though such FF movies have had a bad track record).

aabajian··on The web starts on page four
I'm sure everyone is used to typing:

'VanVan motorcycle starting issue reddit'

'FJ Cruiser clutch chirp forum'

Adding the "reddit" or "forum" tag makes the results way more useful. Is there a way to tell Google to only include forum-like results?

aabajian··on Zillow just gave us a look at machine learning's future
Personal example. I bought a condo in the University District in Seattle back in 2017. It's a 92-year-old, 478 sq ft condo. I paid $279,000 (yikes, I know)...Zillow's estimate? $500,000.

I have no idea how they were so far off. Even right now:

Zillow estimate: $479,100 - https://www.zillow.com/homedetails/905-NE-43rd-St-APT-212-Se...

Redfin: $307,337 - https://www.redfin.com/WA/Seattle/901-NE-43rd-St-98105/unit-...

Keep in mind that Zillow's HQ is also in Seattle. While we may be protective of the data scientists on HN, this is one case where I agree, there's something wrong with their models. The "similar recently sold" condos that Zillow shows are indeed $400K, but those are at most 50 years old and more square footage. I think a simple logistic regression of square footage and age would give a better prediction than whatever model they are using.

aabajian··on What does Google think minimum wage is?
If you're interested in a querying for quantitative data, try Wolfram|Alpha:

https://www.wolframalpha.com/input/?i=minimum+wage+in+the+UK

aabajian··on Rivian S-1
The stock market has always been a bit of a mystery to me. There is no requirement that company revenue be tied to a stock's value. Rather, the stock for a given company is entirely based on human sentiment.

Person A buys stock S at time t0 because he thinks it will be more valuable to person B at time t1.

Person B buys stock S at time t1 because he thinks it will be more valuable to person C at time t2.

...and so on.

Buying a small-volume of stock in company X provides no tangible value. You aren't getting a seat on the board with a few thousand dollars. Obviously, things like dividends and retirement tax-breaks have value, but I'm talking about pure speculative investing.

My question to the investors out there: Why should be there be a person C or D who buys the stock from you in the future? In the extreme case, you could buy stock X, and then never find someone to buy it from you.

aabajian··on A search engine that favors text-heavy sites and punishes modern web design
Gotta say, sometimes the results really are nice. I searched for "Land Cruiser 70." The first result is a simple, short blog post about a couple who traveled across Europe and Asia in their Troop Carrier (http://www.destoop.com/trip/1%20PREPARATION/2%20Vehicle%20sp...).

The first results on Google are Australian site for buying a LC70 (news-flash, I can't buy one in the USA). There is also a MotorTrend article about the LC70...also irrelevant since it's only sold in Australia.

aabajian··on Use of artificial intelligence for image analysis in breast cancer screening
I'm a 5th year radiology resident with undergraduate and master's degrees in computer science...and currently on a mammography rotation.

A couple background facts:

1. Mammographer subspecialist salaries are now the highest of any radiologist (median ~$440K) - https://www.auntminnie.com/index.aspx?sec=ser&sub=def&pag=di...

2. Mammography is one of the easier fellowships to undertake and is "lifestyle-friendly" overall (no call as an attending!)

3. Mammographers are highly in-demand, even in desirable locations.

So it's (relatively) easy, very well-paid, and you can live where you want. What's the catch? Well...

1. Mammographers are the most likely to be sued. Missing a mammographic diagnosis of breast cancer is the #2 reason for medical litigation, with some sources citing up to 10% of radiologists per year being named in a mammography litigation case. See: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3138733/ and https://pubmed.ncbi.nlm.nih.gov/22596034/

2. Mammographers directly interact with patients to relay findings and biopsy lesions, and they must collaborate with referring providers and pathologists. Breast cancer is a highly emotional topic in the United States due to a combination of public personas ('Angelina Jolie effect') and political maneuvering (see the history of the federal Mammography Quality Standards Act and Program - MQSA). You can't be a "behind the scenes, in the dark" radiologist if you go into mammography.

From a computer science perspective, I one-hundred-percent agree that mammography is a "solvable" problem by AI. Standardization of mammography images (CC and MLO screening views) alongside BI-RADS makes for an amazingly well-labeled dataset. I fully expect AI to surpass the abilities of one-or-more mammographers in the near future.

And yet, amongst radiologists, mammographers are the least likely to be replaced. Why? For the same reason mammography is lucrative to begin with - Litigation, emotional patient-doctor interactions, and the social/political climate. Any one of these is a steep hill by itself. Most AI companies want none of the legal risk, and practically zero of them offer patient-facing results. And, I wonder if any of the AI startups have considered hiring a celebrity as their spokeperson. Perhaps politics/legislation might work in favor of AI if it becomes standard of care, but I wouldn't want to bank my startup on a federal regulatory change.

aabajian··on O.mg Cable
I live in Seattle and have had my car broken into three times. Each time, they steal my iPhone's USB cable. I wonder, could they add cellular connectivity to this cable and use it as a GPS locator?
aabajian··on How to Work Hard
Paul Graham makes a few points that show why medical training needs reform. There's a disconnect between hard work and income in medicine:

i. You work hard throughout residency, yet your salary is fixed. The hardest-working neurosurgery resident gets the same check as the coasting primary care trainee.

ii. Trainees are praised for their academic knowledge and their academic output, yet the highest-earning physicians are in private practice.

iii. Physician jobs in desirable areas are scarce, and they pay the least. Hard-work improves your chances of getting a job in a competitive market, but at a lower salary.

aabajian··on Java for Everything (2014)
I learned Java long time ago and have worked on a number of large projects. I agree that it is easier to maintain code in Java. The downside, as the author described, is the verbosity of the language.

Python is a wonderful language and exemplifies the "essence" of programming in that it nearly looks like pseudocode on a whiteboard. The syntactic overhead is minimal. I liken it to solving an algebra problem with pencil and paper - you would never write "int x = 5" on paper, just "x = 5."

It's this simplification that makes Python a joy to use...but also difficult to maintain. Programmers have to get in the habit of following code style guidelines, rather than being (somewhat) coerced into them by strong typing.

A simple example of this is the question: Where does a Python program start? If I provide a Python script, the reader would have to scan from top to bottom to see the first non-definitional line of code. With Java, there is no ambiguity; it is the main function. A company may declare a style guidline that requires Python programs define a main-check function such as:

if __name__ == "__main__": main()

but the developer needs to learn this policy.

aabajian··on AR with SceneKit and Metal
I think AR kit is the most under-utilized API available on iOS. Combined with the U1 chip, you could make a real-world tagging-app. Post a note on your favorite trail or maybe overlay In n Out's secret menu on their real menu. How about AR Zoom backgrounds?
aabajian··on On moving from statistics to machine learning, the final stage of grief (2019)
The author's pie chart showing data science to be 60% data manipulation is accurate. The biggest gap between good and bad data scientist is their comfort level with data wrangling. When interviewing candidates for data science positions, one of the simplest questions is to have them sort a 1 GB tab-delimited file.

1. Poor candidates will try to open the file in Excel.

2a. Marginal candidates will use R or Stata.

2b. Okay candidates will use a scripting language like Python.

3. Good candidates will use Unix sort.

To my knowledge, there are no university courses teaching the Unix toolchain and it remains very much a skill learned through practice.

aabajian··on New slats make the Golden Gate Bridge sound like a David Lynch movie
The expressway I-5 bridge in Seattle is notoriously loud and can be heard for miles around. Interestingly, the sound from the busy topmost lane doesn't seem to make it to the ground. You can verify this during the brief pause when the express lane changes directions. I believe the traffic noise from the express-lane is reflecting off the underside of the upper lane. Seems like a diffusing layer on the underside of the upper lane would reduce the sound getting to the city below.
aabajian··on Ask HN: What startup/technology is on your 'to watch' list?
Geometric algebra: https://www.youtube.com/watch?v=tX4H_ctggYo

It makes a lot of hard physics problems (Maxwell's equations, relativity theory, quantum mechanics) much more understandable and (I'm told) unifies them in a common framework. I think it will help your average developer become comfortable with these areas of physics.

← PreviousPage 4 of 10Next →