HNHacker News
TopNewBestAskShowJobs

tokipin

1,062 karma · joined September 21, 2007

submissionscomments
tokipin··on No One Saw Tesla’s Solar Roof Coming
The timing may be calculated but that's about it. If a product isn't going to work out, that's one thing, but Musk doesn't waste time on vaporware. Notice that Tesla never put a silly will-never-be-an-actual-product car for display at auto shows.
tokipin··on All Tesla Cars Being Produced Now Have Full Self-Driving Hardware
i'm guessing that the data is just plain too different (probably no more mobileye sensors, for example) and not worth adapting to the older ML models (which itself would require extensive testing) when the new system is going to end up with different ML models anyways.

[edit] Tesla previously hired Jim Keller (chip designer) into the autopilot team. considering the kinds of things he may be working on, i'd be surprised if the differences in either the sensors or GPUs aren't significant.

tokipin··on All Tesla Cars Being Produced Now Have Full Self-Driving Hardware
i don't think they ever believed the previous sensors were capable of "full autonomy". the reason they didn't throw in all these sensors sooner is that it would have been a premature optimization considering their lack of experience and also pretty expensive in terms of the hardware (which includes the GPUs necessary to make use of the sensors).
tokipin··on Shame on Y Combinator
hey now, the blame for Guantanomo not being closed falls directly on congressional republicans, not on Obama's lack of trying. this is the case for most of his failed endeavors (nominating court justices ring a bell?). there's plenty of stuff he does deserve blame for. most notable i think is that although he effectively ended the Iraq war, he rescinded his decision to end the Afghanistan war in 2014. regardless, Obama and Trump are not in the same category in any respect, not even as politicians.
tokipin··on Elon Musk on How to Build the Future
According to your logic, Elon's been "in the right place at the right time" how many times? That you think that's due to luck, when it's clear he's a polymath with great design talent and work ethic, says more about him as a psychological Rorschach than anything else, I think.

He does trigger conservative-minded people pretty hard, which I imagine is partly due to how Tesla got politicized during Obama's first term.

I don't know how serious you are about the billion dollars thing (or what exactly your point may be), but I'd bet a billion dollars you wouldn't be anywhere as effective as Elon is with that money. The idea that any two people are going to be equivalently effective given $X is silly. There are just as many orders of difference in effectiveness with money as there are in any other endeavor.

tokipin··on Tesla envy grips Germany’s giants
I think in those cases, it was a matter of a better-designed, clearer-headed product taking space from a not-so-well-designed product. That's not what would happen here, because Tesla is already the creme de la creme. In fact, Tesla executing so well is the only reason EVs have seen their recent "resurgence" in the first place. And remember that Tesla isn't the first company to make electric cars.

I also think people underestimate by a lot how hard making new things is. It's not just a matter of coming in with more money. It's almost deceptive on Tesla's part that the cars seem simple. But if it really were so simple the Model S would already have competition. And when you're talking about someone who has raised a company that lands rockets in the middle of the ocean, competitors are fooling themselves if they think even just a superficial copy is going to be easy. This is something they have to bring their A game to at all levels.

tokipin··on Space: A sudden light
The idea is to have our eggs in more than one basket so that the next mass extinction event doesn't take us all out or set us back hundreds of years. Whether life on mars is fun or not is a separate question. (I do agree with you. We can barely manage to live in Antarctica. Mars will be much more difficult.)

And in becoming multiplanetary, we can start gaining the tools that could help us avoid some of those extinction events.

tokipin··on iPhone, iPad and Mac sales are down, but Apple only cares about services now
The biggest thing for me is speed. I don't know who Apple thinks it's fooling with their slow ass animations. Windows 10 is way snappier.

And I think the OS X windowing/dock system is inferior. It's way harder than it should be to find windows.

tokipin··on Master Plan, Part Deux
I don't think there is a cutoff point in company size or whatever above which writing suddenly needs to become sterile. What you have is actually more a cultural thing I think. People thinking that a "professional" message is needed, when straightforwardness would be a superior goal. (On the other hand, a good reason to aim for "professional" is that everything released by a company has a consistent voice/tone.)

Notice that Musk runs his companies the same way - by trusting what's really important vs what isn't. And anyone who has invested in Tesla is already familiar with his irreverent character. (See this[1] interview for example).

[1] https://www.youtube.com/watch?v=ajP3B0gYJlo

tokipin··on Master Plan, Part Deux
The specifics probably weren't in his plan, but in various interviews he's talked about the consumption of energy (vehicles) and the generation of energy (solar) as being two sides of the same coin. Elon is the one that told his cousins to start a solar company after all.

IIRC he originally saw three major fields to move forward on: the Internet, Transportation/Energy, and Space. By the sounds of it he was thinking of these as early as high school, to at the latest some time after he sold his first company (Zip2). He did the Internet thing pretty well, now he's working on the other two. This is someone who does think on these scales. Keep in mind, as the article itself points out, that you're an idiot if you think starting a car company is a good idea. Energy was the whole point all along. A company for its own sake isn't in Elon's DNA. That would probably just be too boring for him.

Whether this specific merger is a good idea or not is a different question. And I also don't think the emergence of cheap supercomputers in the form of geek-subsidized graphics cards and their applicability to a particularly effective machine learning genre would have been predicted by anyone.

tokipin··on Sharpest ever view of the Andromeda Galaxy (2015)
On that note, the stars all have a 'stellation' along what appear to be the same axes. What is this effect? On the surface of the Earth my assumption was that the stellation/twinkling is an atmospheric thing. In this case is it an effect of gases or some telescopic thing?

Actually if it were telescopic it could probably be analyzed out.

tokipin··on Coursera shuts access to old platform courses
you're probably thinking of "Introduction to Astronomy" by Ronen Plesser. that course was awesome.
tokipin··on Model 3 Unveiling [video]
I think it's because he knows the presentation in and of itself is unimportant vs the takeaways delivered, so he doesn't bother wasting time on it specifically since it has low ROI given his speaking capabilities. And I suspect he dislikes acting/faking, so he wouldn't use precise scripts anyway.
tokipin··on Model 3 Unveiling [video]
Most cars have grills because they need the ventilation to cool the engine. I think they're mostly something we've gotten used to. When electric cars become more common those gas car grills will probably start looking clunky.
tokipin··on Why I think Tesla is building throwaway cars
People used to repair televisions but at some point it became cheaper to just buy a new one due to the tech/manufacturing being sufficiently evolved and commoditized/cheap, as well as the issues involved with repairing more complex circuitry.

Electric cars have the same potential, I think, because of their inherent simplicity. That potential already seems clear given that battery costs will keep falling.

tokipin··on Why Big Companies Keep Failing: The Stack Fallacy
Nice observation. Another way to put it is that induction is harder than deduction.

A related factor is that larger companies tend to be more specialized (formalized processes, specialists, focused teams/departments, and so on), meaning they can be prematurely optimized with respect to new goals and poorly equipped to conduct the necessary roaming.

tokipin··on The Product Engineer
Not to my knowledge. It's very easy to fall into twisted Escher boxes and ego traps when talking about these things so I try to only entertain them lightly (says my ego), and I don't know if this kind of thing is discussed in "official" psychology/sociology.

At its core it is about raw physical "existential modes" and not the more superficial label of "personalities", and it's very hard to appreciate what that means even in the case where you think you know what it means. Like if you're a trained psychologist. Certain drugs might be windows into our fellow humans' worlds. Other than that, strokes and other forms of brain damage are the only cases I know of people that have experienced multiple "existential modes". For example[1].

I think in terms of hiring, the broad strokes are useful but I wouldn't dilly dally with the specifics too much since they can be misused easily, nevermind all of the other variables in play. And as far as I know there isn't really a central place on the internet that describes the scheme as a whole. Dilly dallying through MBTI forums was one of my pastimes in ages long past and that might be the way to make sense of it today, keeping in mind how slippery things get when everyone's self is involved.

[1] https://www.ted.com/talks/jill_bolte_taylor_s_powerful_strok...

tokipin··on The Product Engineer
When it comes to these things people are going to reinvent a vocabulary that already exists: Jung's psychological functions.

What Jung called extraverted intuition (Ne) is going to be the main "product" function in tech, with extraverted sensing not as common in the industry. Ne appears in Ti-Ne/Fi-Ne/Ne-Ti/Ne-Fi (what MBTI calls resp. INTP/INFP/ENTP/ENFP). In other words, the xxxP types are going to correlate with being product people. Steve Jobs was practically raw Ne.

The stereotyped "Silicon Valley Programmer" that every company wants to hire is introverted intuition primary and extraverted thinking secondary (Ni-Te, what MBTI calls INTJ). This archetype is quick-witted and naturally good at verbal/logic challenges.

To really, really, really simplify things, with xxxJ vs xxxP, you essentially have a dichotomy between speed on the one hand and (external) depth on the other. External depth being advantageous because the world itself is external. The current standard for interviews is essentially to judge speed, so of course this has the appropriate consequences. Which, by the nature of the particular elements involved, rapidly runaway.

Speed is quite easy to see and understand. Depth on the other hand is harder to pin down. But it is depth, and the vision that often carries it, that can render fine details onto a product. Those details can be so small that most people don't see them, so general appreciation of this power is rarer, and by its nature harder to test on top of it requiring a longer time horizon to work in.

Many people will think that you can get both qualities in a single person, but it doesn't work that way. Evolution would have done that long ago if it was possible. Instead, what we're dealing with are fundamentally different brain topologies/strategies, essentially characteristics that species can min-max in individuals because of the survivability allowances afforded by their ancestors having been social. Or who knows. Regardless, these are the same types of trade offs you have in data structures/algorithms.

tokipin··on SpaceX launch webcast: Orbcomm-2 Mission [video]
This is in the "to be optimized" bracket. If they can get any kind of consistency landing even heavily strained rockets, they can reframe/readjust/redesign as necessary (like adding more support in areas etc).
tokipin··on A Look into Machine Learning's First Cheating Scandal
To give an extreme example of overfitting, let's say you have a test set which has these examples: B -> 11, H -> 5, M -> 11. Well, you can solve these easily. Just make a function like this:

    function solve(letter) {
        return {B: 11, H: 5, M: 11}[letter];
    }
machine learning 4Head . Of course, the point is to solve for examples you haven't seen yet, so this "solution" isn't amusing anyone. That said, overfitting commonly happens even without trying to overfit, and people use techniques to minimize it.
tokipin··on Why I Still Use Python for High Performance Scientific Computing
No need to kid. Self-described "FP" languages are a pain in the ass for the type of programming scientists do. I don't even know what kind of programming those languages aren't a pain in the ass for.
tokipin··on The New Atomic Age We Need
the obvious solution is batteries, which improve all sources of power generation including traditional ones by serving as support during peak times. in principle you only need <average energy used per house> stored per house during night time, which may be a much lower bar than it sounds. if the utilities serve a role later it may just be them acting as high-volume energy storage facilities, but even that might not really be necessary if individuals can make money by investing in extra batteries.

i think once solar crosses a certain cultural/economic threshold it will just take over everything. it's just way, way too good. there's no other power source that even compares to the potential it already has.

what's interesting is not whether it's going to take over everything, but what kind of economy/world you will have when electricity is orders of magnitude cheaper than it currently is.

tokipin··on VLC contributor living in Aleppo writing about the Paris attacks
i think the key is that christianity had the historical accident of being connected to western development, so it was forced to tone down its shenanigans. a present example is how it is being forced to accept homosexuality. unlike dogma seems to claim, religion is the one that follows the morality of society, not the other way around.

not that religions aren't powerful. the kinds of religion-based stupidity you see in other nations is exactly the reasons why the founders of the U.S. explicitly separated church and state.

tokipin··on VLC contributor living in Aleppo writing about the Paris attacks
well there's a few thousand religions out there in the world, even if you use granular criteria for what you consider to be different religions. and thats not even counting human history.
tokipin··on The Tech Talent Shortage Is a Lie
it's a simple idea in theory, but the reality of these things is a different beast, in particular that these tests form a filter not just in terms of degree, but also in terms of kind. this is why Google is Google and is extreme in both its strengths and weaknesses.

what i'm curious about is how Apple weathers these particular waters, since they have more to lose than any of the other big companies ever had. hehehe.

tokipin··on StarCraft: The past, present and future
There's so many failures to it. The company apparently thinking the professional scene drives the player base, the company apparently thinking it's responsible for 100% of the balance of the game, not realizing "the meta" plays a huge role towards that end. Because of these Blizzard was way too conservative and the game ended up just not being fun. Once the game is not fun, the player base leaves. Once the player base leaves, the professional scene dies.

The most egregious error might be the company not having a genuine designer in charge of the game, which to me seems like a very corporate-y lack of respect for game design. SC2 as a whole was way too conservative. And this is just on the design side, there was greed and short-sightedness all over the place.

tokipin··on Brains Sweep Themselves Clean of Toxins During Sleep (2013)
Anectdotally, I've found that if I drink some electrolyte drink before sleep (Gatorade etc) I seem to sleep better. I haven't managed a consistent sleep pattern/environment to really test it though.
tokipin··on Twitter’s stock falls toward IPO-level prices
I think part of the issue is that Twitter is a great tool for people in media/news as well as PR/celebrities etc, and because of the role these groups play in society they have been doing what amounts to a lot of free advertisement for Twitter. All along this may have been "artificially" increasing Twitter's user numbers relative to the kinds of users they want.
tokipin··on Take-home interviews
[edit]oops. obviously, i meant apples to apples :D
tokipin··on Take-home interviews
for some it's not about introversion/extroversion but simply the fact that whatever part of their brain deals with speech is the same part that deals with thinking through those kinds of problems, so they can't do both at the same time as a matter of physics.

this is to put it precisely. the reality is much messier and varied. the point is, people's brains are not wired the same. and what this means, to make a long story short, is that these interviews are not comparing apples to oranges... not even close.

← PreviousPage 3 of 21Next →