HNHacker News
TopNewBestAskShowJobs

mbenjaminsmith

1,329 karma · joined July 9, 2009

SSEuAAGo60Ssb9UMZhmYW77MLQcz1azm xSTO8k2zwIBX7jfw4PQEDozixMnjkZ8L
submissionscomments
mbenjaminsmith··on How an app with 200,000 downloads led to developer homelessness
I'm currently building a game company so I sympathize with these guys, but it sounds like most of this is self inflicted.

1. Development cycle is way too long. You have to move much faster in mobile, especially for a casual game.

2. Assume financial failure for each game. Doesn't sound very encouraging but you should never count on income from a new title, especially if you're trying out new ways of monetizing.

3. Keep a cushion. If you need to take some time off to contract then do it. Running out of cash isn't an option.

4. Don't leave customer acquisition to chance. It sounds like their customer acquisition strategy was based solely on app store visibility (and hope).

mbenjaminsmith··on NASA TV feed of 10:31 pm PDT Curiosity landing
That's the second time in recent memory that I got chills watching a room full of really smart people erupt into applause. Congrats to everyone involved.
mbenjaminsmith··on Why Apple Hasn't Shipped a Full TV. Hint: 12x12
That's pretty interesting reasoning but it's based on a key flawed assumption: that Apple would offer anywhere near 15 screen sizes. Given Apple's history I think it's reasonable to assume they would sell two or maybe three sizes max.

Also, this is their first product that would be primarily wall mounted. Not needing space for peripherals would open up new display options -- like sticking them on a wall.

mbenjaminsmith··on Ask HN: How much do you pay for cloud service each month?
"The photos themselves go straight to Amazon S3, which currently stores several terabytes of photo data for us."

http://instagram-engineering.tumblr.com/post/13649370142/wha...

That was 7 months ago, so let's say 5TB today. Plugging that into the calculator for AWS says Instagram pays a whopping $500 for photo storage each month.

I don't want to take the time to do it myself but here http://highscalability.com/blog/2012/4/9/the-instagram-archi... someone punches (in the comments) all of their numbers into the AWS calculator. He comes up with 12k per month based on what they've said they're running.

mbenjaminsmith··on Ask HN: How much do you pay for cloud service each month?
I'd be surprised if Instagram was paying that much. I've known at least one person sharing video at 1 million uniques per day and his cost was a fraction of 200k (though I don't remember how much exactly).

The chances of you going outside your 1k budget before you can take on investment or earn some revenue are basically zero.

mbenjaminsmith··on Nobody Ever Got Rich Working for Somebody Else. #horseshit
"I don’t have any exact numbers, but I would guess that there are actually more rich people in the world who got that way working for somebody else than there are rich people who were entrepreneurs at the start."

What kind of argument is that? It's backed by absolutely nothing. It also conflicts with everyone I've ever known.

Also, from the discussion here, the OP's definition of rich is a dubious one: "But generally, I think if you have a couple hundred grand in the bank you're rich."

If you're an employee, and assuming you've already paid taxes on that 200k, and you don't have kids in school or health problems, that's going to get you a few years down the road. It should be enough runway to get a new job with room to spare. As emergency money it's pretty good.

If you're working with 200k in the bank what does that get you? A down payment on a house you can't really afford? Maybe a sports car? Again this is assuming you're not going to put kids through school. That 200k means you're rich if you don't spend it on anything rich.

That 200k is decidedly not an amount you're going to retire on. You could take that 200k and start a business, but then we're in conflict with the idea that starting a business is completely unnecessary for obtaining riches.

Having said that, I'd be very curious to know how many startup jobs made an employee 200k richer after they've moved on. Given that many startup jobs (at least in the US) are in the Bay Area with its high rent and high taxes and that the average startup salary is what, 100k? IIRC, 100k is going to put you at close to a 40% tax bracket in SF. So that's 60k - 24k for rent for the year? That's not a lot left for a lifestyle in that area. Assuming you saved very aggressively you could, at the outside, put 20k away a year? So that's 10 years. What could you have accomplished working for yourself over those 10 (critical) years? Possibly less, though I think the numbers are probably in favor of a 10-year entrepreneur in a field with low barriers to entry.

What percentage of the current crop of startups will ever see employees make significant amounts via their stock? If you want to roll the dice that way certainly the numbers favor the entrepreneur.

Work for someone because you believe in what they're doing or you simply need the money. Don't ever do it to get rich.

mbenjaminsmith··on Newly released UFO files from the UK government
An interview with Stephen Hawking was recently posted to HN. In the interview he states that UFOs are only witnessed by crackpots and that the laws of the universe prevent interstellar travel.

The "crackpot" statement aside, I think it's incredibly disturbing that a scientist of his stature makes absolute predictions like we'll never travel to another star. I believe he is pretty good at his job, but saying something is impossible -- based on our current understanding of physics -- seems silly. No one in the physics community could say that our understanding of the universe is complete. The most prevalent theme in physics today is still the incompleteness of our understanding. We can't reconcile our understanding of the largest things in the universe with our understanding of the smallest. Although we can measure gravity, predict and observe its effects, we have no consistent explanation for it.

I'd say that thousands of years of fiction (speculative or otherwise) remain largely unaffected by hundreds of years of physics. Our understanding of the universe is still in its infancy.

We also have a history of having our deepest beliefs overturned by progress. Why would the beliefs of the late 20th century be any different?

[Edited for language use]

[Edit]

He dismissed the idea of time travel in that article, not interstellar travel. My criticism still applies.

http://arstechnica.com/science/2012/07/steven-hawking-on-tim...

mbenjaminsmith··on Newly released UFO files from the UK government
Another well researched book is:

http://www.amazon.com/UFOs-National-Security-State-Chronolog...

It covers the intersection of military/governmental organizations and the phenomenon, mostly in the latter half of the 20th century. The short version would be to search YouTube for Richard Dolan. He's a great speaker and presents the information in a way that should be palatable for even the hardest skeptic.

mbenjaminsmith··on DC "Uber Amendment" would force sedans to charge 5x more than taxis
"What would there be more of, in total, if you confiscated a bunch of people's cars?"

Road. Fewer cars on the road would mean a more constant speed for the cars on the road.

"Would you advocate banning most people from a popular public beach on the grounds of overcrowding?"

You generally don't have to. Since the benefits of going to the beach are relatively low, people more efficiently self-regulate their usage. That is, if a beach is too crowded to put down a towel, you'll probably leave.

That's not the case with taxis. As long as buying taxies and renting them out is profitable, taxi operators will continue to do so. To make things worse, the profitability of driving a taxi is only indirectly related to the profitability of renting that taxi out. The cost of rental is relatively stable while the money made by the driver varies considerably. The drivers absorb short-term (even medium-term) fluctuations in the market.

mbenjaminsmith··on DC "Uber Amendment" would force sedans to charge 5x more than taxis
Yeah it's called a traffic jam. As the total number of cars on the road per unit of distance increases, the speed at which those cars can travel decreases. It's not linear of course but, well, I think you get the picture.
mbenjaminsmith··on DC "Uber Amendment" would force sedans to charge 5x more than taxis
Pretty clear case of this: http://en.wikipedia.org/wiki/Tragedy_of_the_commons

The shared resource in this case is traffic capacity.

It's proved ineffective to regulate traffic by making cars expensive in this market. That's because cars are not profit centers for people -- they give mostly intangible benefits (increased comfort, status, etc). Taxis on the other hand are cars that are used for profit and therefore will respond efficiently to regulation (if that regulation increases capital or operating costs).

mbenjaminsmith··on DC "Uber Amendment" would force sedans to charge 5x more than taxis
Because favoring the trains would mean fewer cars on the road, something the city needs.

The trains are faster in many cases (provided you're starting / ending close the line) but they're not cheaper. Making cabs more expensive would change this.

Take a look at what Singapore does http://en.wikipedia.org/wiki/Electronic_Road_Pricing to keep cars off the road. It might seem a bit draconian but it's effective.

mbenjaminsmith··on DC "Uber Amendment" would force sedans to charge 5x more than taxis
Bad example. I've never seen a city in more need of stiffer regulation than Bangkok. The single worst part of driving in the city is dealing with agressive or incompetent taxi drivers. They break traffic laws and block lanes looking for fares.

The over-supply of taxis has the opposite effect of a more regulated public transport. It does not reduce the total number of cars on the road. It's useful from a consumer standpoint but the low cost means people are less likely to use the state of the art elevated train and subway, both of which run under capacity.

"Price it like that and a lot less people would own cars." Possibly in other cities -- though I doubt it -- but it certainly didn't help in Bangkok. People don't buy cars because they need them, they buy cars because they want them. The number of news cars sold in the country per day is astounding (though the number escapes me) and that's despite the fact that they cost 50% - 350% more than the same cars in the US.

Bangkok needs to regulate the shit out of the taxi industry and make it more expensive to drive in the city if it wants to sort out its remaining traffic issues.

Uber can probably work in US cities that have the traffic capacity, but it's a horrible idea in congested areas.

mbenjaminsmith··on Corrupt App Store binaries crashing on launch
It was a typo and I really think you should avoid correcting people's grammar on this forum. It doesn't add anything to the discussion.
mbenjaminsmith··on A Gentle Introduction to Algorithm Complexity Analysis
"Yes, there are cases where the 'insignificant' part is so significant it's actually more efficient to use asymptotically slower algorithm for everyday problem sizes. The Big O notation is not so much about running speed but about scaling."

I do understand that. My point was that the analysis is likely too optimistic in some cases. Maybe that's not an issue. I guess the whole thing is fairly arbitrary anyway, and conservative, since we're going to use "worst case" wherever it can be applied.

mbenjaminsmith··on Corrupt App Store binaries crashing on launch
I've seen download corruption effecting apps from the Mac App Store.
mbenjaminsmith··on A Gentle Introduction to Algorithm Complexity Analysis
Correct, I meant n^2.

Certainly there are cases where the "insignificant" parts of an algorithm are not so, correct? In a simple case n^12 dominates a few n^2 sections, but there must be exceptions to that.

Are there cases where a more complete analysis is done -- like comparing two similar algorithms? Or at that point are you actually implementing the algorithms to test and measure?

mbenjaminsmith··on A Gentle Introduction to Algorithm Complexity Analysis
Great article. I've understood the why and what of Big-O but never how to do the analysis.

I have a question for the informed however:

The article says that we only consider the fastest growing parts of an algorithm. So, counting instructions, n^2 + 2n would just be n^2. But why do we do that? Imagine an algorithm where we have n^12 + n^2 + n^2 + n^2 + n^2, etc. Do we really ignore each n^2 section?

[Edited]

mbenjaminsmith··on Show HN: landing pages for iOS apps with interactive, usable demos
Please consider adding Android as well as Mac apps. I'd sign up for an unlimited account today if you handled Mac apps.
mbenjaminsmith··on Stephen Hawking on time travel, M-theory, and extra terrestrial life
I'm surprised to hear someone of his intellectual caliber resort to ridicule when someone brings up UFOs. UFOs "appearing" to "cranks and weirdos" doesn't really describe the phenomena, as anyone who spent more than 30 minutes on the subject would know.

I'd would have loved to have heard him debate the late John Mack (Harvard Medical School, won a Pulitzer) on the subject. It's incredibly arrogant for him to think that he can ignore the research of other academics (or remain in ignorance of it) and still speak authoritatively on the subject.

I don't expect him to champion the field but since he's asked so frequently about extraterrestrial life and/or UFOs it would be nice if his canned response was a bit more intelligent.

It's incredible to dismiss so much theoretical/speculative physics in the interview and at the same time say that maybe M-Theory will tie up the loose ends in the Standard Model, but then again, maybe it won't and we'll be back to the drawing board. Even though we admittedly don't understand the universe with any degrees of completeness, obviously these popular ideas are false based on our understanding of the universe?

mbenjaminsmith··on US Navy X-47B Robot Fighter Jet Completes First Phase Of Testing
1. With in-air refueling, a greatly increased time in the air vs manned aircraft.

2. Ability to outmaneuver any manned aircraft by exceeding the human g limit.

3. Greatly lowered cost in both manufacture and pilot training (or lack of).

4. When fully realized, AI that is superior to human pilots -- increasing overall combat effectiveness and reducing error.

mbenjaminsmith··on RIM announces $518 million loss, 5,000 job cuts, delays BB10 to 2013
I've known exactly one RIM employee. He was fairly high up the food chain. I met him at a party and for some reason I ended up talking to him for an extended period (end of the evening, too much booze). He was racist, sexist, pretentious and convinced of his own ability to understand the world.

This was when iPhones had firmly taken hold of the market and BBs were already yesterday's handset. I talked about that with him and he threw a lot of jargon at me and listed the reasons why RIM would continue to dominate into the future.

That conversation was unpleasant enough that I still think of it whenever I see RIM news pop up. If he was characteristic at all of upper-level execs there then it's absolutely no surprise to me that RIM is circling the toilet bowl.

mbenjaminsmith··on A day in the life of a startup founder
Yeah I just went back to the drawing board with one of my companies. It's in double stealth as well. Had a good laugh when I read it.
mbenjaminsmith··on Firing
As I stated elsewhere in this thread, correcting employee behavior is a normal part of management but is not the same as saying "you may be fired if" or "improve or else". Putting an employee in a defensive state (where they're either looking for another job or trying desperately to keep their current one) is not sound management technique.

Do you really run a business where you threaten employees with termination as a form of motivation? I've built three companies and have managed, hired or fired dozens of people. Under no circumstances can I see what the OP proposed working for me.

mbenjaminsmith··on Firing
Talking to someone about correcting their actions or improving is completely normal -- I've never managed someone who didn't respond positively to discussing their performance constructively.

Telling someone to "improve or else" is a completely different animal. If you want that employee to focus on doing their job (and improving their performance) they have to not be wasting energy looking for another job. From their perspective, moving to another job before they're fired is much better for their career (and ego) than sticking around and possibly getting fired.

I've only been in the situation once where I had to "reset expectations" and that was because the employee was not hired at will and we were hoping that person would resign (they did).

mbenjaminsmith··on Firing
The role of a manager is not to determine the fire-worthiness of current employees. Putting employees in front of a firing squad on a regular basis and having "half-fired" employees working in a state of limbo would murder company morale.

Any good employee would be out the door as soon as they were handed a "second chance". They would use that chance just long enough to find a better job, if they didn't quit outright. Truly "bad" employees would likely stick around (lack of motivation to move, inability to move, spite, etc) and wreak havoc on your business.

Firing is sometimes a business decision, sometimes due to someone actually doing harm to your business but often the last in a series of management failures. It's a last resort but if you've reached that point just get it over with.

mbenjaminsmith··on Stop claiming you’re profitable
What? Profitable is profitable, there is no ambiguity there. The post linked above is talking mostly about opportunity cost (your own) and growth. Is your business sustainable? Is it growing? Is it worth doing full time? Those are all critical questions but the concept of profit does not include them.
mbenjaminsmith··on Intel Reveals Neuromorphic Chip Design
I don't understand why Leon Chua keeps popping up first when memristors are discussed. Williams at HP is the person that made it a reality. Chua postulated the existence of the memristor but as far as I know made no progress on a working example.

That's a bit like giving da Vinci credit for the invention of the helicopter.

The Williams talk on youtube is really good. He discusses the element itself, its use in memory (as vast amounts of on-CPU memory) as well as the possibility of new CPU architectures based on it:

http://www.youtube.com/watch?v=bKGhvKyjgLY

mbenjaminsmith··on How to Be an Indie Game Developer
I'm 6 months into an independent game studio. Right now it's just me and some freelancers.

I'm in the process of finalizing all of the paperwork for the business and hope to have our game in testing by the end of the month. The bottom line is it's way more work than I anticipated. I'm not inexperienced either, this my third company.

I've been running a software business since the beginning of 2011. The amount of work in one game dwarfs the amount of work in a utility app. There's just no comparison. I'm confident we're making a unique and fun game (iOS first, later Android) and I am paying a lot of attention to promoting the company and our game, but I really don't expect to make much money this year on it.

Two years before the company is paying off seems about right. My first company (bricks and mortar) took around two years before we were out of crisis mode. I think the app stores give people a false impression of what it takes to build a business and see your hard work turn into something meaningful.

The upside is there is incredible upside potential in the video game market, especially if you're running a lean shop. Having said that, I'm not doing it for the money as much as I'm doing it to build something that I'll be happy working on 5 or even 10 years from now.

I don't think people should approach it (unless they are just hobbyists) because making games is fun. It's not really fun after the initial conceptualization is done. It's mostly just hard work. Dull, repetitive, never-ending work. I get through it by exercising religiously and paying attention to my diet. I also take off at least every other weekend to relax and hang out.

[promo] If you're interested in hearing about the game's progress, please follow us at twitter.com/rapidfirestudio [/promo]

mbenjaminsmith··on The exercise habit
I've heard countless times that if you're going to start exercising and stay with it you should compete -- for most people that means running and running races on the weekends once in a while or even training for a marathon.

I've done this in the past and I always overtrain and race too early. I have a bad race and stop training shortly after that.

I've found a much more effective way: Between my wife and I whoever is in the gym less in any given week has to pay the other person on Friday -- around $50. There is still competition (neither of us want to admit to losing) but there is also a very tangible penalty.

The combination is an excellent motivator and neither of us have missed a single day since we started it. It's similar to having a partner (a partner got me through 3 years of weight training in college) except that you don't have to work out at the same time, which is much better for busy people.

← PreviousPage 6 of 15Next →