Podcast: Play in new window | Download
Subscribe: iTunes | Android | RSS
Justin and Jason discuss Justin’s recent post about converting his site from PayPal to Stripe, Jason’s idea for a mathematical equation describing the tyranny of choice, the latest on Pluggio, AnyFu and Appignite, why Jason believes war with Iran is likely and how the price of oil will skyrocket if it happens, the story of how Jason’s attempt to sell his house before the market crashed in 2007, 3D printing, RepRap and MakerBot, Colby’s awesome new RC helicopter, the DC taxi commissioner who’s trying to stop Uber, what kind of jobs belong to the 1%, Jason’s idea for crowd-sourcing the identification of logical fallacies in news columns and opinion pieces, whether journalists should be truth vigilantes, how Jason and Guyon built a profiler for Node.js, NASA’s discovery that there are 1,500 earth-sized planets within 50 light-years of earth, upcoming interviews and Jason’s idea for creating a synthetic university education.
I’d like to challenge a few assertions made on the show:
1) Rise in oil price wouldn’t cause an increase of inflation. [Jason]
But surely you remember that a majority of goods is moved around with vehicles powered by oil / gasoline? What will happen to prices of bread, when the cost of delivering flour to bakeries and loaves to groceries jumps by, say, 30%? What will happen to the purchasing power of peoples’ wages, when food (and most other prices) jump by a significant percent? I don’t have to go on, you know the rap.
2) U.S. will go to war with Iran. [Jason]
I think it is essential to first define what do you mean by war.
If you mean a full-scale war, involving ground troops and occupation, like in Iraq (since the latter was brought up during the show), that just won’t happen. Short of detonating a nuclear device in Jerusalem, there’s nothing Iran can do to encourage the U.S. to engage in a full-fledged war. Iraq was started when U.S. economy was in a relatively good shape and yet most people would agree that the cost was too big. Now we’re talking double dip recession, run-away deficits, not to mention that Iran would be a much more formidable foe than Iraq.
If you mean a limited exchange of hostilities, that could happen, I’d give it like 30% chance. And it would have effect only for a limited time, since the Iranian armed forces aren’t really a challenge for the 5th Fleet. The best Iran can do is lay some mines in the Straits, which will take a little time to disarm, and hence block 25% of world oil for a couple of weeks. That might cause some acute pain to our wallets for a limited period.
The scenario from Antiwar that you linked to was a fun read, maybe worthy of a Tom Clancy novel, but too much of wishful thinking to be considered seriously.
[BTW, you guys were talking the role of aircraft carriers on a recent show. Indeed, they aren’t enough to invade a country, but they sure can deliver some punches far away from home. A single Chinese carrier means the Chinese will finally be able to show some muscle in the South China Sea and assert their sovereignty around these tiny, uninhabited islands (which are surrounded by not-so-tiny undersea oil deposits).
http://news.bbcimg.co.uk/media/images/54145000/gif/_54145268__48951920_south_china-sea_1_466-1.gif
So yeah, it’s a kinda big deal, especially for the South-East-Asian nations. But – again – not a big challenge for the 7th Fleet. ;)]
3) MakerBot is a _professional_ 3D printer. [Justin]
Well, not really. It’s an inexpensive home device for DIY’ers. I know $2k sounds like a lot of money, but in the realm of 3D printing it’s pennies. If you want to see some professional stuff, check out the photos from the “factory” of 3D Robotics, the commercial arm of the DIY Drones website (run by Chris Anderson of Wired).
http://www.diydrones.com/profiles/blogs/tour-of-the-new-3d-robotics-factory
And BTW, these are merely mid-range. The hard-core, industrial stuff, used by aircraft manufacturers, can print metal (!) objects by zapping titanium powder with a laser or electron beams.
http://www.economist.com/node/18114221
An increase in the price of oil would not cause inflation – it is important to remember that true inflation, a general decline in the purchasing power of money (it never happens evenly, but it theoretical possible) is a wash for everyone. Everything costs more, but everyone also makes more.
If there were a sudden increase in the price of oil then everything would cost more, but no one would be making more. – which all things being equal, would be worse than inflation – it is a decline in the general standard of living, but not inflation.
@Steve
Well, as I understand it, inflation is defined as a general rise of prices.
It is true that it is usually followed by an increase of wages, and everything evens out (for the most part*). But if that fails to happen, I don’t think you can say it’s not inflation.
—–
* – if inflation is sudden and big enough, even an increase in wages won’t help, since many things in the economy will still be denominated in the “old money”, like savings, long-term contracts, etc.
Inflation is a decline in the purchasing power of money – a wage is the price of labor. A sudden increase in the price of oil would be bad for the average American, but it is not inflation.
When writing profiling tools, you start as you do with enter/leave annotations to your function in order to compute how much time you spend and where. Then at one point you start to realize some drawbacks:
1. It has a large overhead on small functions that executes in a small amount of time
2. You need to instrument your code with enter/leave sections. Even if you have an automatic system that runs through your code and modify it, you may have corner cases like having anonymous functions (callbacks) that don’t get annotated.
To overcome those limitations, the idea is to let your program run and sample where it is (in which functions) at regular intervals like 100 times per second. This kind of profiling is a statistical, because it works by having a probability distribution of where your program is running.
The overhead is small and independant of the kind of functions (big/small) is running. You also do not need to instrument your code, only to have an id of the currently running functions that you use when you sample your program. All your code is included in the profiling system: even anonymous functions gets profiled.
The main drawback is that your program needs to run for a long time in order to have a meaningful result.
@Pierre De Pascale – Thanks so much for the explanation of statistical profiling. Guyon and I were able to integrate the profiler’s enter/exit calls within the Class.extend function, but you’re right that it still isn’t able to profile anonymous functions. But I wonder if it might be possible to iterate over all functions in the global namespace and attach the enter/exit calls that way. The only requirement would be that you’d need to avoid using anonymous functions, which is easy enough – just give them names. It turns out that that is a best practice anyway, so why not. Hmm…maybe I’ll play with that today to see what I can do.
I haven’t done an intensive analysis of the overhead incurred by the profiler, but it seems quite minimal upon casual observation. That said, I do need to run some tests to get some hard data. If it turns out that my observations are indeed incorrect and that the overhead is high, then it might be worth building a sampling profiler as an alternative option.
Again, I very much appreciate your insight. 😉
Hey Jason…Justin, below are some thoughts I had while reflecting on this episode. Please forgive the “just the facts” bullet point format, and feel free to ask for additional detail if anything below isn’t clear.
Real Estate – Yup…your dollar goes a long way in certain regions (Absolutely no endorsements or recommendations by me here….just providing relevant and hopefully balanced information!)
http://www.youtube.com/watch?v=twCjLEXZVSQ
(price update via email I received 17 January, 2012 = $299k
Map for the property referenced above
http://g.co/maps/h69ss
Apocolyptic Warning (This dude is more concerned than Jason….ha!)
(Delivered with enthusiasm!)
http://www.stansberryresearch.com/pro/1110PSITESVD/LPSIN118/PR
…additional data point to be considered (i.e. “secret meeting Fisk reserve currency” search shows the story originated years ago….so realize his premise hasn’t proven true over the past half decade)
https://www.google.com/search?q=secret+meeting+fisk&ie=utf-8&oe=utf-8&aq=t
Stock Market related links (Absolutely no endorsements or recommendations by me here….just providing relevant and hopefully balanced information!)
Israel ETF (Most direct way to take a position on how events in the Middle East will impact Israel’s economy)
Summary
http://finance.yahoo.com/q?s=EIS
Holdings
http://finance.yahoo.com/q/hl?s=EIS+Holdings
Gold
http://finance.yahoo.com/q?s=GLD&ql=0
3x Leverage – Bullish on Gold (If price of gold goes up 1%, you earn 3%)
http://finance.yahoo.com/q?s=ugld&ql=1
3x Leverage – Bearish on Gold (If price of gold goes down 1%, you earn 3%)
http://finance.yahoo.com/q?s=dgld&ql=1
Oil
http://finance.yahoo.com/q?s=USO&ql=0
3x Leverage – Bullish on “Energy” (If price of its energy holdings goes up 1%, you earn 3%)
http://finance.yahoo.com/q?s=erx&ql=1
3x Leverage – Bearish on “Energy” (If price of its energy holdings goes down 1%, you earn 3%)
http://finance.yahoo.com/q?s=ery&ql=1
Conflicting Messages From The Pundits (Just a reminder that there are always 2 sides to every trade)
Brent crude falls after World Bank cuts growth forecasts
CORRECTED(JAN 16)-UPDATE 6-Oil prices gain after Iran warning
Inflation Protected Treasury Bills
— Justin, if your only pursuit is to have the “same amount of money” one year from now……regular ol’ USA Treasury Bills (you can buy directly or via an appropriate ETF) will pay you little if any interest, but will indeed return to you the “same amount of money” at the end of their term (12 month term is available). Even simpler, however, would be to just keep your money in a no-fee savings account at your bank of choice. Right? If, however, you want to have the “same amount of purchasing power” one year from now…..then Jason’s suggestion of Inflation Protected Treasury Bills (or an appropriate ETF) is more suitable for ya’. In general terms, if your investment at the outset would enable you to buy “X” amount of a widely diversified basket of goods and services, the inflation protected investment assures that at the end of its term (e.g. 12 months)….regardless of what has transpired over the investment period viz a viz inflation……you will have enough money returned to you to enable you to buy that same “X” amount of that widely diversified basket of goods and services. So….dollar in now, dollar out in a year…..go for a savings account or regular T-bills. For dollar’s worth of purchasing power now…..maintaining same dollar’s worth of purchasing power in a year……go for the inflation protected variation of T-Bills.
Commentary (please read from “Jason’s personal trainer” point of view……i.e. “tough love”!)
Jason…..Justin…..I think you guys are awesome. Really. I also think that you both acted like weenies in this latest episode…..with Justin expressing “Georgie’s concern because AnyFu hasn’t launched” (instead of just taking accountability for what surely is his own concern), and with Jason making an equally questionable claim regarding the extent of his agency over his ability to give at least reasonably more precise AnyFu launch timeframe guidance. Guys…..nip this sort of crummy communication in the bud. Your audience has too damn much invested in you both and in the future of your partnership to allow either of you to screw it up……hahaha! Man up regarding AnyFu……talk it out…..plan it out…..then…..get it (a minimum viable product) out!
Ahhhh……one final comment…..and it’s actually back on topic. 🙂 In regards to the ramifications of the USA dialing back its expenditures geared toward providing military protection for all its allies and providing military power against all its foes, whose economy suffers more under that scenario…….USA or, say, Japan…..or the E.U., neither of whom have the infrastructure in place (let alone the spare cash) to ramp up military power on an expedited timeframe? Realistically that leaves India and China. I lived in India for a bit…..and currently live in Asia (Italy and Switzerland before that……born and raised in USA)…….and India is in absolutely no position whatsoever to divert the quantity of resources required to overtake the USA. And China? Yes…..their growth trend is impressive (borderline hyperbolic at times)…..but consider that 2 years from now…….January 2014……China’s economy will reach a milestone……it’s economy will, for the first time, reach 50% of the USA’s (That’s total nominal GDP, not per capita). And last…..even if the USA turned completely away from international / global military participation, the mainland USA would not immediately be even 1% more vulnerable due to geographic realities. The same can’t be said for Japan, the E.U., Africa, India, China, or the Middle East……all of whom would be forced to scramble to implement viable defense strategies.
http://webcache.googleusercontent.com/search?q=cache:eV84H2YbFRsJ:en.wikipedia.org/wiki/List_of_countries_by_past_and_future_GDP_(nominal)+&cd=1&hl=en&ct=clnk
Why not try something like this to annotate your functions:
function annotate(f) {
function () {
enter_profiling() ;
f.apply(arguments) ;
leave_profiling(this.name) ;
}
}
// Now go through all the function in the global namespace and add profiling annotations
for (var slot in this) { if (typeof(this[slot]) == "function") { this[slot] = annotate(this[slot]) ; } }
There are still some issues, but it should be a good start…
@David – Holy shit that’s a lot of info. Thanks so much. You’re right about the Anyfu discussion. What I will say is it’s very difficult for us to create any kind of plan or release target date, I’ll let Jason speak as to why that’s the case.
@Jason – have a look at https://github.com/substack/node-burrito – you can traverse the Javascript AST and then wrap function calls in a profiling method. Not sure how it would handle anonymous functions – but maybe it’s a start!
@Both – You guys should get James Altucher back on the show for some investing advice 🙂
Chaps,
Still listening but just wanted to get this feedback over while its fresh…
Regarding getting signed contracts regarding Pluggio – I agree with Jason, having being ‘burnt’ by a trusted client because I did not have any agreements in writing, I would definitely ensure this is the case.
And regarding proceeds from Justins house sale, personally I would just leave it in a regular savings account or invest in another property. Boring? yes, but I’m no expert in financial markets or politics so im too much of a coward to risk anything substantial in areas beyond my domain
Just saw this on HackerNews: it’s an article showing the cost of Stripe compared to different payment providers as costs increase. (Paypal is not on the chart for some mysterious reason).
http://expletiveinserted.com/2011/10/02/stripes-new-online-payments-service-wheres-the-catch/
http://news.ycombinator.com/item?id=3487751
Thank you for posting Axure,
I too was thinking the same things as I was listening to this part of the show, eventually though, I just skipped to the end since I like Jason the entrepreneur much better the Doom and Gloom Jason which he brings to the show at times.
🙂
I’m the one who mentioned kettlebells to Justin in an email. In the email I didn’t explain why I thought they would work for both Jason and Justin so I’ll expand on that here.
Jason often bemoans the monotony of the treadmill/bike so it seems like kettlebells could be mixed into his normal routine for a nice break. And yes, one will break a sweat with them. There’s nothing fru-fru about them — a kettlebell is an iron cannonball with a handle.
I thought the convenience of kettlebells would appeal to Justin along with the combo of cardio and strength training in one workout. It’s a very efficient use of time.
Anyway, kettlebells have worked for me and I enjoy (and look forward to!) my workouts. YMMV