Udacity Wishlist (Part 1)

Holy flip, I haven’t updated this blog in a long time.  Anyhow, yes, I am still alive.  Since I last updated this blog, I actually went to Israel for a ten-day birthright trip and met some wonderful people there, and learned some stuff as well.

I also found a job as a Research Engineer at a small Silicon Valley algorithmic asset manager.  Currently, I’m in training, because there was some stuff I didn’t know about R, and my boss is absolutely adamant about great programming practice.  Yep, I’m still kicking myself for not taking more computer science courses, and not being as serious about comp sci as I should have been (I thought I was going to become an actuary for 3 out of 4 years…worst decision ever!).

And speaking of computer science educations, several weeks ago, I was at a Udacity worldwide meetup, where I was hoping to meet the man that founded the company, Dr. Sebadasstian Thrun (errr, Sebastian…I still call him Sebadasstian, because it’s absolutely badass to have invented the self-driving car, and started what I think is the coolest of the 3 MOOC companies (edX, Coursera, Udacity)).  He was tied up in a meeting with Eric Schmidt, but I did get to meet the COO, Vish.

And of course, Udacity was looking for feedback, and while I do have some small criticisms (please bring back the old vertical checkmark progress method, rather than these tabs across the top that are just that much extra confusion), my main feedback for my most favorite startup in the world was namely…a wishlist–simply because I’d like to see Udacity rock house and revolutionize the world in terms of education.

So, let’s get to my main wish:

Vertical silos leading from no knowledge to jobs at top Silicon Valley firms (as well as algorithmic trading companies =P).  Basically, the way I envision it is this:

Employers state that they simply can’t find enough applicants skilled in computer science.  Colleges charge an arm and leg for results that aren’t stellar for most people.  And applying to jobs, unless you come from the elite of the elite universities with a nosebleed top-tier GPA, is often a crapshoot.  In fact, I can generally envision the algorithms (here’s some R pseudocode…probably not the best language to use for natural language processing but that’s beside the point >_>)

topSchools <- students[students$university %in% c(“Harvard”,”MIT”,”Massachusetts Institute of Technology”, “California Institute of Technology”, “Stanford University”, etc…),]

engineeringMajors <- topSchools[grep(“engineering”, topSchools$major, case=FALSE),]

otherSTEMmajors <- topSchools[tolower(topSchools$major) %in% c(“math”,”mathematics”, “operations research”, “computer science”, “statistics”, etc…),]

acceptedMajors <- rbind(engineeringMajors, otherSTEMmajors)

acceptableGPAs <- acceptedMajors[acceptedMajors$GPA >= 3.5,]

citizens <- acceptableGPAs[which(acceptableGPAs$citizen==TRUE),]

write.csv(citizens, “arrangeInterviewsForThesePeople.csv”)

(Basically what’s happening here is just a lot of subsetting)

At the moment, there are a lot of inefficiencies in the recruiting process, mostly stemming around trying to gauge a candidate’s knowledge in just a single digit amount of hours grilling the candidate after looking at one piece of paper.  And while Github may help with that, for one, I’d think the lion’s share of a person’s time that programs for a living is spent writing code that either A) isn’t flashy (i.e. if I’m a data scientist, most code I write isn’t going to do something profound.  It’s going to be selecting this data or that, subsetting this (as demonstrated by the pseudocode above), doing a basic operation on that, etc…) B) locked away under NDA or C) a “for-fun” little piece of hackery.

Now, when I called Udacity to get some career help, I was essentially told “github, excellent kaggle results, or bust”.  Namely, I felt dismayed that although I had passed Dr. Thrun’s self-driving car course with distinction, I had completed Dr. Norvig’s (research director at Google!), and completed the “building a search engine” course with all gold star questions answered (most correct, I believe?  Obviously the base class was “just” CS 101, but the gold star questions included a bit of NLP–in all fairness, for one of the questions, I actually googled up somebody else’s implementation of Levenshtein Edit Distance in Python =P ), and yet, that still didn’t matter one iota–to the person responsible for Udacity’s career services, no less.  I suppose this is a fair point, as Udacity at the time was (and still is) a very young company.

However, as Dr. Thrun alluded to, he wants to make Udacity’s mission that of helping people advance their careers in technology.  In fact, I really like this article:



However, I seriously disagree with Bill Gates in that anyone motivated enough to learn through Udacity would do equally well learning through a book.  Nuh-uh.  What I personally like (or liked, I think their format’s changed to something hideous lately, with the tabs across the top), about Udacity is the way that Udacity can teach to someone like me, who grew up on games, lights, and sounds, and thus, doesn’t have the greatest of attention spans, or ability to simply read a chapter for several hours and digest everything.

See, while most reference books may be good for just that–reference, in my opinion, most of them aren’t all that great for teaching.  They have no exercises, or simply present information in a deluge. I think one of Udacity’s greatest strengths is that due to the way it’s designed from the ground up to have people do programming online (unlike Coursera), Udacity can teach and assess for learning in small, bite-sized chunks that actually have students coding, rather than simply filling in a fill-in-the-blank, or multiple-choice question every 5 minutes ala Coursera.  Granted, the simple online programming environment may not be enough for more involved coursework, but that road can be crossed as well.

In any case, the fact that there’s a voice, colored text, a moving hand, and small, bite-sized assessments, along with the checkmarks that show progress for the unit is a very innovative, dare I say gamified (hello, Kevin Werbach!) approach to teaching, that in my opinion, makes Udacity’s format FAR superior than reading a book, whether paper or online.  In fact, it’s also far superior to any 90-minute lecture as well.


Now in my opinion, it is wishful thinking that anyone with the drive and motivation to make it through MOOCs will be an easily-employable candidate that commands a salary respectable enough to really live well in Silicon Valley.  Simply, there are (and I know from personal experience) a whole host of other factors that go into getting a job.  For instance, choking at interviews.

Some people just are not great interviewers.  They may talk too much, they may get nervous, they might choke up on a question they’d easily be able to answer if they were allowed to do a quick google search, and so on.  In fact, the reason that I got my first-ever job was that the quant that hired me, instead of putting me under the gun, sent me several scripts just to read, and gave me a week of a deadline.  In other words, no pressure whatsoever.  I found the process so entertaining that I literally asked for more work.  Unbeknownst to me, however, when I arrived for my in-person interview, the offer was already in place, so long as I didn’t piss anyone off in conversations.

In my opinion, this should be the ideal way to do recruitment.  That is, find a way to answer the question of “is this person qualified to do the job” without putting it in the context of “can this person answer a bunch of random technical questions while under the eyes of people who will determine his or her standing in life”?  After all, if a person is sending out resumes, they’re not doing so for laughs and giggles.  They’re doing so because they need either a job, or a new job.  And if they do, odds are, they’re probably under some amount of stress to actually find that job.

Also, there’s the issue of networking.  Although the old adage is “it’s not what you know, it’s who you know”, in my opinion, it’s both now.  Not only do you have to know something, you also have to know someone that can refer you.  Odds are, you don’t need to have a cutting-edge elite university education to do statistical analysis at Google.  But why should Google risk hiring from a non-elite university?  The solution?  Networking, referrals, etc…, as opposed to ability.

And on the part of the recruiter/employer, as I stated before, the problem they have is that they have to be extremely choosy simply due to time constraints.  If you’re a recruiter at Google, you’re dealing with thousands of resumes, and for each candidate to go through an interview process of several hours, as well as the time put into scheduling and preparations and so on, if you have, say, 200 candidates getting in-person interviews, and each candidate requires 10 hours of man (or woman) hours, that’s one year of human-hours gone, right there.  So recruiters/employers are fully justified in doing business the way they do (and why referrals work the way they do).

So, my question is: why doesn’t Udacity make it its mission to replace not necessarily the university–but the inefficient recruiting process (of which a university degree is but one aspect)?

Here’s the logic: if a recruiter judges someone on the quality of the undergraduate college they got into, well, the college someone got into is predicated on their high school performance, up to junior year.  In other words, ages 14-17.  Then, by the time those people apply for entry-level jobs, their GPA is composed of their performance from freshman to junior year, all equally weighted (i.e. if you had a couple of terrible semesters early on, like your freshman spring and sophomore fall as in my case, tough luck!), and from then on, someone’s employment history is most likely comprised of the fact that if they started off as a second-tier candidate, they’re going to be a second-tier candidate for quite some time, since a candidate’s current state is directly dependent on his or her previous state.

In other words, there’s a huge amount of latency on that resume.  In high-frequency trading, millions of dollars are spent on technology to shave off microseconds in execution.  In gaming, latency beyond a fraction of a second is unacceptable.  But in recruiting?  We’re talking months, or even years of latency.

What Udacity is in perfect position to do, in my opinion, is to have capstone classes, that, if passed to a certain acceptable degree, rather than just hand the student a shiny PDF saying “you passed our course with highest distinction!”, instead say “you completed our course with highest distinction!  Here are special, one-time, Udacity-generated links to employers, that when you apply to them, they’ll know you applied because you have demonstrated both the self-motivation/drive, and the technical aptitude, to be a productive employee of this company.”

And from such capstone courses, Udacity can build down, right to the very basics, such that given enough drive and motivation (and perhaps talent/intuition), anyone who simply is willing to put in the time, can fast-track their way from “this is cool” to “I do this for a living”.

And rather than having to roll the dice on candidates, and simply look at titles of courses on transcripts, employers will know for a fact that Udacity itself vetted the candidates, and will know exactly the material that was covered in these courses, and so, can hire with a lot more accuracy.

And as for Udacity?  Well, rather than try and get fees from candidates for “authenticated certificates”, or teaming up with universities to try and collect royalties from referring candidates for online masters’ programs, if it can produce these employment silos, and be both teacher and recruiter, well, if a Silicon Valley company hires a candidate for $100,000, and Udacity collects 10%, that seems a win for the company, a win for the student (now employee), and a win for the company that hired him.

As for my own personal silo wishlist, it’s quite simple:

Two of them–data science/statistics/machine learning/data analysis


Trading (as in, the quantitative, computer-and-rules-based Renaissance Technologies/D.E. Shaw kind).

(No, it’s not just about the money–I think that trading systems are simply cool–just the idea of risk vs. reward, controlling position sizes, the trading signals, etc… are awesome).

So yeah.  Odds are, this post was too wordy.  Which is why I suck at interviews =P

Posted in Education, TECHNOLOGY!, Udacity | Leave a comment

On Suits and MOOCs

First things first: I hope wordpress doesn’t try to become another tumblr with all this reblogging/following nonsense.


Today’s post is inspired both by the latest in educational zaniness, courtesy of Coursera and Udacity, and one of my new favorite TV series, Suits (on USA).  Namely, the idea that sufficiency should not be its own necessity.  To the point, although a degree would imply knowledge, a lack of a degree does not imply a lack of knowledge.  To put it more obviously, consider a wolf.  If wolf, then animal.  Makes sense, right?  Well, what if not wolf?  Does not wolf mean not animal?  Our subject can still be a bird, which is also an animal.

In any case, the entire premise of the show Suits rests on the fact that its main protagonist (well, arguably its main protagonist, played by one Patrick J. Adams) gets hired by Manhattan’s top corporate lawyer (played by one Gabriel Macht) because he demonstrated brilliant knowledge of the law, despite not having a law degree (or any degree, for that matter).  Essentially, this causes complications later on in the show, when, in reality, why exactly must one possess a law degree to even take the bar?  Also, a subplot is that the female that ranks highest on the scale of sweetness to snarkiness (Meghan Markle’s character) is smarter when it comes to research and the practice of law than most of the Harvard law associates, yet due to test anxiety, can’t pass the LSATs, meaning no law degree, meaning dreams are put on hold.  Yep, this is a show about *lawyers*, and in full disclosure, I’m a data guy, so I never thought I’d enjoy a show about lawyers because I thought it’d be all legal mumbo jumbo drama the way CSI is all about solving crimes, and the only forensics show I ever liked was numb3rs (I’ll give you one guess as to why =P).  In truth, it’s a very character-centric show and I love all the snarking.  But I digress.

Now, I’m not sure if anyone’s been reading the MOOC news, but Udacity, aka the brainchild of uber-roboticist and inventor of the Google Self-Driving Car, Dr. Sebadasstian Thrun (Sebastian Thrun), has just partnered up with San Jose State University to offer very cheap college credit (slightly over $100 a pop) for some low-level college courses (comp sci 101, elementary statistics, etc. etc.), while Coursera is having a fee-for-thorough-verification in case someone needs to really claim that they took those courses (I don’t think they give academic credit even with this, though).

But this new branching out of my two favorite MOOC companies has me somewhat disappointed.  Namely in the fact that they’re still adhering to the diploma-mill model, when, given the availability of top flight educational resources, why not try to circumvent that model completely?  I mean, if recruiters know that Udacity/Coursera are offering all of these computer science classes, can’t they just say “well, either you have a degree in computer science, or you will have completed courses X, Y, and Z, after which, you can take this technical assessment, and if you pass, we’ll give you half an hour on the phone, assuming we can verify your identity.”

I think this honestly has more to do with the paradigm of employers looking to hire someone who isn’t bad, as opposed to someone who is stellar.  That isn’t to say that I’d qualify as stellar–after all, if I was this great machine learning expert, I could probably make a living off of Kaggle alone.  This isn’t the case, and I definitely need guidance and training, just like any 20-something would.  But there are probably better ways to see if someone knows their stuff than simply making every Tom, Dick, and Harry (well, I don’t see any Toms or Harrys in here…) spend $100,000 on a mandatory stamp, the vast majority of which will be forgotten from disuse.

Now prior to now, perhaps these other methods haven’t been as readily available.  However, thanks to the power of the internet, shouldn’t people start to really question whether or not the only legitimate way to earn your stripes is to take out the equivalent of a mortgage before even getting your first job?  To go with the Suits show, are the candidates that pass through Harvard Law School really that much brighter than everyone else on the face of the planet?  If, thanks to the power of MOOCs, you’d allow anyone to audit Harvard Law School’s coursework, do you honestly believe that only those students physically sitting in the lecture halls can hack the material?

Well, according to statistics from Sebastian Thrun’s and Peter Norvig’s Artificial Intelligence course, which gave Dr. Thrun such an epiphany that he left his perches at Stanford and Google, the top scoring students in the class were not from Stanford University.  And furthermore, as Dr. Thrun administered the lectures physically, more and more students stopped coming to lectures, citing that they preferred to watch him where they could rewind him if they ever got stuck.

So with all that said–in the face of education that is perhaps more effective than sitting idle in lecture halls, why is it that we cannot fully switch over to MOOCs for obtaining the education that’s necessary for completing the technical competency part of job assessments, or taking the bar, or whatever else it is?  We could massively increase access, we can save millions of people tens of thousands of dollars (right now, student loan debt is greater than credit card debt, and is over a trillion dollars in the U.S.), give opportunities to the economically disenfranchised, the test-anxious, and whoever else, and most of all, we can allow anyone who’s simply motivated enough to have a guided path to the job of their dreams, ensuring a brighter tomorrow.

I wonder what the plot of Suits would look like if the New York bar simply required a collection of MOOCs instead of a J.D. just to take that exam.  There’d still be plenty of other dramedy to explore.

(For those that have never seen Suits, it’s on USA at 10 PM.  I recommend you marathon it ASAP and then watch the new episodes starting January 17th.)

Posted in Education, TECHNOLOGY! | Leave a comment

Problem Solving In Progress…Solution Under Way

Well, I haven’t touched this blog in months…not really much to say these days, actually.  Just working out, treading water at the current job, etc…

So I went to the MoMath today.  And, welp, I definitely set my expectations too high.  I mean after reading that it was started by a RenTec alumni and has RenTec and other ridiculously fkn smart quant directors on its board, I was expecting to have it rock and knock my socks off.  Instead, it’s still in a very start-up-y phase, with lots of moving parts still not completely working (and their under construction sign is absolutely badass, which is why I titled the blog with it).

That said, I am so thankful that we have this new museum to get kids interested in math.  Granted, there wasn’t a lot of explanation–just a lot of “welp, here’s something neat and cool”, with little of the actual “boring” stuff of expressions, equations, and applications beyond “oh hey, that’s cool”.  But when I talked to some staff there, their primary audience is 5th-8th graders–in other words, Dr. Glen Whitney and the others are trying to reach out to the impressionable kiddies and make them like math, as opposed to try and wow the rest of us who have already dedicated ourselves with our university studies and careers.

Still, it felt somewhat good to just play with everything playable.  I felt like a kid again (a little bit) but the knowledge that I was 26, and not 11-14 kind of took the wind out of my sails.  I mean considering I live near Philly and as a kid, went to the Franklin Institute a couple of times, I was hoping to have that kid-in-a-candy-store feeling, and it just wasn’t there for me.  I mean, if I met Dr. Whitney (or any one of the other current RenTec scientists or alumni), I definitely would have felt that whole sense of awe and adulation, but the exhibits themselves weren’t as targeted to my age group as I thought they’d be.  Probably because by the time people get to my age, they know whether or not they’ll be using STEM disciplines in their daily/professional lives.

In any case, I definitely hope this museum succeeds, and succeeds massively, because the more Americans know math, the more innovation we get.  The more innovation we get, the more Googles and gadgets and electric vehicles and whatever else we get, and that makes the lives of tons of people better on a whole.

Oh, and there was also a finance networking party, which was mostly unproductive because nobody wants to hire someone who can just program in R and train them a bit.  R isn’t too popular among NYC.  A lot more popular in Silicon Valley and Chicago.  Ah, well.

Posted in Education, My Life At The Moment, Society | Leave a comment

A Post Dedicated to a Merrittorious Bradass

So, as anyone who regularly reads my blog may know, I like game design.  I might not let it on much, but I think it’s a pretty awesome thing to be able to design something which other people will derive endless amounts of interactive, intellectually stimulating pleasure from.  And in addition to Extra Creditz, which is an amazing show and such, I found a podcast on game design that goes on for an hour each week.

The way I found out about this is by way of the twitter stream of one of the most amazing cosplayers (that’s people who dress up as cartoon/video game characters) ever, Mrs. Lindze Merritt.  She’s a makeup artist that’s part of a two-woman photography studio that makes anybody look absolutely beautiful and is located in the back of a wedding boutique, much in the same way that one of my best friends located his training studio in the back of a health smoothie shop.  (Nice one, Lindze!)

This blog post arose from listening to one of her husband’s podcasts on game design on a particular topic I like to wax poetic about–piracy.

Now, before I begin, let me state that this man, one Brad Merritt, in my opinion (for what it’s worth), is someone who I think has utterly won at life.  As in, when it comes to knowing what’s important, having your ducks in a row, and generally achieving happiness, he seems to have done it.  He’s a game designer for Cartoon Network, is married to a woman who can effectively be summarized as a Disney pixie, has a daughter who’s a miniature of his wife, and seems to be one of the most relaxed, laid-back, everything-is-awesome people I’ve ever had the pleasure of listening to on podcast.  Also, although he majored in something artsy, he’s become an adept programmer in some web languages, to which I say: RESPECT.  So yes, he’s a human being from whom just about every male on this planet can take some pointers from.

Now, as I stated before, this post will be about piracy.  But not in the usual way.  Though first, I do have to get something off my chest:

The one thing that Mr. Merritt said that I disagreed with is that saying piracy was bad was like saying water is wet.  The former is an opinion, and the latter is a fact.  But aside from picking at semantics because wordpress has no word limit, I think that piracy is a force of economics and tradeoffs.  I mean let’s just do a back of the envelope calculation here…some entry-level schlub makes $50,000 a year.  Let’s say $15,000 of that goes to Uncle Sam in some way, shape, or form.  Down to $35,000.  Now let’s say rent is $15,000 a year (a hair under $1300 per month, which seems to be feasible rent in a city).  Down to $20,000.  Assume ~$100 a week for food, down to $15,000.  Let’s take $2,000 a year for utilities+transportation, which brings us down to $13,000.  Now let’s get some $3,000 spent somewhere along the line (haircuts, laundry/bathroom products, new clothes, etc…) so we can get to a nice round number of $10,000 worth of discretionary spending cash.  So, considering that there’s 2,000 working hours in a 9-5 50-week work year (just for the sake of math), we have, essentially, someone making $5 of discretionary income per hour.

So, if we’re talking about a game on the market for the MSRP of $60, if our little hypothetical human being can locate a working torrent or other way to pirate said game in under 12 hours from the point he realizes he wants it, he’ll actually be doing himself a service by pirating it, just going by the numbers.

Essentially, my argument for piracy is that just because a game is downloaded for free, there is also the issue of spending time to find it, which could be spent enjoying other leisure.  If a game can be offered to someone for a lower price than the time they spend looking for it multiplied by the amount of discretionary dollars they earn per hour, a rational person would be wise to just buy the game.

Of course, there are those who would pirate something for the sake of not paying The Man, and those people cannot be reasoned with.  They’re determined not to give their money to anyone, so shouldn’t be catered to.

In any case, one of the more interesting arguments that Mr. Merritt brought up wasn’t about the whole moral preachiness of why piracy is bad and blah blah blah, but about how some people have tried adapting to piracy.

One of the ways was with microtransactions (I’m so bowie!), but the issue with that was, as Mr. Merritt so rightly pointed out, that in order to make money from the whole microtransactions model, the gameplay had to be deliberately painful so that people would pay real money so that they could advance in the game faster.  That a game that would otherwise be finished might no longer be finished because in order to monetize it, the designer suddenly had to think of a whole bunch of items and how to make people grind for them in a game that could be played for hundreds of hours, many of them fruitless, so as to encourage people to pay money.

And that sort of model really flies in the face of creating a short, sweet, memorable, single-player game, because if that game gets pirated, it’s not long enough to have microtransactions in it, or some other sort of revenue model besides “get paid for the game up front” (or so I interpreted).

And here’s the real salient point that he was trying to make: out of fear of piracy, game designers may not follow their hearts, because if they follow their hearts and get pirated, they can’t pay the bills.  I definitely don’t think it’s wrong.  But at the same time, I don’t think that piracy is the only factor, or even the main one.

In my opinion, the risk with a game studio (be it a single man or a whole team) following their hearts and setting out to do something different is that the project can completely flop because it just wasn’t a good game.  I suppose when you’re part of a larger company and can be compensated for experimenting and failing, that’s one thing–but if you’re an indie shop and one too many failures means closing the doors, you’re going to want to imitate what sells.  Would piracy affect such a studio?  I think it may, but I’m not sure if it’s the main detriment (though I’d love to see concrete data).

My take on it is this: if your game does well, it will be pirated because it’s popular.  By that point, you’ve probably made enough on the game to pay for everyone involved and hopefully get the chance to swing for the fences a couple of times.  If the game doesn’t do well, I don’t think it’ll be because it’s an amazing game that just got pirated, but because it was a bad game to begin with.  I’m sure there’s some region for which piracy is the difference between make and break, in that a game was marginally profitable, or marginally unprofitable, and just a few more people who pirated the game had paid some amount of money for it did, that the results would have been that much more promising.

In my opinion, perhaps a better way to fight piracy is to simply use a different model of payment–that is, not from customer to game developer, but to acknowledge that a single-player game will ultimately be free, and to take advantage of the fact that free means anyone can play the game who wants to, meaning more general consumption of said game, meaning more eyeballs with the game.  Perhaps the game can use a square screen, so if played on a laptop, can say, off in the right hand corner, sponsored by XYZ, without actually covering up the screen.  It’d be a harmless ad, but it’ll catch eyeballs, and if the game itself is good, then the game players might think “well who was nice enough to sponsor this game developer so that we can play this game for free?”

Another model may be to release a demo and price the game at a low enough point that it wouldn’t make much sense for someone to scour around for a pirated copy.  Essentially, get someone hooked on the game, and say “well, for $10, you can get the whole thing, just click here”.

IMO the big issue, though, is that most games aren’t all that good, such that when you get to the “well, this concludes our demo, you can buy the full version here for $10”, most people will think “why should I buy this game?”

Now while I can’t speak for others, I like for my games to be memorable experiences.  A game developer can develop the most skinner-boxed game ever and try to use all sorts of psychological tricks  such as the sunk cost fallacy (well, I already invested time/money into this, so I have to continue) to make someone continue playing a game long after it’s stopped being fun, but if they want any form of positive attention or compensation, then their game should be memorable.  And while I can’t speak for others, what makes a game memorable for me are A) the story (do I give a damn about what happens in this game) and B) the music.

Because long after a game’s graphics have become outdated, long after the actual mechanics of its gameplay have become simplistic by another time’s standards, the story will still be remembered (for instance, Last Scenario and RPG Shooter: Starwish), and, if you’re talking about music…well, anyone remember Street Fighter 2 from the good old Super Nintendo?  Youtube up Guile’s Theme.  The first two videos combine for a good FIVE MILLION views.

So, here’s my take on things:

Price games cheaply enough that people don’t think it’s worth their time to try and spend hours looking for a torrent.

Have a memorable game, with an excellent story and music.

Let players play a demo for free so that the music and story hooks them in.

There will always be those people who want to pirate something for the sake of pirating it, for that sake of triumph that they stuck it to someone who wants their money and got the product without paying for it.  Don’t move heaven and earth trying to fight against those people, as the very mechanics of digitized distribution are against you in that fight.  Lastly, I hope that game developers follow their hearts in creating their own vision that others will want to play, and hopefully, pay a reasonable price for.

Posted in Games | Leave a comment

Starting to Have Doubts About MOOCs

Recently, I saw a class go up on Coursera which is part of the University of Washington’s computational finance master’s program.

Namely, this one.

Now for anyone completely new to computational finance, this is, like the AI course taught about a year ago, what looks to be like a fingers-in-ten-different-pies sort of thing.  Sample the very basics from a bunch of different areas, get a taste of the foundations of things, and come away with next to nothing immediately relevant to be applied in the real world.

Now considering that I interface very closely with one of the individuals with a hand in almost everything current in the R financial universe, I was told that this course would really teach me nothing considering my master’s in statistics, and my background and work experience in quantitative finance already.

I emailed the professor releasing this course and asked him about placing the rest of the program on Coursera, and he told me that he only plans on putting three courses up to Coursera, since the program has to make money.

And what I was told by this man I work closely with is this (not verbatim): “would you rather give something away for free, or charge $25-$30k a year for it?”.  I answered “give it away for free and charge recruiting commissions for placing the better students into work, which Udacity plans to do” to which he replied along the lines of “good in theory, but I think Udacity would run out of venture funds before it could get such a model up and running.”  He also mentioned that MIT doesn’t have its advanced engineering courses released yet, which matches my experience in that the class releases seem to just be haphazard, disorganized, and don’t really seem to release courses that would allow someone to suddenly be highly employable whereas he wasn’t before taking the course.

Essentially, I’m left to ponder whether MOOCs will be treated not as a real education, but as just a way to expose yourself to some new material, but not really have a place to go with it compared to the more involved projects at junior/senior levels of undergraduate classwork, let alone involved master’s level classes, and probably not PhD-level coursework, especially ones that require proofs, unless you go the Udacity route (fill in the blank proofs, but that sort of defeats the purpose of thinking of which properties to use).

I for one am hoping that MOOCs succeed.  Frankly, I think the whole admissions process resulted from the fact that there are only so many students you can logistically fit into a campus, lecture hall, whatever.  However, as the AI course evidenced, of the top 400+ scorers who got a perfect score…none were from Stanford.

The issue is what business model supports sustaining all of these free courses, and why universities would part with cash cow programs such as the aforementioned computational finance program at the University of Washington to release these sorts of classes for free, when they can instead print out some fancy letters on a $10 piece of diploma paper, have a professor take 30 seconds to sign it, and charge between $25,000 to $100,000 for it.

In my opinion, ultimately, the proof is in the pudding, and the pudding is in the jobs that candidates will get because they made the personal commitment to enroll and complete MOOCs.  Until that happens, I feel everyone’s just spinning their wheels.

Posted in Education, TECHNOLOGY! | Leave a comment

News update

So I’m finally getting some work at my new job.  Call center data analysis.  Not learning too many new things.  But I work from home so the after-expense pay is quite nice :D–that said, I’d like to get my own place eventually.

Now, onto some reading I’ve done lately.  I believe I blogged about reading Greedy Bastards, by Dylan Ratigan, about how the problems in our society stem from the acronym BOUGHT–Banking, Oil, Universities, Government, Healthcare, and Trade.  Essentially, all these processes are very much corrupt in his opinion by Greedy Bastard behavior–aka something that while innocuous most of the time, will inevitably lead to a huge disaster–like a candy bar with a rock inside.  Each bite might be sweet and chewy, but at some point, you hit the rock and need to get your teeth repaired at the dentist’s.  I think it’s absolutely a terrific book.

Next, I read Good Self, Bad Self, by Judy Smith, a crisis resolution expert, which just seemed to me to be a worse version of Dale Carnegie’s How to Win Friends and Influence People.  I think the latter book is much better than the former, in which Smith writes about (in big fat text), a bunch of qualities and how they can be both good and bad (things such as ego, denial, ambition, patience, accommodation, and two other ones I forget).

Next, I read Unintended Consequences by Edward Conard.  I really, really liked this book, namely because I think it provides a strong counterpoint to a lot of the progressive Keyensian schtick these days.  Namely, that our GDP is composed of C+I+G+NX, consumption, investment, government spending, and net exports.  Essentially, Conard’s argument is to try and maximize the investment aspect of GDP, since risk-taking equity investments, such as done by venture capitalists (who he implicitly though never really explicitly) lauds, leads to world-breaking innovation such as the Microsofts, Googles, Facebooks, Ciscos, and other life-changing technology companies of our time.  And that these innovations aren’t the result of just one genius out of the blue having an amazing idea that just waited to come to fruition at the right time and place, but that these companies owe a lot more to the laws of large numbers than we think–a bunch of failed attempts at innovation beforehand, and that investors need more capital to take more risks so that they could fund more failed experiments so they could get that one good one.  And that when these innovations come to fruition, everyone benefits.

And in a way, he’s absolutely correct.  After all, even our lesser-well-off individuals have a better absolute standard of living than medieval kings.  We have heating, air conditioning, refrigeration, electricity, the internet, flight, and many, many other great innovations that increase standards of living for everyone all around, and even if the occasional innovator turns into a multibillionaire, society as a whole benefits.

My one nagging issue with his book, after taking a step back to digest it all, is that he seems to have a rosy vision of just how easy it is to retrain when creative destruction sweeps through a sector.  While Silicon Valley employs quite a few people, you can’t honestly expect that 50-year-old manufacturing worker to go and become a software engineer.  Although Conard just says such individuals will find jobs in the service economy, to go from a paid pension plan and benefits at GM to a waiter is a pretty steep drop.  And furthermore, I don’t feel he addressed why despite the fact that effective tax rates (especially for investors) are the lowest they’ve been in a long time, that people just below the cusp of the superstar Silicon Valley employee have a much more difficult time finding jobs.

For instance, one of my friends who lives a block down, despite a UPenn CS degree, has been effectively unemployed since quitting his consulting job at Deloitte to try and start a start-up (none of them took off) two years ago (though he was trying to go about it from the business end since he hates coding–yes, a CS major that hates coding, and his resume says CS, so the interviews he gets more easily are technical ones–where he gets burned every single time from what I can tell).  One of my best friends is in Columbia dental school, with his parents paying 100% for his education and costs of living (which my mother did for Rutgers, at least I have her–but my friends got undergrad, UPenn, and now Columbia paid for, while my father was trying to eject me from college the whole time…and was a genetic train wreck.  Yes, I am freaking jealous, but that’s a story for another time and why ladies should wait until their suitors have a lot to show for why they’re a good match), my other best friend is a fitness entrepreneur that’s running to and fro to get his business running, and IMO, he’s utterly amazing (I dropped 5 pounds in one month, 10 pounds in another under his training, and can only maintain weight working out alone).  Here’s his site.  So my conclusion is this: while we have an amazing system for getting the superstars to work, anyone below superstar status, even a notch or so, has a ton of trouble finding work, which just doesn’t seem to fall in line with Conard’s theories.

Also, one other issue I have not with Conard per se, but with a lot of free-market extremists is their glossing over of how some government intervention really spurred innovation.  Airbags were the result of regulations.  The space race (and it was just that–a race between the US and the USSR to the moon over military high ground) prompted massive innovations.  DARPA led to the internet.  And I’m sure there was plenty of material and other engineering research that went into the F-22, which is still so advanced that it’s still being debugged, and may not even have a proper foe to face, hence it got capped at 187 jets.  Furthermore, the NSF funding basic research fills up the well from which applied projects can come from, often decades later.  I know I personally get a distaste in my mouth thinking that somewhere, some professors are just going back and forth writing theoretical papers about theoretical equations and contributing little to society otherwise and think that such brainpower would be better applied to real-world applications, but come to think of it, basic research is also necessary.  Government agencies such as the NSF, DoE, and EPA IMO are necessary to fund ventures with a nonprofitable outlook, or one that will take an enormous amount of time while waiting for multiple paradigm shifts.  Things like desalination technologies (only 3% of the world’s water is drinkable currently), sustainable energy (currently nascent) research, development, and deployment, infrastructure repair and innovation, basic research (we could use another Bell Labs, and Google seems to be taking that calling with Google [X].  Oh what a dream it would be to work there), and so on.


Now currently, I’m reading The Price of Inequality, by Joseph Stiglitz.  And after reading five chapters, he really irritates me.  Now while there’s definite inequality in the country and some people are at the bottom of the barrel (someone has to be), and a statistically insignificant portion of the population have a large amount of wealth, I really hate how he just peddles the typical redistributionist propaganda.

First and foremost, yes, I agree that there should be a minimum standard of living, and that people shouldn’t be starving in the streets or going naked.  This isn’t the same thing as saying that everyone should be eating shrimp and pizza every day.  Health insurance should be a basic human right, but the overconsumption of end-of-life care is absurd.  Right now, most of our spending comes from:

A) a far overinflated military budget that is definitely not just used for national defense (look up the newsroom rant on youtube)

B) medicare/medicaid, 50% of which goes to 5% of consumers, and lots of that for expensive/unnecessary tests since doctors overcharge for service, and for end-of-life care, to buy a few more overmedicated months in this world strapped to who knows what kind of apparatus, while freelance employees, even young people like me, would find it prohibitively expensive from getting individual insurance plans.

C) Social security, which is solvent so long as there are more man-dollars being paid in than consumed out with what seems to be little return on those investments (in short, a classical ponzi scheme that more or less is solvent so far because life expectancy was shorter when it was implemented, with people living less than a decade post retirement, and now possibly fifteen years, not to mention the demographics issues of all the retiring baby boomers and people who were forced into retirement early because of the crisis)

D) interest on the national debt, which is around 100% of GDP.  This was probably run up thanks to the above three factors.

What part of that really benefits the rich?  The military-industrial complex by going on adventures into the Middle East, which pays out today’s oil barons at Halliburton, Exxon Mobil, et al. in more ways than one, as well as the defense contracting.  But beyond that…do rich people need medicare when they have their much better insurance policies, or social security, which is a pittance compared to their IRAs?  What about the very-much-underperforming publicly funded K-12 system?  Nope.  Private schools.  Or college grants and loans?  Nope, they can sneeze and pay their kids’ ways through school.

Infrastructure?  Which is a huge rhetorical point for the “rich paying their fair share”?  Hah.  The combined cost of the above four is $4.5 T, while infrastructure spending is $300B.  So when President Obama says “if you have a business, you didn’t build those roads!”, I find it highly dishonest.  Infrastructure spending is tiny compared to military, health, and social security spending.

The issue is that for all of the things that our politicians and lefty economists say the rich should pay their fair share for because it benefits them (an educated populace, infrastructure, and so forth), none of those things have to do with the four elephants in the room (military, health, SS, interest on debt).  So to say that the rich owe the government that much more for those things is nonsense.

What created the deficit?  Two unpaid wars, immense tax cuts during said wars which could afford to be rolled back, and emergency measures to make sure our economy didn’t go completely off a cliff.

What bothers me about the national conversation is the whole painting with a broad brush. Some wealthy members of society may not just be greedy for the sake of greed, but given how ineffectively Uncle Sam spends that money (terrible k-12 despite some of the highest per-student spending in the world, ineffective healthcare for the spending, etc.), they’d be correct to say that they could multiply their dollars and create more value than the government.

Furthermore, I feel there’s a case that needs to be made for honesty.  Rather than say that the taxes on the wealthiest should be raised to pay for infrastructure and better schools, why not just say “as you may know, most of our spending is on the military-industrial complex, social security, healthcare, and interest on the national debt.  The latter arises as a result of the three former, all of which are loaded with waste, but we want to raise your taxes to pay for them anyway because we don’t believe in letting people die on the street,” and then let people vote based on real issues.

My problem with Stiglitz though is that so far, he has yet to make a case (at least through the first five chapters) for how it would be beneficial to the wealthy that the 99% retain healthcare benefits as they are, or a social security system in probable need of reform.  Edward Conard used statistics to show how policies focusing on maximizing the I of C+I+G+NX would benefit not just the investors, but everyone in a better way than redistributing from wealthy to poor.  Stiglitz so far has yet to make this case, despite reading half of the book, instead focusing on, effectively, how wrong it was for wealthy bankers to get fat bonuses despite driving institutions into the ground, despite the fact that pay at the top of large institutions is a large chunk derived from values of call options and restricted company stock.

I mean yes, we get it, Joe Stiglitz is an “economist devoted to the study of inequality.”  That euphemism can be summed up quickly in one word: redistributionist.

My entire problem with redistribution is this:

If the poor knew how to manage their financial house, they’d be middle class.

If the middle class knew how to invest their money more effectively, they’d be rich.

You’re always going to have a left tail of a distribution of people.  The people there will not be happy in that position.

Digging and filling ditches does not work.  Throwing more money into a broken system such as education or healthcare to give people better education or healthcare, respectively, also does not work.  Trying to implement a one-size fits-all solution among a whole melting pot of people with different cultural values (African Americans who listen to rap and shoot hoops, Caucasian Jews who write code while participating in debate team, Asians who do calculus in their head while playing the violin or piano, yes I’m deliberately stereotyping to hammer the point, etc.) is not going to work.

All in all, so far, I’m on the economic side of people like Ed Conard and Peter Thiel, simply because in its current state, the government has shown that it is terribly bad at handling money.  If the government were better at allocating resources, I’d say to tax the crap out of Paris Hilton, Kim Kartrashian, and others (you know, maybe there can be a marginal tax rate of 95% for incomes above $250,000 for socialites just to get bimbos like that out of the national spotlight), but currently, I say there’s a huge chasm between identifying a problem (wealth disparity) and providing a workable solution.


Posted in Society | Leave a comment

CS 212 Wrap Up and More Udacity Stuff

With a LOT of help from one of my best friends out of Lehigh, Paul Berruti, I was able to answer all six questions on Dr. Norvig’s final, despite not being able to complete the unit 6 homework (was sorta busy that week, assembling a costume that I wore to the Philly Comic Con this past weekend–more on that another time).  Not that he gave me answers, but greatly helped in directing my thinking, without which I may not have gotten the questions submitted on time.

In any case, CS 212 was by far the most challenging online course I’ve taken and not dropped (because I felt I wasn’t learning enough in those courses).  Dr. Norvig is a very demanding professor, in my opinion, because he gives you only what is necessary and sufficient, but nowhere near enough to simply regurgitate a concept.  He forces one to use his company’s best product (well, he is the head research director at Google!), and really really think hard about the problem at hand.  And even after that, he’ll still make one look silly by coming up with a solution that takes half the lines of (if not an even better fraction).

But after finishing his course, I definitely feel like a better computer scientist for it, in that I feel I can better solve the types of challenging problems that even a professor as demanding as Dr. Norvig would pose.

On a technical level, I feel that this course really pounded into me the ability to use dictionaries, and really forced me to think about details of implementing an algorithm, even when I knew the idea of what I wanted.  So far, I feel Dr. Norvig, of all other Udacity professors, knows exactly how to put students’ teeth to the fire, and motivate them to actually stick with it, even when it would be so easy to drop the course.  Of course, in my own personal opinion, it  greatly helps that he’s the head research director at Google, because when the head research director at Google has something to teach, it would be a mistake of the highest order not to try one’s hardest to learn.

Which is also why I signed up for Dr. Sebastian Thrun’s stats 101 class.  Now, while I have an MS in this stuff, I know I can learn something about implementing this in Python, and at the same time, I will make an honest attempt to “TA” this class from a distance by assisting with questions people may have.  I’m not sure who decided to have Dr. Thrun, of all people–the head engineer at Google[x], AKA the second coming of Bell Labs (I can hope, right?), teach this class.  I greatly enjoyed Dr. Thrun’s class on self-driving cars, and was hoping he’d teach something else really advanced, because his teaching style is really, really amazing (trademark: Dr. Thrun himself), and it’s a huge pleasure to learn something beyond my realm of familiarity from this amazing, enthusiastic man.  Plus, he’s a German engineer.

In terms of Udacity’s other upcoming courses, I’m not quite sure I have a need for classical physics, though I would wholeheartedly and without hesitation recommend it to anybody who hasn’t had any physics exposure yet, as the way it’s taught is that the TA for CS 212 and CS 373 (Andy), went through Europe, and replicated the classical experiments of the pioneers of mechanical physics.  Which is perhaps the most amazing way I’ve ever seen of motivating the learning of basic physics since ever.

I am strongly leaning towards enrolling in the algorithms course, since while I’ve had a course on it before way back in Lehigh (IE 170), my performance in it was a disaster of unmitigated proportions and I’ve become a far better programmer since then, and want to give it another go–especially in the context of understanding social networking a bit better.

Speaking of social networking, I hear that Coursera has some course in the future on social network analysis.

Which is why I’m going to take this opportunity to go off on a tangent and compare Coursera and Udacity.

In my mind, there’s no comparison between the two.  Udacity may not have name-brand universities explicitly backing it, but in my opinion, it is by far and away the better product.  Why?  Two reasons:

1) A much better teaching method.  While Coursera’s method may have worked fabulously for Dr. Andrew  Ng when he taught Machine Learning (The ur-Coursera course, just as AI was the ur-Udacity course), the reasons it worked were that A) Dr. Ng is an amazing professor and B)  the material was distilled to the point that it was very manageable to implement in Octave.

However, in my experience, Coursera still feels like listening to a lecture in college.  Same powerpoint slides, same quick-to-get-past questions, and after exposure to Udacity’s methods, this just feels antiquated.  Furthermore, Coursera’s treatment of computer science leaves much to be desired in my opinion, since they don’t take advantage of the fact that students are at a computer by having them write code, until a massive homework assignment at the end of the unit–while giving little treatment to actual syntax over the course of the lecture.

In my opinion, Udacity is perhaps the first institution to truly take advantage of the fact that the students taking the class are in front of a computer.  Which means that they can write code as class progresses.  Now while more challenging problems quickly demand the user’s own Python interface in short order (I’m using Pyscripter, but really should look into IDLE since I don’t know how to access the call stack in Pyscripter), such as in CS 212 and CS 373, the very fact that the user is at the computer means that even without the specialized interface, any professor can say “since this is a computer-intended course, please pause the video now to solve this coding problem”.  Udacity has taken clear advantage of this.  Coursera, in my experience, has yet to in anywhere near the same extent.  Furthermore, in my experience, while the Coursera lectures cover the theory, the actual implementation gets very little treatment in lecture, whereas Udacity isn’t just about teaching theory–but teaching the implementation, and also, better implementation.  In fact, that’s what CS 212 was entirely about–being better at implementing the details of an algorithm.

2) Secondly: the people.  Peter Norvig.  Sebastian Thrun.  First of all, I think Sebastian Thrun is just an amazing professor, full stop.  Second of all, there’s something to be said about being taught not by university professors, but some of the greatest practitioners of computer science in the world.  Dr. Thrun taught the course on the self-driving car.  But he also invented the self-driving car.  This isn’t just some well-studied professor teaching from a textbook–this is the man who won the first DARPA grand challenge, who placed 2nd in the urban DARPA challenge, and who will more or less be credited with dramatically reducing vehicular fatalities in a generation once self-driving cars prove just how good they are (they’re already legal in Nevada).  Let me reiterate: he invented this stuff.  He’s also lead engineer at what I hope will be the second coming of Bell Labs–Google[x].  While it’s known that Google[x] is already working on scouters (I mean uh, augmented reality glasses), there’s also speculation that it may be even working on a SPACE ELEVATOR.  A freaking SPACE ELEVATOR.  And the man teaching courses to us mere mortals is the head of that place.  What next?  Google will reveal that within that lab, they’re hiring 100 PhDs to build up an algorithmic trading machine and create the next Renaissance Technologies overnight?  Would not surprise me.  Oh, and he’s a German engineer.  You know, like Werner Von Braun, the guy who created the Saturn V rocket?  Except Dr. Thrun may be creating something even more important and historic than that, in my opinion, if you can believe it.

And then, the man who just finished teaching us how to be a better programmer is the research director at Google.  The research director.  Do I need to say more?  Yes, actually, I do.  Because for him to take time out of his schedule to educate the world is something that blows my mind.  And considering that if anybody, he’s the one best qualified to teach Natural Language Processing, and even told me personally in an office hours session that he’ll be teaching it in the future, this is going to be amazing.  Learning NLP from the head research director of the most important natural language processing machine on the planet.  Who can possibly ask for better?

See, Coursera and EdX have professors teaching courses, and that’s all well and good.  But Udacity has not just professors, but the individuals that are literally the go-to person in the world–the world–teaching this.  It’d be like Jim Simons, Ed Thorp, or David E. Shaw teaching computational finance.

Now while I dropped out of the web design course, the fact that the founder of Reddit, Steve Huffman, was teaching it, also speaks volumes.  Now while the man is young enough to be my older brother and could certainly improve his pedagogical skills, it nevertheless goes to show that when it comes to exactly who teaches these courses, Udacity does not screw around.

(Oh, and for the record, I do hope that they tap some alumni out of Renaissance Technologies or D. E. Shaw’s quant desk (maybe Jane Street or Two Sigma?) to teach computational finance and make all of those masters programs irrelevant overnight.)

So, overall, an other hexamester of Udacity, and again a very satisfied Ilya.  This is an out-of-this-world amazing company.  I also think that for this, Sebastian Thrun is going to be immortalized by history as the man that vastly improved standards of living for millions of Americans–first by practically obliterating the cost of education, and secondly by allowing cars (perhaps buses, maybe trains as well) to cart people around without all of the flaws of a human being.  And I am watching history unfold.  This, to me, is awesome.

Also, I’m beginning to wonder whether or not I should start a new blog dedicated solely to Udacity.

Posted in Education, TECHNOLOGY!, Udacity | 1 Comment