# Ridiculously RADiculous

**Posted:**May 25, 2010

**Filed under:**Uncategorized 9 Comments

**EDIT:***An user with the name gwoo corrected an incorrect assertion I made in the comments section:*

Your last example shows how an active record can be created and has very little to do with your assertion that it is to “use late static binding inside of closures…” The actual code to use late static binding in an anonymous function is…$class = __CLASS__; $lambda = function() use ($class) { return $class::invokeMethod(); }

—–

I have seen my fair share of absolutely atrocious PHP code, having being called way too many times on projects at a fairly late state only to discover I should have asked for a bonus (in the form of a personal masseur or something similar) just for the headaches.

I’m presently doing some research for a series of paper I plan to publish on bad design decisions plaguing the PHP code base. To this end, I started studying the Lithium 3 framework, which deemed it appropriate to crown itself as *the most RAD framework for PHP 5.3+*, hence the name of this blog post. I learned about the existence of this framework whilst attending a talk Joël Perras and Nate Abele gave at the Tek-X conference last week, which ended up getting a mitigated reception because of their *we’re right, you’re wrong* attitude.

Well, at the end of the presentation, I knew something was wrong. I knew I would be surprised if I ever opened the hood to see what kind of black magic happens in there. Boy, was I right !

I won’t enter in the detail of my fundamental disagreements with the design of the framework (for now, this’ll come later on the form of a detailed analysis) but I simply wish to share what I think is one of the funniest piece of code publicly available on the interwabs which actually is distributed shamelessly, to say the least.

public static function invokeMethod($method, $params = array()) { switch (count($params)) { case 0: return static::$method(); case 1: return static::$method($params[0]); case 2: return static::$method($params[0], $params[1]); case 3: return static::$method($params[0], $params[1], $params[2]); case 4: return static::$method($params[0], $params[1], $params[2], $params[3]); case 5: return static::$method($params[0], $params[1], $params[2], $params[3], $params[4]); default: return call_user_func_array(array(get_called_class(), $method), $params); } }

That code resides in the lithiumcoreStaticObject. You can browse it online here.

It was love at first sight. Oh dear ! This is so great. Pure genius. Of course, I’m being a bit cynical here. Interestingly enough, Tek-X’s keynote speech, given Josh Holmes, had the title *The Lost Art of Simplicity* (if I may, I want to mention that it was one of the best talks I have attended in a long time). Well, this code is *exactly* the opposite of simple. Is the invokeMethod() method called so often that it necessitates such a micro-micro-optimization ? Probably not. And, well, isn’t it a bit like solving a problem that would not have needed to be solved, had sane solutions been applied, i.e. stop using static methods all the time ?

**Yes.**

From what I could see (and maybe I am wrong), the Li3 guys use this method inside closures to do some **black** magic.

/* Why is this method named so ambiguously anyway ? */ $self::_instance(); $badAssLambdaFilter = function ($whatever) use ($self) { return 'OH HAI '. $self::invokeMethod('getEndOfLine'); }

But wait. Wouldn’t something along the lines of:

class Test { public static function aMethod() { echo 'OH HAI !'; } } $lambda = function () { Test::aMethod(); };

do the trick ? Ohh wait, no ! Because of Li3’s obsession for anything static (a strange obsession indeed), black magic needs to happen; the *static* keyword happens to not work inside of a closure, hence, they had to hack their way around this limitation.

Here’s yet another piece of code to entertain you; it’ll show you exactly how they do that:

protected static function &amp;_instance() { $class = get_called_class(); if (!isset(static::$_instances[$class])) { static::$_instances[$class] = new $class(); } return static::$_instances[$class]; }

So, in order to use late static binding inside of closures, there’s actually a method that creates “fake” instances which are then passed to the closure as a parameter. This baffles me. Li3 is full of such “anomalies” and cool stuff that does nothing more than being cool. This is broken beyond repairable. I often have reached a point where it seemed my code was broken, not because I knew exactly why it was broken, but because of all the weird hacks I had to use to actually make the code work. In the end, I always end up finding the missing link. In this case, it seems that it has never crossed the minds of the Li3 masterminds that something was evidently wrong.

But in the end, I’m still glad I save a few femtoseconds each time I call invokeMethod(). Hurray !

# The CES Production Function – Addendum

**Posted:**March 13, 2010

**Filed under:**Uncategorized 1 Comment

In the last post, I presented the *CES *production function and mentionned that its elasticity of substitution (which is constant), is:

I already have received a question about how to proove this so I thought it would be a good idea to share my answer with everyone. First of all, for a many-input production function,* *the elasticity of substitution must be measured between two inputs and, as such, it was imprecise of me to say that the elasticity of substitution is constant and I should’ve rather said that *all elasticities of substitutions are constant and equal. It is interesting to note that there are, in fact:*

distinct elasticities of substitution. We note the elasticity of substitution between the *i-th *and the *j-th *production factor:

where is the production function, and . First, we calculate the first degree partial derivatives, for all :

Hence, the ratio of the first degree partial derivatives, for and :

From this yields the result, which is now obvious:

QED.

# The CES Production Function

**Posted:**March 5, 2010

**Filed under:**Uncategorized 1 Comment

In many fields of economics, a particular class of functions called *Constant Elasticity of Substitution *(CES) functions are privileged because of their invariant characteristic, namely that the elasticity of substitution between the parameters is constant on their domains (for a definition of elasticity, take a glance at this post).

A standard production function, linking the production factors to output is of the form:

where *Y *is the total production, *L *the quantity of human capital (labour, measured in a unit like *man-hours*) used and *K *the quantity of phyisical capital used (measured in a unit like *machine-hours*). Generally speaking, the production function for* ** *production factors is of the form:

Where is the quantity of factor *i *used. The factors generalized *CES *production function (also called the Armington aggregator) is:

With . First and foremost, to study the returns to scale (refer to the link if you are not familiar with a formal definition of returns to scale) of the function, we shall study its homogeneity.

In other words, the *CES *function is homogeneous of degree . Hence:

- If , the returns to scale are increasing.
- If , the returns to scale are decreasing.
- If , the returns to scale are constant.

The other parameters are: ** **

**Relative weight:**The parameter associated with each production factor represents its relative distributional weight,*i.e.*the significance in the production.

**Elasticity of substitution:**The elasticity of substitution, as the function indicates, is constant. It is: (see the addendum to this post).

What is so interesting about this function is that one can derive special cases from it:

- If , we obtain the
**perfect substitutes production function**: . - If , we obtain the
**Leontief production function**, also known as the perfect complements production function: - If , we obtain the
**Cobb-Douglas production function**, also known as the imperfect complements production function: .

Proving such assertions is far from trivial and my next blog post will be dedicated to deriving the Cobb-Douglas production function from the *CES *function.

# The Elasticity Conjecture

**Posted:**December 10, 2009

**Filed under:**Uncategorized 1 Comment

If you know economists, you probably have noticed that when they discuss microeconomics, they almost certainly end up talking about *elasticity. *In fact, I have summed up this simple observation in my very own conjecture:

The longer a discussion about microeconomics between two economists is, the probability that elasticity is mentionned exponentially approaches 1.

Of course, if you are not acquainted with this notion, it may seem far-fetched to talk about the elasticty of a curve and discard the comments about the demand for insulin being perfectly inelastic as yet another surrealistic comment about the strange nature of markets. Well, not really.

But what *exactly *is elasticty ? Let . We define an operator which we call *-elasticty of parameter * as:

Simply put, it is the product of the partial derivative in regard to parameter and the ratio of and , which, you probably have noticed, means elasticity has *no unit *and is exactly the reason why its use is so widespread in economics. Elasticity is the measure of the *relative *effect the change in a variable has on another variable, regardless of the units employed. The elasticity of a parameter is classified in the following categories:

- : perfectly inelastic: a change in the parameter has no effect on the other
- : inelastic: a change in the parameter has a small effect on the other
- : unit elastic: a change in the parameter has a proportional effect on the other
- : elastic: a change in the parameter has a more than proportional effect on the other
- : perfectly elastic: a change in the parameter nullifies the other

To better illustrate the notion, let us take a very straightforward, thus bogus but instructive exemple. Let us imagine the demand for wheat is described by the following demand curve:

We can calculate the price elasticity of demand:

Thus, at point , elasticity is:

Which means that the demand is quite inelastic at this point, *i.e. *that a change in price will only have a small effect on the demanded quantity. Insulin is a very good example of an almost inelastic good: whatever the price, someone with diabetes will pay this much to get his dose, as it is of vital importance to him. *In contrario*, an example of a very elastic price demand is the demand for leisure goods, such as DVDs and books.

Microeconomists study many kinds of demand elasticities (*e.g. *income elasticty of demand, cross-price elasticity of demand, *etc.*) and it is of peculiar interest to observe the empirical measures of such elasticities. In a future blog post, I will introduce you to a selected collection of elasticity data taken from litterature.

# Of Ice Storms and Candles

**Posted:**December 2, 2009

**Filed under:**Economical science Leave a comment

In 1998, the province of Quebec was affected by the most important ice storm in its recent history, cutting electricity supply to over 1 million customers for a period of time varying between a few days to a whole month. During that time, I was living in a small town in the middle of the most affected area, sometimes referred to in the media as “The Triangle of Darkness”; sketchy, ain’t it ? Being only a decade old at the time, it is of very profound interest to me to reanalyze these events and put them under the scrutiny of my acquired sagacity. In this blog entry, I shall explain a particular event that caught my eye eleven years ago and that I ended up reanalyzing in a completely different manner after understanding the basic principles of economics. By extension, my conclusion also serves as a specific example of how the allocation of ressources through a market works and why it is an efficient system.

One of the most important things to have in an electricity outage is lighting, and candles provide a cheap and reliable alternative to flashlights or other alternatives to lamps. After a while, it seemed apparent the outage provoked by the ice storm would be very long. In my hometown, the suppliers of candles could not anticipate such a high demand and the stores were quickly depleted of their candles by eager customers. Furthermore, distribution was paralyzed by the outage, putting an ever greater pressure on the supply. All suppliers ended up running out of candles except one who made a sound business decision whilst saving the town of a complete penury: the grocery store.

It appeared that the owner of the store, a very clever and opportunist businessman purchased the remaining stock of candles in all the stores in the area and sold them for a much higher price. In fact, the price at which he sold the candles was around three or four times what one would normally expect to pay for such a good. This, at the time, appeared to my relatives and myself as a disgusting act of greed. In fact, on a strictly personal and motivational basis, it was probably an act of greed. Seeing the stocks of candles dropping, the manager quickly understood that the rarity of the good was skyrocketing, in a time where its main substitution competitor, electricity, was unavailable. Although one can question the ethical implications of such a move, the actions of the manager prevented, most certainly involuntarily, an even worst problem: a penury of candles at the time when they were the most needed.

In the case we study, the supply of candles from suppliers is constant since the usual supply routes are disrupted and the studied amount of time is very short, whilst the demand varies according to price, as it usually does. The situation is shown in the graphic below.

The two blue curves are the demand curves. The higher the price is, the lower the demanded quantity is. The curve with the steeper slope is the one representing the ice-storm situation. In this situation and compared to the normal situation, customers will always buy more candles as the number one substitute good, electricity, is unavailable. The initial price is and the supply is . If the price is not adjusted following the new situation, the demanded quantity will be . Since the supply is fixed and , there will rapidly be a penury, meaning that candles will be sold at but could have been sold at , preventing a penury.

The fact that we, as customers or producers, are all “greedy”, or said more gently, “economically sound”, is what permits a distribution of resources that is *optimal*. There is a finite set of resources on Earth and infinite needs; we thus need rationing methods. The simple fact is that there was an upcoming penury of candles in the town, aggravated by the fact that the electricity outage was anticipated to last for a very long time. People were buying candles like crazy, probably even irrationally so. This is a perfect example of customer greed. After all, how is it possible for an atomic customer to have the whole picture? There is no electronic board in the store detailing how many candles are left and how many each person should buy so nobody is left without candles. And even if there was one, how could anyone assess, without a prohibitively costly system, the metrics associated with obtaining such information ? Furthermore, everybody values candles in a different manner. If I already have a flashlight home, a candle is of very little use. I might be ready to pay $1 for a candle, but certainly not $3. On the other hand, if I have no alternatives, I would certainly be happy to pay $1, but I am still ready to pay $3.

In a sense, the grocery’s manager’s greed, combined with the greed of customers, is what prevented an even less enviable situation,* i.e.* unused or incorrectly used resources and customers left with pressing needs. By setting the price to , the demanded quantity was adjusted appropriately. Sooner or later, there would have been no candles left and many people really needing them would not have been able to get them, simply because they did not anticipate the candle-run or were not fast enough to evaluate their need in candles. On the other hand, some would have had too many and they would not have felt it was necessary to use them with parsimony, even if there was a critical situation of penury in the community.

All this seems beautiful but I will temperate my enthusiasm with a very important remark. On paper, the system works very well and prevents waste or badly allocated resources. Unfortunately, the price increase might mean some people cannot afford candles anymore, meaning they are unable to obtain a good that is essential. When this happens, government subsidies are the best solution. Wealth redistribution is, in theory, not efficient in a strictly economical sense (*i.e*. it is Pareto-inefficient), but it is simply the right thing to do. Collectively, we cannot accept that a certain portion of the population is deprived from what is essential, be it water, food, health care or candles, in this particular case. Without directly subsiding the purchase of candles in a specific town (which would have required a degree of flexibility way above any bureaucratic organization’s means), the Gouvernement du Québec provided an emergency financial aid during the crisis. Although I do not remember and have not researched the specifics of this plan, it is basically an aid aimed at the poorest to surmount the hardships associated with such a prolonged lack of electricity, including sudden hikes in essential goods.

In the end, one could say the store’s manager was greedy. One could say he was a heartless person profiting from other people’s misery. What is often forgotten is that customers are also very greedy, intentionally or not. This would never have been mentioned as such if there would have been a penury of candles; the stores would probably have been blamed for a lack of supply. After all, how is it possible to blame micro-actions when it is their aggregated effect that causes macro-problems?

# The Frivolous Theorem of Arithmetic

**Posted:**November 25, 2009

**Filed under:**Mathematics, Uncategorized 2 Comments

Steven Pigeon recently introduced to us the Frivolous Theorem of Arithmetic in his *Harder, Better, Faster *blog. As a complement, I shall present a rigorous, rather useless but quite educative proof of this theorem which will serve as a base for more inquiries in the internal structure of natural numbers.

The theorem reads as follows:

Almost all natural numbers are very, very, very large.

First and foremost, this yields the question of what is considered *very, very, very *large. In fact, the notion of a number being large or very large is inexistant in mathematics and irelevant to the internal structure of numbers. Hence, for the purpose of this proof, we shall choose a *überlargeness *threshold that we shall call *M*, a natural number. In fact, you can be an eccentric fellow and say 1 is already large enough for you and so be it. On the other end, maybe the excruciatingly, humongously large number:

is a number you envison as almost big (the Google calculator accepts to calculate 10! but dutifuly refuses to calculate 10!!; how unfortunate !). Regardless, the proof still holds. Let *A *and *B *be two sets such that:

Where A is the set of natural numbers smaller than or equal to the threshold and B the set of numbers we consider to be very, very, very large. To prove the theorem, we have to compare the cardinalities (the number of elements) of both sets. First, we note that the cardinality of of *A *is:

And the cardinality of the set of natural numbers is:

Which reads *aleph-null*, the first transfinite cardinal number who denotes the cardinality of countably infinite sets, *i.e. *sets with elements which can be counted individually, even if there is an infinite number of them. Since there exists a one-on-one relation (bijectivity) between N and B, which is of the form:

And according to the properties of cardinalities:

Meaning there is an infinity of numbers larger than *M*, a number chosen to be *überlarge* and we can thus conclude that almost all natural numbers are very, very, very large. This proof led us to an important insight on the internal structure of natural numbers: since each natural number has a succesor which also has a succesor (1 has succesor 2, which has succesor 3, *ad nauseam*), no natural number can be used to express the cardinality of this set. We can’t simply say there is an “infinite” number of natural numbers, as this is too imprecise. In fact, there is an injective relation between the reals and the naturals but there can be no bijective relation between the two sets, meaning that the cardinality of the naturals is “smaller” than the cardinality of the reals. Intuitively, the cardinality of reals is also infinite, meaning we have *two or more *types of “infinity”. This will be the subject of a follow-up post.

# Call Center Vendors

**Posted:**September 30, 2009

**Filed under:**Economical science 1 Comment

Until recently, I worked for a customer service vendor called Nordia as a customer service agent. Although I did not particularly like the job, I should say I spent a few good months there making good money. Nordia was created in 1999 as a partnership between J-Telecom Interests and Bell Canada (who now owns 100% of the company) for outsourcing customer services, such as the 411 directory assistance. Doing so reduced substantially Bell’s cost structure and increased overall productivity. Transferring the responsibilities from inside the company to an external vendor has many advantages, both financial and technical, because of the way the contracts are built and allocated as well as how the external companies are managed compared to the main company. In this blog post, I will explore what is behind this cost difference.

First and most obviously, the vendor’s agents are usually paid far less than their counterparts working for the company, sometimes 3 to 4 dollars an hour less, probably more if we consider bonuses and social advantages. Hence, each call made to the company’s customer service costs far less than what it would be expected to cost if it were handled by agents of the company. The salary difference does not make sense on a strictly comparative point of view. What justifies such a difference? Vendors usually hire their employees at market price. The skill set associated with the job is worth, market wise, about the hourly price they pay to their employees. That is, **they are able to hire enough employees who match the necessary skill set to fill all open positions without increasing the salary they offer**. The vendor does not have any incentive to increase the salary since it would mean a) an useless increase in applications and b) an increased cost. They do not have any incentive to lower the salary since it would mean not enough applicants for the job. Simply put, offer for work matches demand for work.

In the situation representend by the graph above, the vendor hires *Q *employees at an hourly rate of *S*. This is in contrast with Bell and big telecom companies who are usually unionized and do not have the flexibility vendors have, which means they pay more for a comparable skill set. The conclusion we can extrapolate from these simple facts is that it is the main company’s employees who are overpaid and not vendors’ employees who are underpaid, at least according to market economy principles. Below is a graph that represents this situation. *S’ *is the salary fixed by the collective agreement, meaning the work offer for the job will be *Q’*.

If investors feel that they are able to make a profit while paying their employees at market price, they will go in business provided there are vendor contracts available on the market. All this seems easy, but it’s really not. Paying employees an hourly salary is not all you need to do to keep a call center running. If a call center obtains a poor performance, such as a high wait time for clients or low quality of service, the contract linking the company and the vendor stipulates it will be paid less, or sometimes nothing. This is a clear incentive for a culture of excellence as the difference between a vendor losing and making money is the performance of its agents. This incentive is not present, or at least far less present in the company’s own structure. Even though the company can pay performance bonuses to its best agents, the extra money is far less valued since they are already paid a higher salary than their vendor counterparts. Furthermore, on an administrative level, there is clearly a bigger incentive in keeping the company profitable since its only source of revenue is determined by the call volume it handles and how well it handles it. The fact that there is competition in the vendor field also forces vendors to get the most out of their employees if they want to stay in business.

The clear end result is that the company is spending less for the same service provided to its clients, whilst losing less money if poor service is provided to its clients, making it a situation where it always wins, even though it has to keep a close eye on the quality provided by its cheap vendors to ensure customers are still getting acceptable service. It should be noted that this level is not necessarily the level of service deemed acceptable by the customer, but the level of service at which the company makes the most money, but that’s a whole different story.