CausticMango

Given:

• Oil is $103/barrel (as of today)
• Gasoline is $3.13/gal on Average
• One Barrel of Oil Produces 19.6 Gallons of Gasoline

Therefore, there is over $2.12 unaccounted for in the retail price of gasoline at the pump. Actually, more since there are additional costs beyond the raw materials — processing, refining, warehousing, transportation, etc each of which add to the price.

Now, we know the current retail price is averaged; the price reflects an aggregate of previous prices. One year ago, oil was about $55/barrel which gives a base price of $2.80/gal, again before any profits are taken. If we assume a linear price increase over the last 13 months (which is roughly correct), that gives an average price of $79/barrel or $4.03/gal of gas, again before any profits are taken.

If gasoline is like most other goods, the raw materials are a fraction of the price of the finished goods. In the case of furniture, for example, they make up roughly 25%. For prepared foods it’s even lower, something like 10%. Let’s be very generous and say that gasoline production and sale is super efficient and only has a 25% markup through the whole chain. The retail price of should then be $5.04/gal.

So, my question is, where the heck did the extra $2 go?

My friend who knows a former oil company executive says emphatically that there are no subsidies in gasoline production/sale. Given the amazing profits these companies make, I really hope that’s true, but the numbers don’t seem to add up.

Update (We Found the $2)

A friend pointed out the obvious that since other products are produced from the oil, you can’t use the full $103 in the calculation, even though gasoline is the most valuable product. At 47% of the value, that gives a base price of $2.47/gal and 27% markup, which is more realistic. If were assume the relative value of the gasoline is 60% of the oil value and use the average of $79/barrel, that give a base price of $2.42/gal and a 29% markup at the pump.

My guess was the next release of Ubuntu after Hardy Heron was going to be Intrepid Ibis, but it turns out it’s going to be Intrepid Ibex. So, I’m just going call the next now: I predict the 9.0x release of Ubuntu will be Jaunty Jackel.

So there, or whatever.

This is wonderful — I’ve never even gotten Windows to be this flexible. It turns out I had some trouble with my beautiful, 20″, widescreen, external Samsung SyncMaster 205BW monitor and my hand-me-down Gateway laptop after upgrading to Gutsy Gibbon.

Although Gutsy recognizes it fine using the default (non-restricted) driver for the ATI Mobility Radeon 9600 and mirrors the display to it, the GNOME desktop still thinks the resolution is limited to the resolution of the built in LCD on the laptop (1280×800). This causes the panels and various screen masks to be constrained the upper left of the screen (the external monitor’s native resolution is 1680×1050).

The “Screens and Graphics” system preference applet only succeeded in hosing the xorg.conf file and making the display fail completely.

I had kind of given up on it until I happened across some information stating that the applet was simply broken in Gutsy but that the xrandr command line tool was working fine.

After some poking around it turns out that my external monitor can be enabled on the fly (no need to log in/out) and the built-in LCD disabled manually with these commands:

xrandr –output VGA-0 –auto
xrandr –output LVDS –off

This also gets GNOME to recognize that the primary resolution has changed and resize the panels to fit.

The reverse sequence disables the external monitor and enables the LCD, resizing appropriately:

xrandr –output LVDS –auto
xrandr –output VGA-0 –off

To make life easier, I put these into two scripts in my home folder named “at-my-desk” and “away-from-my-desk” and created launchers on the desktop for them. Now all I have to do is double-click the right launcher to “dock” and “undock”.

Here’s hoping the next rev or so of Ubuntu fixes the “Screens and Graphics” applet, makes GNOME a little smarter about mirrored displays with differing resolutions and maybe gives us some hotkeys for what I did manually.

The iPod really is a great device. The hardware works well, the interface on the player is simple and useful, and it does what you want a player to do, especially with podcasts:  it identifies new episodes; remembers where you left off; and excludes them from your music collection on shuffle.

Unfortunately, it’s really designed to be an extension of iTunes. We’re able to get it to work with other software, like Rhythmbox, Banshee, gPodder and GtkPod, through the Herculean efforts of smart people who spent the time to reverse engineer it and build a support library and API around it.

But, that’s the thing. It works inspite of Apple’s unwillingness to provide open access to the device. No matter how well the hardware looks, I just can’t rely on Apple. You have to basically look at it as orphaned hardware with homebrew drivers (which really is great, actually, but doesn’t help you if you are looking for something state of the art with actual support).

Thus begins my quest to find an alternative player that:

1. Supports MP3 and OGG at the least.
2. Understands what a podcast is and treats it appropriately.
3. Can sync metadata as well as files (rating, playcount, last played position, album cover, etc).
4. Has an open API or sync standard.

I have this awful feeling I won’t find one, but, we’ll see.

Here’s a gotcha — when using XSLT files in .NET 2.0 with the XslCompiledTransform class, if the XSLT file contains an external reference like the one below, you must supply an instance of XmlResolver (like XmlUrlResolver) or you’ll get a nasty “Resolving of external URIs was prohibited” XmlException when you attempt to load it.

<!-- external references like this can trip you up -->
<xsl:include href="some-external-file.xsl" />

The reason this happens is because the internal implementation of the XslCompiledTransform class uses an instance of an inner class called XmlNullResolver which will simply refuse to resolve any external reference and throw an exception instead.

The solution is to provide invoke different signature of the Load or Transform methods and pass in a new XmlUrlResolver instance.

XslCompiledTransform transform = new XslCompiledTransform();
transform.Load(xslt, new XsltSettings(), new XmlUrlResolver());
transform.Transform(xml, args, writer, new XmlUrlResolver());

It’s unclear why the API designers decided to make this change in 2.0; most likely it’s a security concern, possibly an application of the Principle of Least Privilege.

Well, in yet another case of stupid DRM it looks like the Network or Cameron or someone has convinced TiVo to mark the Terminator series premier as protected in some way so that it can’t be transferred between TiVo units you own.

How asinine is that? We have two TiVo units, one upstairs and one down specifically so we can: 1) record shows with overlapping times; and 2) so that we can watch TV upstairs. Since the downstairs unit was already recording something else, we had the upstairs unit record it. When we tried to transfer it downstairs to the main TV, it refused stating that it was copyrighted and couldn’t be transferred.

I thought I hated the iTunes DRM scheme, but this takes the cake. It nearly ruined the show for me, I was so upset that this is even allowed to happen.

No more TiVo’s for me. In fact, they can all take their steaming pile of DRM horse manure and shove it.

It’s fascinating how powerful language is. Just invoking the right words can practically create a new reality.

One particularly powerful word is science. The average person, especially in the USA, has a complex relationship with it. To most, it’s a word of power infused with notions we used to call magic. It can be good or evil, light or dark, and the dark side is strong, maybe stronger than the light. Those that dedicate their lives to it live outside of normal society and are to be pitied and maybe a little feared, like the witches and shaman of before. In any case, it is a thing of great power, best not meddled with and respected if not fully trusted.

With such a perverse perspective, it’s no wonder that it can be so easily misused. Just call something a science (whether it really is or not) and, presto, it has mystical powers. Like, say, creation science. Or economics.

Economics is especially interesting. Politicians and legislators invoke it constantly; in fact it’s probably the most respected of the “sciences” among politicians. People with practically zero knowledge of real science, who don’t hesitate to speculate on and question the validity of biology, geology, astronomy, physics and mathematics latch onto and become acolytes of Economics, invoking it at will and never daring to question it. “The Laws of Economics dictate that such and such and so and so shall be done.”

It’s not that surprising, really, if you look closely. What is economics the science of? Given the name it must the science of, well, the economy, right? OK, well, what’s “the economy”? That’s Wall Street, Big Business, the flow of money and goods, buying and selling crap, and so forth and so on. Technically, it’s defined as “the system of production, distribution and consumption” by the people who study it, though Webster’s definition is probably more accurate and useful:

The management of domestic affairs; the orderly arrangement and management of the internal affairs of a state or of any establishment kept up by production and consumption; esp., such management as directly concerns wealth. (emphasis mine)

That makes the science of Economics “a social science concerned chiefly with description and analysis of the production, distribution, and consumption of goods and services.” But that’s a nice innocuous, modern and pretty definition. Let’s go back a bit to the time of the industrial revolution and we get this more honest definition, again from Webster:

The science of the utilities or the useful application of wealth or material resources. (emphasis mine)

Now it starts to make a whole lot more sense. If economics were really the study of how populations adapt in response to limited resources, as one source would have it, that science already exists. It’s called biology. If we narrow it to how humans behave, that science already exists, too. It’s called anthropology. In fact, “economic science” doesn’t really seem to contribute much of anything useful in the field it purports to study.

What are some of it’s primary principles? There is a thing called “the law of supply and demand”. It’s pretty obvious, really; the relative value of something is inversely proportional to the ratio of supply to demand (when something in short supply relative to its demand, it becomes more valuable). Astounding observation. Self evident to any three year old. Except, of course, it doesn’t really seem work in the real world that easily.

For example, we are told there is a critical shortage of IT professionals and that we must allow companies to both import workers and export work. Of course, it just so happens that the places we export work to are those without many of the basic human rights and worker safety laws or protections we’ve worked so hard to acquire so the labor costs are a fraction of ours. When we import the workers from those places they don’t have the protections or freedoms of citizens so they are more easily constrained to unfair contracts, thereby driving down local wages, too. So, here we have a case in the real world where a supply shortage in the face of increased demand has lead to stagnate or even depressed wages for the workers and a fewer students entering the field by enrolling in computer science classes. Economics tells us this is impossible.

Now, one might argue that is not a fault of economics, per se, but rather the disingenuous use of it. However, that’s typically how it is used. Let’s take another stellar example of economics at work — supply side economics, or as it was foisted on us, Reaganomics. The (overly simplified) principle being that encouraging production by creating incentives and benefits for the producers and owners will benefit all actors. Of course, it doesn’t, it only benefits the producers and owners which was probably the goal all along. It was just dressed up with the trappings of science to dazzle the masses with Laffer curves and all sort of official sounding terms.

Or, let’s look at another example, that of “free markets” and privatization. The basic idea is that mercantilism is the final state of human society. All problems can be solved when stated in the context of a “market”, where things can be bought and sold and profits can be made. This is so fundamentally cock-eyed that it can only be discussed seriously when dressed in the mysticism of scientific-sounding jargon. For examples of what it leads to, just look at Enron, the Russian economy or even the US healthcare system.

Finally, let’s look at the way economics leads to thinking about the human condition. In particular, that “worker uncertainty” is good for society. Think about that for a second. The premise is that if your life is too comfortable it’s bad for society as a whole. “Society” needs you to worry about losing your job, being thrown out on the street and not being able to feed your children. What kind of society actually wants that to be the normal state for the majority of it’s members? A truly fucked up society, that’s what. Or one that exists primarily for the benefit of the wealthy elite. Then again, we have names for those kinds of societies: oligarchy and plutocracy. But you can’t convince a majority in a democratic society to support a plutocracy unless you dress it up in fancy names, impressive graphs (like, say, a Phillips curve), and invoke the power of science. The science of Economics.

At its best, Economics is just amateur psychology, kind of a real world attempt at Asimov’s Psycho History. More often than not, it’s practical application is really about wealth and power and using it to control society for the benefit of the powerful.

So, it seems that merchants own the commons still as OGG support has been removed from the HTML5 draft spec.

Brilliant.

I hate being critical of someone’s podcast because I know how much work goes into them, but I do believe that Windows Weekly has officially jumped the shark.

Episode 39, the one with Ed Bott, was truly a WTF? experience, but at least I made it through the whole episode.

The most recent episode, 42 with Dave Fortin, was so tragic I couldn’t make it past the 20 minute mark. I was expecting to hear what’s so great about Windows Live they kept going on so much about (sorry, I just don’t get it), but all I heard was 20 minutes of “north stars” and “giving the customer what they want the way the want it” marketing/MBA babble. I still know absolutely nothing more about Windows Live or why I should even care, but I just couldn’t slog through it.

Guys, please, get back to something real. This is worse than the obligatory 10 minutes of syrupy “I-just-had-to-tell-you-about-how-Spinrite-cured-my-erectile-dysfunction” GRC testimonials in every Security Now episode.

Often when writing adaptation logic for SOA integration you’re faced with converting between similar XML formats with schemas that don’t seem to have been created with “transform-to-like” style operations in mind. (I’m thinking if you, MISMO!)

What happens when a business analyst / data analyst pulls up the schemas in a transformation tool like MapForce and starts trying to build the map? Extreme complexity in the map is what happens. This leads to excessive development time, unpleasant test cycles, and brittle transformations that break whenever they are touched.

The answer to this problem is to simplify the transformation by leveraging your SOA platform to enhance and pre-transform the schema into one that makes the adaptation map simple.

Here’s an example:

Suppose a source schema with an element, let’s call it “APPLICATION”, that has several types of child elements one of which we’ll call “BORROWER” that may appear one or more times where the ordinal is important (i.e. the first appearance is primary, the next is secondary, etc). Now, suppose a target schema that has a single element, let’s call it “APPLICATION”, too, with a fixed set of attributes named in the fashion, “BORROWER_1_NAME”, “BORROWER_2_NAME”, etc.

Remember, too, that the BORROWER elements in the source will be ordered relative to each other but in arbitrary order as children of APPLICATION (i.e. mixed in with several other sibling elements in no particular pattern).

This can be a nearly insurmountable problem without dropping to code for analyst mapping tools.

An alternative to handing this task over to a developer is to produce an interim “enhanced” schema that adds an Ordinal attribute to the borrower, pre-transform and enhance the XML to populate the new attribute, something like this:

<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">

  <!-- identity transform (produce the same xml output as input) -->
  <xsl:template match="*|text()|@">
    <xsl:copy>
      <xsl:apply-templates select="*|text()|@" />
    </xsl:copy>
  </xsl:template>

  <!-- enhance the borrower select with an ordinal that counts its position
       relative to other borrower elements in the same context -->
  <xsl:template match="/APPLICATION/BORROWER">
    <xsl:attribute name="Ordinal">
      <xsl:number count="*[name() = 'BORROWER']" />
    </xsl:attribute>
    <xsl:copy-of select="@*" />
  </xsl:tempalte>

</xsl:stylesheet>

This is not only trivial for the developer and simple to deploy into the SOA, it makes the map trivial, too. After all, this is the goal — drive out unnecessary complexity.

The point is, you don’t have to do everything with one tool or in one big step. Decompose the problem into several simpler and smaller problems and set yourself up so that the adaptation map, where all of the real business logic resides, is as straight-forward and simple as possible.