Musings on the Human Condition

Queen's Gambit is the first series I've binge-watched end-to-end in a very long time. Not since Anthony Bourdain's Parts Unknown have I found a series worth investing so much time into. And it's fitting, considering that although I've been playing chess since my father taught me the game as a kid, and I've dabbled in it off and on ever since, it's the now 9-months long COVID-19 social isolation that pushed me into the next level of obsession with chess.  What follows is a list of all of the factual chess references I could keep track of while watching this riveting series. I have attempted to omit anything that appeared to me to be fictional references, of which there were many. Episode 1: Openings Scholar's Mate Sicilian defense  Levenfish variation  Najdorf variation Queen's Gambit  Modern Chess Openings (book) Reti opening  A simultaneous  Episode 2: Exchanges Backward pawns Forks Knight-rook combination  My Chess Career (Jose Capablanca's aut

  I'm a Mac (and iPhone) diehard. But occasionally there's no getting around the need to install a Windows app. Such was the case when I wanted to play around with Adobe's Campaign Classic (ACC). The Adobe Campaign Classic Client Console application is only available for Windows. There are options such as Parallels and BootCamp for running Windows. But dual boot is too intrusive an option for my liking. And Parallels is not free. So, here's the process I followed to install ACC on my Mac for free with just a few tips to keep in mind. Before going into the steps and tips, here's a quick note on terminology . The following terms are synonymous.  Host (in this case Mac OS X) Guest (in this case Windows 10) is also known as virtual machine or VM or simply machine Now let's get to the steps and tips . Download and install VirtualBox Create a new VM for Windows (this YouTube video covers the basic steps nicely but doesn't include some of the additional tips below

  Covid lockdowns have significantly altered consumer behavior and buying habits. The desktop is Queen. For example, as noted in Gartner's Digital IQ Index 2020 report released recently, traffic originating from desktops is understandably growing (relative to mobile traffic) as more consumers access digital properties from the comfort of their homes. For details, check out Gartner's Digital IQ Index 2020 report for big-box retailers in the US (executive summary). Big-box retailers are superstores generally housed in large buildings that look like a big box (hence the name), for example, Wal-Mart, Ikea, and Home Depot. E-commerce. Another trend has seen the shift to e-commerce (relative to brick and mortar) even more so than before the pandemic. This is an expected result of reduced traffic in shopping malls and consumers being forced to buy online. Companies that had a solid e-commerce and omni-channel presence prior to the shift have clearly benefitted whereas those that did

Gartner recently published a study " Marketing Data and Analytics Survey 2020 " that finds that most clients aren't getting the value they expected from their analytics implementations. The results of the study are not entirely surprising and represent a huge opportunity for analytics experts. I have first-hand knowledge of multi-billion dollar companies that have purchased Adobe Analytics (AA) but are grossly under-investing in qualified resources to implement and support the platform.  They hire resources that are not required to be experts in AA with the expectation that they can learn on the job, just to save a few dollars in hourly rates. Big mistake! It’s like buying a Ferrari and then filling it up with regular gas and taking it to the local mechanic for service.  Upskilling in-house talent can only work if you have core expertise in-house (or contracted). But we’re often seeing clients who are lacking that in-house expertise and leaving their resources to fend for

When a pawn promotes (usually to a queen, but could be any piece, i.e. not a pawn), the arithmetic on points seems to go awry. If you were +5 when your opponent promotes the pawn, you should then be +6, since your opponent lost a pawn. Yes, they also gained (let's say) a queen. So they're up 9. So the correct arithmetic would be +3 for your opponent. But, I believe most chess engines calculate it as +4 because they don't deduct a point for the pawn your opponent lost. Not fair!

The COVID experiment has shown not only that corrupt, selfish societies fare worse, but also that a multi-tier healthcare system is a symptom of such a decaying society and results in higher deaths per capita. Elites like Bollywood actor Amitabh Bachchan get hospital care for "minor symptoms" while those who actually need hospital care are left to die. Non-egalitarian societies do worse in countless ways. Here's a link on death rates per capita by country (look at the second tab on the graphics). Leaders include some obvious culprits: Russia, US, Brazil, India. All nations with "cult of personality" leaders. The list of countries by wealth inequality has the same leaders.

I recently watched an interview with Python creator Guido van Rossum where he describes feeling the need for Python while working with the great Andrew Tanenbaum on a project called Amoeba.  He clearly states that he intended Python as a language/tool that could fill the gap between scripting and enterprise workhorses like C and Java.  He said scripting was good for up to 50 lines of code, after which scripting became unwieldy. And he designed Python as a language suitable for 100 to 1000 lines of code.  So, he clearly wasn’t intending Python to be used for industrial strength ML programs. It’s a bit scary to think that a lot of the ML libraries written in Python might not be as bulletproof as they ought to be.

Disclaimer: I am a Canadian citizen by birth and a US citizen by naturalization. So, I've experienced both systems first-hand.  If the COVID19 pandemic doesn't give Americans the push they needed to pass universal healthcare, I don't know what will. Bill Clinton, Hillary Clinton, Barack Obama, Bernie Sanders, and many others have been advocating for the right policy for decades now. And I struggle to understand what is preventing the US from legislating universal healthcare into law other than apathy and sheer ignorance. As reported in this article , Canada has fared significantly better than the US in terms of almost every per capita statistic, including cases, deaths, testing, unemployment, and more. What this pandemic has proven is that when those without healthcare or with inadequate healthcare hesitate to seek medical care it impacts all of society. Those without adequate access to healthcare are more likely to spread the virus as they delay seeking medical att

I've recently seen a lot of folks struggling with this, so I figured I would document the approach I'm using. Firstly, the best way to install most programs on the Mac is obviously Homebrew . However, whereas Homebrew works fine to install OpenJDK versions of Java, the Oracle versions are not available on Homebrew (or any of the Casks) as I write this post. So, you will need to download the installer from Oracle's website (requires registration). Make sure you pay attention to whether you need the JRE (Java Runtime Environment) or JDK (Java Development Kit). The JRE is sufficient if all you need to do is run a Java application or another application that runs on the Java VM. The JDK is needed if you need to compile or build Java programs. JREs and JDKs are designed to allow multiple versions to co-exist peacefully on the same machine. If the installer asks you to uninstall an older version, that's usually a recommendation aimed at preventing you from continu

Oh Henry! (Canadian Packaging) I was watching the documentary Catfish and was suddenly reminded of the Oh Henry! chocolate bar that I used to love during university days in Canada. I've always had an obsession with peanuts, all the way from taking peanut butter sandwiches to school, to loving chocolate peanut butter ice-cream (especially Haagen-Dazs). Turns out the Canadian version (pictured here) isn't available in the US. No wonder I've been missing it! Oh Henry! Oh Canada!

An instance can contain multiple web sites. Each web site can contain pages for multiple languages and locales (i.e. regions). OSGI configurations are global (i.e. they apply universally across the entire instance) and are used to control things like how many old versions of a page to keep as a page goes through modifications or edits. Most OSGI configurations can be as loose or as tight as you want them to be. If they are loose, they allow more control at the individual web site level. If they are tight, they allow less control at the individual web site level. All remaining elements of an AEM setup, such as user groups and users, assets (images, videos, PDFs), OSGI bundles, code, templates, components, rollout configurations, workflows, integrations, etc. can be seamlessly made available across an instance. Individual web sites can choose to reuse them or create their own. Reuse is easier across a single instance but also achieved easily enough across multiple instances via

The Denominator Problem There was an interesting op-ed in the WSJ a couple of days ago (see link below) by Eran Bendavid and Jay Bhattacharya from Stanford. They ask if the cost of a lockdown is perhaps higher than the cost of no lockdown. The op-ed is behind a pay wall, so I'll provide a summary of their argument. The authors claim that the commonly assumed coronavirus (CV) death rate of ~1% is inaccurate (i.e. too high) due to the lack of antibody testing to determine the true denominator for the ratio. (The death rate is calculated by dividing the number of deaths, i.e. the numerator , by the number of infected, i.e. the denominator .) In other words, we don't have an accurate count of everyone who has the virus until we do antibody testing of everyone. That's because many people who have the virus develop antibodies and very mild symptoms and never get tested and never get counted as part of the denominator. I suspect when the they calculate the death rate

There's a lot of confusion out there about the terms virtualization, containerization, etc. and the related products such as Vagrant, Docker, VirtualBox, etc. When do I use what? I like to think of it this way. The physical machine could be a Mac or a Windows PC or laptop. VirtualBox is an Oracle product that provides the virtualization layer, thereby allowing you (for example) to run Linux on a Windows machine. You can configure the resources (disk, CPU, RAM) you want to allocate to each virtual machine (VM). Vagrant is for consistent environment management paradigms across all platforms/OSs. You can use it to install base software for you application, e.g. the Apache web server. Docker is for containerization and only runs on Linux because it uses Linux's container facility to do its thing. So, it used to be that if you're using Windows, you had to run VirtualBox or some virtualization software to stand up a Linux machine first before you could install D

I've recently run into several posts critiquing Adobe for their licensing practices as well as their product suites, which often consist of products with overlapping functionality wherein the products are inconsistent with each other in terms of the UX, thereby making it look more like a hodgepodge than a suite. For example, the Adobe Create Cloud (CC) suite reminds me of Bloomberg, the financial trading platform. The Bloomberg platform has multiple ways of doing the same trade (e.g. let's say you want to go an FX trade) and it can be impossible to figure out the pros/cons of each way of doing it and pick one way to do it. I know this because I worked in the wealth and asset management sector for a long time in a previous life. In Adobe's case, this is the result of growing by way of acquisitions (as opposed to growing organically). Adobe's digital marketing suite, aka Marketing Cloud, is starting to suffer from similar issues, e.g. there are multiple ways to manage