Tea Time – 9-12-1015 — Big Herb

I have not yet figured out why there is a human propensity to distrust peer-reviewed studies and basic science. Sugar is a case in point. High fructose corn syrup is villianized but agave syrup is good for you? A quick use of a search engine and a restriction to .gov and .edu provides information describing the chemistry of both. (https://www.nlm.nih.gov/medlineplus/ency/article/002444.htm) Agave syrup (nectar?) turns out to be fructose. Corn syrup turns out to be glucose syrup.

Dig further and one finds that high fructose corn syrup is what you get when you use an enzyme (xylose isomerase) to convert the glucose in regular corn syrup into fructose (An enzyme is a protein that serves a particular biochemical function such as breaking or combining molecules). Thus, the main sweetener in agave syrup is chemically the same as high-fructose corn syrup. But, is it bad?

Too much of anything is bad. Sugars (anything that ends with ‘ose’) are empty caloric units and thus can be problematic. If you consume 3500 extra calories (technically kilo-calories) above your normal daily requirements, you have the makings of an additional pound of energy stored as fat.

So, why do we crave sugars? Why do we crave fats? Well, from an evolutionary point of view, we are the descendants of hunter-gatherers who were most likely to survive if they could sense and consume scarce high-calorie food sources would seem to be the most likely reason.

So, back to agave syrup. Why is it marketed as natural and better for you? Well, because it is a great way to extract money from a consumer. In the United States, the Food and Drug Administration (FDA) does not regulate supplements nor provide guidance on what is natural. (http://www.fda.gov/Food/DietarySupplements/default.htm) Thus, with it being under-funded, few of these products are tested unless someone complains.

So, no one controls dietary supplements.  To me, that means be afraid.  Many people fear “Big Pharma,” the companies who make approved pharmaceuticals.  Few people seem to fear “Big Herb,” the companies who don’t back research on their products as they may discover they are worthless or worse — dangerous.  (http://www.sciencemag.org/content/349/6250/780.summary?sid=e0a31d95-c418-4ef7-a82b-75cd460101f2 You may need to go to the library to read it as it is a subscription service) .

While “Big Pharma” must go through years of clinical trials to prove the safety and efficacy of their products, you or I could harvest our weekend crop of poa pratensis, dry it, put it in gel caps and sell it as a supplement “as effective as echinacea” since there does not seem to be any body of research indicating echinacea is other than a placebo nor poa pratensis as other than fiber.  Oh,  in case you didn’t look it up, poa pratensis is the botanical taxonomy for, “Kentucky blue grass,” which is what grows as a lawn in many American yards.

So, one lump or two?

Tea time – 9-8-2015

What a strange fascination we have with making robots that replace humans. Bina48 is a talking head (bina48) .  The Japanese are really into it (just go to youtube.com and search for ‘Human like robots’ and you will get an eye and ear full.

Why?  Are there not enough people in the world?  I’m pretty sure they eat more coal powered electricity and are less efficient than the 100 W bio-powered systems they imitate.  Perhaps “Actroids” are cheaper than humans but you must have to pay the voice over folks right?

How about AI?  There seems to be a great resurgence in fear of AI.  If you have ever had to work on over a million lines of code with little documentation and an expansive team you know that no one person knows how each section actually works.  Thus, you end up with people who specialize in portions of a code base.  Or, how about something as complex as a space station?  No human has the whole thing in their head.  However, if you could make an AI with a much larger and more available memory that could take the whole thing in and understand every nuance of every system.

Here is another example.  Biology is VERY complex.  Not only do you have to deal with ATC and G patterns for replicating proteins but you also have to deal with the methylation of start sections that disable or enable portions via environmental influence.  Genetics + epigenetics = exponential complexity.  Ok.  So none of us can fathom the entire thing at once.  But a very cool AI could.  It could fathom the entire thing at once and assist with the development and evaluation of the exabytes of genetic data and it wouldn’t even have to sleep.  Probably eat a lot of coal though.  Can’t be helped.

So, why do some people fear AI and why do some people insist on making silly human-like robots?

Software development processes

There is a very nice article on Quora at : http://www.quora.com/Agile-Software-Development-1/Why-do-some-developers-at-strong-companies-like-Google-consider-Agile-development-to-be-nonsense

Steve McConnell is my favorite author on software development processes.  His book, Rapid Development ( http://www.amazon.com/Rapid-Development-Taming-Software-Schedules/dp/1556159005/ref=sr_1_1?ie=UTF8&qid=1441682139&sr=8-1&keywords=Rapid+Development&pebp=1441682140131&perid=0DV2A3BNGGAG9W0SRYT7 ) goes through the long history of development practices prior to Agile.  It is my true hope that Steve updates the book to include his views on what Agile means.

Per the article, Agile methods work very good when one has a moving target and you need to rapidly change.  However, the author explains that Agile and Stable are two different worlds.  One can see where being agile with web page content is a good thing.  Perhaps not so much with the software in a cardiac pace-maker? Maybe not the software controlling your anti-lock brakes or your accelerator in your car either.

All that being said, I cannot wait for someone to come up with a term that replaces “Agile” since it is, to me, just another marketing buzzword like so many I have seen in the past.  After all, one cannot be expected to develop good software without having a large number of high-dollar consultants about to tell you what you should do.  The binary Agile vs Waterfall is one of the most irritating.  Are there no other processes in the past rather than Waterfall?  How about Agile vs the Spiral Development Model?  How about Agile vs the Rational Unified Process?

Jira is another tool to be abused.  It seems to be a ticketing system that is slowly trying to evolve into something as useful as Microsoft Project.  Very much like the Job Shop model from operations analysis, Jira builds a backlog of software items to be manufactured and they are placed into various worker’s production queue.  The developer works through the items in their queue.  It is a pull model.  The worker pulls the next piece of work from their queue when they are ready to work on it.  They break down the partial delivery times in “sprints.”  If something doesn’t get done in sprint it is just “technical debt.”  People love buzzwords doen’t they?  Perhaps someone would be so kind as to fix the Wikipedia article on Agile Software Development?  (https://en.wikipedia.org/wiki/Agile_software_development)

Thus, my opinion is that the word “Agile” associated with “Software Development” should be abolished and a new term used with concrete process oriented definitions.  I mean, isn’t it supposed to be a development process?  We could call our new method the “Nimble Software Development” methodology.   Of course, we would have to market it heavily and make it as ambiguous as possible so that everyone thinks that whatever they are doing it is wrong.  Oh.  Wait….  Hmmmm.

Tea time observations – 9-6-2015

It is interesting that gasoline is not a single chemical. Rather, it is a mixture of hydrocarbons with a lengths between 4 and 12 carbons long. UNLEADED GASOLINE (UNBRANDED) MSDS No. APPC975 Ver. 1. Print Date: 05/19/2003 lists some of the contents as:
BENZENE, CYCLOHEXANE, ETHYLBENZENE, METHYL TERT-BUTYL ETHER, TOLUENE and XYLENE.

Most people do not realize the fantastic energy density contained in a liter of gasoline. (http://www.eia.gov/todayinenergy/detail.cfm?id=9991) EIA states that even gasoline diluted with 10% ethanol still provides about 120000 BTU per gallon (31700 BTU per liter – 9.3 KWH per liter). Compare that to our highest density, publicly-available lithium-ion battery which boasts a mere 0.25 to 0.675 KWH per liter.

Now suppose that there is only a 25% energy conversion efficiency between the gasoline to electricity function.  That still leaves 2.3 KWH per liter of gasoline.

Ok.  So you buy an electric car.  Now you are burning coal (9.8 KWH/Kg) with a conversion efficiency of 33 50 45% (EIA again http://www.eia.gov/tools/faqs/faq.cfm?id=107&t=3) then assume a 40% power transmission loss to heat generated in the power lines.  On top of that, you can only get an 80 to 90% charging efficiency for LiPo batteries and that doesn’t include the loss due to AC power rectification at the point of delivery.

Interesting isn’t it?