Thursday, June 25, 2015

Khan Academy

Earlier, I touted Codecademy (here and there). There are plenty of places where one can go and learn computer languages on-line. I liked that they were free, that they had an interesting collection of languages to look at, and that they provided the testing desktop in the browser.

Aside: Of course, my case is not the norm. If you search on 10 programming languages that will not die (Lisp, Pascal, PL/1, ...), I have worked in 7 of those (plus many dozens of other languages and environments in situations that could be called production - meaning, post release with real users doing actual work that relied upon the results provided by the computer system - err, app - the modern wannabe).

---

Khan Academy logoToday, I was searching on a technical topic and ran across Khan Academy (KA). I had seen them before as I was watching mathematics lectures at Stanford, Harvard, and MIT. That is, today, all subjects of note are accessible.

Aside: When I earlier ran across a KA video, it was interesting. But, at the time, I wasn't so much trying to learn something; no, I was watching the classroom situation (2012, Osgood's lectures - note the Remarks during the period) across different milieus (to be discussed) and how it influences delivery/discussion, etc. (2012, Wildberger).

Aside: Before going further, let me tout Wikipedia which is the first place for me as I search. Actually, it's Google, but, for the most part, a Wikipedia article pops up early in the stack. Wikipedia is getting some press. For instance, the Economist showed, recently, a tapestry that had been made from the Wikipedia page on the Magna Carta.

But, after looking at KA today, I am going to go there often. Why? It is a great place to jump to a "101" (not meaning the highway on the West Coast - love it) view. Earlier, I probably bounced back from its K-12 framework. But, there are plenty of new topics that are beyond high school.

Too, I looked at the "Computing" pages. Very nice. Businessweek just did a special issue on code. It was written/edited by one person, Paul Ford. Very unusual tactic. But, it worked. I actually bought the paper copy after seeing the issue on-line.

---

Okay. What does KA bring that is more than the others? It covers a slew of disciplines. And, we all know that it is not shameful for one to look at 101 material for any discipline that is outside of one's usual work. In fact, going to that view can lead to suggestions about additional material (as we see on Wikipedia).

At the top line, click on Subject, and you will see a fairly good representation of things that ought to be in one's mixture of knowledge. So, the material may seem elementary to the expert, but any, but a node-it-all, ought to find something of interest.

There is another perspective. All of these topics have been studied by those who teach them; the teacher-to-be ran the gamut from 101 to the end of the learning process as defined by the particular discipline. So, you might look at the material to get some notion of what is involved.

If you think of yourself as learning about the basics of a subject, then you will have the right mindset for enjoying KA (that is, if you're older and set in your little knowledge sphere). As well, you will see the general view of the topic; albeit, those who might have evolved and adopted a worldview over a period of time might benefit from going back and seeing the fundamentals, from time to time.

Actually, science, in its proper mode, would require that. Axioms and other assumptions seem to get hidden, almost as a general rule.

---

Any enthusiasm that one can perceive in this post's message came about from two areas: computing and math. Computing has some examples, with code, that ought to avail one of successful understanding even if one must provide one's own workbench. In the math area, the collection of post-high-school topics is nice (say, multivariate calculus and pick gradient - same subject, at Wikipedia - someone put a link to the KA site - same subject at Wolfram's collection by Eric Weisstein), albeit sparsely, somewhat, covered right now.

One can easily look toward a continued expansion on each subject. And, extending out toward the first couple years of college would be a boon. Of course, it has taken Wikipedia years to collect their material (all entered by volunteer authors). So, KA will be interesting to watch and to review from time to time.

Remarks:   Modified: 06/25/2015

06/25/2015 --


Tuesday, June 16, 2015

Node it all

For now, see some motivation (FEDaerated).

We will start to lay out the problems and our means for handling. In general, we can think of what we know and do not know. Let's use a simple chart.
    Know what we know - this is the basis of success and goodness
        Do not know what we [think that we] know -- all sorts of variations
                 (operational idiocy, for one)

    Know (as in, think that we know) what we do not know - again, lots to discuss
                 (fools rush in)
         Do not know (cannot know?) what we do not know - basically,
                 the infamous unknown-unknowns
                 (see 7oops7; to the wise, the largest set - listen up, cosmologists)
All sorts of modern workarounds exist, with statistical reasoning being the most insidious (many times). Think compactification (other analogs can be used) as one culprit (when applied without due concern for ramifications).

We will start from scratch. Charles Sanders Peirce, for one, will figure.

Remarks:  Modified: 07/24/2015

07/24/2015 --

Thursday, June 11, 2015

Rework

This blog has been an accumulator of posts since July of 2007. So, that is, now, almost eight years of dribbling "text" bits into the cloud for presentation to browsers on the subject of "truth" (any and all) and the engineering (including our handling) of such (hence, truth engineering - a companion blog dealt with operational aspects - 7'oops7). Along the way, there have been side roads taken, as the world has been through a lot the past decade. Don't you think?

However, the world is always thus. As I look back, every decade had its issues. As one ages, some of the older problems are resolved (or go away - like the gnats after a storm), but there is always a crop of new issues coming to fore. Perhaps, one life lesson is better handling of the change brought about by some problems.

By the way, "change" may be many things, including mental adjustments from knowing more, etc. One problem with such, though, is that one can far outdistance those around them (no hubris there; everyone learns at their own pace (whatever that is); standard pushes (yes, like STEM) may turn out to be counterproductive, as there is left little time for exploration - I wrote (and published) an oped on this subject way back as a young man in a college paper).

---

We're on the cusp, again, just like we were back in the 2007 time frame. I ought to use "cusps" as there are things boiling in several areas. One constant, though, is the growing use of computers, which includes the cloud. We are far beyond what was the state of the art in 2007 in several ways.

In many ways, we are not. We have argued for more recognition of quasi-empircal notions. One subject, of importance, would be undecidability. Yet, many do not appreciate the problem. For one thing, numeric processing sort of steps around the problems.

via the Friesian School
But, to use numbers, you have to number-fy; then, you need to de-number-fy (BTW, yes, this is a kludge based upon the fuzzify/defuzzify inverted scheme of fuzzy logic). It is in those areas where we really get to the problems. "halting" may have a lot to do with how well some result matches up with the expectations.

In my experience, for any bit of numeric processing, a whole lot more time went into setting up the problem and ensuring that things ran correctly. Then, there were the added tasks of making sense (in many cases, this could be a larger problem). ... People of intelligence, let me ask you? That whole sequence is now subsumed within some processes on a computer, and we are to believe such tripe just because the marketing folks think that it's okay (because the money rolls in?, Lord, help us) or because it has the mathematical basis of operational statistics?

---

Aside - things to look at further: Halting problem not important, ...

In a discussion of the importance, or not, of undecidability, there was the use of "intractible" which implies hard (several definitions). In one of my application foci, we solved the problem by having a human finish the work. That is, computers can chase their tails (much like dogs). Approaching some solution state that is well-defined can allow the expertise/intelligence of humans (yes, expect that we will get into the qualities related to problem solving that are not replicate-able - now (and, perhaps, not ever) to complete the process and check results. Simple statement; all sorts of implications (infer as you wish).

But, we know the other side of the story, namely that computers can run circles around humans (yes, numerically - you see). But, can the computer de-number-fy? Of course, I have to define, better, what that might mean. ... Want a metaphor? Vertigo is not far off the wall. Vertigo? Yes, a better way of saying the above - computer chasing its tail - actually, being lost/unbalanced is quite apropos.

---

Aside: The most atrocious of all is the modern manner of driving all sorts of human activity by computer (especially, when it tends toward solely). A bifurcation comes about where those with the numeric sense rise to be little gods (ah, the mischief they impart on the hapless - do I really need to start a litany here?) while those with other talents (far too much under-appreciated - you see, the numeric-folk'd minds see robotics as their savior - oh yes, get one to wipe your arse when you soil yourself - err, water-blasting, sand-blasting, etc., is not equivalent to a gentle touch). ... The solution, intelligent people, would be some recognizance of the peripatetic sense's necessity (oh, why do that when the mind (and its computational assistants) can run off on cosmological tangents - again, who will take care of the arses?).

---

So, even with its little shaky basis (and, intelligent folk, I am not talking GIGO), we allow these mechanisms to play a serious role at the heart of our economy. Oh, can it be any worse? -- perhaps, we could learn from listening to Made-off. How many of his ilk are there right now (milking the system under the guise of creating/applying Adam's little hand - this post is old, but it'll be upgraded)? Gosh, Adam rolls and rolls over in his grave.

Remarks:  Modified: 07/19/2015

06/14/2015 -- In the Feb. 2015 ACM Communications, Landwehr wrote that we need a build code for software. Is it not atrocious that we have allowed these all encompassing software disclaimers (May 2015)? With the proliferation of apps everywhere densely, this issue needs some attention. How? We have mentioned this before. However, given the hard problems, we would have to agree that we cannot ever have a node-it-all (that is, some position that can reach all) without raising the view (oh, yes, the bit of controversy there deals directly with the blind/delusion paradox). Even if we can seek the higher-view, what limits remain (as if, by residue)? Yet, carte blanche, a priori absolution of sins is very much at the heart of a whole lot of mischief.

07/19/2015 -- Today, I did a post after a few days of waffling (Many truths).  What took so long? Well, a lot of posts are in a conversational style. That implies some interchange. The epitome of that is twit-ville where the message is limited to 140 characters (or so). What would intellectuals from the past think of such nonsense that ensues? ... So, today, I pulled the content to my article editor; the effect was immediate as I have published six articles last year and have several in the works. It has to do with structure (conversation has none, essentially - too, conversation descends to the bar and barracks - you do not agree? why does Obama use the teleprompter?). As well as firming up the post, it was obvious that the thing had to split. So, a large part was cut out to spawn the next post on the subject. ... Related to this change in technique will be a re-look at posts by categories. These have been collecting over the period of the descent to hell (from which many have not escaped) and the rise of the elite (they did not go to ashes - too bad). ... The intent will be to rephrase, especially those that exhibited sophomoric presentation - mind you, there is nothing lacking in the content - albeit, many times things were left dangling, but, again, that comes from the styles reinforced by the idiotic approaches). ... Besides the 1% inflated set of goodies, the mesh has grown taut (oh, Lord).