Look, guys, this thread and its article are making a big, gigantic, HUGE mistake. Why? Heavily because the 'culture' here is limited almost entirely just to c

Look, guys, this thread and its article are making a big, gigantic, HUGE mistake. Why? Heavily because the 'culture' here is limited almost entirely just to computer programming. Also, the business examples are limited almost entirely to Microsoft, Apple, Google, Facebook, Twitter, etc. For software, here call that hardly more than load-store, calculate, if-then-else, do-while, allocate-free, read-write, and call-return. Okay, we are all now on the same page for 'software'. Uh, let's get in a little info helicopter and rise up to about 10,000 feet and get a wider view. So, the view we want is of 'information technology' and its entrepreneurship. So, for the entrepreneurship, as in Web 2.0, we want to please the users and paying customers, e.g., the advertisers. The advertisers are mostly on a monotonous diet: They want clicks. But they are cheap: They want the most clicks per ad dollar. So, we need to please the users and get lots of clicks per ad dollar. One way to please the users is to give them 'information' they will like. So, in both cases, we have some data, process it with software, and get out more data that we regard as the information we want to please the users and get lots of clicks per ad dollar. A big question is how to process that data. Or how to convert data into information. We want powerful means for doing that. So, where do we find out how to process input data to give desired output data? I.e., where do we go for such recipes for such 'secret sauce'? Sure, so far in computing, the processing has mostly been just intuitive heuristics or automation of what people understood well how to do in principle just manually. Can there be more? Actually, not much from computer science: E.g., lists, queues, trees, parsing, etc., in a cooking analogy, just grind grain into flour but don't say how to make, say, a Sacher Torte. Here is a hint from the word information: Or, as we know from Shannon, there is information theory which is really just some applied math. So, really, it's math that is the academic source of knowledge and techniques for processing input data to give desirable output data. Uh, it ain't high school math, and college calculus doesn't go very far. What can be done with that material is far beyond what can be done intuitively or manually. Can use this material to construct 'secret sauce' for the processing. This secret sauce need not be at all easy to duplicate: Sure, the US National Academy of Sciences could put together a team, but nearly no one in business would or could. So, if the secret sauce is powerful and its application valuable and is sufficiently obscure not to be easily duplicated, then it can be a "competitive advantage". "But everyone has access to the same math". Not in any important practical sense! Might need a well designed undergrad major in pure math. What fraction of the computer programming community has that? Getting it takes ballpark four years. Then might need a few years of carefully selected graduate mathematics. Now what fraction? Some of the most important material may be rarely taught. Now what fraction? We're getting down to a convention in a phone booth, guys. Then do some original derivations. Now we can be down to one guy and the guy he sees in his mirror. Kleiner Perkins, Sequoia, etc., write all the checks you want, but you will have a tough time finding anyone who can make sense out of this, find the right rack in the right section of a research library, read the right books and papers, and redo the original derivations, in less than years. You'll be "digging in the wrong place". The commonly claimed "deep domain knowledge" will leave you walking randomly in the stacks and getting exhausted. There can be little more discouraging than an unguided tour without prerequisites of some of the more important books and papers. Are there lots of examples now? In business, no. Thankfully for US national security, yes: For decades, the US DoD has seen these things very clearly. Can it work in business? Hmm .... "But it's not the mainstream; that is, it's at most just tiny, niche stuff." Uh, a movie starts out, "The early history of America is a tale of great first times." Well, we're still trying to have "great first times". So, don't try to copy Apple, Yahoo, Google, Facebook, etc. Or, to see farther than others, stand on the shoulders of giants. Fortunately, on selected shelves of the research libraries, actually can find such shoulders and no one standing on them, in that library or any others. Then can do some original derivations, translate that work to if-then-else, do-while, etc., have at low costs lots of great infrastructure software, processor cores, main memory and disk bytes, and Internet bandwidth. What a country! What a time! But it ain't just routine computer programming, guys.