From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: To: 9fans@9fans.net Date: Thu, 25 Mar 2010 13:31:36 -0500 From: blstuart@bellsouth.net In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit Subject: Re: [9fans] quote o' the day Topicbox-Message-UUID: f21224a8-ead5-11e9-9d60-3106f5b1d025 > On Thu, 25 Mar 2010 14:31:20 -0300, wrote: > >> It's this kind of intellectual ugliness that makes the >> teacher in me hang my head in shame. How could >> we be managing to produce a whole generation of >> programmers who actually buy into that stuff? And >> ... > > I assume you haven't studied human behavior much... Not much in the collective. My AI work was focused on learning in individual organisms. > sadly we are socially > dependent, we will follow even if we know something is harmful. Look at > fad diets and superstitions. I have seen that, though I still fail to grasp it. > I always have to laugh at the code reuse crap; I though that was what a > library was. Yep. I try to put it in perspective in the classroom by saying that we've had code reuse ever since Grace Hopper put a roll of paper tape for the Mark I in a drawer and labled it square root. (Slightly fictionalized, but not too far from reality.) > These hackers using > techniques from the 1970's to program machines in the 2000's .. 2010's; > all the while these techniques have decayed and warped to "fit" the modern > era. I also see the other extreme where trendy seems to be preferred over straighforward. There has to be a balance. We have to keep open minds about new techniques, but we also have to be versed in older techniques so there can be some reasonable judgement about when to apply which. Using the ABC technology (for any of a million values of ABC over the years) because it was written up in some management magazines and some marketing type said "you have to or we won't sell any" is neither engineering nor science and has no place in Computer Science or the Art of Comptuer Programming. (And for the record, no marketing person I've ever encountered who said something like that had the slightest clue about either the technology or the *market*. It has always amazed me how much the story is different when you bypass marketing and talk to the customer directly.) (And don't get me started on "Software Engineering." That's a whole 'nother soapbox I don't have time for today.) > May be I've studied too much history and may be I've studied too much > psychology, but, I think the only way for things to change for the better, > is for what we have now to collapse. I do hope I'm wrong. I hope you are too, but I must admit that I've often had the same thought. It's been observed, even semi-rigorously that the set of programmers falls into a bi-modal distribution with a small group that's about an order of magnitude better than the large group, and very little in between. When you probe into these bizzare techniques, tools, processes, etc, the rationale always seems to boil down to preventing the large, lower calibre group from doing too much damage, and to allow them to have a modicum of productivity. Unfotunately, the tyrany of the majority seems to prevent the examples of the Kens and Dennises of the world from guiding the progress of the art. I have a hard time envisioning how to make it better short of some kind of collapse that quashes the majority. In the mean time, I just try to hide away in a corner and affiliate mostly with the upper node of the distribution. But I should get back to real work. So, I step down from the soapbox again. BLS