Dejavu

Post #154,621 by FuManChu 5/11/04 12:44:10 PM Reply	Dejavu Just had a raging thread on comp.lang.python and c.l.lisp about this: [link\|http://groups.google.com/groups?threadm=7xznfspmjf.fsf%40ruckus.brouhaha.com\|http://groups.google...ckus.brouhaha.com] ..wherein both sides had their say. You said: The problem with letting syntax govern the rules of abstraction is that as the level of abstraction gets higher and higher, due to the complexity of the program, the syntax can act as a barrier to dealing with the increasingly complex interaction within the program. ...which is fine. But the vast majority of code being written today has a common, low "complexity threshhold", IMO, which doesn't warrant a new syntax for each application. How many people are writing general ledger applications at this moment? Most cubby-farm programmers benefit from regularity. The less that programming language fetters you in the higher level of abstractions, the easier it will be to maintain and extend your code base. ...assuming you are the one maintaining and extending it, and have a good memory. For the rest of the world, we enjoy a "lingua franca" when possible to ease maintenance across brains and time.
Post #154,657 by ChrisR 5/11/04 4:19:46 PM Reply	I like big apps and I can not lie. Most of the apps that I write are years in the making, and I do wear most of the hats associated with the software lifecycle. But the larger issue, that I'd take exception to is that software development is not a monolithic industry. Yes, there are many jobs that are mundane, with fill-in-the-blank programmers. Many business don't like being dependent on individuals, so they try to promote mediocrity. In such businesses, software is not seen so much as an asset as it is an expense. But then, is that really the kind of business we want? Well, certainly if it makes a difference in getting a paycheck. But if one has the luxury of working for (or starting up) a company that values innovation, I'd run with flexibility. I'm not interested in the whether a mediocre programmer can follow. I'm interested in providing unique and compelling software solutions, at a reasonable cost, with reasonable speed, and timely delivery. If I don't deliver on any one of those three aspects, then worrying about the person that comes behind me is purely academic, as the software will have no need to be sustained.
Post #154,659 by ChrisR 5/11/04 4:26:02 PM Reply	Language de jour Almost forgot, since the purpose of this particular thread is to pretty well ignore Bryce, and just discuss Lisp in general, I thought I'd mention that what I see a lot of in the business today is multi-language solutions. And along those lines, I kind of like the following quote from [link\|http://lemonodor.com/archives/2004_03.html\|Richard Cook]: I look at it this way. A lot of the new, interesting software is being written in a combination of two languages; a high-level, dynamically typed language with good built-in data structures and an environment for rapid software development, and a statically typed, heavily optimized language for performance. For a Python programmer, these two languages are Python and C. For the Common Lisp programmer, the two languages are Common Lisp and Common Lisp.

Welcome to IWETHEY!