On Thu, Jul 24, 2008 at 10:01 AM, Brian Lloyd <[EMAIL PROTECTED]> wrote:
> Yes, that works just fine. Of course you are very likely to find > yourself in a vacuum -- which may be the intent and may, in fact, be > the most efficient way of proceeding. Actually writing and debugging code is a fairly hermetic activity. However... > OTOH you have to be really, really good to not want to see the outside > ideas. Granted 90% of everything you get will be of little or no use > (including what you get from me) but that last 10% (1%? 0.1%?) might > prove surprisingly insightful and valuable. You have to dig through a > lot of rock to find the occasional jewel. You find more jewels if more > people are digging... I think the issue here is *not* a resistance to outside ideas. The problem is that so much of the design and implementation that's happening behind the scenes is essentially a refinement and a critique of a large body of closely-related work that's already very mature and very refined. Astonishingly little of the "new architecture" is in fact new at all. A lot of the design job has amounted to reviewing and culling the best available common, Open technology for our general family of applications. For example, much of the VR is different in no important details from a modern high-performance, mission-critical musical sound engineering application. The art and science of such applications is *very* ripe, and the documentary trail is decades long. Similarly, Erlang is the culmination of both a couple generations' worth of system development combined with shrewd predictions concerning the coming round of application-level networking innovations. It strikes me as the height of profligacy to think we can do better, given the resources available to us, at least as a practical matter. Designing this particular VR, for the application arena we're concerned with, requires above all a lot of woodshedding and absorption of all the pertinent preceding work. We've tried pretty hard to publicize what the necessary background is and how it can be acquired. Many of the issues that keep coming up for discussion are old, old discussions that have already seen their outcomes decided by the real world. One of the most conspicuous lessons is where the boundaries need to be drawn to keep a system coherent and maintainable. If I in particular am resistant to public discussion of many design concepts, it's because in any practical sense there *aren't* any issues to discuss. At least -- and I stress this, again and again -- at least, *not without a tangible prototype to criticize.* We're standing on the shoulders of many, many, very smart and productive developers and users *already*. Where we're at now is in exploiting what's already there. Once there's something to criticize, then it's time to bring out the knives and axes and see what we need to learn to do it better. Until then, the discussion amounts to little more than psychotherapy sessions devoted to how our (technical) ancestors decided to bring us up as children. They don't help getting the house built. 73 Frank AB2KT -- Travelling by airplane in the US is nothing more than mass training of Americans to the requirements of the coming police state. The whole point is to make you learn to acquiesce without question, en masse, to completely absurd directives by dull functionaries wearing uniforms. -- Digby _______________________________________________ FlexRadio Systems Mailing List FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archives: http://www.mail-archive.com/flexradio%40flex-radio.biz/ Knowledge Base: http://kb.flex-radio.com/ Homepage: http://www.flex-radio.com/
