> And I think Bart skipped over Pernod's central point, which is that you acquired
> emulation-relevant knowledge out of necessity when working with classic computers.
> You don't anymore, and you haven't since Windows 95 finally displaced DOS. You have
> to explicitly seek out exposure to things like assembly language and memory maps and
> A20 gates. So obviously a lot fewer people end up knowing it.
Good point; though remember there were far fewer people using computers back then, and they weren't networked, by-and-large. Whereas it was once only the lucky kid on the block whose family owned a standalone computer, these days your average 13-year-old is running around with a multi-gigaherz Linux workstation and eight different comm stacks on his phone - whether he truly realizes it or not. Most of them would never care to explore the core of their system, but I'll bet some eventually would. So, even given the lower percentage of younger programmers with a deep understanding of computer architecture, you would certainly expect a higher volume overall. This fact, coupled with the notion that the Internet binds the global pool of talent ever-closer, seem to suggest that emulation development should be stronger than ever.
With that said, the reality of the situation with MAME and other emulation projects previously discussed clearly paint a different picture, more along the lines of your converse assertions. Being the optimist, I'd love to write this off as a temporary lull based on other reasons we haven't explored; and we will see higher levels of development activity in the future. Perhaps due to nothing more than an increased fascination with computer heritage in general.
We may be in a period where there isn't so much interest poking around in, say, the Apple II or the Zilog Z80 - but one day that may change dramatically. If for no other reason, deeper understanding of preceding architectures that were far more visible and less encumbered by API cruft could be highly educational to future generations, and even assist in the creation of better hardware and software. Perhaps by actually using older emulated systems, one could develop new skills and interest in making improvements to them. After all, the goal of emulation shouldn't be to just archive these systems with a tag and store them away for dust, like the Indy Jones Ark; they should be USED, studied and learned from, as the important educational tools they are. There should be classes centered on MAME and MESS, to teach the fundamentals of computing through accessible and hands-on examples of past architectures, comparing and contrasting various platforms and operating environments - or at the very least, exploring the evolution of computing through a rare perspective. Without these tools, such a method of instruction would be unavailable and inconceivable.
|
Index 
Flat 
Re: Does anyone know how many dev are there work for Mame?
Reply
