I thought that was what you were saying, I just wanted to verify. As far as I know there aren't many (if any) generic function systems that are this flexible. One of the big problems of Common Lisp (even though it has CLOS available) is that some of the core functions are just regular functions, and as such aren't easily extensible to new data. PEAK-Rules in this regard seems pretty ideal.<br>
<br><div class="gmail_quote">On Fri, Feb 15, 2008 at 12:30 PM, Phillip J. Eby <<a href="mailto:pje@telecommunity.com">pje@telecommunity.com</a>> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div><div></div><div class="Wj3C7c">At 10:40 AM 2/15/2008 -0600, Aaron Lebo wrote:<br>
>Wow. This is all very informative.<br>
><br>
>Back to the practical applications of Peak-Rules. Should I design my<br>
>applications using a mix of functions and generic functions, or only<br>
>generic functions? If the former, what is the best way of<br>
>determining 'this looks like a good place for a function' and 'this<br>
>looks like a good place for a generic function'?<br>
<br>
</div></div>Er, I thought we already explained that.<br>
<br>
If you're using PEAK-Rules, you can just write functions -- and add<br>
rules as needed. There is no up-front decision to make, because<br>
*all* functions are generic, as far as PEAK-Rules is concerned.<br>
<br>
Of course, some functions, you'll know in advance are generic --<br>
anything that you would use a visitor pattern for, for<br>
example. However, it still doesn't matter, because with PEAK-Rules<br>
you still just define a regular function, and then add rules as needed.<br>
<br>
</blockquote></div><br>