It's more a question of the stability of the language specification. I don't doubt the code is solid enough for production sites; it's just that keeping up with Arc's evolution might be difficult. But as I pointed out, you might not need to do that. (I edited the original comment to be more clear.)
It depends what you mean by templating language. Arc already provides a means to create different reusable layouts and wrap output in them. The HTML-generating macros act a lot like templates. You can see how they work in the blog example that comes with Arc.
I actually much prefer Arc's method. Templates of the traditional variety are just so clumsy. It's hard to build abstractions using them; it's hard to document what you're doing with them (what local variables does this partial template require?); they end up being scattered around in a gazillion files the names and locations of which are hard to remember; and the coupling among templates and various other parts of the framework makes it really hard to test them in a REPL.
In Arc, just use functions. You'll have better abstractions, better testability, and better organization. A function that generates one line of HTML will just be a function in a library of functions rather than a tiny template in its own file.
The seperation of logic and presentation doesn't belong between code and HTML, it belongs between HTML and CSS. HTML is best controlled and generated programatically. Arc does it right, Rails does it wrong.
I'm not sure what your professional experience making websites is, but mine has shown me time and time again that making good looking sites is an iterative process full of workarounds for real-world considerations.
The method you're suggesting would make it excruciating to converge on a good design with any real designer. Look at the markup any good looking and well-implemented web page (especially one with dynamic content), and you'll see a huge amount of extra tag work to accommodate the real world of cross-and-backwards-compatibility between browsers, CSS limitations, browser quirks, and other concerns that people making professional sites require.
If every one of these changes and workarounds required changes to the code, which would require my time instead of the designer's time?
The Rails way is just copying a method that nearly every web toolkit uses, and they use it for a reason. Arc's way may be cleaner, but without a templating language it will be at a disadvantage.
- If you get the layout, including both how it should look and how things should work, from an external source, or if there is substantial external input, use templates. It will save time and headache.
- If you are just about the only one responsible for the layout but support skins (for lack of a better word; the CSS part), use the HTML tag functions and macros embedded in Arc.
Since Arc is at this point targeted for exploratory programming, the former is an unlikely case.
What I would like to see, though, is a templating engine akin to HTML::Template  in Perl 5. It has the minimum of control structures and extra syntax on top of standard HTML, just enough to make it possible to use the template for dynamic content. All templating frameworks suffer from the same fault, though: they define a new but severely restricted embedded language that's mixed in an unpleasant way with the HTML source. The language embedded in Arc is a much cleaner way to do things.
This is actually a really useful feature. The Ramaze Ruby framework does this by default and I like it a lot.
One thing that makes it even more convenient is automatically detecting what files to reload. Ramaze does it by using Ruby's $LOADED_FEATURES and $LOAD_PATH variables. You can get the same conveniece with Arc with a simple modification to arc.arc:
Nice. I was trying to think of an easy way to do this.
In order to not have to muck with the internals of load itself I wonder if you could wrap the saving path name functionality but then call the original load. Maybe this would be some new synatx / ability to add to the language? Of course this is just aspects which I thought Lisp could do so - I'm probably missing something.
Yep, thanks. It would be cool if there were an abbreviation for that, since it's fairly common. Eg l.2%1 (% randomly chosen) => ((l 2) 1). Some may say that goes too far toward a parentheses alternative for all sorts of things, but on the other hand data structure access is one of the clumsier parts of Lisp.
Seconded. Reading the source code is a lot of fun. It's well organized and really makes the case for Arc's approach to HTML. I know I'm a convert; having stuff in so many separate files, MVC style, feels like sludge compared to this. Even if you do have to start breaking things into separate files as the app gets more complex, being able to start out like this is a serious gain.