"It looks nice" isn't really enough of an argument

It works for roses: what is their purpose?

I tend to have very different reasons for making a typeface: intersting form I've found, and then trying to incorporate that into a font; an existing typeface that I like, but which has some forms that irk me (I made Kaapeli for this reason); mechanical or technical features, like ink trapping. For this I made two families: Tang with huge, exaggerated ink traps, and Tar with round corners, and so controlling the trapping.

And also just because it's fun.

And also just because it's fun.

Agreed. Why do I need any more purpose than that if I’m not working on a commission?

Adam, if you are new to type design, and have caught the "obsession", may I suggest you start by trying to recreate something you like by not looking at it. Put a type design you admire in your mind. Look at it, study it, contemplate it, and then go about recreating it without going back for reference. You'll create something new and have fun at the same time. And how interesting would it be to compare your "inspired" design with the original.

You will then be doing what most type designers are doing, using your "sense memory" of a typographic form to direct a new design.

Thanks for the advice and encouragement all of you.

It's nice to hear that 'it looks nice' is actually a valid end goal as well!

@jmontalbano - could you explain a little more on what you mean by "your 'sense memory' of a typographic form"?

I think I may give your suggestion a go.

I was also contemplating giving some of the past 'type battles' a shot, just to get a feel for designing for such a variety of challenges!

Thanks again,

Adam

@Tomi from Suomi "an existing typeface that I like, but which has some forms that irk me (I made Kaapeli for this reason)"

I'm a web developer by trade, and the lowercase 'x' in Consolas (an otherwise great font) has been getting to me the last two days... on it. ;)

If I have to explain typographic "Sense Memory" then you don't have it. Sorry.

@jmontalbano - wow, ok.

Sorry for that comment, too much Tryptophan last night.

Sense Memory for a typeface, How does it make you feel.

Strictly speaking, Sense Memory is a phenomenon associated with Constantin Stanislavski's teachings, normally referred to as "Method Acting." The basic premise is that emotions are tied to experiences, and remembering the experience will spontaneously produce the emotion. Hence, a Method actor doesn't need to "act" to portray a specific emotion: he or she need only recall a life experience during which the emotion was felt, and the feeling will return.

We also have somewhere in our total memory and experience a sense of what the "Essence" of a letterform is. This is not a style or classification issue. It is more about developing a form which will be reasonably interpreted as a particular glyph. It has "B-ness" or which ever form we mean. (I dare not say "A-ness ;-)
This is not necessarily a skeletal form and may require the rest of the glyphs in the set to make its case. If we just take a skeletal form "B" and clothe it in a suit of serifs, we may see it approaching Roman Letter but it may be a poor example. Essence requires enough to elicit the desired response; "Yes, this is a Roman letter B". I think of it more in Plato's terms of "Seeing Forms" than just a Fontstruct kind of thing.
If, as type designers, we have acquired the vision to "see forms" in the typographic sense, then we do not need to examine existing fonts (although there still is an education in it). This may only be my preferred way of working but I find it valuable as a way of arriving at the specific by beginning with the general and avoiding using a specific known solution without first banging my head on it a while.

Many of my fonts are commissioned so there's always some kind of purpose involved. When I'm not working on commission, I still have a purpose in mind such as heavy, compact web headliner or narrow, antihistorical typewriter. At the very least, a font is designed with either headlines or text in mind. When testing a headliner, I don't need to worry so much how it performs in paragraphs at small sizes. When testing a text font, I don't need to worry as much about how it will look on a billboard.

"It seems to clear to me that every new typeface has to have a purpose, a reason to be created, or has to be meeting a need."

You can make up your own hypothetical needs too. I've done that quite a bit. Silicone is a font I designed to look good embossed in silicone. Nobody commissioned it for that purpose, it was just done on a whim. Having a purpose in mind makes it easier for me to make a unified design.

Not every typeface is made with purpose. I made Voivode with no purpose in mind. I opened up Fontlab, started drawing glyphs and let myself go. Voivode sold almost zero copies so that might tell you something. Some of my earlier fonts (Larabie Fonts) were experiments with no purpose in mind other than releasing them.