Every new font is a new challenge, a new way to go off the rails. I love that blank first screen, as the bricks load for a new attempt. I always start out with high hopes, and then the inevitable compromises come. How I love Fontstruct. It's so damn hard.

Nah, it's a faux-italic, courtesy Photoshop.
I was in a bit of a funk over this one. Looking at it this morning it does look rather better than I thought it did. Press on with the dreaded caps.

Here's the brick (in four rotations) I most lust after. In this font it would get round having to decide upon a too-small but correctly-shaped cap or a better-proportioned height but ugly shoulders.

Drums fingers on desk...

Case in point: cap S: isn't it a shame to have to do horizontal rather than diagonal? If I had those bricks I could do it! Winge! Bleat! Double moan!

How long! How long o Lord!

Yes, brickstacking. It's a technique first discovered by Williaum, since departed our ranks. It exploits a Fontstruct saving glitch, and it's mindbogglingly useful for smoothing off harsh transitions.

Align two different bricks one directly above the other. Highlight both, and move them both up one brickspace. Deselect. Highlight the top brick, move it back down one place. It will obliterate the brick under it. Control-Z twice. If you've done it right, both bricks will magically reappear, occupying the same brickspace.

It can be done vertically or horizontally, and takes some trial and error to figure out what will and won't work. You can combine three (possibly more) bricks, but that's deep magic I won't go into here.

Only two drawbacks: how your browser is configured can affect whether you can do it or not. On my work computer, I can't copy stuff to the clipboard and paste back in; I suspect that's also why brickstacking won't work on the office machine. And probably a good thing too!

Second drawback: It's not possible to copy a brickstacked glyph and paste it back in again. The stacks go haywire. Which means every glyph that contains stacks (most of the glyphs in this font contain six, some more) must be laboriously reconstructed. Suits those of us who tend towards the obsessive/compulsive...

Happy stacking!

It's instructive to ponder why this design has given me grief. Of course there are unfortunate choices: I will insist on attempting things there just aren't the bricks to do! Hence the problems with cap J and Ww in both cases. X is always problematical if you're going to flirt with diagonality. (Is that a word? If it isn't it needs to be invented, so there.)

I'll be happy when there are enough (straight edge) bricks to cope with a design that's start point is in the brickspace or traverses it. Hmm. It's hard to express what I mean. These thick-thin designs cause problems where they traverse the grid. Perhaps that's the way to put it. It shows up in the glyphs I've mentioned. Nearly all the thick-thin fonts I've made have had to adapt to one brickless chasm or another. The brick set I'm proposing will go a long way towards filling the gaps.

I know I'm molesting a scab I've picked at before, but scabs just have to be picked, don't they.

Meek unleashed mayhem when he added the half-bricks. Speaking of Pandora, I wonder if he knew the grief he was bringing upon himself....

The immediate problem he has is how to add more bricks without the toolbox becoming unwieldy. The "my bricks" bit doesn't work because it doesn't increase in size as you add bricks. I use every brick under the sun (well, not quite) and the personalised brickbox quickly becomes useless. Besides which, the bricks appear in the toolbox in whichever order you lay them down. I rarely start two designs the same way, so they are always jumping around all over the place.

My vote is either to dump the personalised brickbox or make it expand dynamically as you add bricks. Which may require a tricky bit of a programming for an indifferent result.

Maybe the sub-menus off tools method used by Adobe in Phshop InD and Illustr could be the way to go, if the idea hasn't been copyrighted to death.

That way you could have a main brick and then all its rotations would be available in a sub menu off it. Once they are used on the grid, perhaps these sub-menu bricks could appear in another palette (groan!) which would become a free-floating version of the current personal brickbox -- with the important proviso that you could hide/reveal it with a hotkey.

I can see why (to us impatient types) Mr Meek is dragging the chain. He doesn't want to gum up a sweetly running engine. Except that the engine is running on 96 gasoline and it should be running on ethanol.

@em42: I can understand your reluctance to fix an unbroken engine. But I'd be ecstatic if I could generate serifs and whatnot that consistently cross the borders without having to change shape.

I can see you think about this topic as much as I do, thal. Interface design is a very interesting area, isn't it.

I do a lot of things by hotkey in my work life, but it requires a fair level of geekrancy to learn them all. Some of my co-workers almost never use keys to access functions, and I don't know how they manage, but they do. (Slowly!)

For them the mouse-driven commands are their access into the programs.

In Fontstruct I've only bothered to learn a few of the hotkeys. The interface is so simple to use I haven't (yet) felt driven to nerd myself up. I'm sure it's that simplicity Mr Meek wants to retain. So that somebody entirely new to the site can quickly and easily figure out what to do.

It's a not-insignificant challenge: how to increase the features of Fontstruct without eating up the screen or making it hostile to newbs.

No wonder the upgrade is taking forever.