I found the requirements somewhat intimidating.
You have to be a paleo-linguist, software encoding genius, and be able to draw nice-looking glyphs to boot.
Sounds like a job for a Reading graduate, not me!

I found the requirements somewhat intimidating.

I know that feeling. Tiro might be able to pull this one off, but not me!

Yes, they don't actually ask for ability to walk on water, but it seems to be implied. In terms of background knowledge I'm pretty much a perfect fit for this (I've even published an article about a pseudo-Byzantine coin), but I'm no graphic designer and this is too big a project to take on in the hopes that I might quickly turn into one.

This isn't suuuch a heinous set of requirements. I can think of several Reading MATD grads who could do this. It would also be a good fit for John Hudson and Gerry Leonidas.

Easy peasy!

Not - easy peasy.

But in the realm of the do-able for some.

The tone of the request is what strikes me as unfortunate. Maybe I am reading too much into it but it feels like the folk you would be dealing with are somehow... quite uneasy.

Also this is the realm of programmers: "We may also ask for an easy-to-use keyboard utility, either separately or in conjunction with a commonly available utility, for scholars to find needed characters easily." I know a few people who could build this but it doesn't seem like something to casually "toss in".

Seeing how it will be used though, I can see why they would hope for such a tool. The best thing to do is probably to use the InDesign glyph pallet.

It might be cool to be able to get access to variant forms by making an opentype script that allows you to typing the primary glyph and a number and have it substituted by the relevant purple variant.

Often RFPs are created from lists of wants rather than a list of needs produced from a formal needs assessment process.

Eben, you're both presuming a lot about the kind of workflows and publishing environments that might be used and overlooking the need for transcriptions to be searchable, which isn't possible if you're using keyboard input sequences to access glyph variants. Keyboards input characters; glyph variants are, well, glyph variants.

I don't think I share James' view that this is project is ‘easy peasy’. Or, rather, the drawing glyphs and sticking them in a font is the easy part, but that alone isn't going to resolve Dunbarton Oaks' publishing needs.

I was thinking that the extra purple glyphs would not be used for searchable text, actually that would probably become tedious for users in a hurry. But perhaps they could be used to make the link between the text and the image more apparent - eg The text says "Alpha" and then in parenthesis the kind of Alpha which is depicted could be shown as well. It might be besides the point to do such a thing because the real users of such documents are are savvy enough already.

When you talk about workflows in this case you mean that this font is likely to be used in MS Word - not In Design. Correct?

What do you think the main issues in meeting their needs are likely to be?

If I understand the RFP correctly, what they want is a simple way of dealing with a situation where glyph variants are semantically significant to their user community. From experience of working with very similar material, I'm sure they get text submitted to them using all sorts of random kludges - for instance, I worked for a while on a book that had a lot of Greek coins, where the author had typed Cyrillic che to represent the Hellenstic-Roman shape of Greek koppa, and the element symbol from set theory for lunate epsilon. It's time-consuming to fix a book-length manuscript full of this kind of thing, and it requires knowing more about epigraphy than most editors/typesetters do. So I think they want to create an integrated system that channels people into doing things the "right" (i.e., standards-compatible) way.

One large problem, as John says, is that even if all the pieces of what they want are individually doable, making the whole thing work well may not be. If the workflow involves an author typing her/his text in a word-processing program, and then a publisher moving this into a page layout program and/or into a website management system - all the while maintaining control over minor variations in glyph forms that are of stylistic or chronological relevance - well, how do you manage all the variables? Is the author using Mellel on Mac or OpenOffice on Linux? Or do you create a workflow that only allows Word 2010 on Windows? And my limited knowledge of TEI does not suggest to me any easy way to make the whole thing compatible with TEI XML while retaining all glyph-variant information: perhaps if I actually read the spec I'd find something, but it seems at first glance like it would require a fair bit of scripting to get glyph-level information out of one system and into another.

What they need is a team, not a type designer, and I really wonder if they budgeted in accordance with what they're asking for.

Eben, the point of having a transcription font like Athena Ruby is to transcribe text in a way that represents the physical appearance of that text. But at the same time, you want that text to be searchable based on the underlying character. Eleven different shapes of the letter Alpha are still all the letter Alpha. Ideally, you want users to be able to do two things: to search for text and to search for uses of particular variant letters, since both of these are of interest to students of numismatics and sigillography: what a text says and what a text looks like. A scholar wants to be able to say ‘Show me all the coins/seals that include the following word’ and also to say ‘Show me all the coins/seals that use this particular form of Alpha’.

And as I hope this makes clear, no, I am not talking about using MS Word. :)

___

David: ...as John says, is that even if all the pieces of what they want are individually doable, making the whole thing work well may not be.

I don't think I said that. I said that the actual type design bit was easy, but I don't think the rest of it is not doable. I do think that it will take some time to pulls all the pieces together, and fully expect the process to involve things like proposing new characters to the UTC, extending the TEI XML to include glyph variant information, mapping this information to OTL features via style sheets. All doable, but yes, more than just designing a typeface.

John: I don't think I said that.

Sorry about that - I think I conflated your comment about the font itself not resolving their needs with my own doubts about their having thought through the practical issues.

For an in-house system, it's just a technical problem, and I think that is all soluble. But when they say they want it to be designed for scholars to work with easily, I assume they mean to distribute it, not just use it in-house. And that is a larger problem.