Though I didn't read much, I remember reading the UF Core sections a long time ago and followed up a little on the newer installments. I'd like to know how that's going and what new works you've put in as a result (and in addition, which ones you've set aside like with Buffy).

The major reason why I got into your fic is because I've been working on similar projects of my own. If at its highest potential, it could even rival yours. It's mostly in the form of video games, but I'm looking forward to writing a few fanfics, starting from the chronological beginning for my main character.

I wanted to ask some tips about how you made the massive background that is Undocumented Features. You mentioned how you despised the first few installments and would've scrapped them if it was for the positive reception.I wanted to ask for some tips in relation.
- What did you find wrong with your first few works? How did you learn and fix these problems in the later installments?
- What other tricks did you learn to create better crossover writing material?
- What other suggestions do you have?

Since we're here, I'd like give some details of said project. I call it Megacosmos (last initial in my alias) and it is named after a massive circle of parallel universes that serves as the setting. Each universe contains a fictional universe and behaves like an organic cell. Currently, it has 4 million Earth years of history and has divided into ten political bodies. Though there was no theme to begin with, it now revolves around the idea of our interaction with fiction. The main character is Jamey Jason (also part of my name), who represents the typical otaku fan, and one of the deuteragonists is his sister Tammy Jason, who wants to live normal and has little clue about the worlds her brother knows.

Well, I'll await your reply and if you want, I can tell more about my series. I also have many original projects worth sharing.

Gosh. That's... not really answerable as asked, I mean, it's a big dang project, and it's evolved fairly organically as opposed to following a predetermined master plan. I doubt we on the EPU team even know, comprehensively, what-all is in here any more. Often, when rereading older pieces, I come across things that I know are references, and I know I wrote them, but I can't for the life of me remember what they're actually references to.

As for what's left out, that's also a much wider topic than can really be covered in any depth on the Forum, though I suppose in that case a distinction can at least be made between stuff that's not in here because one or more of us don't know anything about it (which is... well, a lot of stuff, I mean, I'm sure I'm unaware of most of what's going on in pop culture howadays, I'm an old hermit) and stuff we're deliberately leaving out.

You mentioned Buffy the Vampire Slayer, which is one example of the latter. Another is Firefly, and for the same basic reason, that being that Bad Robot has lawyers and a pretty-well-documented track record of using them, and we don't want that. There's a handful of stuff we steer clear of because we either don't like it or feel it's been done to death (Ranma 1/2 is both of those). But, again, there's no master list we keep around for reference. It's just a thing that happens.

>You mentioned how you despised the
>first few installments and would've scrapped them if it was for the
>positive reception.

Well, sort of. I mean, I didn't despise them at the time; only in retrospect, after enough time had passed that my sensibilities had changed and I started to perceived things as clunky, or mishandled, or just not very elegant. I think that happens to every writer who lives long enough. I'm sure there are passages in Neuromancer that make William Gibson wince now.

(On a cinematic level, you see this mechanism at work every time George Lucas releases a new even-specialer edition of Star Wars. Which also makes a fairly strong case for not obeying the impulse to make repairs; after a certain time, these things have lives of their own. If I tried to "fix" the original Undocumented Features that way, the result would probably be woeful.)

>- What did you find wrong with your first few works? How did you learn
>and fix these problems in the later installments?

It's mostly just a matter of polish and experience, I think. It's like any kind of craft output, I suppose—the first sweater someone who takes up knitting attempts isn't likely to look as nice as one that same knitter makes after being at it for 20-some years. In some cases, I'm not very happy with the technical execution; in others, looking back on them, I feel like I hadn't really found my voice yet. Which is not to say today's output is perfect, by any stretch! Fallible humanity and all that. The downside, if you want to call it that, of experience is the opportunity it presents for the development of bad habits. This is why you occasionally see really amazing, creative, daring first novels—they come from writers who didn't know they can't write like that. :)

>- What other tricks did you learn to create better crossover writing
>material?

I always have trouble answering questions like this, for the same reason that I had trouble with the ones above: I don't really work in a way that's... I don't know, considered enough for there to be identifiable steps in it. The crossover stuff in UF largely just happens, because it's the way my head works. Stuff goes into this mental database I have that's constantly trying to relate them, even though I know they're not actually related, and sometimes surprisingly non-accidental-looking convergences come out. There's no trick to it that I can relate. I know this is a disappointing answer, I've been told it is often enough, but I can't help that. This is how it really works for me.

>- What other suggestions do you have?

Well, since I don't really work to a set of conscious guidelines, I don't have much in the way of process tips to offer, so all I can really suggest is to do with mindset. This stuff can get discouraging, especially when you're deliberately setting out on a project that has the kind of scale you're describing here. I think the thing that saved UF from that sort of problem at the start is the fact that it wasn't intended to be this great sprawling thing, it developed into one of its own accord, layer by layer, like limestone. Now, of course, it's big enough that stepping back for a wide shot causes a sort of creative vertigo. In a case like yours, where you've got a master plan in place when you start out, just the size of the thing could be intimidating.

At times like that, I've found that the only really useful offset is investment in the characters. When the universe itself is too large and complicated for my poor old brain to be able to cope with, there are always the people in it to spur me along, because I'm invested in them and I don't want to let them down. (Upon reflection, this is probably why most of the more recent "imports" to UF started with the people and then brought along bits of their original settings and whatnot in with them as creative space allowed.)

Also, there's a variation on the same mental trick that can be used to make a very long drive bearable, which is to break it down into smaller pieces. I'm not writing a gigantic, interconnected, multi-time-period-spanning metafictional saga! I'm just trying to get poor Corwin out of this mess he's gotten into with the two submarines and the Earthforce Navy. :)

Anyway, I hope this helps, even though it's a bit of a muddle and a lot of shrugging. I'm never any good at answering process questions, because I've come to believe that "process" in a creative sense is such an intensely personal thing that knowing how someone else's works isn't really helpful, apart from satisfying curiosity.

--G.
-><-
Benjamin D. Hutchins, Co-Founder, Editor-in-Chief, & Forum Mod
Eyrie Productions, Unlimited http://www.eyrie-productions.com/
zgryphon at that email service Google has
Ceterum censeo Carthaginem esse delendam.

I'm going to try and answer your questions as if you had asked them about software development, because that's my field and I think there are a lot of similarities. By harnessing the power of analogy and metaphor, you might get something out of it.

- I'd like to know how that's going and what new works you've put in as a result (and in addition, which ones you've set aside like with Buffy).

When choosing new libraries to work with, you need to consider a few things. Is it a good, consistent piece of software? Does it have at least one useful part that you really want/need to use, or are you just thinking about it because it's new and shiny? Is it going to be a easy enough to use or work into your project? Often, you'll end up choosing just based on how fun it looks like working in -- which isn't a bad metric, since you'll be more productive if you're happy. But beware of adding complexity that doesn't solve the problem at hand.

Sometimes you have to exclude the software outright because of licensing concerns. Like if a code has a GPL license and that's incompatible with your existing license -- Stallman is well known for hiring lawyers to enforce copyright.

- What did you find wrong with your first few works? How did you learn and fix these problems in the later installments?

To be honest, everyone thinks that their first program sucks. As you keep learning, you find your standards change. Myself from six months ago was *such* an idiot. Like Rumsfeld said, there are "unknown unknowns", and you can't fix things that you don't know are broken yet. Still, the best way to learn is to keep writing more code.

In the best case scenario, programmers like to "build one to throw away" -- a basic system that gets you started, from which you find out the real problems and build a much better version. Real life seldom works this way. I'm still working on code that my CEO wrote in 6 months, 10 years ago, to get clients served ASAP. It has rough edges, but it works pretty good, so we keep adding to it.

If I use an analogy from megacrossover fiction, Gryphon originally intended UF as a throwaway story before the real writing began. And then the whole thing took on a life of its own, despite some decisions that, with retrospect, were probably not great for the project. Still, it's a testament to how good that base actually was that CORE still anchors much of the story.

But back to software! Fixing mistakes is something that get very difficult if other parts of the program depend on the badly designed part. Often, you just try to route around the damage by avoiding calling the bad section, or encapsulating the code in something better. You can probably replace it completely (retcon) if you have tests (sorry not applicable to fiction), but this takes a lot of work.

- What other tricks did you learn to create better crossover writing material?

Read a lot. Read manuals on how to write good code, and read examples of code you want to write like. Make sure what you're writing is well documented, so that you don't forget why you did something.

It's easy for things to sprawl all over the place, so make sure that you don't ramble all over the place (GOTO considered harmful, etc.) The hardest part of writing software is choosing the correct abstractions. Human attention is limited, so try to keep a few things in focus at any one time.

- What other suggestions do you have?

Use a version control software like git to keep track of your changes, so you don't lose anything you want later. I do this for my creative writing too, though a deletions file can work in a pinch. Even save work-in-progress.

Never implement your own crypto(-lesbian subtext).

- Have you ever thought of giving your works original motifs and themes as well to your works?

You will always have your own writing style, and that will show in the code. But in general, choose the approach that will best let you solve the problem (i.e. use themes that will best help you tell your story). Choose between object-oriented, functional programming, simple scripting, or whatever based on the task at hand -- don't try to force your code into something that it's not. Unless you're going for a really cool hack; beware that hacks don't always work as well as tried-and-true methods but sometimes work much better.

I hope this helps you. Sometimes its useful to look at a problem from a different point of view.
-- ∇×V