Really great tips in these links, and heaps of other useful resources hanging off them. Thanks for posting.
I found Sarah Waters' and Joss Whedon's lists the most helpful (copied below), and John Steinbeck's sixth rule for writing dialogue is solid gold:
6. If you are using dialogue—say it aloud as you write it. Only then will it have the sound of speech.
Sarah Waters
1. Read like mad. But try to do it analytically – which can be hard, because the better and more compelling a novel is, the less conscious you will be of its devices. It's worth trying to figure those devices out, however: they might come in useful in your own work. I find watching films also instructive. Nearly every modern Hollywood blockbuster is hopelessly long and baggy. Trying to visualise the much better films they would have been with a few radical cuts is a great exercise in the art of story-telling. Which leads me on to . . .
2. Cut like crazy. Less is more. I've ¬often read manuscripts – including my own – where I've got to the beginning of, say, chapter two and have thought: "This is where the novel should actually start." A huge amount of information about character and backstory can be conveyed through small detail. The emotional attachment you feel to a scene or a chapter will fade as you move on to other stories. Be business-like about it. In fact . . .
3. Treat writing as a job. Be disciplined. Lots of writers get a bit OCD-ish about this. Graham Greene famously wrote 500 words a day. Jean Plaidy managed 5,000 before lunch, then spent the afternoon answering fan mail. My minimum is 1,000 words a day – which is sometimes easy to achieve, and is sometimes, frankly, like shitting a brick, but I will make myself stay at my desk until I've got there, because I know that by doing that I am inching the book forward. Those 1,000 words might well be rubbish – they often are. But then, it is always easier to return to rubbish words at a later date and make them better.
4. Writing fiction is not "self-¬expression" or "therapy". Novels are for readers, and writing them means the crafty, patient, selfless construction of effects. I think of my novels as being something like fairground rides: my job is to strap the reader into their car at the start of chapter one, then trundle and whizz them through scenes and surprises, on a carefully planned route, and at a finely engineered pace.
5. Respect your characters, even the ¬minor ones. In art, as in life, everyone is the hero of their own particular story; it is worth thinking about what your minor characters' stories are, even though they may intersect only slightly with your protagonist's. At the same time . . .
6. Don't overcrowd the narrative. Characters should be individualised, but functional – like figures in a painting. Think of Hieronymus Bosch's Christ Mocked, in which a patiently suffering Jesus is closely surrounded by four threatening men. Each of the characters is unique, and yet each represents a type; and collectively they form a narrative that is all the more powerful for being so tightly and so economically constructed. On a similar theme . . .
7. Don't overwrite. Avoid the redundant phrases, the distracting adjectives, the unnecessary adverbs. Beginners, especially, seem to think that writing fiction needs a special kind of flowery prose, completely unlike any sort of language one might encounter in day-to-day life. This is a misapprehension about how the effects of fiction are produced, and can be dispelled by obeying Rule 1. To read some of the work of Colm Tóibín or Cormac McCarthy, for example, is to discover how a deliberately limited vocabulary can produce an astonishing emotional punch.
8. Pace is crucial. Fine writing isn't enough. Writing students can be great at producing a single page of well-crafted prose; what they sometimes lack is the ability to take the reader on a journey, with all the changes of terrain, speed and mood that a long journey involves. Again, I find that looking at films can help. Most novels will want to move close, linger, move back, move on, in pretty cinematic ways.
9. Don't panic. Midway through writing a novel, I have regularly experienced moments of bowel-curdling terror, as I contemplate the drivel on the screen before me and see beyond it, in quick succession, the derisive reviews, the friends' embarrassment, the failing career, the dwindling income, the repossessed house, the divorce . . . Working doggedly on through crises like these, however, has always got me there in the end. Leaving the desk for a while can help. Talking the problem through can help me recall what I was trying to achieve before I got stuck. Going for a long walk almost always gets me thinking about my manuscript in a slightly new way. And if all else fails, there's prayer. St Francis de Sales, the patron saint of writers, has often helped me out in a crisis. If you want to spread your net more widely, you could try appealing to Calliope, the muse of epic poetry, too.
10. Talent trumps all. If you're a ¬really great writer, none of these rules need apply. If James Baldwin had felt the need to whip up the pace a bit, he could never have achieved the extended lyrical intensity of Giovanni's Room. Without "overwritten" prose, we would have none of the linguistic exuberance of a Dickens or an Angela Carter. If everyone was economical with their characters, there would be no Wolf Hall . . . For the rest of us, however, rules remain important. And, ¬crucially, only by understanding what they're for and how they work can you begin to experiment with breaking them.
Joss Whedon
1. FINISH IT
Actually finishing it is what I’m gonna put in as step one. You may laugh at this, but it’s true. I have so many friends who have written two-thirds of a screenplay, and then re-written it for about three years. Finishing a screenplay is first of all truly difficult, and secondly really liberating. Even if it’s not perfect, even if you know you’re gonna have to go back into it, type to the end. You have to have a little closure.
2. STRUCTURE
Structure means knowing where you’re going; making sure you don’t meander about. Some great films have been made by meandering people, like Terrence Malick and Robert Altman, but it’s not as well done today and I don’t recommend it. I’m a structure nut. I actually make charts. Where are the jokes? The thrills? The romance? Who knows what, and when? You need these things to happen at the right times, and that’s what you build your structure around: the way you want your audience to feel. Charts, graphs, coloured pens, anything that means you don’t go in blind is useful.
3. HAVE SOMETHING TO SAY
This really should be number one. Even if you’re writing a Die Hard rip-off, have something to say about Die Hard rip-offs. The number of movies that are not about what they purport to be about is staggering. It’s rare, especially in genres, to find a movie with an idea and not just, ‘This’ll lead to many fine set-pieces’. The Island evolves into a car-chase movie, and the moments of joy are when they have clone moments and you say, ‘What does it feel like to be those guys?’
4. EVERYBODY HAS A REASON TO LIVE
Everybody has a perspective. Everybody in your scene, including the thug flanking your bad guy, has a reason. They have their own voice, their own identity, their own history. If anyone speaks in such a way that they’re just setting up the next person’s lines, then you don’t get dialogue: you get soundbites. Not everybody has to be funny; not everybody has to be cute; not everybody has to be delightful, and not everybody has to speak, but if you don’t know who everybody is and why they’re there, why they’re feeling what they’re feeling and why they’re doing what they’re doing, then you’re in trouble.
5. CUT WHAT YOU LOVE
Here’s one trick that I learned early on. If something isn’t working, if you have a story that you’ve built and it’s blocked and you can’t figure it out, take your favourite scene, or your very best idea or set-piece, and cut it. It’s brutal, but sometimes inevitable. That thing may find its way back in, but cutting it is usually an enormously freeing exercise.
6. LISTEN
When I’ve been hired as a script doctor, it’s usually because someone else can’t get it through to the next level. It’s true that writers are replaced when executives don’t know what else to do, and that’s terrible, but the fact of the matter is that for most of the screenplays I’ve worked on, I’ve been needed, whether or not I’ve been allowed to do anything good. Often someone’s just got locked, they’ve ossified, they’re so stuck in their heads that they can’t see the people around them. It’s very important to know when to stick to your guns, but it’s also very important to listen to absolutely everybody. The stupidest person in the room might have the best idea.
7. TRACK THE AUDIENCE MOOD
You have one goal: to connect with your audience. Therefore, you must track what your audience is feeling at all times. One of the biggest problems I face when watching other people’s movies is I’ll say, ‘This part confuses me’, or whatever, and they’ll say, ‘What I’m intending to say is this’, and they’ll go on about their intentions. None of this has anything to do with my experience as an audience member. Think in terms of what audiences think. They go to the theatre, and they either notice that their butts are numb, or they don’t. If you’re doing your job right, they don’t. People think of studio test screenings as terrible, and that’s because a lot of studios are pretty stupid about it. They panic and re-shoot, or they go, ‘Gee, Brazil can’t have an unhappy ending,’ and that’s the horror story. But it can make a lot of sense.
8. WRITE LIKE A MOVIE
Write the movie as much as you can. If something is lush and extensive, you can describe it glowingly; if something isn’t that important, just get past it tersely. Let the read feel like the movie; it does a lot of the work for you, for the director, and for the executives who go, ‘What will this be like when we put it on its feet?’
9. DON’T LISTEN
Having given the advice about listening, I have to give the opposite advice, because ultimately the best work comes when somebody’s fucked the system; done the unexpected and let their own personal voice into the machine that is moviemaking. Choose your battles. You wouldn’t get Paul Thomas Anderson, or Wes Anderson, or any of these guys if all moviemaking was completely cookie-cutter. But the process drives you in that direction; it’s a homogenising process, and you have to fight that a bit. There was a point while we were making Firefly when I asked the network not to pick it up: they’d started talking about a different show.
10. DON’T SELL OUT
The first penny I ever earned, I saved. Then I made sure that I never had to take a job just because I needed to. I still needed jobs of course, but I was able to take ones that I loved. When I say that includes Waterworld, people scratch their heads, but it’s a wonderful idea for a movie. Anything can be good. Even Last Action Hero could’ve been good. There’s an idea somewhere in almost any movie: if you can find something that you love, then you can do it. If you can’t, it doesn’t matter how skilful you are: that’s called whoring.
My latest story is a racy little piece about what happens when someone cute from work invites you over to watch Netflix and Chill. 
