On filmmaking and software development: Part 1

For many years now, I’ve been intrigued by some of the more unusual similarities between filmmaking and software development; so much so that in 1997 I wrote a paper called Fast Iteration with Learnt Memory (or FILM), looking at how filmmaking techniques could be used in software development.

Things have moved on since then, both in filmmaking as well as in software development. If anything, the similarities betweern the two have become more pronounced over the years, catalysed and accentuated by agile techniques. So I thought I’d revisit the topic afresh.

Wikipedia defines five basic stages of filmmaking:

  • Development
  • Pre-production
  • Production
  • Post-production
  • Distribution

Here’s what Wikipedia has to say on the subject of development as a phase in filmmaking:

This is the stage where an idea is fleshed out into a viable script. The producer of the movie will find a story, which may be from books, other films, true stories, original ideas, etc. Once the theme, or underlying message, has been identified, a synopsis will be prepared. This is followed by a step outline, which breaks the story down into one-paragraph scenes, concentrating on the dramatic structure. Next, a treatment is prepared. This is a 25 to 30 page description of the story, its mood and characters, with little dialog and stage direction, often containing drawings to help visualize the key points.The screenplay is then written over a period of perhaps six months, and will be rewritten several times to improve the dramatization, clarity, structure, characters, dialog, and overall style. However, producers often skip the previous steps and develop submitted screenplays which are assessed through a process called script coverage. A film distributor should be contacted at an early stage to assess the likely market and hence financial success of the film. Hollywood distributors will adopt a hard-headed business approach and consider factors such as: the film genre, the target audience, the historical success of similar films, the actors who might appear in the film and the potential directors of the film. All these factors imply a certain attaction of the film to a possible audience and hence the number of “bums on seats” during the theatrical release. Films rarely make a profit from the theatrical release alone, therefore DVD sales and worldwide distribution rights need to be taken into account.

The movie pitch is then prepared and presented to potential film financiers. If the pitch is successful and the movie is given the “green light” then financial backing is offered, typically from a major film studio, film council or independent investors. A deal is negotiated and contracts are signed.

What does all this have to do with software?

Software is about stories.

As software becomes commoditised, what matters is the customer experience. And what the customer experiences is a story. As with films, stories are made up of scenes. Each scene has a one-paragraph description with clear directions to all participants in the scene.

Stories are about visualisation.

How does the customer experience the story? By visualising it. By seeing it unfold. By feeling part of it. The better the visualisation, the more engrossed the customer becomes. So it is with films, so it is with software. Customers respond to stories.

Visualisation improves with iteration.

That’s why the beta culture works. People know what they want and what they like when they see it. They also know what they don’t like, what doesn’t work for them, and they can point it out when they see it.

What gets iterated is the script for the story.

The centrepiece of a film is not the set. It’s not the special effects. It’s not even the actors, except insofar as they bring the story to life. It’s not the director either. It’s the script. That’s what makes the story. So it is with software. Get the script right and you will wow the customers.

Only successful scripts should get financed.

That’s what angels want first and foremost. Good scripts. Because they know that they can get good producers, good directors, good actors, good everything. But good scripts are harder to come by.
BTW, have you ever wondered why angel investors exist in only two investment genres, filmmaking and software development? Now you know.

There’s a lot more I will be covering in future posts over this weekend, including, in no particular order:

  • The central role of the script
  • The role of the producer
  • The role of the director
  • The cast
  • The set
  • Stars
  • Reuse and stock shots
  • Vendor independence
  • Iterating through scenes
  • Calling a “wrap”
  • Getting certified
  • Localisation
  • Subtitling
  • IPR and DRM issues
  • Revenue windows (e.g. from cinema to rental to purchase to free-to-air)

Let me leave you with a taster. Here’s the wikipedia definition of development hell as in the filmmaking context:

Development hell is media-industry jargon for a film, television screenplay or computer game (or sometimes just a concept or idea) getting stuck in development and never going into production.

In the case of a film or television screenplay, the screenwriter may have successfully sold a screenplay to a certain set of producers or studio executives, but then the executives in charge change, and these new people raise objections to all the scripts and casting decisions they oversee, mandating rewrites and recasting. As a director and actors become “attached” to the project, further rewrites and recasting may be done in order to accommodate the needs of the new talents involved in the project. Should the project fail to meet their needs, they might leave the project or simply refuse to complete it, causing further rewrites and recasting. Worse still is when a finished project (for example, a television pilot) is sent back for rewrites and recasting, which can often force a project to begin again from scratch. This process can last for months or years, and a project trapped in this state will more often than not be abandoned by all interested parties or cancelled outright. This process is not naturally an element of filmmaking. Many times, this “Hell” occurs simply due to the lack of foresight and competing visions of those parties involved. This revolving door in the film industry happens most commonly with projects that, to some, may have multiple interpretations and affect several points of view.

Sound familiar? And so to bed.

Community participation rule-of-thumb: a follow-up post

There were quite a few interesting comments on a recent post of mine, where I’d outlined a community participation rule-of-thumb I’d been using for a while, which I repeat here for the sake of convenience:

  • For every 1000 people who join a community:
  • 920 are lurkers, passive observers
  • 60 are watchers, active observers capable and willing to kibitz
  • 15 are activists, actually doing something
  • …and 5 are hyperactive, passionate about what they’re doing, almost to a point of obsession

David Tebbutt wanted to know whether this was about public communities or business ones (by which I’ve taken him to mean within-enterprise ones), and bemoaned the implications for enterprises if it held true within-enterprise. David Churbuck, who’d been working on a similar issue serendipitously, felt it reinforced what he’d learnt at Reel-Time since 1995; John  Dodds was in similar vein, pointing me at a post he’d written on the same subject, quoting Jakob Neilsen on participation inequality and suggesting a 90:9:1 split from lurker to passionate. Mark Cahill, who also has experience of Reel-Time, warned about something I’d apparently missed, while endorsing the core: he felt that 2 out of a thousand were monumental jerks who need to be disciplined “pour encourager les autres”.

It’s actually quite a lot to take in, at least for me, and so I’ll keep my response brief. For now.
I think there are two dimensions to community that we must take into account if we are to understand community dynamics in this kind of analysis.

The first is time. People do not stay static in the community roles I’ve suggested; over the right period of time, yesterday’s lurkers may become tomorrow’s hyperactives; there is a gentle churn in roles as people leave and others join, and individual motivations change.
The second is topic. As communities grow, they tend to stay together on some overarching principle or shared ideal, yet clearly disaggregate into a number of subcommunities where the lurker-active-hyperactive roles shift again. One man’s meat is another man’s poison and all that jazz. In fact, as a community scales, I think its very existence depends on its ability to evolve these overlapping narrow-interest subcommunities. Chris Locke used to call it “organic gardening” whenever we spoke about this during the early Cluetrain days; Amy Jo Kim made clear reference to such behaviour in her seminal book  Community Building On The Web.
And I think, above and beyond all this, there is a Long Tail effect as well, where lurkers have their Warholian fifteen-minutes-of-intense-activity because a particular debate or issue really got them. These are healthy things. Not to be derided or pooh-poohed. No lurkers no community.
I am reminded of what Milton ended On His Blindness with:

They also serve who only stand and wait.

As far as Mark Cahill’s total-jerk problem is concerned, I’m not quite sure. I can see participation being blocked where there is gender, race or similar abuse, or even strong language, something that offends communal sensitivities.

But the extreme freerider is not one I’ve felt I could block. It is part of the price that can be paid by social software, and the only real way I know of curbing such behaviour is by the use of peer pressure.

Those are my early thoughts. I’m very grateful for the comments I’ve received, good food for thought.

Enterprise Nil, as in Stenhousemuir Nil: a passing comment on microformats

I had a strange postcolonial upbringing.

Imagine this: Calcutta, 1965. The Dalhousie Athletic Club, in the middle of the wonderful green lung that is the Maidan. (In fact I think the Dalhousie Athletic Club’s maidan branch was colloquially referred to as the Maidan Club, but I could be wrong).

I’ve just had my first tennis lesson, and, luxuriating over a Coke with ice and lime cordial, I watch this tranquil very-British scene unfolding in front of me. Gentlemen and ladies bowling in whites. They have their hats on, but the sun hasn’t. Occasional sounds of leather on willow in the background. A bar that would do any rugby club proud, shored up by drinkers who would also do any rugby club proud. Regaling all in their regalia. And, almost surreally in the midst of all this, there’s someone wandering around collecting pools coupons. Yes, pools coupons, while Jackie Charlton (the before-combover version) scores for Leeds United in a Pathe-like newsreel in the background.

And a few days later, true to type, I saw everyone crane and stretch to hear the evening’s results and check their coupons.

My fascination with soccer results probably began around then, aided and abetted by the apocryphal story of that wonderful team Stenhousemuir Nil, and by the delivery of James Alexander Gordon. [Note to people who don’t follow UK soccer. Stenhousemuir is a venerable and much-loved Scottish soccer team, affectionately called Stenhousemuir Nil, apparently because of the regularity with which that particular phrase got used during Final Score, often as announced by James Alexander Gordon].

Which brings me, somewhat circuitously, to the point of this post. I wanted to write about Enterprise 2.0, but realised that it now has so many connotations, anchors and frames that it was about as meaningful as Enterprise Nil. Whatever label gets you going. It’s only a label.

Chris Messina makes some interesting observations in a post about IE 8 and microformats. I quote from him:

With discussions around support for microformats in Firefox 3, and Apple showing strong support for microformats as well, it’s only a very short amount of time before we can move on from the “so who’s using microformats?” question to “okay, so now what can we do with them?”

And there’s the rub. Now, with microformats, all this Web 2.0 (or Web Nil, if you prefer) stuff is really getting into the enterprise. People may have wondered about social software; they may have wondered about mashups and videos and music. But with microformats, those arguments will fade away. Even troglodytes will begin to see the value, hidden as it is amongst amongst the WMD.
I’ve tried (and failed) to get much traction for microformats in the past; success was rare other than in the simple identity space, which is not a bad place to start, along with calendars and contact details. But there’s so much that can be done with low volatility “reference” information in pretty much any industry, covering product and inventory codes, location information, customer information, currency and value information, you name it.
So thanks, Chris, for reminding me.

The time has come. Enterprise, be ready. Microformats are go. Watch this space.

And for once a standard has emerged that hasn’t already been sacrificed at the altar of vendor whoosis. For once it has been about market adoption. Which is why it will work.

None of Us is As Smart as All of Us

I had the pleasure of meeting Tom Malone shortly after he’d published The Future Of Work, at Supernova 2004 (thanks to JD Lasica for the coverage), where we were both speaking. [If you’re interested in what a 21st century firm will really look like, you must read Tom’s book, especially in conjunction with that of John Roberts, The Modern Firm.]

Last Friday, Tom, along with a few others, kicked off a large-scale experiment. Called We Are Smarter Than Me, it brings together MIT, Wharton, Pearson and Shared Insights, seeking to confront and confound a particular paradox. In their own words:

  • A few books have recently been written on this topic, but they all fail to confront one central paradox. While they extol the power of communities, they were each written by only one person. We’re putting this paradox to the test by inviting hundreds of thousands of authors to contribute to this “network book” using today’s technologies.

We’ve had experiments where people have collaborated on books online before (Dan Gillmor’s We The Media comes to mind) but I’ve never seen anything of this scale. I understand that over 1.5 million invitations have been sent out; no I did not get one of them :-) but that did not deter me from signing up. Check it out for yourself, barriers to participation are remarkably low. Which is a good thing.
Currently I’m just lurking, taking a look at what’s happening, as the structure and content of the book morphs around me lazily. Insane, like a smoke-ring day when the wind blows. Soon I will figure out where and how I want to participate. Follow the links to We Are Smarter if you want to know more.
I’m also intrigued by something else about the cultures forming in this giant Petri dish. I’ve always believed in a simple rule-of-thumb about opensource communities:

  • For every 1000 people who join a community:
  • 920 are lurkers, passive observers
  • 60 are watchers, active observers capable and willing to kibitz
  • 15 are activists, actually doing something
  • …and 5 are hyperactive, passionate about what they’re doing, almost to a point of obsession

I shall watch with interest as the numbers grow, to see how close my rule of thumb is to reality. Any views out there, or shall I start a prediction market on this?

One final thing. One of the trends the book is meant to pick out is that of microfinance. I wonder. I just wonder. Take the network book and attach it to network funding, make it available on-demand whenever, wherever, however. I wonder.

Time for The Wren Gap

For those who remember where this conversation started, I used the term Four Pillars to describe search, publishing, conversation and fulfilment, and asserted that it was only a matter of time before enterprise software consisted of these four “pillars” and not much else. If you weren’t around at the start, reading this and this might help you. I’ve probably written a hundred posts on the subject as well, dealing with different facets and aspects of Four Pillars. [And, once I’ve finished cleaning up the categories and tags, you might even be able to find them!].

But in the meantime. I was doing my usual waltzing over the blogosphere, and got to this post by Dave Winer via Doc Searls. [Ever since I got myself a decent RSS aggregator, netvibes, I seem to do this often. Read person A and move from there to a story by person B before I’ve got to person B’s blog. Does that happen to you as well?]

Anyway, there is much I have to thank Dave and Doc for. There’s a crucial four-pillars point being made in what they say. And for now I want to call it The Wren Gap.

Let me explain. There’s a Guildhall in Windsor, the town where I live. This is what it looks like:

800px-Windsorguildhall.jpg

There were guildhalls in Windsor before this one. And there will probably be guildhalls in Windsor after this one, it’s a Guildhally kind of place. This particular Guildhall was built in the late 17th century and, following the death of the original architect, the work was continued by Sir Chriptopher Wren.  And at this stage let me pass you on to Wikipedia:

  • The story is widely told that the borough Council demanded that Wren should insert additional columns within the covered area, in order to support the weight of the heavy building above; Wren, however, was adamant that these were not necessary. Eventually the council insisted and, in due course, the extra supporting columns were built, but Wren made them slightly short, so that they do not quite touch the ceiling, hence proving his claim that they are not necessary! Whatever the truth of this story, it is undeniable that, to this day, there is a small gap between the columns and the ceiling.

And so the Wren Gap was born, as shown below:

wren.thumbnail.jpg

And so to the point of this post. Dave Winer, in the post that Doc referred me to, has this to say:

  • Many years ago, when the Internet was still the domain of geeks, researchers and college students, the smart folks often said that the opportunities for new software companies were over, it simply required too much scale to compete in an industry dominated by Lotus, Microsoft and Ashton-Tate. Now it’s clear how ridiculous that was, even though it was correct. The next layer comes on not by building on the old layer (a trick, the guy you’re building on will eat your lunch), or re-doing what they did (what the naysayers correctly say you can’t do), but by starting from a different place and building something new, and so different that the old guys don’t understand it and don’t feel threatened by it.

My italics. My emphases.

The councillors of today want us to connect to the old layer. We need to understand this. They want us to connect to the old layer. We don’t need to, but they want us to.

The way they want us to connect to the old layer is often via that terrible space variously called DRM and IPR and licences; these badly-thought-out things form the cement and mortar they so badly need.

We need to be wise, wise like Wren. And create Wren gaps everywhere. Opensource software, open SDKs and frameworks, open communities, open licensing, all this is here and now. And we can use them to create the Wren gaps.

People, you will be told that your edifices will not stand up unless you connect to the old layers.

Balderdash and piffle. Time for Wren gaps.

A coda. I loved this story from Cory at BoingBoing, but via Smart Mobs: another example of my reading feed A before feed B.

Tagging things as DefectiveByDesign, using the tag tools provided by the retailer. Oh frabjous day.

[Note: Portions of the Wren story have appeared in an earlier post, but I felt it was appropriate to repeat it and expand upon it.]