I love participatory media, collective knowledge systems, user-generated content and the like, and spent much of my life and career participating in them and making them. As I say in this post from 2005, the internet is built on a culture of generosity -- the first web page I built was when I noticed there was no page on Nabokov and realized I could just make one. Amazing! And it dawned on me that every other page on the web -- this was 1994 -- had come about for the same reason. Then the dotcom thing happened. And then Web 2.0 brought us back to the web's roots -- communication and contribution. That is why I love participatory media and must defend it.

There are so many things wrong with Jaron Lanier's recent manifesto in the Wall Street Journal (excerpted from his book) that I hardly know where to begin. The self-proclaimed "father of VR" believes that people who don't get credit or compensation for their work are lesser, humiliated beings without dignity -- the work in question being such activities as saving a bookmarks to delicious, correcting spelling errors on Wikipedia, exposing one's listening history on Last.fm, and the like. As the "father of VR" and a musician who has had various other occupations he has a particular lens through which he is viewing our latter-day participative media. He seems not to have built participative web sites, hailing from a Mondo 2000-era view of the world. For a non-participant all these new types of media would naturally appear to be what he calls a great "global mush". To discerning users of social media, you see what you deliberately select. The point is to filter out the noise, the mush. Obvious, no?

Systems such as Wikipedia, Flickr, Delicious, Facebook, Twitter, Hunch and various parts of the open source movement are based around small contributory systems, bodies of work in which there are incremental improvements by multiple contributors, or exposing small actions that would be insignificant in isolation, but are meaningful in the aggregate. These types of software and platforms are specifically designed for conversation and contribution. That is the point. There is no final product such as a book, movie, song or album. This method of creation would be pretty poor for designing a space shuttle or an ad campaign or writing a biography. There is no final product to which the epithet "design by committee" might apply. He is misconstruing goals.

He also appears to believe that quality is a zero sum game. A bunch of amateur musicians singing in someone's living room take nothing away from Lady Gaga.There's a lot of tilting at windmills in this excerpt. I've never heard anyone assert, as he appears to think everyone in the digital arena is constantly asserting, that "collectives make the best stuff" -- quite the opposite. Everyone agrees that 99% of everything is crap, and no one is claiming Wikipedia's entries are better written than those of Charles Lamb or Edmund Gosse in the 1911 Encyclopedia Britannica (my favorite). But really, who cares? By sharing my (admittedly crappy) snapshots on Flickr, I'm not claiming to be Margaret Bourke-White. And my sister *likes* to look at photos of my dog. Who am I hurting? Should I charge a penny to look at my photo? Do I need a photo credit? No. If someone other than my sister admires my cute dog, they are welcome to do so for free.

Additionally Lanier does not understand that people do things for reasons other than bolstering their egos and making money. You shouldn't need a motivation or justification to correct spelling or factual errors on Wikipedia -- a certain desire for orderliness, good grammar, or truth should be sufficient. Those who enjoy correcting spelling and grammatical errors online -- I do -- are they thereby "robbed of dignity" as Lanier would have it? Of course not.

From David Weinberger's post, The opposite of "open" is "theirs":

The biggest myth about starting a tech company in NYC is that it's difficult to hire top tier programmers. At Hunch, based in New York, we've hired a team of incredible MIT engineers and are constantly interviewing for new talent for the "pipeline". Being on the board of Etsy, I've also help with hiring engineering talent.

The reality is it's hard to hire top tier engineers no matter if you're in CA, NY or Vancouver, where Flickr was based. Good engineers come from all parts of the world -- Cal Henderson was living outside London when he started working on Flickr and moved to Vancouver. These engineers are often recruited to move to a startup hub. Those are primarily CA and NY, with some other cities having their own smaller scenes (Boston, Austin, Tel Aviv, Stockholm, Pittsburgh...)CA has a much larger tech scene and much better weather. NYC is a more exciting, diverse city with a strong tech scene. And there's plenty of back and forth -- John Allspaw just moved out to New York from SF to work on Etsy, joining Chad.

But what NYC is actually missing is not engineers. In NYC you can find lots of great engineers, visual designers, and great publishers and contributors to social media. But in CA I seem to find far more people with multiple skills - engineers who blog and dabble in design, designers who can do great UI but also great UX, etc. These multidisciplinary people are the ones who hack together brilliant new stuff, can innovate across the board, see various avenues of attack, and are indispensable at startups. It is these hybrid people that we are always looking for at Hunch and for whatever reason find them much more often in CA than NYC.

I think the problem in NYC is primarily cultural -- a lot of MIT talent is hardcore eng. and the designers often come out of print (publishing and advertising are huge out here) NY web devs are more than capable of adapting to CA-style, multidisciplinary approach. At Hunch, we want everyone to embrace the social and technical tools that they are using and building -- blogging, tweeting, contributing to open source. This is what I think what NYC needs more of.

When I hear about startups that have made some amazing technical innovation in NLP (natural language processing), OCR (optical character recognition), image recognition, speech recognition, I am somewhat skeptical. The government, especially DARPA, many university research departments, Xerox PARC (and Interval Research, where I worked briefly in 2000) have sunk a lot of time and money into this kind of research, in some cases billions of dollars, and a lot of this research not only available but free to all comers. Improving on them is a vast and difficult task, and even a 5% improvement on the existing state-of-the-art is invisible to the end user - the innovation has to be an order of magnitude better to 'feel' better.

There were dozens of startups that approached us at Flickr with tech that could do fantastic things with image recognition software. But we were pretty good at getting people to tag photos with what were in them. The image recognition software could tell us if a photo was a photo of Mom, but not if Mom was smiling, or looking puzzled, or dancing on her 50th anniversary. People adding tags, descriptions, titles and comments turned out to be better sources of metadata for those photos, and that metadata could then be used in interesting ways.

Some easier, faster, less expensive and I think, more satisfying ways for startups to innovate using these technologies are:

1. Find a new way to use the tech that serves a pressing consumer need (this is hard)
2. Find a new source of data (the original PageRank algorithm's use of links as its primary data source)
3. Create a new source of data and apply the tech to it (in the example above, Flickr using "social engineering" to create a vast amount of human-added metadata)

To have credible tech, it's a bit of a Goldilocks problem. If you build something too easy, it's not defensible, and you'll be easily copied. Too hard, and it's a job that requires government research or academia. Startups whose strength is technical should try to something of middling technical difficulty. Some startups, such as Twitter or Daily Booth, can be successful without a lot of tech. Nothing wrong with that! But I see startups all the time making tech claims that I find dubious.

Previously, I wrote about how working on the right thing is more important than working hard, and a lot of people asked how you know you're working on the right thing. And the only way to gauge that is instinct, gut feeling, or the tingle in your spine that Nabokov says he feels when he encounters a great work of literature. But there are methods of arriving at the right thing, and generally that requires exposing yourself to a lot of ideas, fleshing out a few, ruminating on them, and throwing almost all of them out.

Yesterday, Rob sent me links to Ira Glass's advice on Storytelling, which are yes, about Storytelling, but also applicable to any creative endeavor, whether that be producing a movie, writing a book, or creating a piece of software. You should watch all 4 videos in the series, but the second one talks about the importance of *finding the right story* -- the equivalent of an entrepreneur finding the right idea.

Repeated "failure" and throwing things away -- even throwing away good but not great things -- can't be overemphasized, and are often forgotten when people are talking about grand successes. This, from Steve Jobs:

In the comments on Chris's blog (great post today about search and social graphs) -- Sean O'Rourke mapped out some interesting points regarding Social Graphs when used for Search, and I think he's nailed the numbers:

My friend Todd Lappin sent the following about his new program with the Market Street Railway:

Which vampire franchise is for me? - make thousands more decisions on Hunch.com

I am reading The Collected Stories of Lydia Davis and came across one of my favorites, which I had forgotten, Happiest Moment:

Monsieur Proust is the account of Proust's final decade, as told by Celeste Albaret, his servant and confidante. He lived his final days, as is well known, in a cork-lined room with the curtains drawn, awake mostly at night, writing his masterpiece. Albaret kept to his strange hours, fixed all his meals, took his clothes to be laundered, and sat and listened to his accounts of his childhood, his social life and the many people he had known. After Proust's death, Albaret lived for 50 years in obscurity, refusing any inquiries for interviews from journalists and biographers. She finally agreed to narrate this account in her 80s, just to correct the many myths and inaccuracies that had grown up around Proust, to tell his story as best she could. It's quite an engaging read, and I was deeply impressed by Albaret's devotion to and love for Proust. Halfway through the book, I flipped to the back cover, and was a bit taken aback by the review by Angus Wilson printed there:

"Suicidal" "monstrous" "cold-blood" "revulsion" "deforming" "madman" -- wow. I was reading it as a study in devotion, service and sacrifice. I tried to figure out where such revulsion had come from, and it seemed to me that people in Western society have a horror of selflessness, and what they perceive of as subordination and subservience. In some ways this book was a perfect and remarkable complement and contrast to the expendable warrior theme of my prior post on Dogfights and Gameness with their triumphs and heroics -- here was a quiet, modest life, lived in the service of another. She took pleasure in warming Proust's bathwater to the perfect temperature, fixing his coffee just so, and knowing exactly which hat he wore on which occasion.