Starting Up: Nobody Go Near A Bus

In Cute-Fight, Startups on 4 September 2012

We’re working feverishly on Cute-Fight right now, so I’m thinking about startups. This is one of those thoughts.

There’s a lot to say about how hard startups are. They require an enormous investment of time and energy. And even when you go in prepared, there are still moments when you say, DAYAM this is a metric fuckton of work. (Cussing helps relieve stress. Science said so.)

So why do it? Here’s one reason.

There are four of us working on Cute-Fight right now. If any one us were hit by a bus, we’d be fucked. (Again, cussing helps.) We are all doing several jobs at once, and every one of them is absolutely critical.

Devin is responsible for everything on the backend. He’s programming, administering the servers, and doing tech support. James is the designer, but he’s also writing serious frontend code. Chris is drawing as fast as he can, as well as thinking through how those illustrations work on web pages. I’m kind of the conductor and the cleanup crew. I’m art directing, designing the game mechanics, talking to investors and mentors, writing the site text, and responding to feedback from our members. (Side thought: All CEOs should have to respond to support email. It’s impossible to maintain any illusions in the face of an inbox full of people with the same feedback.)

If any one of us was hit by a bus right now, the whole thing would fall apart. This is not the way “real” businesses work, but that’s also why people like us start companies. When is that last time you felt completely irreplaceable at your job? Like the company’s life or death depended on you doing your very best work?

The insanity of starting up is crushing, but so is the boredom and monotony of “regular” work. If you’re someone who feels replaceable and extraneous at your job, try a startup. It’ll make you crazy, but you’ll never feel unimportant.

Starting Up: The Love

In Cute-Fight, Startups on 1 September 2012

We’re working feverishly on Cute-Fight right now, so I’m thinking about startups. This is one of those thoughts.

If you go to business school, they teach you about indicators and predictive analysis. You’re taught to identify a market opportunity and exploit it. Your personal affinity toward the particular market doesn’t factor into the equation.

I did not go to business school. I learned everything I know about business from making websites, working at startups, and experimentation for kicks. I learned the most watching startups fail. I’ve had a pretty good education.

I once half-started a company. I identified a market opportunity, built a business plan, bought some domains, and started building. And then one morning I woke up and realized that, if everything went perfectly, I’d spend every day doing something I hated. The market opportunity was awesome – I just wasn’t the guy to do it. I had no love for it.

One way of looking at entrepreneurship is this: What will you look forward to doing every morning? You should start a company around that.

Because, truth be told, startups are hard. Like, really hard. So if you don’t have The Love for what you’re working on, you’re going to fail. Or, put another way: most startups fail, so you might as well spend the time working on something you enjoy, just in case you succeed.

If Cute-Fight is successful, I’ll wake up every morning to look at adorable pets. Not bad, as dayjobs go.

Starting Up: Skate to Where the Puck Is

In Cute-Fight, Startups on 30 August 2012

We’re working feverishly on Cute-Fight right now, so I’m thinking about startups. This is one of those thoughts.

There’s an old quote from Wayne Gretzky: “Skate where the puck’s going, not where it’s been.” This has been widely adopted in startup circles to describe thinking ahead. The idea being, skating to where the puck’s going is good, and skating to where it’s been is bad.

The problem with this conventional wisdom is that Wayne Gretzky never launched a website. People are more difficult to predict than a hockey puck, especially in groups online. I’ve been a part of many startups since 1995 that have tried to skate to where the puck was going, only to wind up in the right place too soon, dead before any puck arrived; or wind up in the wrong place entirely, wondering where everybody went. Like I said, startups are all about timing.

When it comes to community-based websites, I prefer to find places where people already congregate and build a playground there. People like personal stories? Fray. People tend to complain online? Kvetch. People show off their best photos? JPG Magazine.

And those are just the ones I started. Blogger* may have created blogging, but it was just a tool to do something people were already doing. Blogger just made it easier and more fun. Same goes for Twitter, which is just a much easier, much more fun way to blog. People forget that there were things called “facebooks” before Facebook – they were just on paper.

So Cute-Fight came from this observation: People with pets like to share photos of them. Everybody likes to see photos of cute pets. Let’s build a playground around that. We’re still in private alpha, but so far the feedback has been amazing. (I can’t wait to show it to you. Sign up if you haven’t already!)

The best part about skating to where the puck is that, when it works, your community will tell you where they’re going. You don’t have to guess. You’re building something together.

* I was Blogger’s Creative Director in 1999.

Introducing Cute-Fight

In Cute-Fight, Fun, Startups on 29 August 2012

I’m on the phone with my dad, telling him that I’m winding down all my paying clients in order to work on a website idea. I’ve convinced some friends to join me and we may even have a little angel investment.

“What’s the site?” he says.

I tell him that it’s a game where people create profiles for their pets, challenge each other, and the community will vote to determine who wins. The site is called … Cute-Fight!

There is a long pause.

Finally my dad says, “People do that?”

Introducing Cute-Fight, an online game you play with your real life pets.

If you have a pet, you can create a fighter profile and challenge other pets to Cute-Fights. If you don’t, you can vote and cheer for your favorites. The goal is to create a game that’s as fun to lose as it is to win, as fun to watch as it is to play.

Cute-Fight is designed to be surprising, adorable, and most of all fun. It’s full of amusing random delights. The web feels so damn serious these days. We just want to make people smile.

It’s also an experiment in game mechanics. Game playing is an essential human trait, but “gaming” in community spaces has come to mean something negative. We want to make a place where playing the game results in a positive experience for everyone.

The site is in private alpha right now. (You can sign up here.) It’s being built by a dream team: Devin Hayes on the backend, James Goode on the frontend, and Chris Bishop on the drawing tablet. We have a little bit of investment and are looking for more.

Cute-Fight is the result of years of community experience by all involved. People love to share photos of their pets, so we built a playground around that. It’s silly, but that’s the point. It’s fun, and we’re having fun building it. And if it succeeds, our reward will be waking up every morning to look at photos of cute pets.

Do people do that? We’ll see.

Startups Are All About Timing

In Cute-Fight, Startups on 13 August 2012

When people talk about startups, they mostly talk about the idea. Somebody thought of something new, started something up, and then … boom. They’re on the cover of Wired.

But the truth is, startups are really all about timing. Lots of people have lots of ideas every day. Ideas aren’t the hard part, timing is. Good timing won’t guarantee success, of course, but you can’t succeed without it.

And not just market timing. In my experience, it’s the personal timing that makes all the difference.

Personal Timing

When I started Fray in 1996, it was not the first website about true stories. But the timing was right for me. I was young, underemployed, and had something to prove. I was ready to work for it, so it worked for me.

When Heather and I started JPG Magazine in 2004, the timing was right for the market (inexpensive DSLRs were new, people had high-res photos to share), but more importantly the timing was right for us. We were newly married and wanted to do a project together. Plus, we’d both established ourselves in the photoblogging community and wanted someplace to feature our talented friends.

But timing cuts both ways. When I started Pixish with a small team in 2008, the market wasn’t ready (crowdsourcing is still controversial today). But more importantly, I had just come out of a horrible startup experience, so I wasn’t up for another firestorm. And the team had other jobs they had to attend to, so when we got a pile of negative feedback, we just couldn’t deal. I still think we could have righted the ship, but none of us were in a place to make that happen. It was easier to just learn some lessons and move on.

Confluence

So after all these experiences and more, I’ve become very attuned to the whispers of timing. That’s why I noticed, earlier this year, when three things happened at once.

1. Devin Hayes needed a job. Devin and I worked together back in the JPG days, and I remembered him as the rare coder who’s talented, fast, and mellow. One of my greatest regrets was not being able to work with him longer.
2. James Goode, who I’ve worked with on everything from MagCloud to Pixish to Fray, had some time available. James is my favorite designer because he’s talented as hell, but also flexible and friendly. In my experience, being able to talk about design is just as important as putting pixels on a screen.
3. I was ready. Four years after my last startup blew up, I was finally ready to try again, and hopefully not make the same mistakes again (I’m looking forward to all-new mistakes).

The three of us with open time and ready to work? Good timing.

The List

Anyone who makes web stuff has a list. “Wouldn’t it be cool to have a website for XYZ?” I’ve been keeping a list like this since 1995. So I dusted it off and put each idea through its paces. What was the potential audience? What could we bring to each idea that felt like something new? And most importantly, which idea would excite all three of us?

I have a garden site in me, but the other guys aren’t into plants, and many established gardeners are still not online. Not time yet.

I really want to start a television community site. The time is soon – the industry is about to go through an enormous change. But soon is not now. Not time yet.

The Idea

One idea kept coming to the top. It has a huge potential userbase, it’s built to take advantage of our hyper-connected realtime world, it’ll be a good arena to test out some of my community/game design theories, and best of all, it’s fun. After years of trying to reinvent things (blogging, publishing, search), maybe it’s time to just do something that’ll make people smile.

I brought the idea to the guys and they were in, each adding their own spin to it. Then, more good timing. I went out for drinks with a friend who offered us an unsolicited bit of angel investment. The investment allowed me to pay the guys (paychecks make it real, even when they’re far smaller than they should be). My friend Chris Bishop, who served as the Illustration Editor for Fray and is a crazy talented artist, joined us to define the illustration style of the site.

Cue the montage. Me working in San Francisco, James in Sydney, Devin from a nomadic road trip, and Chris in DC. We collaborated in email, Basecamp, and chatrooms. Building building building. For months.

Present Tense

Which brings us to now. We’ve honed the idea, drafted a huge list of features we want, and then scaled that way back to a minimum viable product. (It’s amazing how complicated simple ideas can get when you start actually building them.)

We’re now this close to opening up the site to a limited first round of testers. If you’d like to be one, you’ll need to have a real live pet in your house, you’ll need to be in possession of a computer, a camera, and a sense of humor, and you’ll need to go to the site and sign up. We’ll let you know when we’re ready.

The site is called Cute-Fight. And what’s Cute-Fight? That’ll be the subject of another post soon. I can’t wait to tell you all about it. Hopefully, the timing will be right for you.

On Seeding Communities

In Community on 2 August 2012

I wrote this in response to this question on Branch but I’m having technical difficulties posting it there, so…

Greetings from the Lake of Bays in Ontario, Canada, where I’m on a family vacation. I’m taking a moment out to write this and will post it when the (very intermittent) internet connection appears. Life is slower here on the lake, so forgive me if I ramble.

I’d like you to hold two slightly contradictory thoughts in your head at once for a few minutes. They are: 1. One of the best things you can do for a new community is seed it with good stuff. And: 2. One of the worst things you can do to your community is lie.

People make their first impressions quickly and hold them forever. This is no surprise and there’s all kinds of social science to back it up. What this means for a virtual community space is that the first “content” the user sees will form the user’s definition of the place forevermore. This first-viewed content will do more to drive future behavior than any interface decision, any set of rules. Starting off with excellent example content is the single biggest factor in predicting the quality of future contributions (at least, that you can control).

As the same time, my number one rule for community building is: Do Not Lie. The internet is very good at ferreting out liars. Community building is all about trust, and once you lose trust, it’s gone forever. Do not lie to your community. Ever.

So the question becomes: Is fabricating members to seed a community space lying? Can you lie to a community when there’s no community there yet? And how bad is that lie compare to the good of having example content?

I find that when I’ve been asked questions like these, the person is usually just seeking permission to lie. As in: It’s okay to lie in the beginning, right? To which I respond with a question of my own: Would you like to find out one of your relationships was based on a lie? Communities are sets of real relationships, and building a relationship on a foundation of lies is never a good technique.

So my opinion is, yes, you have to seed content, and no, you should not make up fictitious members to do it.

Instead, curate your early members. Invite people to join. Tell them exactly why they’re being invited, what you’re trying to build, and how important their participation is to you. Do not invite Robert Scoble – invite your mom, your non-tech friends, people you know have something to contribute. Go to where they congregate already and participate. If you’re not a member of this community, hire someone who is.

Then edit their contributions. Revise. Perfect. There is absolutely nothing wrong with working directly with your first members (no lying involved). My background is in journalism, so I’ve always been comfortable telling people what we need and editing their work when it comes in. If you’re not comfortable doing this, hire someone who is.

The benefit of involving real, not-made-up people as your first users, in addition to not having to lie, is also that they’ll give you real feedback. You’ll find out quickly where people are getting lost, or misunderstanding the tools.

The “ghost town” problem is a real concern – a site with many social tools that are all unused becomes a barrier to participation because no one wants to go first. That’s why you start with your friends, relatives, and anyone else you trust. They’ll go first because you asked nicely (and it’s nice to be invited).

The other way to avoid the ghost town problem is to start with an extremely limited set of community tools. Any good startup will have a long list of social tools they’ll want to enable eventually, but start with one. Just one. It’s okay if it makes the site look thin. Start with one and let people start to use it. Then add in others one by one. By starting with a shallow toolset, you’ll avoid the ghost town problem and help early users know what to do.

When Heather and I started JPG Magazine, we put all of this into practice. We participated in the photography community ourselves, so when we needed our first users, we asked our friends and photobloggers we looked up to. The first version of the site had no public tools – it all happened privately. Slowly, we made more and more public, until finally we had a thriving site with lots of community tools. The core thing we did right, I think, was really being participants in the community before starting the site at all, so its growth was seen as something we were all doing together, instead of some kind of interloper.

My nephew informs me that working while on vacation is “wrong” so I’m afraid I must go now. Hope this helped.

Someplace Else

In Stories on 23 July 2012 tagged Albany Bulb, Peace

A few weeks ago on a sunny Saturday, Heather and were walking around the Albany Bulb. It’s a man-made blob of “land” created from nearly a century of dumping trash in the bay. The trash stopped in the ’80s, and now it sits in between San Francisco and Berkeley, some of the most expensive land in America.

It’s a strange place. Totally unnatural, since only water is supposed to be there, but now it’s set free and wild. Palm trees sprout from in between graffiti-covered concrete. Animals I’ve never seen in the city scurry past. It feels like you’re in the middle of nowhere, but the city is always just over your shoulder.

It’s removed enough that there’s a kind of anarchical community there. Not exactly homeless, since they’ve built small camps, but not exactly legal, either. (Can you even own land that’s not technically land? Not the point, I guess.)

We were just walking around, exploring the place. As we passed a fairly established encampment made of wood scraps and plastic tarps, we realized there was a someone emerging from it. He came out, a man in his 50s, with a deep tan and a wary expression.

There was a moment of awkwardness. Were we invading his space? Was he going to yell at us? Maybe he was wondering if we were going to give him shit for being there? We all looked at each other with apprehension. And then:

“Peace,” he said.

“Peace,” we said back.

And that was it. Such a simple moment. Such a simple word. There was no need for fighting. Fighting was for the rest of the world. This was someplace else.

What If Social Networks Just Aren’t Profitable?

In What If on 12 July 2012 tagged Business, Facebook, Metafilter, Social Network, Tumblr, Twitter

Here’s the short version: Every community-based site in the history of the web has essentially been a stab at creating a social network. Most of them fail as businesses, with the rare exception of small, lucky communities that become self-sufficient but not exactly prosperous. What if that’s just the way it is?

Here’s the longer version. Let’s start with some seemingly unrelated bullet points. I was dreaming when I wrote this, so forgive me if it goes astray.

• Before the web, I worked for alternative weekly newspapers. There was conventional wisdom even then that the business of running a weekly paper sucked. But we weren’t in it for the money, we were in it because it was important to the community the weekly served. We made enough from advertising to print the paper and deliver it to the readers. It was very rarely profitable. In the alt newsweekly world of the early 90s, breaking even was considered the success case. We did it anyway.
• This week it was announced that Digg, once valued in billions, had been sold for 500k. An inglorious end to a once beloved social media darling. Digg was attempting to scratch a particular community itch. It tried to make sharing newsy links social, and you could follow friends, which is the basic element of any social network. It worked, for a time, but it was never profitable.
• Last month Facebook, certainly the biggest player in the “let’s monetize a social network” game went public and their stock price took an immediate flop and has been bouncing around like a fish out of water ever since. The question on everyone’s mind: How will they make money from all those free members? Without souring the milk, of course.
• Twitter and Tumblr, both incredibly successful at cultivating their communities, both yet to prove how exactly they’re going to survive as businesses.
• Last week it was announced that the WELL, an online community that predates the web, was to be sold by its present owner, Salon (a business relationship thats’s always been a head-scratcher to me). The community is currently rallying to buy itself.
• When I wrote a book about community sites 11 years ago, I included many examples of sites doing it right. Almost all of them have died since. One that hasn’t: MetaFilter, a small community company supporting a small staff that makes money through advertising and membership costs.

Can you see a pattern here?

The flow, as I see it, works like this.

1. We want to be a social network. The more people in it, the more “value” it has, so we need everyone to join. Because we want everyone to join, we cannot put up a pay barrier, so we have to make money another way. Let’s say advertising. (Note: Most never make it this far.)
2. Our advertisers want as much data about, and contact with, our users as possible. We want to only allow limited engagement. Either advertiser interest wanes (Flickr), or we coast on our investment (Twitter, Tumblr), or we give in and let the advertisers run the show (pretty much everyone else).
3. Members become angry at us because we’re selling them out. The exodus begins. There’s always somewhere else to go (see Friendster, MySpace). Go back to step 1.

See it? The bigger you go, the harder the road. Meanwhile, small, focused, and yes, exclusionary community sites flourish. Matt Haughey made several key decisions in the formation of MetaFilter, but the most important one was to limit growth. Hell, for years you couldn’t get an account if you wanted one. After that, they started costing money. When it costs money at the door, that means you don’t have to sell out your members to advertisers. It also means the community stays small, which – surprise! – also leads to healthier communities.

What if we all realized that social networks are a societal good (at least as good as a local alt weekly) but not necessarily good businesses? We’re all desperately hoping that Twitter or Facebook or Tumblr will figure out the secret ingredient that turns a large-scale community of free members into a cash machine. What if we’re all just waiting for the impossible? Like a business that turns water into gold? We’ve got lots of water, we just need to figure out the gold part….

What if we eventually realize that, like the alt weeklies, these are things we do because they should be done, because it’s fun, to make our little community a better place … not because they’re going to be great businesses.

Because so far, when you look at the numbers, that’s just what they are: not great businesses.

The one truly great business born of the web is Google, and not their self-driving cars and the other nonsense that accounts for zero percent of their income. It’s putting small, self-serve ads beside their search results. You and I create those search results with our behavior online, but not directly on Google. And that line between where I’m using my voice (you’re soaking in it) and where it’s being monetized (*cough*) is enough of a separation that it doesn’t bother me. The problem happens when the content creation happens in the same place as the ad deployment. So, of course, that’s exactly what Google’s trying with Google+, to less than stellar results.

My point with this thought experiment is this: What if we designed a social network to be small, self-supporting, and independent from the outset? How would it look, work, and feel? I bet it would come out looking nothing like the ones we’ve got now, the ones still trying to turn water into gold.

+

UPDATES:

• Looks like Digg sold for more than reported. Seems like a complicated deal.
• I forgot to mention the original social network: the phone book. Not a sexy business.
• Maciej Ceglowski writes convincingly about the power of charging at the door. See also: Clive Thompson.

How To Spot A Liar

In Apple, How to, Journalism on 20 March 2012

Like most people, I’ve been lied to a few times in my life. I’ve even told a few (though, as my wife will attest, perhaps not as many as I should). No one likes being lied to.

That’s one of the reasons I was attracted to journalism and photography – they’re about telling the truth. It may be a personal, subjective truth, but it’s a truth nonetheless.

It’s also why I started a magazine all about true stories. We even held a series of storytelling performances not unlike Mike Daisey’s theater show.

Yes, this is about Mike Daisey.

But I’m not going to get into the subjective nature of truth, what’s appropriate in theater, or Mr. Daisey’s seemingly pathological need to nail himself to a cross. I just wanted to share this one observation.

Like anyone who’s been lied to, I’ve made an effort to become more aware of when the person talking to me may be a liar. This can get complicated. You can get caught up in lots of micro-expressions, language patterns, and telling gestures. But in my experience, there’s one sure-fire way to know when you’re dealing with a liar: they tell you.

In Mike Daisey’s original, now retracted, This American Life story, he says:

In context, he’s saying this to his interpreter, in an effort to convince her to help him pose as a businessman (an effort that, we now know, was also largely fabricated because his translator needed no convincing and planned to do this from the outset).

When I heard him say that in the original story, I was driving down Divisadero in my car, and I blurted out: “Then how can I trust anything you say?”

Here was a man telling an extraordinary story and admitting that he was a liar. If he lied to those businessmen, how could we really know he wasn’t lying to us?

There’s a reason that journalists are trained not to do this, and it’s not just highfalutin professional ethics. It’s far more practical: If you lie to get the story, it throws the entire story into doubt. Tell the audience you’re a liar and they stop believing you. Or, at least, they should.

If you traffic in true stories, you can’t lie to your audience, period. I don’t care if it’s in a book or on a stage or over the radio. When you tell a personal story, you absolutely know if the words coming out of your mouth are true or not.

Mike Daisey knew he was lying to us. I’ll never believe another word he says.

Listen to This American Life’s retraction here.

Flickr: Domini Domini Domini, You’re All Christians Now

In Christmas, Flickr on 22 December 2011

There’s an old Firesign Theater sketch called “Temporarily Humboldt County.” Some Native Americans are sitting around enjoying nature when the Spanish conquistadors show up with a priest. The conquistadors claim the land for Spain and Father Corona adds, “Oh! By the way, Domini Domini Domini, you’re all Catholics now.”

I was reminded of this sketch on Tuesday when Flickr decided I was a Christian.

Since Tuesday, if you visit any Flickr member’s photos with a modern browser, you’ll see three little snowflakes beside the Flickr logo. Click them and you’ll be treated to a cascade of snowflakes over the page and all its photos, as well as a row of blinking Christmas lights at the top of the page. For an added treat, you can roll over the lights with your mouse and they’ll pop, complete with sound effects. Click the little “[x]” beside the logo and it all goes away … at least until the next page load when the three little snowflakes show up again.

It’s a cute little diversion, created without Flash, which is pretty clever. Flickr engineer Scott Schiller, who I contacted about this but did not reply, recorded the audio himself. This is obviously a long-running project of his.

Flickr is the community website that’s closest to my heart. The site’s founders are friends of mine and my wife worked there for five years. But more important than that, it’s a community that I love. I’ve uploaded gigabytes of photos there. My photostream has become a virtual home for me. Our virtual homes are just as important to us as our brick and mortar ones, if not more. I’ve lived in my real house for a few years, but I’ve lived on Flickr since 2004.

So it’s distressing when someone puts Christmas lights on my virtual home. I’m not a Christian. I don’t care how secular the holiday is nowadays. I know about the holiday’s Pagan roots. None of that matters. The fact is, Christmas lights on a home are a signifier that the occupant is a Christian, the same way a mezuzah is a signifier of a Jewish occupant. These symbols have power, which is why we use them.

It’s not just that Flickr is smearing Christmas “cheer” all over itself. As a non-Christian in a Christian country, I’m grudgingly used to that. (Though it would be nice if clicking that “[x]” set a cookie that prevented it from loading on the next pageview.) It’s that my Flickr stream is my personal identity in the Flickr community. That’s my face there at the top. Flickr has added a Christian signifier to my virtual home and I have no way to remove it. In the eyes of the rest of the community, Flickr has turned me into a Christian.

Flickr has done other Christmassy things in the past. For a while, you could add a string to a URL to make it snow on the page. Other years, if you put a note on a photo with a special phrase (“ho ho ho hat”), a Santa hat would appear. But these were all secret easter eggs. (Easter! We can’t even talk about this without more Christian holidays coming up.) And in the case of the notes, I could easily remove them and control who has the power to leave notes on my photos. But this year’s festivities are unavoidable. Don’t like people seeing Christmas lights on your virtual home? Too bad.

When you begin a virtual community, you’re building for yourself. You can safely assume that most of the community is a lot like you. But as it grows, the community becomes more diverse. If you’re extremely lucky, some of your members will invest themselves so much, they’ll come to view the site as a kind of home. This, by the way, is the success case. It’s what you want to happen.

Flickr is now a truly global community. A huge set of their members don’t celebrate Christmas. Heck, it’s summer in half the world right now, so I’m not sure what they’ll make of the snowflakes. Flickr should know this better than anyone.

The decision to put Christmas lights on all of their members’ virtual homes shows a profound lack of understanding for who their users are and what those symbols mean. It’s the kind of decision you make when you assume the rest of the world is just like you, or you’re so enamored with a technology you forget to think through the social ramifications of its implementation. Making your members feel unwelcome in their own homes is the first step in the decline of a community.

The lights and snowflakes will go away after Christmas, but I’ll still be incredibly disappointed in one of my all-time favorite sites.

What Flickr Should Have Done

It’s undeniable that the snowflakes and Christmas lights thing is a cute technology demo. So what should they have done with it? Here are my top five suggestions.

1. Don’t. Not all your members celebrate Christmas.
2. If you must, limit it to pages with multiple voices, like the Flickr blog, search results, tags, and the homepage. That way you’re not accidentally converting individual members.
3. Really, don’t. People see their pages as their homes. Would you put Christmas lights on someone else’s house?
4. If you must, make it an easter egg. Trigger it with a search, like Google’s “do a barrel roll” or other hidden behavior. Let people discover it and pass it along on Twitter and Facebook. It’ll be seen as cool and special by those that find it, and it won’t annoy the others.
5. And for Christ’s sake, give those of us that don’t celebrate Christmas a way to turn it off and never see it again.

A version of this story also appeared in Gizmodo.