Stubborn Dreams

“A Tale of Three .Orgs” (Talk given by Andy Baio at 6/04/2008 PDX Web Innovators meeting)

Bram Pitoyo’s writeup: http://linkenfuego.wordpress.com/2008/06/05/portland-creativetech-event-review-pcter-16/

It started with a script that Andy wrote. He published the result of the script which used a dictionary word list to find available domains on his website, and he registered the three .org’s he liked (Upcoming.org, Waxy.org, Meaty.org).

Waxy.org

A couple of months before Meetup.org launched, Andy started Waxy.org. He said that the process of starting a blog was the most important decision he ever made (and he didn’t know it at the time…).

In this first post on Waxy.org dated April 14, 2002, he set three rules for Waxy.org:

1. No journaling. He believes that you have to be a really good writer to make your personal experiences relevant to people who don’t know you. [I believe this is why most blogs of the personal nature aren’t successful]
2. No tired memes (at the time, there were memes going around like what kind of smurf were you – this is still prevalent today with the 50 questions)
3. The most important. Be original. Don’t write stuff that other people are writing. [When I look at waxy.org today, the fact that you can deep-dive into his archives and be entertained and fascinated for hours is proof Andy creates interesting, original content.]

The purpose of these three rules? Add value.

According to Andy, starting Waxy.org was of immeasurable value.

1. It raised Andy’s visibility beyond that of his personal social networks. Within one year after Waxy.org launched, Andy was cited in the NYTimes five different times by five different writers. Waxy.org became an influence on mainstream media
2. It was a platform for launching his future projects. The blog post on the upcoming launch of Upcoming.org: http://waxy.org/2003/09/upcomingorg_lau/
3. His blog connected him to like-minded people.
4. Through Waxy.org, Andy has pretty much been able to meet everyone he cares about and admires.

“So many people are not producing, they have it backwards. Don’t try to develop an audience. Blog what you love. Write what you care about.” – Andy talking about blogging

Meaty.org or the genesis of Upcoming.org

Meaty.org was Andy’s first (of the three) .org site. It was going to be a place to get like-minded people to meetup.

Andy had been meeting with a group of geeks in the LA area. He liked the whole thing of meeting people from virtual communities. He wanted to make software to enable that experience. He started working on Meaty.org on the side. He had a nightmare dealing with recurring dates. Then, Meetup.org launched (June 2002). When Meetup.org launched, he was beaten to market. He abandoned Meaty.org.

Lesson #1 of startups from Andy: “Finish it. Don’t take time off from it.”

Upcoming.org

Started January 2003. Following his first lesson, he did not stop until it was ready nine months later.

The Ebay/pez-like inspiration: He always loved live music but he was terrible remembering.

Friend [day after interesting band played]: “That was awesome”

Andy: “Why didn’t you tell me about it?”

Friendster was hot at the time. The social network du jour.

Andy’s novel idea: Treat the event like a blog post. Let people connect on it and build their friend network. Result: Events would be automatically known by friends, sharing of events.

The first version of Upcoming.org was pretty rough.

Side note: Andy is big on high-resolution, pixel-perfect mockups done in Photoshop. He cannot imagine a site from a rough wireframe.

[The wireframes he showed looked very good. Like what you would expect from a major design house. I believe Andy’s design sensibilities are way way above the average developer and probably was a big factor for upcoming.org’s success.]

Andy did the design of Upcoming.org’s logo himself. He commented that it looks great on old school baseball jersey tops (which they gave away/wore).

Upcoming in 2003: http://web.archive.org/web/20030919021857/http://upcoming.org/

2003. Rough site launched. Barely functional. He held a beta month before.

Point: It was good enough. Put it out. The feedback that came back so validated the September 2003 launch.

[Side note: question from audience about the “watching feature”. Andy said people were confused from the beginning (His intent: this event looks cool, will let friends know, not actually attending]

Then the project stopped completely in June 2004. He killed the project development. His son was born.

He said pick two out of three:

1. Day job
2. Side project
3. Baby.
   

As a result, all work stopped for eight months, but Upcoming.org kept on growing slowly.

There were more people using it, but the use was not really exploding.

The Infoworld Article fires momentum

Link: http://weblog.infoworld.com/udell/2005/03/21.html

John Udell wrote an article about upcoming.org. He loved it but said it was lacking, suggested an open API.

After Udell article, Andy publicly committed in his Upcoming blog post. He promised all stuff out within a week. Andy’s blog post announcing his commitment: http://upcoming.yahoo.com/news/archives/2005/03/21/jon_udel/

Andy decided to bring on two friends, Gordon Luk and Leonard Lin, as partners to help get the stuff done. The partners split the commitment-based work. One crazy week later, he and his two new partners launched with an API, tagging, email, SMS.

Andy’s blog post one week later: http://upcoming.yahoo.com/news/archives/2005/03/28/huge_cha/

“Evite was the old (Web 1.0), Upcoming was the new (Web 2.0)” – Tim O’Reilly Web 2.0 white paper

White paper: http://www.oreillynet.com/pub/a/oreilly/tim/news/2005/09/30/what-is-web-20.html

People were building applications on the API.

Upcoming was a micro startup. But Upcoming’s influence was disproportionate to its size. Heavy SV influence. Very active in SF. It had Robert Scoble writing about it all the time.

Yahoo acquisition process

In July of 2005, Andy got an email from Caterina Fake, co-founder of Flickr. Flickr had just been acquired by Yahoo. Caterina had been tasked with bringing the cool back to Yahoo.

The Yahoo Acquisition via Caterina Fake was an outgrowth of connections Andy had made through Waxy.org. Andy originally started connecting with Caterina via Metafilter and various blogs

“Friends doing great work/able to bring/pull up others with them” – Andy’s musing

So, Andy and his two partners were brought in to meet with Yahoo Local.

The acquisition process

Patentable IP, just a website (easy for the acquisition lawyers on the Yahoo side)

Side note: patentable IP – qualifies for taxable long-term capital gains!

Back of envelope calculations – Andy decided, partners agreed (cannot discuss split)

Negotiations – they doubled money they were asking for.

Did they have a lawyer? Of course! Same lawyer that del.icio.us used later.

Three meetings later, acquired October 2005.

At the Web 2.0 conference at which Yahoo announced its acquisition of Upcoming.org, Andy was actually already there on a press-pass (via his Waxy.org writer credentials)

Becoming part of Yahoo

Entrepreneur naivety: You sell site, then do something else

Reality: You sell site, get a new title, maintain site

Good parts:

1. Brilliant people hidden in the nooks ‘n cannies of the rank-and-file. Even rank-and-file were great/smart (del.icio.us, mybloglog, Yahoo pipes)
2. Full-time focus – so wonderful/great to go full-time on upcoming
3. Platform technology
4. Able to replace sketchy roll-your-own geo-locating technology (e.g. ‘same event listed in multiple SF locations’) with “WhereOnEarth”. The same geoplatform that powers the Flickr geo stuff

At the time, Yahoo’s standard platform = PHP. Del.icio.us written in Perl -> lots of changes

Andy wrote upcoming in PHP, easier. He admitted he was not the best programmer in world – PHP code he wrote not the best. But it did the job.

Since the Yahoo acquisition, every lick of code rewritten.

The Not-so-great parts

1. 3-person company -> 14,000 person company (bureaucracy)
2. Technical integration with other parts of Yahoo: very hard, time-consuming,
3. Drew away from things/features? (for organic growth). The time-consuming integration with Yahoo News/search (was not external traffic, was sort of artificial traffic – coming from normal Yahoo visitors)
4. Complexity – tremendous amount, once acquired by Yahoo they had to switch to their APIs/platform

However, everyone benefits from the acquisition. Upcoming.org became more stable, more powerful than ever would have been just with Andy’s limited resources. At the time of the acquisition (2005), he had just doubled servers. To 2 leased servers.

How Upcoming works behind the scenes at Yahoo:

1. Yahoo – integration with search and local – Upcoming became the event infrastructure for all of Yahoo
2. Yahoo! page inline widgets – Upcoming powers the event listings
3. Yahoo Auto – auto events, powered by upcoming

Upcoming had good APIs (benefited integration). Occasionally, had to expose private APIs but in general, API-level integration.

Not every event had to be added by Yahoo user -> event feed auto-pulled stuff in

Upcoming metro – replaced by Yahoo GeoSearch (no more double-booking, one booking in wrong metro area)

Upcoming was never intended to be a business. Wasn’t even a company, never even incorporated (LLC). Yahoo M&A guys, doing due diligence, it was easy for them. 3 guys, leased servers, website (IP)

Lesson #2 of startups – “Build for yourself. Build something you love. Build with smallest team possible. Bootstrap yourself.”

Andy didn’t leave his day job at financial firm until acquired (link to Dimensional Fund Advisors).

No overlap with firm – conflict/of/interest. Andy worked nights and mornings on upcoming.

Yes, of course, occasionally he had to tweak the site while at work.

On the other hand, much bigger problem if work at an Internet/technology firm. Yahoo: One of Andy’s partners wanted to start Pizza shop – couldn’t do it – real/potential Yahoo food+local conflicts

In fact, Pixar anecdote: Pixar one of best offices to work in (Andy’s opinion – he got to tour the SV landscape) they have right of first refusal – you have to tell them what you’re working on. They invariably say no – go ahead as soon as they do

Advice: if any conflict, don’t do it. Try not to do something employer would want.

Andy’s big BUT (even though acquired) “not to do it to seek acquisition exit”

Lesson #3 of startups – “Do it well. Find an audience. At the very least, people will like you (provide something of value)”

Post-Yahoo

Post-Yahoo, post-integration, when contract up last Nobember, Andy was offered to work on some really interesting things (can’t talk about because of NDA) at Yahoo.

Feeling “working on something you built for someone else” <- not the same

Andy left – walked away with more flexibility + options

After leaving Yahoo, Andy moved to Portland – able to buy bigger home (was in Palo Alto)

Andy loves Portland’s DIY culture. It is different from SV: Classic Silicon Valley mindset like the 6th & Sunset strip in Hollywood in the 80s: “We’re going to get signed with a major label” mind-in-clouds/ mentality

Why are you building it <- Silicon Valley mindset/commercial goals

In the time since his son was born, the whole waxy.org site has been neglected (time to devote focus on waxy again)

This year (2008), Andy went to GDC. He loves the Indy Game movement. It was a mind-expanding experience (GDC)

He believes Gamedev lags web 2.0.

His Startup idea??? Something about games and dropped a clue: Providing a way for creators to make money doing what they love

Andy has been meeting with Rael Dornfest (former CTO of O’Reilly Technology). They have “Bottle Cap labs”: every week pick a project/launch it

Question and Answers

When was the tipping point?

It happened early on. There were some bizzare uses of upcoming.org

• College town – spike – only 2 or 3 individuals to get anonymous college town on upcoming. Just a few dedicated people
• China – Americans in China following soccer, using upcoming for gatherings

The core users shine through. To do this day, enough people using site. PDX WI event top event in Portland. Yahoo! brass annoyed populist event like Britney Spears concert #2 billing to some tech gathering. However, what other way can measure activity vs implict/explicit activity

Andy wanted website to become more popular/more events get advertised. Very large audience who likes to lurk – 95-98% of Yahoo just going to look. While at Yahoo, he tried to get Upcoming to become more Yelp like (reviews, consumed by Lurkers).

Inappropriate events

“Class of frustrating behavior” -> flat out abuse

Nigerian spammers discovering upcoming (using private msging feature)

Fought back with heruristics to do analysis of incoming private messages/event listings

Great hack : Send all private site messages through spam assasin

Noman’s land – other banned users – only other banned users can see what banned users do

The Yahoo acquisition, transitioning the users

Comparing how Blogger did it with how Flickr did it:

• Blogger: Big redesign, new features, just use new Google account)
• Flickr (no new features, sent emails month in advance)

Upcoming (like Blogger approach, old school people free t-shirts)

“The deck”

Hates most forms of web advertising. But loves “The deck”. Different sites, no CPM love it. network of high-profile sites