The new worldwide startup

I’ve been traveling around the world studying how startups get formed. Yesterday I visited Bootup Labs in Vancouver. Last week I was at Startup Riot in Atlanta and while at the Olympics I’ve been hanging out with Saeed Amidi. He was recently profiled in Business Week and is CEO of Plug and Play, an incubator in Silicon Valley (he invests in, and rents to, more than 280 startups in Silicon Valley and now owns the building where Google and PayPal, among others, started up). I’ve been at YCombinator and Techstars events recently. I’ve visited London, and Paris in the last year to meet startups there. In April I’ll be visiting Israel again to study startups there in more detail.

There are some common trends.

1. Most everyone outside of the valley complains that they can’t get access to enough capital.
2. Most everyone outside of the valley complains that they can’t get access to enough PR.
3. Most everyone outside of the valley complains that their best startups get dragged to Silicon Valley once they get big and need more talent (Flickr, for instance, moved from Vancouver to Silicon Valley when it sold out to Yahoo — they sold in part because they needed help dealing with scaling issues, I’ve heard that story over and over from other companies and communities, too, like Atlassian who moved to the valley from Australia).
4. Most everyone outside of the valley complains that they don’t have the business infrastructure that they need to succeed. Mark Zuckerberg told me he moved Facebook to the Valley to have access to mentors, lawyers, PR people, and other people a fast-growing tech company needs.
5. Most everyone outside of the valley complains about lack of entrepreneurial culture. In Europe, for instance, failure is stigmatized. In other places there just isn’t the kind of culture that values wacky weird ideas. Go to a local coffee shop in your neighborhood, for instance, and ask people what Foursquare is. I guarantee you that in most Silicon Valley coffee shops you’ll find someone. Not so in most other places in the world.

But you already know these problems, among others. So, what’s changing? A lot.

1. The infrastructure needed to start up a tech company is now decentralized. You can use cloud servers from Rackspace, where I work, or Amazon or other companies. That infrastructure didn’t exist five years ago and before then if you wanted to start a web company you would need to build your own data center. Not every community has datacenters, but today everyone has access to the same cloud hosting services.
2. PR is being decentralized. Thanks to blogs, Skype, YouTube, Twitter and Facebook you can get onto TechCrunch no matter where in the world you are.
3. Costs are coming way down. Associated with first point. No longer do you need hundreds of thousands of dollars in servers to start up, you just need a few hundred bucks on a credit card to buy cloud servers.
4. A ton of startup accelerators/incubators have formed in past few years. I’ve listed a few on this post. They provide money, offices, mentoring, and other services you need like legal help.
5. Tech talent is growing around the world. Silicon Valley used to have a lock on geeks. That no longer is true as many universities around the world have educated tons of computer scientists and engineers.
6. Tax advantages. In Vancouver government officials told me this week that they are seeing a widening corporate tax rate gap. They expect that in 2012 their rate will be 25% while USA’s rate will be 40%. Other countries, like Ireland, have even lower rates. Plus, tons of countries want to help form tech zones. In Vancouver Bootup Labs officials told me they are working on getting some R&D subsidies from the Canadian government, (I’ve heard similar things from other countries, which can help even more businesses startup around the world).
7. Lower costs of living. In San Francisco it’s expensive to buy housing and health care needs to be purchased at sometimes great costs to families. Not so in many other communities around the world.

Even Silicon Valley folks are seeing these trends and are looking to capitalize on them. Saeed told me he’s looking to build Plug-and-play facilities in many communities in the world. I’m seeing other incubators/accelerators like YCombinator and TechStars do the same, spreading outside of their original communities to get deal flow from startups around the world.

I expect over the next few years these trends to speed up. I’m looking forward to it.

Are you noticing the same?

Palm's small-screen bet doomed the Pre

Some of you might not know, but if I like a mobile phone I buy it. I have purchased several iPhones for my family and I own a Droid as well (I don’t recommend buying that one, instead I am telling my friends to get the Google Nexus One, which is a better device due to its speed).

Yesterday Palm announced that its smart phones are selling disappointingly poorly.

I’ve been doing a lot of thinking about why Palm didn’t get my money or get most of my friends excited. Yes, my friend Luke loves his Palm, but he just hasn’t been able to convince me.

Why not?

I believe Palm made a fundamental market miscall. They assumed that people would adopt a small phone with a decent experience and web browser.

They bet against the geeks. They bet against the web.

They bet wrong.

As I walk around Vancouver’s airport you can see why. A phone is no longer just a phone. People walk around holding their gadgets in front of them. Some, like Blackberry users, do email. Every Blackberry user I know wants a bigger screen.

But more and more I’m seeing iPhones and Android devices in airports. Most of the time these users are not on the phone, but are stabbing at the screen with their fingers doing various things.

Palm bet against these users by putting a small screen in their Palm Pre. What Palm didn’t realize is that users who actually go into stores and buy phones now need more than just a phone, they need a Web device.

By betting against the geeks they made a HUGE market misjudgment because the market follows the geeks. People get this wrong all the time.

It’s really a shame, too, because Palm has a very nice OS and a great stance toward developers.

But until they give me a device with a glorious huge screen with super high resolution they aren’t going to have a chance with the new users.

Compare to what Microsoft is showing off with its new Mobile 7 devices. Huge screen. Android devices? Huge screens. iPhones? Huge screen. Nokia N900? Huge screen.

These are the devices that are pushing the industry forward. It’s too bad that Palm’s CEO Jon Rubenstein made such a fundamental misjudgment. Why did he make that misjudgment? I think it’s because he probably did customer research and the research kept telling him that people wanted a great phone first.

See, customers lie about what they really want. Truth is, they don’t know what they want until you show it to them.

Remember what Henry Ford said? He said that if he asked people what they wanted they would have told him to build a better horse-drawn carriage. Well, we all know how that worked out.

Rubenstein shouldn’t have listened to the marketers. People want big screens with easy to use email and web. Palm didn’t deliver that and now it’s the loser.

Second Life's new player builds toward Web

Linden Labs has just released a new player for its virtual world, Second Life. This is important because it makes significant moves toward the Web and shows a new strategy: one of integrating into social networks (much of that shift has yet to come, Linden Lab’s CEO, Mark Kingdon told me in an interview embedded here):

I’ve linked to the new player, a FAQ, and started a discussion on Google Buzz, join us here.

In the interview Mark reveals quite a bit of other info about how Second Life is doing, including how many visits, minutes spent, and other info. It might shock you how well Second Life is doing, even though many people are reacting to my early tweets with messages like “I thought it was dead.”

Is Second Life about to enter its "second life?"

You probably have forgotten about Second Life (the virtual world from Linden Lab), right?

Remember, that’s that virtual world that got a TON of hype back in 2005/6. It was on the cover of magazines. On CNN and other TV shows. It looked like it was going to be THE new thing of the decade.

What happened?

Well, a few things.

1. Corporations figured out that they’d need to spend a lot of money to build an island in Second Life (Microsoft spent somewhere around $100,000 if I remember right back then) but soon they figured out that each island could only hold 100 people. Not a good ROI.

2. It had game dynamics. Games are fun for a while, but eventually people get bored of playing games. That’s what happened. People who were very excited and evangelistic about Second Life eventually moved on.

3. It lost its “new and shiny” patina. That’s most of why the press forgot about it. We only pay attention to new and cool stuff. Heck, just look at Techcrunch. Do you read about older technologies there? No.

Anyway, one thing happened that I find very interesting: it continued to grow in users, time spent on the site, and dollars spent in it.

On Friday I sat down with Mark to find out why.

First, the users remained very evangelistic. Second, corporations like IBM found other uses for its islands and kept investing (they now use these islands for training and replacements of expensive conferences). Third, the technology has been steadily improving. Fourth, the company has found new ways to bring new users in and make the experience easier to get into.

But he admitted that they had been pretty quiet and avoided doing more PR work until just recently.

Why is that changing this week? You’ll see why tomorrow morning at about 11 a.m. on building43.

But to tease a bit, I find that their new direction, the first part of which you’ll see tomorrow in the video I did with Mark, is interesting and represents a new life for Second Life and its host Linden Labs.

To wrap this up, have you used Second Life lately? Even if you haven’t, see you on building43 tomorrow morning for more.

Coming soon: the disruptive molecular age of information

Now we’ve seen what Google has had up its sleeve with Google Buzz. I expect this is the last tool of the atomic age. No, not the energy field, the real-time content field.


Before I start, tomorrow I’m giving a talk to Stanford University MBA students with MC Hammer and Loic Le Meur, founder/CEO of Seesmic (he wrote about his part of the presentation on his blog tonight) about what social media is doing to our marketing, and I’ve been working with a few companies on products that will come out over the next year that will move us from an atomic age of information streams to a molecular one, so wanted to talk about it, both here, and tomorrow at Stanford to see what bubbles up.

Look at Google Buzz. Each status message there is an information atom. You can’t easily grab two of these status messages and join them together.

Or look at Twitter. Tweets are information atoms. They stand alone. You can’t really combine them with other tweets.

Same with YouTube videos. My videos stand alone. You can watch one embedded on a blog and you aren’t even aware that the others exist. Atoms.

How about photos? Atoms. Try to join a photo from SmugMug with one from Facebook with another from Flickr with yet another from Picasa. You can’t easily. Yeah, yeah, geeks can by copying and pasting URLs but not in any nice way. Atoms.

Go to Facebook or Google Buzz. Each status message there is an atom.

Joining information atoms takes a LOT of work and a LOT of energy. Sort of like with nuclear energy, isn’t it?

Let’s discover what this molecular age of information might look like and what it might enable.

Before I start, though, can you find the original tweets that were made WHILE the Haiti earthquake was happening?

I bet you can’t. Go ahead and try. Go to Google search. Go to Twitter search. Search all night long if you want. You won’t find the original tweets.

Want them?

Me too, but they are hard to get to. Why? Because today we live in an atomic age of real-time streams. Once those atoms (er, tweets) streamed by they are almost impossible to pull back out. Why? Because our search systems don’t have the kind of metadata they need to make searching for them possible. No one linked to these tweets. You probably didn’t even know about them. They were on my screen for a few seconds and then, well, they were gone. Luckily I saved them for this post.

Here are the tweets that were made by people in Haiti DURING THE EARTHQUAKE: The first Tweet I could find was from Michelle Maura who wrote “Earthquake right now.” Within a few minutes a bunch of others had Tweeted similar things. My favorite was Ivon Bartok who wrote “The place rocked like a mofo.” While I was writing this post he also said he felt a 5.9 aftershock. About five minutes after the quake, MSNBC was the first news outlet with this breaking news Tweet.

How did I get these? Easy, I opened up WordPress’ editor and copy and pasted them where they sat unavailable to anyone except me until right now.

Now you notice that a blogger CAN make an information molecule. But look at how hard this is. I had to copy and paste URLs and if I wanted them to look like tweets I’d have to open them up, take a screen capture, upload that screen capture somewhere, link the screen capture in here, then link the tweet up. Whew, a lot of work. Like I said, making molecules takes a lot of energy.

But what if you had an iPad with a new tool? One that had a column that looked like Seesmic or Tweetdeck’s columns? One with a middle column where you could simply drag tweets into a molecule. One with a third column that would be an outbound column where I could add a text block, a video, an audio clip, and then distribute it out to Twitter, Facebook, Google Buzz, LinkedIn, or whatever else will be a place that humans want to read real-time streaming information?

Wouldn’t that enable a new kind of information curation?

Think this is just for info geeks like me? I’ve interviewed normal users about this and they keep complaining that it’s too hard to make a page of all their baby photos from their kids’ first birthday party, for instance. Think about it. You invite 30 friends over to your house. Some put photos on Facebook. Some on Flickr. Some on SmugMug. Now you have to join them all. I dare you to try. Yes, my audience can because we’re all geeks who understand HTML and copy and paste. Now put yourselves in the shoes of people who don’t have those skills or patience. They want a new system and someone over the next eight months will deliver it to them.

When the molecular age does arrive, it will have deep impacts on corporate social media. On traditional media. On us all.

Finally we’ll be able to share the patterns in the streams we’re seeing with other people.

Finally the search engines will have enough metadata to find those bundles of tweets, blogs, photos, videos, audio clips, and other info atoms that are actually important.

So, how do we get from the atomic age to the molecular age?

First, ask yourself these questions to put pressure on the industry to answer these questions:

1. Why can I tag a photo in Flickr but we can’t tag tweets or Google Buzz items?
2. Why do I need to come to a blog tool to join Tweets, blogs, photos, videos, etc together? Why can’t I do that where I read real-time streams like at, Google Buzz, or in tools like Seesmic or Tweetdeck or Tweetie?
3. Why can’t I build real-time information molecules simply by dragging and dropping these atoms into a molecule builder? Why do I need to copy permalinks and paste them into a blog editor?
4. Why haven’t we seen a real-time reader system that lets us see Tweets, Facebook status items, Flickr photos, Yelp restaurant reviews, Google Buzz items, YouTube videos, and other items all in one place?
5. Why hasn’t the web evolved so I could drag a tweet from into’s editor and have it linked up automatically?
6. Why can I favorite tweets (I’ve faved about 12,000 into this stream in past 10 months alone) but I can’t bundle them together?
7. Why can we work collaboratively on Wikipedia to build an encyclopedia of the world’s information, but we can’t work collaboratively on people’s profiles to add data we know about each other onto our profiles?
8. Why can’t brands join their tweets to your tweets about them? Or, why can’t I do that myself? When I write about a Ford car, for instance, I’d love to join tweets from @scottmonty, who works at Ford, in with mine, especially since I might be responding to a tweet of his.

Someday soon these questions will be answered and then you’ll know we’re in the molecular age of information.

I want to mix content together to make even more powerful content, but no one is giving us tools to create real-time molecules.

When will the molecular age of information start? Tomorrow I’m visiting Stanford University where Google and Yahoo started to see if the students there have any answers.

Marc Benioff's enterprise ambitions (first look at Salesforce Chatter)

Marc Benioff, CEO of, has big ambitions for the enterprise. He no longer wants Saleforce to be seen as a service just for salespeople. He’s going after the whole pie. Today he is releasing a new social service that looks sort of like Facebook for the Enterprise, called Chatter. Here’s a video first look at the latest beta.

He just might get it.

First, what pie is he going after? Microsoft’s Sharepoint. Now, he’s not alone in those ambitions. Jive,, Zoho, SocialText and other companies have the same ambitions. The problem is they don’t have a big stick to get into the CTO’s office. Salesforce already has a relationship with almost every company’s CTO. The other players have had some good wins, but they won’t be able to make an entire marketplace pay attention the way Benioff can.

When I interviewed Benioff I compared him to Steve Jobs. He didn’t like that comparision, but the comparison is apt. No one is positioned to become the Steve Jobs of the Enterprise the way that Benioff is.

OK, today Salesforce has firmed up the position in the enterprise as THE service for salespeople. It’s how salespeople track their contacts, sales pipelines, and much more. But here’s the rub: Salesforce is releasing a version of Chatter (which is really a copy of Facebook) for the entire enterprise. How will Salesforce convince the non salespeople to use it?

A few ways.

1. Pricing. Chatter will be free to non-salespeople. So, they will be able to join in without getting approval to buy something new. Other players like Jive and Socialtext don’t have their feet into enterprises yet (at least not most of them) and they don’t have the ability to offer a version of their product for free the way Salesforce does.
2. Salespeople now will badger coworkers, like our friends badgered us to join Facebook. Salesforce already has a strong position in most enterprises with salespeople. Those salespeople will now badger their coworkers to join in. How will they do this? “Hey, John, can you join in this Chatter thing because we are pitching XYZ company next week and we need your engineering input.” Other companies don’t have a built in badgering force the way Salesforce does.
3. Integration into already existing enterprise systems. In the demo I filmed you’ll see that Chatter isn’t really just a copy of Facebook in that it allows developers to integrate already existing enterprise databases and systems like SAP to shove information into Chatter’s feed. This is very powerful and will cause enterprises to adopt it wholly, while other systems will have troubles.
4. Salesforce is positioned to take advantage of a continuing anti-Microsoft force (and the continuing charge into Web services). When I speak at conferences you can see this anti-Microsoft force. How much marketshare has IE lost to other browsers? Do you use Microsoft Word anymore? Have you tried to find ways to reduce your reliance on Microsoft? Many have and you can see this force. Salesforce led the charge when it came on the scene 11 years ago with its “no software” chant. If you buy Sharepoint, for instance, you have to load that software. Someone has to setup a server and keep it running and update it when Microsoft inevitably comes out with updates. Plus, Microsoft is behind in its adoption of Facebook-like gestures. I know workplaces that have already adopted things like Jive or Yammer because they are far better than Microsoft in bringing social gestures to the workforce. Who is better positioned to take over than Salesforce? You might say Google, but, sorry, Google Buzz just isn’t even close to the enterprise trust and thinking that Salesforce has. So, I see two choices: stick with Sharepoint (Rackspace, my employer, is helping make that choice more palpable through its hosting of Sharepoint) or go with Salesforce. Which one will your company choose?
5. This sticks another nail into Lotus Notes. Enough said.

Anyway, it might seem weird for a Rackspace guy to be helping out Salesforce, which is really now a competitor to Rackspace’s enterprise efforts. But great companies aren’t built by companies ignoring their competitors. They are built by being honest about the marketplace and Benioff has big ambitions that’s gotten our attention.

Has he gotten yours? Can anyone stop Benioff? Here, listen to Benioff in his own words describe his ambition for the enterprise in an interview I did a couple of weeks ago.

Google Buzz copied FriendFeed's worst features, why?

OK, now I’ve had a bit of time to play with Google Buzz and everywhere I look I see a badly-executed copy of FriendFeed.

With two important exceptions:

1. Google Buzz actually has a lot of users and much better information flowing through its veins. There’s a reason that FriendFeed doesn’t have many users: it has some very anti-user features that retard user adoption (back when I was excited about FriendFeed I kept hoping that FriendFeed was going to fix some of their issues).
2. It has pretty nice location features built in, especially if you use Google Maps on Android.

But, they made some horrid mistakes. Let’s detail them:

1. They are infatuated with real time flow (items flow down my screen) but unlike FriendFeed they didn’t give you an option to turn that off. For users who are following a lot of people, like me, that makes Google Buzz unusable.
2. They love comments, but that leads to the chat room problem I talked about earlier. But Google Buzz actually made comments worse than FriendFeed did. They didn’t give us moderation capabilities. They don’t let you block people right from comments like FriendFeed did.
3. Unlike FriendFeed they don’t let you group your friends into lists. This makes using it with more than small groups very frustrating and pushes me back to Twitter where list support is an important part of my experience now.
4. They didn’t give us any filtering capabilities. FriendFeed got very close to having good filtering with its real time search engine that let you do things like “show me all items with the word ‘Obama’ but only show me items that have four likes or more.” That was very powerful, Google hasn’t even tried yet to do real time search or filtering like this.
5. Adding friends is frustrating, especially for someone who is getting added by hundreds of people a day. Google could really have innovated here, but they didn’t add anything Twitter or FriendFeed doesn’t have.
6. On Google Buzz, as with FriendFeed, A-list users get too much engagement and attention and that keeps putting their posts at the top of the feed. Google copied FriendFeed’s worst feature, that where items that get engagement keep popping to the top. I desperately want a strict reverse-chronological feed so I can just see items once, and not 300 times as they get engaged with.
7. They copied FriendFeed’s like feature, but didn’t do the most important thing: let me see all the likes by a single person. Over on FriendFeed I can see what Mike Arrington has liked, or commented on. I can’t do that in Google Buzz.
8. I turned off bringing Tweets and FriendFeed items into Google Buzz. Why? Because unlike on FriendFeed my readers can’t turn off just my Tweets. That makes Google Buzz very noisy.
9. They copied FriendFeed’s second worst feature: that comments bring people into my view that I don’t know, don’t care about, and Google Buzz gave me even fewer ways to control or moderate that. I wish there were a way to hide all comments until I want to see comments, for instance.
10. They also copied FriendFeed’s lack of curation features. For instance, Google Buzz has no way to bundle their items together into a single URL.
11. They copied the worst part of FriendFeed’s UI: the boring look but didn’t copy FriendFeed’s best part of the UI, the clean and simple feed. I keep getting boxes and other crap in my feed which reduces the number of items on my screen.
12. They copied the photo feature from FriendFeed, but, like FriendFeed, didn’t give me any control over size of images or anything else, either.

As I type this I’m at 37,000 feet and watching CNN. On CNN they showed off that Google Buzz is hard to use. Well, duh. They copied FriendFeed’s worst features and didn’t innovate.

And I won’t even get into all the privacy issues others had with earlier versions of Google Buzz.

So, Google, why did you copy FriendFeed’s worst features? Is there any way we can work together to find a better system? I really find Twitter and Facebook lacking too, and keep hoping someone brings out a dramatically better system but haven’t seen it yet.

The world needs a curation system, for instance. But until you stop copying other systems’ worst features I don’t feel you’ll even understand what users want.