Moonwatcher

In Mobile, It's Shaping Up to be Rainbows vs. Pinstripes

MacWorld reports that Google is developing native iPhone applications. Google's Vic Gundotra calls iPhone "the pre-eminent phone in the world" and Mobile Safari "a better mobile browser than anyone had ever delivered before". Android notwithstanding, BusinessWeek's two Most Innovative Companies are clearly aligned with each other in mobile, at least for the moment.

But other tech giants are lining up on the other side of the ball. BlackBerry-maker RIM recently announced its plans to develop native applications for SAP, IBM, and Microsoft.

On the surface, this looks like a simple partitioning into consumer and business segments. But RIM is touting the music-playing, photo-viewing, movie-watching, Face-booking abilities of its otherwise buttoned-down handsets in TV spots like this one:

While at the same time Google's enterprise group has signed up 10 million active users for Google Apps and Apple is set to launch iPhone 2.0 with a laundry list of enterprise-friendly features. Here's the announcement (with the relevant part starting at around 2:30):

Things getting a lot more interesting in the mobile world. Should be fun to watch.

Why Doesn't Apple Face The Innovator's Dilemma?

Clayton Christensen's seminal work The Innovator's Dilemma¹ describes how incumbent vendors typically compete by continually adding performance (typically in the form of new features) eventually overshooting the requirements of their target market and providing an opportunity for upstart competitors to provide a lower-cost "good enough" solution.

But John Gruber at Daring Fireball writes:

No one’s going to beat Apple by being "good enough". The only way to beat the iPhone is by creating something better.

Intuitively I agree with John that Apple doesn't face this problem with the iPhone. But why don't they? I think the answer lies in the reason people buy Apple products in the first place.

Apple doesn't sell products based on laundry lists of features. In fact, their products often lack features considered standard by competitors. For instance, the iPhone has a substandard 2 megapixel camera with no flash, no video recording capabilities, no voice dial, etc. Many of the features of the iPhone are in fact only "good enough".

Instead, Apple differentiates itself on design. Their products feel right. And to their customers, "good enough" design will never be compelling, regardless of how high Apple sets the bar.

By choosing to compete on design instead of technology alone, Apple seems to have found a loophole in the Innovator's Dilemma.

Update: Thanks for the many excellent comments to this post.

---

¹ If you haven't used Google Book Search, click the link to be freaked out.

TimeTube: Awesome YouTube/Timeline Mashup

TimeTube, from the increasingly badass guys at Dipity.

Two Quick SAPPHIRE Endnotes: ByDesign and Ecosystems

Tuesday I got to meet with SAP Business ByDesign chief Peter Zenke and ecosystem czar Zia Yusuf. And while I don't have time today to compose the full blog post that each meeting merits, I do want to describe my impressions in broad strokes.

Mr. Zenke was quick to address reports (including mine) that ByDesign has been delayed. As our meeting began he said, "The product is out. The product is not delayed. We are [only] rolling back the accelerated deployment." So it's the big marketing push and worldwide rollout that has been delayed, not the product itself.

Zenke went on to discuss the rationale behind the no-longer-accelerated launch, saying that SAP wants to bring more automation to the in-flight product upgrade process, which should eventually only require a few hours of downtime for the customer and one or two days of work behind the scenes by SAP. He said, "There were some surprises. We thought it was a little easier."

He reiterated Léo Apotheker's comment on Monday that, "What you can never fix is the wrong impression on day one," and said clearly that, "There is no pressure on SAP. We want to do what's right. We want happy customers."

It's fascinating to watch such a mature company with such deep competence in so many areas enter a space that is totally new for them, in this case SaaS. Some companies might gloss over the complexities and subtleties of the new space, and ignore the gaps in their competence. SAP seems to be carefully identifying the gaps, learning from them, and proceeding only once they're filled, even if it means scaling back global launch plans.

The delay, er, rolled back acceleration, makes for juicy headlines. But the resulting increased product quality should make for happy customers. It's pretty clear which is more important to SAP.

I also got to sit down with Zia Yusuf, who runs SAP's far-reaching ecosystem efforts. We didn't get to dive as deeply as I would have liked into the nuts and bolts of how SAP manages its relationships with its ecosystem partners, but I did get a glimpse into SAP's sophisticated portfolio management process and how it helps address the platform problem.

SAP looks to partners for "innovation at the edge", and goes so far as to publish a product roadmap on SDN and explicitly call out the "white space" opportunities around its platform—areas where SAP has no plans to develop competing solutions that are therefore ideal for partners. Zia also briefly described the governance structure within SAP that addresses potential partner conflicts on a weekly basis.

The wild and wooly "Web 2.0" world could learn some things from buttoned-down SAP. I specifically asked Zia if he would be willing to give a talk about managing ecosystems (he's in Palo Alto), and he said he'd be happy to. Google, Facebook, are you listening?

Technorati Tags: sapphire08

SAP's iPhone Amnesia

This morning RIM co-CEO Jim Balsillie took the stage at SAPPHIRE to expand on last week's announcement that his company is partnering with SAP to natively integrate SAP applications with BlackBerry devices. When asked whether SAP would also integrate with Treos and iPhones, Henning Kagermann said that the RIM deal wasn't exclusive, but that he didn't have any plans to integrate with BlackBerry competitors.

Which is indeed strange, given SAP's announcement of an iPhone version of its software six months ago:

SAP is breaking with precedent by introducing versions of the new software that are compatible with the iPhone ahead of programs for mobile devices businesses traditionally use, such as Research in Motion's Blackberry, Palm's Treo and devices that run on software from Microsoft Corp.

"The iPhone has become such a popular thing," said Bob Stutz, a SAP senior vice president who is responsible for developing customer relationship management software. "Everybody wants the ease of use of the iPhone."

The first generation of the iPhone software will load business contacts, information on sales prospects and account data onto the device, Stutz said.

Stutz said SAP decided to introduce the iPhone software ahead of programs for other devices at the request of its sales people, saying they prefer using iPhones to the other devices.

Mr. Stutz, who runs CRM and industry verticals for SAP, just walked past me in the press room. His iPhone was ringing. Once he gets off the phone, I'll ask him to clarify.

Technorati Tags: sapphire08

SAP's On-Demand Suite Faces Operational Issues

The Third Era of SAP was supposed to have started by now, but the world will have to wait a bit longer: SAP's forthcoming on-demand business suite Business ByDesign has has been delayed.

In private sessions with the Enterprise Irregulars yesterday, SAP co-CEO's Henning Kagermann and Léo Apotheker blamed the delay on unexpected scalability and operational issues and, interestingly, labor costs. Apparently the on-boarding and upgrade processes are largely manual, making them not only labor-intensive but also error-prone. Also intriguing was Dr. Kagermann's assertion that the next version of NetWeaver will help alleviate some of the problems.

I'm eager to get a better understanding of operational issues facing ByDesign. We'll be digging into the specifics of these issues with Peter Zencke, who oversees the development of Business ByDesign, later today.

Technorati Tags: sapphire08

False Dichotomies in Technology

A couple of years ago I was at a content management conference and the speaker asked if anyone could define "enterprise" content management. I suggested it just meant "expensive". I was half joking, but no one could come up with a better definition.

So when I read about the millions of dollars being poured into "enterprise social software" it makes me wonder what the "enterprise" moniker actually means. LDAP integration? Delegated administration? On-premise deployment? Auditability? Or just a six-figure price tag for a software license, plus a bottomless pit of consulting?

In the context of security, Bruce Schneier defines dual-use technologies:

The traditional term for technology the military shares with civilians is "dual use." Unlike hand grenades and tanks and missile targeting systems, dual-use technologies have both military and civilian applications. Dual-use technologies used to be exceptions; even things you'd expect to be dual use, like radar systems and toilets, were designed differently for the military. But today, almost all information technology is dual use. We both use the same operating systems, the same networking protocols, the same applications, and even the same security software.

It might be useful to think of many modern applications and technologies—think mobile devices, productivity apps, and social networks—not as "personal" or "business" (or, heaven forbid, "enterprise"), but rather "dual-use", that can be used in multiple, sometimes overlapping contexts.

For surely, customers, partners, and employees are people too.

Questions for SAPPHIRE

Like fellow Enterprise Irregular Jason Bush, I'm hoping to learn some things at next week's SAPPHIRE conference in Orlando, but in contrast to Jason's SRM-specific questions, I'm after a better understanding of SAP's position on some broad industry themes.

First, I'm interested in learning more about SAP's approach to ecosystem cultivation. I've written before about the extremes of how tech companies deal with their ecosystem partners, and am eager to pursue the subject with Zia Yusuf, who leads SAP's platform ecosystem development efforts.

I'm also interested in finding out more about SaaS and PaaS as they relate to SAP. Delays notwithstanding, SAP is investing heavily in on-demand applications with Business ByDesign, and like everyone else, I'm interested to know more about that particular product line. But I'm also interested in the concept of "platform as a service", which was recently debated by Marc Benioff and Hasso Plattner and is being pushed hard by the likes of Salesforce.com, Amazon, Google, and others.

I also hope to get some reaction from SAP executives to the continuing cloud computing partnership between Google and IBM. As Larry Dignan writes:

IBM CEO Sam Palmisano and Google CEO Eric Schmidt joined forces to talk up cloud computing, but their broader message was that consumer and enterprise technologies will merge.

The big lesson: Tech giants such as IBM, Microsoft and Google [and SAP? -cw] will increasingly have similar technical architectures and possibly business models.

I've asked various people at SAP before about Google and how their business might one day compete with or complement SAP's (in enterprise search and productivity apps, respectively). Those questions seem more relevant now than ever before. Maybe they'll wind up integrating "at the glass" (albeit the 3.5" glass)—Google and SAP and are both making special versions of their applications for iPhone. Strange bedfellows indeed.

Finally, this may be the last time I get to talk with Henning Kagermann in his current role as SAP's CEO—his term expires in May 2009. I'd like to hear his thoughts on how the personality of a company's CEO influences the company itself, and how SAP might change under the leadership of his successor Léo Apotheker. Hopefully I'll have the opportunity to ask Mr. Apotheker the same question.

Technorati Tags:
sapphire08

Uphill, Both Ways

Brad Feld clarifies that when he says "email" he means "messaging". But it's always been email.

It used to be a .plan file, now it's lifestreaming.

It used to be talk, now it's IM.

It used to be USENET, now it's blogs.

It used to be trn. Now it's NetNewsWire.

It used to be rec.humor.funny, now it's Digg.

It used to be Rogue, now it's World of Warcraft.

But it's always been email.

Bill McDermott, CNBC, and the Rashomon Effect

SAP Americas, Asia Pacific, and Japan President & CEO (whew!) Bill McDermott recently went on CNBC to talk about SAP's quarterly results and the strong euro's negative impact on them. His performance drew this stinging criticism from ZDNet blogger Michael Krigsman:

In a moment of “irrational exhuberance,” McDermott described SAP as the only software company able to offer these things to customers. I suspect some SAP competitors will take issue with his comment.

One interviewer jokingly asked McDermott whether he had taken media training, because he spun virtually every difficult question into a positive for SAP; the journalists wanted to hear a more balanced presentation.

When Mike sent a note about his blog post (with an even harsher critique) to the Enterprise Irregulars mailing list, I decided to watch the clip myself. Personally, I thought McDermott came off as confident but honest and very well-spoken. The reference by one of the commentators was about his giving media training to other executives given the strength of his performance.

It's amazing how two people can witness the same event and see two totally different things. Cinephilic psychologists call this The Rashomon Effect. Here's the clip in question, so you can decide for yourself.

PS. Hey CNBC, people might want to embed your videos without having to View Source then cut and paste HTML. I'm just saying.

Historical Timeline of Internet Memes

Dipity brings the memes! I hadn't seen the Coverflow-esqe "Flipbook" view before. Nice!

No Meetings

It just occurred to me that in the 20 months or so Spanning Sync, Inc. has been in business, we've never had a meeting. No internal meetings, no partner meetings, no meetings with prospects or customers. I don't think we've even had a conference call.

Of course Larry, Byron, and I do communicate with each other, but it's almost always over email. We've had maybe 30 IM sessions, a handful of one-on-one phone calls, and maybe a half-dozen video chats. Larry and I had lunch together once, but we didn't really talk about business.

That may sound bizarre, but for us at least it's insanely productive.

We do have access to a shared conference room that looks pretty nice, with nice, comfy-looking chairs and a projector. Maybe we'll go in there some day and play X-Box or something.

Preview: Spanning Sync v2.0 Adds Contact Sync

(Cross-posted from the Spanning Sync Blog.)

People know Spanning Sync for its ability to sync Apple iCal with Google Calendar. In fact, more than 70,000 people have used Spanning Sync to do just that. But calendars are only one part of the equation.

Our users tell us that they also need to be able to sync their Mac Address Book with their contacts in Gmail and Google Apps. So Spanning Sync 2.0 does both—calendars and contacts.

Here's a quick video preview. Click here for a high-quality version.

Like all Spanning Sync upgrades, version 2.0—with Contact syncing—will be free to all paid subscribers. Watch this space for the announcement of the public beta, coming soon.

Amazon Introduces Service Health Dashboard, Premium Support Options

Amazon has just announced its new Service Health Dashboard, which "provides access to current status and historical data about each and every Amazon Web Service."

Similar to trust.salesforce.com, the dashboard gives developers and users some idea of how well the Amazon platform is operating. But unlike the Salesforce system, Amazon's also provides RSS feeds to allow easier tracking. It looks like the feeds are meant to be read by people (as opposed to automated processes) but they're a start. I'd like to see Atom feeds with quantitive data about the availability and performance of the various services.

This type of status information is important for users and developers alike, and is unfortunately lacking from both Google Apps and the new Google App Engine. We've considered biulding such a system for Google Calendar to let Spanning Sync users know when Google Calendar itself is down, as it will be most of this morning, but obviously we'd prefer Google offer it themselves.

(As an aside, why is it that PayPal operational issues so frequently coincide with Google Calendar outages? Ugh.)

Amazon also introduced AWS Premium Support, available at both "Silver" and "Gold" levels, both of which include guaranteed response times for support incidents, client-side diagnostic tools, and named support contacts. The "Gold" plan also includes 24 x 7 x 365 coverage and telephone support.

As Om Malik says, "Amazon Web Services has decided it’s time to grow up and play nice with business." Kudos to Amazon for continuing to address the needs of developers looking to build on their platform.

Persistent Storage for Amazon EC2 (!)

From the announcement on Amazon CTO Werner Vogels' blog:

Persistent storage for Amazon EC2 will be offered in the form of storage volumes which you can mount into your EC2 instance as a raw block storage device. It basically looks like an unformatted hard disk. Once you have the volume mounted for the first time you can format it with any file system you want or if you have advanced applications such as high-end database engines, you could use it directly.

Here's the money quote:

And the great thing is it that it is all done with using standard technologies such that you can use this with any kind of application, middleware or any infrastructure software, whether it is legacy or brand new.

Google, are you paying attention? Amazon is eating your lunch.

It looks like I'll be choosing between Joyent and Amazon to provide the infrastructure for the new service Larry and I are working on. Don't worry, I'll document my selection process here.

Update: Jeff Barr offers more detail on the Amazon Web Services Blog. Plus, here's an early review.

Update 2: Mark Fletcher lays it out:

This makes Amazon Web Services even more interesting, because it's now easier to run a normal MySQL instance without having to do something like running some kind of replication just to deal with the non-persistent local storage. And it scales up.

Salesforce for Google Apps Announcement and Coverage

The announcement: Introducing Salesforce for Google Apps

Some of the coverage:

• Phil Wainwwright: Salesforce and Google team to conquer the enterprise
• Mashable: Google Takes Aim at the Enterprise
• VentureBeat: Desktop applications beware: Salesforce integrates Google Apps
• Silicon Alley Insider: Google-Salesforce Deal Shows Google Not Insane, Microsoft Disruption Continues
• TechCrunch: More Details On The Google-Salesforce "Enemy Of My Enemy Is My Friend" Alliance
• Larry Dignan: Google and Salesforce.com: Why don’t they merge?
• Vinnie Mirchandani: Why just target Microsoft?

I'm glad to see Salesforce and Google Apps becoming more integrated. I'll have more to say on this later, but today is a coding day, so you'll have to bear the suspense.

Update: The Google Enterprise Blog has more detail on the announcement.

Austin Ready for 3G iPhone - Me Too

According to the Statesman, AT&T is spending $200 million to improve its Texas network, which is good news. But the better news is that, "AT&T already offers 3G service to nearly 350 cities across the country, including Austin."

C'mon, Apple. It's been almost a year since I bought my iPhone, and almost two weeks since I bought my last $400 worth of Apple stuff. I need a fix. Gimme my 3G iPhone, man. C'mon!

Dopplr and TripIt: Two Travel Startups Kicking Ass

This morning I got an email newsletter from Dopplr announcing their new Google Contacts integration, which uses the Google Contacts API to help you figure out which your contacts are also using Dopplr—without Dopplr needing to know your Google password. Keep in mind that Google only introduced its Contacts API last month. Nice work, Dopplr!

Also cranking out the new tech, TripIt yesterday announced their new mobile service, available at http://m.tripit.com, which works with BlackBerry, Treo, or any other mobile device with a web browser, but really shines on iPhone.

I'm rooting for both of these companies to succeed. Based on their quality and pace of innovation, I think they both have a good shot.

iPhone SDK Beta 3: Like Beta 2, but 1 Better

Joyent Responds to Google App Engine with Free Infrastructure Offer for High-Volume Apps

Yesterday Google announced App Engine, which (so far) is a hosted environment for running Python applications, currently limited to 500MB of storage and 5M pageviews per month.

Joyent CEO David Young this morning responded to Google's announcement, and in a big way. Here's his blog post Joyent's Garden of Eden for Python Web Applications in its entirety:

Joyent is pleased to announce our “Garden of Eden” program offering free infrastructure for high-volume python web applications. If your python web application has 25,000 or more unique visitors every month, Joyent’s “Garden of Eden” program provides you, the python application developer, with unlimited on-demand compute, storage, memory, bandwidth for your python application (besides the nude, and buffed people everywhere). Joyent only asks that you provide Joyent unlimited access to your customer information and clickstream data. Grow your python application to 1,000,000 monthly uniques, all the infrastructure you need to run your application is on us. All we ask for is unlimited access to your customer information and clickstream data. This is no joke. We’re serious. Angel with a flaming sword guarding the garden serious.

Interested parties can get started by contacting Joyent’s developer concierge at playersclub [at] joyent [dot] com.

In contrast with Google's proprietary platform, Joyent provides the kind of non-proprietary, open platform-as-a-service I'm looking for. I'm glad to see that they're challenging Google head-on.

Google Will Host Your Apps, for a Price

As a web application developer I'm interested in any new on-demand platform, but so far I haven't seen one that didn't have an Achilles Heel. Salesforce's Force.com is proprietary. Amazon's EC2 isn't, but doesn't offer any simple way to host a relational database. [Update: this is changing.] Now Google has thrown its hat in the ring with Google App Engine. Here are my initial reactions to the technical details.

The Good:

• Persistent storage (with queries, sorting, and transactions)
• Automatic scaling and load balancing
• Integrated authentication with Google accounts
• It's free to get started.
• Your apps will run on the same infrastructure as Google's apps.

The Bad:

• Your apps will run on the same infrastructure as Google's apps.
• Application code must be written exclusively in pure Python.
• The datastore is proprietary.
• The web server isn't Apache, and apparently doesn't support configuration options like URL rewriting and redirection.
• Application code only runs in response to a web request, and must return response data within a few seconds. A request handler cannot spawn a sub-process or execute code after the response has been sent.
• Developers must choose between authenticating against Google accounts or Google Apps accounts. These two forms of authentication can not be used with the same application.

So if you're willing to write your application in Python, target Google's proprietary platform components, and forget any functionality that doesn't fire in response to a web request, you can get huge automatic scalability benefits. For me, the benefit doesn't justify the cost.

What I'd like to see is simple: a pool of Intel-compatible machines (virtual or real, I don't care which) onto which I can load whatever OS, tools, and apps I like, plus automatic load balancing, scaling, and failover. The on-demand platform that comes closest to meeting my requirements is Amazon EC2 plus Scalr, but even that combination lacks the persistence I'm looking for, SimpleDB notwithstanding. I should give Joyent another look.

BTW, I'm sure it's just me, but doesn't the App Engine logo look a little like The Doomsday Machine from Star Trek? :-)

MacBook Pro Surgery II: The New Hotness, er, Coolness

Today I performed the second and hopefully final surgery on my MacBook Pro—this time to upgrade the wireless card to 802.11n (extreme in Jobspeak) and replace a dead fan.

The network is predictably faster, but more importantly, the right side of the machine is noticeably cooler.

This machine is almost two years old and has held up remarkably well. The only time I find myself wishing for more speed is when I'm ripping DVD's to make them playable on my tv, which I've been hesitant to do with a dead fan anyway.

OK, enough horsing around—back to work!

Will iPod Touch Get GPS?

Let's assume iPhone will get GPS¹ capabilities. After all, the hardware part itself reportedly costs only around $4, and the market for location-based services (LBS) is apparently growing like mad and is supposed to be something like $10B within the next couple of years. (But then again, aren't they always?) Add to that the fact that Apple has included the CoreLocation framework in the iPhone SDK and I think it's safe to assume location is a relevant area of focus for iPhone.

But will iPod Touch will get GPS too? I tend to think so.

While making new features exclusive to the iPhone would help differentiate the products, it's not clear to me that that's what Apple wants to do. CFO Peter Oppenheimer is on record as describing iPod Touch as "the first mainstream Wi-Fi mobile platform running all kinds of mobile applications".

I suspect that Apple doesn't see iPod Touch as an iPhone minus the phone, but rather sees iPhone as iPod Touch plus a phone. That may sounds like a subtle distinction, but it's not.

With the more-accurate positioning data that GPS would provide, Apple could do to the LBS market the same thing they've done to the mobile web, which is to say, turn it upside down.

---

¹ Here I'm using the term "GPS" broadly, to include not only the US GPS system but also similar systems like Galileo in Europe and GLONASS in Russia (where iPhones are reportedly status symbols), but not so broadly as to include iPhone's current cell-tower-plus-hotspot-triangulation technologies.

Will Your Next PC be an iPhone?

Yesterday I had lunch with Susan Scrupski and Brian Magierski, and the talk turned to the iPhone, specifically what it will still be missing after v2.0 ships and a 3G version is available.

Off the top of my head, here are some things I'd like see:

• HDMI out for HD video
• Integrated GPS
• Better still camera
• Video camera
• Voice dialing
• Bluetooth stereo audio out
• Obviously, more storage

and here's the crazy one:

• The ability to run regular Mac OS X apps when connected to a full-sized display

Ultimately, modulo Moore's Law, there's no reason I shouldn't be able to drop my iPhone into a dock with power in and HDMI audio/video out, pair it with my Bluetooth mouse and keyboard and IR remote, and use it like a full-sized Mac—including watching movies and playing games in 1080p HD video—then undock it and carry it with me, using many of the same apps but in their iPhone-native modes, with additional mobile features like GPS.

It's astonishing to me that Apple is expected to sell more iPhones than Macs this year, and maybe 3-4 times as many next year. (Or, if you take into account both the iPhone and iPod Touch product families as does Ross Garber, maybe 8-10 times as many.) Combined with the fact that only 25% of iPhone users are Mac users, that means Apple is making incredible inroads into the Windows PC customer base.

If Apple gets serious about adding features like those I've listed here, who knows? Your next PC might not be a PC—or a Mac. It might be an iPhone.

The Platform Problem: Same as It Ever Was

Chad Dickerson has written a great post titled Facebook and platform complementors: history repeats? in which he puts the recent story of a Facebook developer getting stepped on by Facebook into perspective. But I disagree with one of his points.

In his post, Chad asserts that, "the overall velocity and number of participants in the new platforms makes the game today inherently different." While there may be some difference in how quickly the dynamic between a platform provider and its complements can change, I don't think the dynamic itself has changed.

There's a fundamental tension between any application platform provider and the developers that build applications for that platform. Clearly, the platform provider must be free to innovate or their platform will stagnate and become less competitive over time. But they must also maintain the trust and good will of their developer community, which can be severely damaged by apparent acts of betrayal.

Often the innovations platform providers seek to introduce are the very applications that have been developed by third parties. Cisco and Apple illustrate two contrasting approaches to the problem.

Cisco is the canonical example of a platform company that excels at innovating through acquisition. Their approach is well-known, so venture capitalists are eager to invest in startups that provide novel complements to Cisco products. A recent BusinessWeek article sums it up:

Cisco, meanwhile, has a corporate strategy of letting a whole bunch of venture capitalists capitalize companies, waiting to see what works and buying up the companies that succeed.

Apple on the other hand is notorious for its cavalier attitude toward third-party developers. In 2001 Dan Wood (no relation) released a handy Mac application called Watson, which was featured on-stage at Apple's developer conference and won an Apple Design Award for "Most Innovative OS X Application". Soon thereafter, Apple released a nearly identical application called Sherlock. When Dan complained, he got a call from Steve Jobs:

"Here's how I see it," Jobs said—I'm loosely paraphrasing. "You know those handcars, the little machines that people stand on and pump to move along on the train tracks? That's Karelia. Apple is the steam train that owns the tracks." So basically the message was: get out of the way, kid; this is our market.

Maybe this is why the Mac developer ecosystem consists largely of hobbyists and small independent developers while Cisco's ecosystem is comprised of well-funded companies with loftier ambitions.

Clearly, the trick to keeping any platform ecosystem healthy is to selectively incorporate such third-party developments into the platform itself instead of running roughshod over them. Mastering this process is critically important for any burgeoning platform provider.

June 21, 1996: The Austin Internet Restaurant Guide Debuts

This weekend while cleaning out an old file cabinet, I came across some color laser prints of a project I worked on a few years ago. OK, more than a few. More like twelve. Anyway, it got me thinking about how much has changed since then—and how little.

In 1995, a buddy of mine and I decided to leave our Mac software development jobs to start an Internet company. What that meant we didn't exactly know, but we had both been on USENET for years, had spent too many hours playing Netrek, and had started using this cool new thing called the web.

"That," we said to ourselves, "is going to be big."

So we incorporated First Internet, Inc., leased a T1 from AlterNet, bought the requisite network hardware (it's amazing what people will send you with nothing but a purchase order), hooked up a couple of PowerMac 6100's, and starting hosting web sites for several local business (Shiner Beer among them) at exorbitant rates. It was a great business while it lasted, meaning it covered the monthly cost of the T1.

But a few months later we lost our two biggest customers and my partner decided to bail out. I was left with a too-big office, a $1,600/month ISP bill, and virtually no revenue. As I began the process of maxing out my credit cards to cover the bills, I looked desperately for what to do next.

At the time I was 26 years old, and my biggest problem was trying to figure out where to go to lunch every day. So I decided to solve that problem, which was surely shared by lots of other people, and put together an online restaurant guide. Remember, these were the days when Microsoft was just starting to think about "content", and eight years before Yelp was even founded.

This was also a time when the components of the now-ubiquitous LAMPP platform were in their infancy. I built the application using hand-coded HTML and a database package called Lasso on Macs running System 7.5, all fronted by the well-meaning but tragically buggy WebSTAR web server software.

The Austin Internet Restaurant Guide won a respectable number of customers but was never a commercial success. It did however help me understand the issues involved in creating database-backed web sites. And it was this understanding that helped me see the tremendous value in what a young local company called Vignette was about to start selling. I joined Vignette in September 1997, which turned out to be a good move.

A lot has changed since 1996. In fact, just about everything has changed, at least with regard to the Internet. It's become a fundamental part of business and of daily life, and its business and technology structures have matured to reflect that.

But some things haven't changed. When I look for a business to start (or join) I still look for solutions to the problems that I face personally. That's how I decided not only to create the Austin Internet Restaurant Guide and later join Vignette but also to create Spanning Sync with my partner Larry Hendricks.

Democracy is Not Pretty

As a delegate to the Travis County Democratic Convention yesterday, I got to see the workings of democracy up close. It wasn't pretty, but in the end it seemed to work.

I arrived at 8:30am, stood in line for an hour and a half to sign in, waited six more hours to vote (or more correctly, to "caucus"), and then spent five minutes voting. Er, caucusing. Almost 7,000 other delegates did the same thing.

The result: Obama will likely win Texas, but the final results won't be in until the Texas Democratic Party state convention wraps up June 7.

Even Working at Google Can Suck

Google Calendar has been down for hours, and a lot of people are understandably upset about it. Unfortunately it looks like there's only one person fielding all the complaints, and her name is Becky. She's the "Google Guide" assigned to monitor the Google Calendar Help groups.

I'm sure Becky is incredibly bright, accomplished, and hard-working, or else she wouldn't be at Google. But I bet there are a few places she'd rather be this morning.

On the bright side, this morning's outage means that Becky gets to take a break from responding to the flood of complaints about Google Calendar Sync.

Hang in there, Becky! We're rooting for you!

Today's Fortune

I found this in the fortune cookie that came with my lunch today:

Words to live by.

Google Developer Day Not Dead After All

Writing on the Google Data APIs Blog Jackie Bodine brings the happy news:

For those of you can't make it to San Francisco, or are wondering what happened to Google Developer Day from last year, don't despair! We'll also be holding many Developer Days around the world (more info to come later).

Hopefully Developer Day won't carry the same $400 price tag as Google I/O. And maybe, just maybe, there will be one in Austin. (After all, at least from the street, it looks like work is progressing on Google's new Austin office.)

:-)