Jason Devitt

The Ghost Freeway

My first company, Vindigo, is still best known for its mobile city guide. (We released almost twenty other applications, some of which made far more money, but like actors companies get typecast.) Consequently a lot of people ask my advice about how to deliver local content online as well as on mobile phones. I tell them all the same thing: it doesn’t matter how good your UI is or how many great ideas you have, the biggest problem in local content is the data. There are no reliable sources of local content. I was reminded of this at least four times in the last two months.

1) A breakfast meeting at BOCA in San Francisco. My friend Steve sends me a link to a site I’d never heard of, Ovahere. I double-checked with Yelp. BOCA has closed down. No problem, we reschedule.

2) A lunch meeting at the Penn Club in the New York. I check Yelp again, which sends me here. No sign of any business called the Penn Club. Silly me, I should have looked for the Penn Club of New York. This time I am thirty minutes late for my meeting.

3) We are staying in Hayes Valley in San Francisco. Our rental car has a Garmin navigation system. But the system doesn’t know that a large section of the Central Freeway was demolished four years ago and that there is a celebrated new Boulevard in Hayes Valley that leads directly to 101. As we enter the freeway, a robotic female voice keeps telling me to do a U-turn.

4) To get another perspective on our new home town, I fire up Google Earth. But though Google seems to have up to date images of sensitive military installations all over the world, here in San Francisco the Hayes Valley section of the Central Freeway is still standing, and still filled with cars.

Believe me, I understand the problems that these companies face. For example, restaurant chains like McDonald’s or Subway may open a new branch somewhere in America every few days. (At its peak rate of growth, McDs was opening a new branch somewhere in the world every five hours.) The yellow pages data that most web sites ultimately rely on is updated once a year.

My point is, don’t tell me that you have a great idea for how to deliver local content online or on the phone. Tell me how you are going to fix the data.

Via Tom Evslin

This is the last in a series of posts analyzing the forces that might break the hold that carriers have on the mobile market. See also parts I, II, and III.

(10) Regulation

I saved regulation to last, although not to build suspense. Like most entrepreneurs, I recoil from the subject. But wireless spectrum is a commons. There may be better ways of managing the commons than the frequency allocation chart above, but even in a world of open spectrum and software-defined radio we’d still need a licensing regime for new devices.

I’m interested in what kind of regulatory changes we might see in the next five years and their impact on my business.

Network Neutrality

I missed most of the kerfuffle over network neutrality while traveling this year, but now that Congress is in the hands of Democrats legislation seems inevitable. Partisan lines have been drawn.

The central issue is how to prevent a wireline carrier from deliberately degrading service to a content provider that refuses to pay a fee: is self-regulation enough or do we need an umpire? No one is even suggesting that a cable company or DSL provider should have the right to block access to a content provider completely. But wireless carriers do this all the time. Many mobile content providers in the US cannot get their content onto any network. The more people use the mobile web, the more this becomes a free-speech issue - a smart mob can’t use other media to communicate. How long will regulators let this go, not just in the US but in Europe? Or will mobile operators offer network neutrality themselves in exchange for eliminating their liability for content that goes over the network? 

Unlocked Phones

Carriers control the market because they control the phone. One way they do so is by ‘locking’ handsets to their network; they sell you a phone at a discount and one of the many strings attached is that you can’t take that phone to another provider.

But what if there is no contract? What if the contract is up? Why can’t I buy a CDMA phone directly from LG and demand that Sprint activate it on their network, so long as I pay for service? If I buy a phone from Sprint and switch to Verizon when my contract is up, I can take my phone number with me. Why can’t I take my phone?

A few weeks ago the Librarian of Congress (regulators are everywhere) granted an exemption under the DMCA covering software for unlocking cellphones. So far only Tracfone is seriously affected, because they subsidize handsets without forcing customers to sign a two-year contract. But regulators could go much further. Nothing would do more to increase competition in the US wireless market than legislation requiring carriers to accept other handsets on their network. Right now GSM providers can’t stop you; Verizon will allow it but makes it so hard that no one bothers; Sprint just says no. There is a clear precedent for change: the Carterfone decision.

Full Number Portability

Give someone control of a namespace and they will charge a rent for it. What would NetworkSolutions charge for a domain name if you couldn’t switch to a different registrar? IM interoperability wouldn’t be an issue if you could take your AOL screenname to Yahoo. But Wireless Number Portability is old news, right?

Not in Japan, where it was just introduced. Not in Canada, where they are still waiting. And we are all still waiting for Full Number Portability, between fixed and mobile networks. FNP makes it harder for carriers to charge a toll for terminating a call on a mobile network, a big issue in Europe if not in the US, because the customer can’t know in advance whether the number they are calling is mobile or not.

Restrictions on Subsidies

One way that carriers keep control of the handset is to lock it, but the way that they persuade us to give them that control is by subsidizing our purchase in the first place. Korean regulators have banned and encouraged subsidies at different times, to favor the adoption of certain technologies. Finland has always banned subsidies, to avoid favoring any. Neither country has a shortage of mobile phones as a result.

Banning subsidies makes operator pricing more transparent, and forces them to compete on price and service alone. If they use the savings to cut the price of voice and data, consumers are better off in the long run. If they shift the dollars into marketing instead, we just get more TV ads and expensive handsets. I don’t expect to see a ban in the US, but regulators in other countries may try it.

New Entrants

Regulators play a large role in deciding how many operators there will be in each market, by granting new licenses and approving or vetoing mergers. In the US there is not much on the horizon. Clearwire is rolling out a nationwide Wimax network in the US and European regulators will start auctioning Wimax licenses soon; we have yet to see whether Wimax can match the performance of cellular networks for mobile voice, but dual-mode GSM/Wimax handsets should be more reliable than the WiFi versions, and so cut into the revenue of the incumbent carriers.

Open Spectrum

In the next five years? Not a chance. Be grateful that the FM gadget for your iPod is finally legal in the UK. Hope that somebody can make Bluetooth easier to use. And be grateful for WiFi. Until everybody on your block gets it.

Apparently there is a crisis in the world of venture capital. Those who write about it see two distinct problems, one on the way in and one on the way out. Technology has so lowered the cost of starting a business that traditional VCs are no longer needed; on the way out the number of IPOs and big-ticket acquisitions has obviously dwindled, MySpace and YouTube and maybe Facebook being the exceptions. Some entrepreneurs seem quite smug about VC troubles. I think that the ‘crisis’ is largely confined to consumer Internet services, and that it is more of a crisis for entrepreneurs than it is for VCs.

True, part of the problem on the IPO side has nothing to do with technology or VCs. It’s a piece of over-zealous legislation called Sarbanes-Oxley that has raised the cost of going public. Personally I think the way to prevent corporate crimes is not to make companies swear to their accountants that they are really really not lying about their EBITDA, but to make them disclose as much information as possible to investors. SarbOx didn’t reveal that dozens of corporate executives had back-dated options; it was Erik Lie, a finance professor from the University of Iowa who crunched numbers that were publicly available.

But mostly the problem is that while the technology and the concepts behind Web 2.0 are exciting and important, the companies that embody them just aren’t very valuable. Pioneering sites like Flickr and delicious sold for less than $50 million each, and even YouTube and MySpace didn’t think that they could survive as independent companies. In total it’s just not enough money to get venture investors excited.

No problem, we are told, since cheap Intel boxes and Amazon S3 and Ruby on Rails and the rest of it mean that you can launch a company with $6,000 and a bowl of ramen.

Wait a minute. All over this country, across every sector, small business owners use an average of $10,000 to start their businesses. Even for this year’s Inc 500, a list of the 500 fastest-growing companies in America over the last four years, the median amount of start-up capital was only $75,000.

The key word is small. Like their peers in construction and retail and transportation, most of the consumer-facing Web 2.0 businesses are small today and will stay small. What is different is that most of the founders of the Web 2.0 companies believe that they are destined to be big.

VCs are adjusting to the fact that just like railroads, automobiles, fast food, bowling alleys, wall-to-wall carpet, fabless semiconductors, and personal computers in days gone by, the consumer Internet market has begun to mature and no longer presents many interesting opportunities for venture returns. (Yes, there will be exceptions; but every so often there is still an opportunity for a venture-backed retail franchise.)

A lot of the entrepreneurs that I meet have not made that adjustment. That’s the crisis.

I learned from Fred Wilson that Flickr has just added geotagging, the ability to add latitude and longitude to each of your photos. It’s a good design - you just drag and drop photos onto a map of the world. I quickly tagged 30 or so photos from our travels over the last nine months, plus a few from previous trips. The results are here.

I would like to have a camera (or cameraphone) with integrated GPS so that all my photos could be tagged with time, date, and place. It’s one of the most obvious location-based services, but I am still waiting for it.