Back on October 3rd of last year, I got invited to a roundtable to talk to the White House about startups. I almost didn’t go.

I imagined it would be an auditorium-style set-up with some low-level staffer on a stage giving 200 of us locals a 90-minute slideshow presentation outlining all the incredible ways that the government was here to help.

Well, yeah, there was that, but that was later.

What I got to sit in on was an actual round table. And I was sitting directly to the right of Todd Park, U.S. CTO. To his left were five fellows (actual fellows, not dudes, although four of them were dudes) charged with making the federal government more accessible to startups.

To my right were six Triangle folks who were heavily involved in the startup scene, including entrepreneurs like PlotWatt’s Luke Fishback and Relevance’s Justin Gehtland.

For the record, I was repping Automated Insights and our work making actionable human-sounding narrative out of big data.

Park is an amazingly genial guy with a lot of energy, and instead of going down that tired “what can we do to help the entrepreneurs here” road, he spent most of his time talking about the five initiatives that each of the fellows represented – everything from more streamlined ways for startups to respond to Federal RFPs to the Open Data Initiative.

The latter program was by far the most interesting, in my opinion and, soon to find out, the opinion of the others gathered around the table as well.

In short, the ODI purports to take data that is collected by the federal government at the national (and sometimes worldwide) level, and make it available it to smart people to use for their own purposes, including commercial.

Datapaloozas … Data what?

Talk quickly moved to the White House’s series of what they called Datapaloozas (that name just… could be better), where they expose the ODI program to a group of gathered entrepreneurs in a selected city, with explanation, instruction, and guidance. The hope is that the smart people take this data and make products, companies, and even entire industries out of it.

Can This Work? Sure.

To a point.

First of all, I have what I like to think is a healthy skepticism when it comes to startups getting help from big things like corporations, foundations, and governments. It’s not true in every case, of course, but more often than not, you get a big rush of everyone wanting to help, which is usually followed by an event or two, a lot of talk, tons of tweeting about how clever and diverse and passionate the participants are, and then, when the rubber meets the road, crickets.

The idea behind the Datapalooza is this: The feds have all this data, but they shouldn’t be the ones building the GPS devices or weather.coms (both actual success stories), because the government would be, to quote Park, “terrible at it.” What they can do is keep the GPS satellites in the air and collect and deliver the weather data, and then make that data available to the public.

So this is what I said then – somewhat paraphrased because it was a long time ago and I don’t write down a lot of stuff: You can’t make startups out of data. You can’t even make products out of data. But data helps.

I still believe that.

Lots of Data ! = Lots of Products

At Automated Insights, we use a lot data, with data points in the billions in certain cases. We certainly couldn’t exist without data, but our product isn’t the data, or even the aggregation or the actionable insights contained therein. Rather, the data complements our product, which is a proprietary set of algorithms, code, and fairy dust that spins content out of that data at up to 1,100 unique and totally insightful reports per second.

Yes. I’m a bit of a homer.

In fact, I’d say that we have the best algorithms, code, and fairy dust there is, and that’s our product. If we can introduce new data sets to that product, preferably merging them with existing data sets or customer data sets, our product can produce better results.

In that same vein, weather.com isn’t weather data – it’s the delivery mechanism for that weather data. For every weather.com there are thousands of companies in a deadpool somewhere who tried to make a go at a business with weather data or GPS data or some other kind of big data. They just didn’t have the best product.

You Can Still Palooza It

That’s not to say that the government shouldn’t be doing this or that you shouldn’t get involved –- one only needs to look to Ed Tech and how quickly that sector has rocketed forward with InBloom, a standardized database of education data.

I’m just trying to keep everyone’s eyes open about it. There are two hard parts to building a startup around data.

1) Get the data – which is where Park and the feds have done an admirable job and

2) Build the product that makes the best use of that data.

The Datapalooza is an in-between step, but still a vital one.

Jammin’ Data at The HUB

That step gained further traction last Monday, some six months after the initial roundtable, when the Datapalooza began taking shape at HUB Raleigh (HUB’s Chris Gergen, also at the roundtable, spearheaded it back in October). In a bit of a twist, this is the first regional Datapalooza (as opposed to a single city, but I think that has more to do with the Triangle’s geography than anything else), and it’s the first multi-vertical Datapalooza, with effort being focused on health, energy, and education.

The April session was dubbed the NC DataJam, and brought out about 80 people to a half-day session to discuss the forthcoming 90-day Datapalooza, the timeline, a walk-through of some of the data, and then a breakout into working groups to discuss next steps and ideas.

Park was not there, but rather appeared via a technology-hampered Skype. He laid out the expectations himself, saying: “We have no money. We have no resources. We want to make you famous.”

Which sort of plays into my fears about this kind of program. Building a product, let alone a startup, out of this data will take exactly that: money and resources.

But beyond that, the Datapalooza will work like a standard competition, only somewhat decentralized. Applications get filled out (August 1st), teams are selected and incubated (coaching and “technology” so I assume there must be some resources) and a demo day is held (September 12th).

Ultimately, I’m a big fan of opening government-collected data to the masses, especially if it’s clean and structured. In fact, I’ve got people looking at data.gov as we speak. But… I’ve got people looking at data.gov as we speak, and while I believe promoting the existence of this data is a good and necessary effort, I can’t yet see what tying it to a Startup-Weekend-style competition might produce.

Of course I hope (and I wouldn’t be surprised if) I’m just ignorant to the potential value. In fact, Chris Gergen put it best himself at the DataJam.

“This is only the beginning of the conversation.”

Editor’s note: Joe Procopio is a serial entrepreneur, writer, and speaker. He is VP of Product at Automated Insights and the founder of startup network and news resource ExitEvent. Follow him at @jproco or read him at http://joeprocopio.com