Month: June 2009

Lyza and Tableau according to Mako

Back in February when I heard about Lyza, I thought right away of Tableau. Despite each one’s different strengths in data discovery and analysis, each appeals to the same broad group.

It’s an old group that’s getting new attention: creative analysts, or “cowboy analysts” to some. The like their data raw, not aggregated. They ask questions, forage, synthesize, analyze, and publish.

Joe Mako is one of them. Tomorrow, he’s launching a website for people like himself who use both Tableau and Lyza. Makometrics will publish every Monday morning and sometimes more often.

Joe is a network engineer at a Midwest ISP. He started at the tech support desk, where he saw how much help people needed looking at their data. “They didn’t understand exploring data,” he says. “They just don’t care.” But Joe cared enough to help with data analysis, and pretty soon someone gave him a tag line: “Make it happen with Mako.”

Posts he’s lined up so far:

  • He’ll walk through data analysis problems from challenge to resolution. “I’ll be practicing something akin to the cycle of visual analysis.” (See the Tableau video “The Zen of Visual Analysis.”)
  • Analysis of strengths and weaknesses of Tableau and Lyza
  • Analysis of his “Visualizing Rambo Kills”: how he approached the dataset, and how he created the final result.
  • Demonstrate sophisticated techniques in Lyza and Tableau. He’ll go into detail on such things as combining Lyza’s “previous” and “if” functions and the basics of summary functions like “sumcolumn” and “avgcolumn.”

Check it tomorrow (Wednesday, July 1):

Storage goes node to node

The bad guys do it. Malware and terrorists distribute themselves, hide, and wait for a call from Mother. Now one company is doing the same with backup storage. As the bad guys have known for a long time, the strategy is cheap and reliable.

“It’s disruptive as hell” for the data storage industry, says up-to-now Tableau vice president of business development and “disruption junkie” Kevin Brown. He’s leaving in-the-pink Tableau for for VC-funded Symform. Giving the news to Tableau CEO Christian Chabot was painful, he says. “I had the siren song in my head, and I just had to chase it.”

It sounds like he’s jumping onto another good product.

As he described the choices for data backup at small- and medium-sized businesses, I glanced at my removable 1TB hard drive an arm’s reach from the keyboard. It’s the center of my backup system. I could take it with me at night, but I usually don’t. A megabyte or two of crucial data does get backed up every hour over FTP, but cost and speed limit me there. After a major earthquake, fire, or burglary, I might be out cold. I’m probably typical.

Instead, I could trade storage with other nodes — indirectly, arranged automatically at a central point. I keep bits of others’ data, and they all keep bits of mine. Each of us uses as much storage as we contribute. Instead of putting my own data on that external drive across the desk from me, I put someone else’s there.

Symform encrypts each file and breaks it up into 64 fragments. They’re dispersed randomly into the cloud. An algorithm known as Reed-Solomon ensures error-free transmission.

But what if one of my fragments goes to, say, Kabul and the computer gets blown up? No problem, he says. The control center monitors the fragments constantly. When one goes missing, the system regenerates it and moves it somewhere else on the globe.

“By an order of magnitude, we’re more secure,” he says.

After a disaster, retrieving the backed up data would go faster, he says, because you’ll be pulling on 96 locations at once instead of just one over one throttled pipe.

They’re charging $30 for unlimited storage. That is, up to the limits of your “contribution folder,” the space you make available in your office. As the sign said at a buffet in Reno, “Take all you want, but eat all you take.”

This is disruptive for the storage industry, perhaps, but not for Kevin’s lunch and golf habits. The Symform office is just a “four iron” away from Tableau headquarters.

Another night on Earth

In the film “Night on Earth,” Italian comic Roberto Benigni plays a taxi driver tooling around Rome one August morning at four. His flag’s up, he’s bored, and the streets are empty. “Dove sono i romani?” he asks himself, “Where are all the Romans?”

Where were all the BI people last week? Did they all go to Munich or Tehran? Or were they just resting up for the recovery?

Then I checked Twitter. Many of them had been there all the time. Still, I was bored.

Then on Saturday, Ken Rudin’s email came in announcing his new address. On Monday, LucidEra’s sad news.

At 4 on recent mornings, I suppose LucidEra execs were wide awake and staring into the dark. Mark Madsen emailed that he had an inkling, but I was surprised. This wasn’t supposed to happen. LucidEra was one of the interesting companies, one of the bright lights. In January, I had written a column for TDWI about possible expansion of the Pipeline Healthcheck. Others would go down, but not them.

On the other hand, it was a startup. They often fail, especially now. And I’m sure that chief marketing officer and co-founder Rudin and vice president of marketing Darren Cunningham will land well.

The movie is more fun. Someone finally hails the cab, a priest. The driver feels no reverence. He invents a long “confession” to amuse himself, and it works so well on the drive across town that he doesn’t notice the priest’s fatal heart attack.

Recession reactions: hunkerers and builders

There are the hunkerers and the builders. Those are the two basic reactions to the recession that Jill Dychè observes. She’s a principal down at Baseline Consulting, and she reports seeing more builders.

The hunkerers “are using the economy as an excuse for a fair bit of inertia,” she emails. But the builders are using “the temporary lull” to ready themselves for the recovery. “Strategy” comes up a lot more these days.

One type of builder is revisiting infrastructure. For example, some want to re-platform their data warehouses, and others want to acquire master data management hubs. “When things get really busy again, they’re positioned to be out in front.”

A second type of builder she sees has better user functionality in mind. For example, one specialty retailer has decided to improve analytics and to make users more self-sufficient. “We’ve been helping them load new data into the data warehouse and to build in more ability to drill down.”

Naturally, many clients’ boom-time issues remain. One is the temptation to choose newer, faster and cheaper tools without having prepared the data. But that’s for a post next week.

This is one in a series of posts on the recession and BI.

What have you observed of the recession’s effects on BI? Tell me here.

Mark Madsen’s three indications of uselessness

If you dropped into an organization, how could you tell who did real work? Mark Madsen has developed clues.

Most people probably know him as the insightful and entertaining creator of “Clues to the Future of Business Intelligence” and more recently of “Using Open Source BI in the Real World.” But when he’s not on stage or preparing for it, he’s consulting on BI projects.

“On almost every project,” he emailed me recently, “I find people whose task can be perfectly explained by watching Office Space, the scene where the two Bobs interview Tom and he explains what he does.” Tom’s the one with “people skills” who “deals with the goddamn customers” by supplying reports to them. His secretary does the actual carrying.

Mark has developed a few clues to indicate a Tom.

1. They simply pass information along. They don’t synthesize anything, nor do they have any use for the information themselves. This is often a VP or director, usually with one or two direct reports and usually connected politically.

2. If they do anything with the information, someone else does it for them. For example, Mark asks, “What tools do you use to analyze the information?” The answer “Excel” indicates usefulness. But, he emails, “When I get ‘somebody else does it for me,’ I know I’ve found $100K to shave off the bottom line.”

3. No clear answer to “what does so-and-so do?” People usually know the organization’s problems. So when someone says, “I have a meeting with Tom,” there’s usually a wink or smirk.

Have you got indicators of your own? Please post a comment.