Monthly Archives: January 2014

Saved from smart technology by antiquated technology

In order to test something for a contractor, I needed to make a successful call from my Droid to a contact whose name was in Unicode in my contact list.  


I installed a multilingual keyboard and added Cyrillic, then added a new contact.  My housemate was standing right there, so I filled in his number and hit the “call” button.  And of course, my phone immediately displayed his real latin-character name and his photo, because I have 7 entries for him in my contact list (!), several of which are linked to that number.  For a user, this is a great feature.  For a developer, it was a little distressing as I tried to think of someone whose number I had who didn’t have social network entries automatically syncing their contact data to my phone, thus ruining my test plan.


Suddenly, I remembered an antiquated technology that could help: automated services that you can call for free, very popular during the era when even dumbphones were rare, but verging on useless as smartphones dominate.  I just plugged in the number of a regional weather service and was able to place a successful call.  Yay!

Grace Hopper talk options.

A friend suggested I propose a talk at Grace Hopper this year.  She was thinking I could do a sort of introduction to data visualization (which I already outlined for my Analytics department shortly before I left).  I feel like there are more suitable women for giving a lecture on that, but maybe I’m not giving myself enough credit.  I’m sure “not giving yourself enough credit” is a common theme at Grace Hopper!

Here are some topic ideas, one educational, one personal-inspirational, and one idea-inspirational.  What do you guys think I should talk about?

Introduction to Data Visualization (Data Science track)
I’ve been paid to work on the programmatic visualization of abstract data since 2005 (doing it for personal use since…1999?), so I can give people some solid advice, but I’m not always up on the latest research.  Most of the time in industry, though, the latest sexy thing is not useful.  This is almost as true in data visualization as it is in machine learning (Seriously…the regression model or equivalent is the best choice at least 90% of the time.)  So I’m sure I can cover all the crucial pieces for an intro course.  The upside of this option is that I’ve got an outline lying around somewhere from a lecture I was going to give at my previous job.

Straight From Lower-middle Management to Tech Founder (Career track)
This would be a my personal story with observations, anecdotes and advice.  I can’t guarantee it will be a complete success story, but it’s been a very interesting process for me, and puts my previous experiences in a different light where my gender probably made more difference than I thought.  Quite relevant for the conference.

Social Analytics: So Much More We Can Do (Career or Data Science track)
Many of you know I’ve been thinking about personal social analytics for a long time, and I spent a lot of time thinking about social analytics for the dating sites I worked on at my previous company.  I can talk about the popularity of social media and the minimal analytics they provide, and of course mention Klout and its strengths and weaknesses.  And I can talk about Google Analytics and CRMs.  And then, I can talk about what isn’t on the market.

My open source project only got so far before I put it on hold in favor of other work, but I’ve been brainstorming a lot of possibilities for analyzing personal conversation history in the last decade and a half.  These ideas have all been feasible during that time, and they’re getting faster to compute and easier to scale all the time.  So why isn’t anyone making it happen?  In my experience, social media users have been very excited with any level of navel-gazing: finding the list of people who follow them that they don’t follow, or the other way around (drama!); seeing who talks to them the most on a given platform; etc.  There is a market that would pay just for increased analytical power! And they’d definitely pay to have all of their communications data integrated into one location for analysis or even plain review.

And a lot more can be done with the same data.  Merely extend it with some labels and charts and BAM! Social productivity tool.  New target market.  Extend that with some very basic recommendation capabilities and BAM! Customer management tool.  Another target market.  If I thought of all this at least a decade ago, straight from college and before I ever heard of social analytics, others must have too.  Why isn’t the market filled with options?

sudo make me a proper social analytics platform!

A few good reads

First let’s catch up on some of the great stories other people have been posting while I’ve been AWOL!



If you’re curious whether to join a venture, consider whether the executives have been able to attract their former coworkers.

When should you delegate?  It’s a harder question than you might think.

10 ways to make sure your best people quit.  And some things you can do to retain them.

Tips for organizing your small business.  Most of these apply to independent contractors too.

61% of web traffic is bots…and half of those are scrapers, spammers, and other “bad guys”.



Teaching kids to program before they can read


Information Visualization

Redesigning a visualization to convey the story faster and more clearly



Holidays and important personal business kept me away from you guys for a while, but I have things to say!  And I will start saying them again very soon.  Thanks for waiting!