15 things I’m doing after OpenVisConf 2017

Today was the last day of the incredible OpenVisConf in Boston. I’m still digesting everything from the two days of talks, but here is my general plan.

1. Taking screenshots of my work in progress. So many of the talks had in-progress shots that showed their design and thought process and it made everything easier to follow.

2. Do more medium sized dataviz projects. I might not be able to crank out one every month, but I should do more smaller viz projects like I used to.

3. Turn my network graphs into scatterplots

4. Do at least one map project this year

5. Figure out what REGL is before the author is killed by a volcano

6. Try out WTFCSV

7. Build a particle viz for support tickets with Trello data.

8. Do more of those “draw the chart” tests that the New York Times puts out.

9. Use simulation.find() in D3!

10. Learn more about veroni

11. Do some text data mining on Shakespeare using tidytext

12. Put chart colors/styles in our styleguide

13. See if I can use Vega-Lite to generate all possible chart types at once for a data set, such as automobile deaths.

14. Make some generative art based on http://bengarvey.com/bounce/gravity.html

15. Build some color palettes using Colorgorical

BONUS 16. Hope I get into the d3.express beta

Spoil Your Meetings

You just got a calendar invite your boss. 

What’s going on? Are you getting fired? Your last performance review was fine! 

You walk to your boss’s office. HR is sitting there with her.

Kermit being nervous

Oh crap, oh crap, oh crap.

HR walks out as you walk in. You sit down, and your boss says, “Hey I just wanted to bounce an idea off of you.


Meetings shouldn’t have surprises. Spoil them.

Tell everyone what you’re going to say before you go. Give away the ending. Put all the good parts in trailer. And the bad parts. And the song that’s going to play in the montage while your team builds the A-Team-esque vehicle that is needed to do the job. 

If your meeting doesn’t have spoilers, then it’s probably not worth having.

Real Work

I want to talk about real work and fake work.

Does this sound like your work culture?
Does this sound like your work culture?

We have 1:1’s, standup meetings, team meetings, check-ins, all-hands, and serendipitous meetings at the coffee machine. How long did you spend writing google docs, emails, slack chats, gchats, hipchats, and wikis? What percentage of your day comprised of tuning your vimrc/emacs/bashrc and organizing your trello lists? Are you patting yourself on the back for reaching Inbox Zero?

How many times today did a customer notice something you did?

Sam Altman writes about fake work for startup founders:

In general, startups get distracted by fake work. Fake work is both easier and more fun than real work for many founders. Two particularly bad cases are raising money and getting personal press; we’ve seen many promising founders fall in love with one or (usually) both of these, which nearly always ends badly. But the list of fake work is long.

I tell founders to consider how directly a task relates to growing. Obviously, building and selling are the best. Things like hiring are also very high on the list—you will need to hire to sustain your growth rate at some point. Interviewing lots of lawyers has got to be near the bottom.

So now we have rules for founders/CEO’s:
Real work: Building/Coding and sales, probably hiring.
Fake work: Interviewing lots of lawyers.

Does Sam think lawyers are unnecessary? No, but he realizes picking the 1st or 2nd best lawyer is indistinguishable from choosing the 12th best when it comes to your business. Finding the best employees is a better use of a CEO’s time.

But what about everyone else? How do we distinguish between real work and fake work?

Toyota’s legendary production system strives to eliminate waste everywhere. When talking about waste, Toyota’s Shigeo Shingo observed that only the last turn of a bolt tightens it.

"Only the last turn of a bolt tightens it" - Shigeo Singo
“Only the last turn of a bolt tightens it” – Shigeo Singo

That final turn in software engineering is shipping code. There are other activities so closely tied to deploying code that they are undoubtedly Real Work, such as programming and reviewing another engineer’s code. What about everything else?

I define Real Work as
A. Work that creates an immediate and positive change in your product/service
B. Work that is a direct constraint to the completion of A
C. Work you have been uniquely hired to do.

Type C has to be included because many jobs don’t touch A and B, but someone has to do them. Google famously hired a masseuse, Bonnie Brown, when they had only 40 employees. Was she improving our search results in 1999? Maybe not, but Larry and Sergei thought it was critical to their success. For Bonnie, she should be confident that shipping each great massage (type C) was Real Work as it had been identified by the business as type B work.

So why make the distinction between A, B, and C at all? Why not just say Real Work is what you were hired to do and leave the question of whether it’s valuable or not to the person who hired you. It’s because most of our jobs have a degree of autonomy in the the tasks we take on. We should prioritize type A tasks and have a high standard for B tasks. We should eliminate any B tasks that don’t enhance our ability to complete A tasks. Fake work is anything that doesn’t live up to that standard.

Things you can do tomorrow to minimize Fake Work
1. Skip a meeting you don’t need to attend
2. Turn your weekly meeting into a bi-weekly meeting, or even monthly
3. Be concise in communications. Be polite, but formality in email is dead.
4. Don’t jump into conversations just because you can. Comments can be useful, but they also add cognitive load to an issue.
5. Eliminate, automate, or delegate menial tasks.
6. “Table it!” Brian Sloane introduced this to me. If two people are having a drawn out discussion in a meeting, put your hand on the table (or in the air) and say “Table it” to indicate that this discussion should continue in another channel at a later time.

Things you can do tomorrow to maximize Real Work
1. Figure out what real work means for you in your job.
2. Schedule time specific tasks. Treat them like appointments.
3. Don’t check your email until you’ve been at work for an hour. This is an old Tim Ferris trick. If you’re feeling brave, forego Slack, too.
4. Observe the habits of people who get things done. Shamelessly steal them.

Minimizing fake work doesn’t mean you should eat lunch at your desk everyday. The advantage of focusing on Real Work is so you can perform at a high level and still live your life.

New Job

Garvey Corporation was started long ago as a gas station in 1926 by an Irish immigrant named Gordon Garvey. He chose a location on Route 73 directly between Philadelphia and Atlantic City, since it would be an ideal place for people to stop for gas coming from either direction. The company moved away from being a service station over the years and evetually got into making sheet metal parts. By the 50’s and 60’s, Gordon’s sons Fran and Bud took over the company and developed their own line of modular conveyor systems to sell to the local glass making industry. In 1980, Mark and Bill Garvey, sons of Bud and Fran, took the helm of the company and expanded its reach into more extensive and complex product handling systems with customers all over the world.

As of last week, I’m the new General Manager.

Infinity Rx

It’s a role I never would have expected to come this soon if at all. It’s nerve racking and exhilarating at the same time and an immense test on my abilities. I like the work I have been doing far more than I ever thought I would and that helps. Wish me luck, as I am responsible for the livelyhood of 80 people and their families. I take that 100% seriously and think about it every day.

Not that I don’t have help! We have a great group, including my brother Jake.

If you’re curious about what we do, check it out: http://www.garvey.com

Also, here’s a case study I wrote for a wine magazine about what our equipment did for Rodney Strong Vineyards.

What I do

I don’t make too many work posts, but here is a quick explanation of what I do every day. I’m the Engineering manager for Garvey Corporation, a company started in 1926 by my great grandfather, Gordon Garvey. We make product handling equipment for high speed packagers in the food, beverage, pharmaceutical, and consumer products industries. Mainly that means conveyors, accumulators, loaders, unloaders, inspection stations, etc. Our customers are mostly big food companies like Kraft, General Mills; goods producers like Proctor and Gamble; beverage makers like Ocean Spray; and pharmaceutical companies like Merck and Wyeth. We also do a ton of work in the wine industry for Rodney Strong, Don Sebastiani, Mondavi, etc. which was spurred on by my Dad’s work in creating a new type of buffering system to accumulate wine bottles between the filling and labeling machines. We called this new accumulator the Garvey Infinity and have sold hundreds of them since 2001.

The reason for this post is that we’ve come up with a breakthrough accumulation machine for the pharmaceutical industry that I’ve been working on for months. It’s finally finished and it turned out even better than I hoped. It applies all the lessons we learned in the beverage industry and focuses them on super tiny pharmaceutical bottles, sometimes as small as 15mm wide.

Here’s a video of the Infinity Rx accumulating and single filing 2ml 15mm vials at over 850 vials per minute. It’s for a show, so that’s why it loops back into itself.