Chasing Ghosts

One of the most annoying type of technology problems to battle are the ghosts. Ghosts are problems that appear, are easily fixed for a specific situation, and then later re-appear. Ghosts are persistent and they haunt you. It is incredibly challenging to debug ghosts because their trigger is usually unknown. My first step to handling ghosts is to list them.

For most of my ghosts I have a formula that fixes them. The majority of the time this formula involves a device reboot. For sufficiently infrequent problems this remedy is not a big deal. Once or twice a week is fine. Some issues recur daily or even multiple times a day. These are the ones that are the most painful.

A few things I try to deduce are whether the issues are related to hardware or software. If I suspect it’s a hardware issue then I try to go through device support to get the unit replaced if it is under warranty, or I give in and purchase a new device. Software is more challenging. A new device is not likely to fix it. I need to report the issue to the manufacturer through their support channel. Sometimes they have ideas on how to fix it. Other times I am just trying to get their engineers enough information about the issue to debug it for a software update.

Unfortunately most of the time I just decide to live with the issue. The time to debug it ends up being longer than the amount of annoyance to deal with each individual occurrence. I wish I had more time to fix some of these issues, but ultimately life is too short to keep chasing ghosts.

Media Friday - Casey Neistat on Vergecast

I really enjoyed this interview with Casey Neistat by Nilay Patel. It is rare for an entrepreneur to be so open with sharing the challenges around something that did not work out. In Casey’s case he talks about how his former company, Beme, was acquired by CNN. After a while the organization was absorbed into CNN proper and Casey and his chief technology officer parted ways with CNN. While I enjoy Casey Neistat’s video blogs for their optimism, the realism of discussing failure makes him a more relatable person.

Appreciating Art

Yesterday I attended Art in the Age of the Internet at the ICA in Seaport. They held an event for the technology community in Boston and the fun part was that in addition to the regular exhibit they had some pop-up VR demos and offered talks to contextualize the work. Art has two components: the work itself and then the story behind it. Getting that story can help you understand and interpret the art itself.

One of the biggest challenges with art and art museums to me has been accessibility. Folks who are into art seemed to already know and understand how to view and appreciate it. I did not. When I was at Berkman Klein I had the privilege of having some artists as colleagues and began to better understand it. The ICA, by contextualizing art through talks and facilitated conversations in addition to the written descriptions, has taken the initiative to make art accessible to everyone.

Muscle Memory

I have spent a lot of my life building new organizations and teams. When I was in college I started WPI’s chapter of the College Democrats. I spent a summer going into my sophomore year helping build the canvass team for now Gov. Malloy’s primary campaign against John Destefano. After I graduated law school I was a campaign manager for a mayoral race in New Haven. All these roles involved helping organizations build muscle memory.

Muscle memory is the difference between the first and hundred and first time you do something. You get on a bicycle the first time. It feels weird. You fall down. You get back on. You move forward a little. You fall down again. You get back on. You go farther. Eventually you stop falling down. Your brain and your body know what to do. The tension melts away. The thing changes from stressful to easy. You go farther than you ever imagined.

There are two types of situations I encounter: no muscle memory and bad muscle memory. No muscle memory is easy. You notice a lack of process. You apply some thought. You pick a process. You implement it and improve. Bad muscle memory is hard. You try something new. You have to identify and stop yourself from applying your old habit. The unfamiliar fights the familiar. Everyone else engages in behavior that encourages the old habit.

If you try to build too much muscle memory at once you will get overwhelmed. No person nor organization has the mental bandwidth and capacity to change everything at the same time. You have to prioritize a singular thing and build the muscle memory around that. Then once that has been mastered you can move on to the next thing.


One of my most and least favorite part of using a computer is picking my tools. I have developed a suite of tools that work well for me. Some tools fit you like a glove. Other tools you adapt to and learn to stop hating after a while. As a computer user I have spent years using a computer and learning my tools. The result is that a large number of things other people struggle with are solved problems for me.

The limitation of tools is they cannot fix human problems. The increasing volume of email might be more easily processed by GMail’s interface than Microsoft Outlook, but it does not change the fact that it linearly takes more time to process more email. OmniFocus gives me a great place to store my personal task list, but it does not have the power to make me do my tasks. We must be cautious with imputing too much power and promise to our tools.

The other down side is we risk becoming tightly coupled to our tools. For example I have installed the zsh shell in the terminal in my Mac. It makes me a large order of magnitude more effective in using the command line interface on my computer. However when I login to a remote linux machine it is not available unless I set it up. The benefits outweigh the cost, but when you have used something that works better for you, your tolerance of other things goes down.

Follow posts: RSS Feed
This work by Matt Zagaja is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.