When a side project takes on a life of its own, you feel both intensely gratified and frightened all at once.
This has been my experience with all the side projects I’ve been involved with anyway. On the one hand, the reason you build product at all is to solve a problem users have, ideally in an enjoyable way.
On the other hand, you’re faced with support concerns. Before you know it, your users come to depend on your product, and when it’s not working, they want answers.
Connect, OraTweet and the iOS People app have all gone this route.
Most of the time, they run untouched. People happily get work done, and you pat yourself on the back for building something people actually want to use. Every once in a while, something breaks and people freak, and even if it’s not something you can control, it’s your fault. Because to the end user, your stuff isn’t working.
This happened to Connect and OraTweet recently. They went down due to some database issues, and due to time zone constraints, they stayed down for about 12 hours. Noel (@noelportugal) and I were off grid at the time, but over in APAC and EMEA, people were frustrated because they couldn’t do work.
When we realized the problem, we had to track down the right people in the datacenter to fix it because even side projects need IT sometimes.
The People app hit something similar last week. IT made a change that broke it, but because the app is Clayton’s (@cdonley) side project, no one thought to notify him.
All these projects have spread well beyond the small group who realize they aren’t fully supported by IT, so users were understandably frustrated because they couldn’t get answers from the usual IT channels. IT was also frustrated since they couldn’t find the right contacts.
This is a classic startup problem, but it’s a different breed when you’re spinning up projects within a company that has standard IT practices. Users expect the support they’ve come to expect from IT, but they also want useful products, which sometimes don’t come from IT. Not judging, just saying.
There’s a reason. IT is designed to support users, not for users. I’ll explain.
In the mid-90s, I did PC, server and printer support for a few hundred people, mostly in person, but sometimes over the phone. When something broke, I fixed it. I managed my own queue of requests, which essentially means I tried to help everyone as quickly as possible. Niceness was appreciated, bribes were accepted, hierarchy took precedence, etc.
Other people in other offices did the same, and we were loosely affiliated but not officially a team. At some point, someone realized this model didn’t scale very well, and that IT needed to be fully centralized and automated to do its job best. So, we started carrying pagers, logging tickets, automating the request queue, accounting for time, etc.
Users hated it. If someone stopped me in the halls with a problem, I has to have them file a ticket with the helpdesk first, which would route to me or to another person locally. This was highly inefficient compared to the old model, especially for quick fixes, but it helped the larger IT organization track its resources and costs and provide faster, more efficient support.
Users eventually got used to the system, and they got better at self-service support. IT controlled its costs and improved its support metrics.
Win-win, right?
Maybe, maybe not, but my point is about what IT’s purpose is, i.e. support.
Let’s use home support as an example. Many of you support your family and possibly friends. When they ask for recommendations, you suggest what you know, and admit it, you factor in how difficult it might be to support. I know I do.
So, your recommendations aren’t always based on what the user wants, but rather what you can support.
Therefore, IT is not for users; it’s for supporting them.
This sometimes creates a gap between what the user wants and what IT provides, a gap that has allowed social and mobile to flood into enterprises like a tidal wave.
To be clear, I’m not arguing for or against IT. I’m just musing on the ramifications and reasons behind what I’ve noticed with side projects and how they fit into enterprise IT.
I’m sure you have thoughts. Hit the comments.
If all my family would buy iPads and have done with it, Captain Support could retire. A simple phone message saying, “If you can’t find an app for that, it can’t be done!” would suffice… 🙂
If only it were that simple, and even then, they’d manage to find holes and break it somehow. Device loss would be problematic for you too. Captain Support’s work will never be done.