I was thinking how I like to be a swiss-army knife tech guy. I am studying for the LPI Linux exam, and wondered suddenly why employers don't value an "answers guy" (or gal). You know, a person you can go to with questions and problems and he (or she) can figure out or just knows the answer. Then it struck me that Valve's model kinda does that. I've been reading a lot about the company and I wish there was a case study about them. Their flat management style seems extremely effective and unfortunately unique. Being a private company, it can run its ship however it wants so Valve runs it in an interesting way. They don't hire people for particular jobs. They hire people and basically let them figure out where they can contribute the most. Desk have wheels because on a project per project basis you may be moving around forming different teams. It sounds like an Agile-dream workplace. But a key here is not the way they work but how they hire. While they look for people with certain skills, by not hiring them for a specific job, their "answer guy/gal" abilities are fully tapped. Is your sys-admin really good at marketing? In a traditional company you'd have no way of knowing and even worse no way of using that talent. Defined roles segregate and also limit the scope of the possible contribution. I always felt ham-stringed in my previous roles, having to play politics sometimes successfully and sometimes not to contribute to different areas than my little circumscribed job responsibilities. The traditional way also treats people like machines, do this, get that, and so makes the automatization of their job at least by the rough 10,000 feet view possible. So weirdly by enforcing the limits of the job title you get the value that a machine would produce or could produce and you get less of that human magic that no machine could substitute.
While I worked in Quality Assurance, I wrote programs (with the help of tech guys like Jason Rowlands or Roupang Wang) that allowed my job to be more automated, even though this reduced my "job security" it increased quality by automating certain tasks. I'm thinking of dub-card creation, where me and Jason hashed out a Javascript script for Photoshop that allowed you to output as many perfectly formated cards as you wanted out of one source psd file. Together it took us around thirty minutes to hash out what I hadn't been able to do on my own. This actually made me more replaceable and I had to seek out Jason out of his normal work duties, and create space in my schedule to tackle the creation of the script. Yet it saved massive time and increased quality in the future. The sort of step that betters productivity and quality that businesses so crave. And it came from my own time. I basically had to slightly subvert the system to do it.
At another time I was teaching myself programming in the break-room. I had my massive book(I use it for weightlifting now) 'Learning to Program Python' plopped open while I ate. Roupang asked me why I was reading that. I told him I needed a simple program that would reverse letters in a line but not the line order to convert Hebrew and Arabic text from the input order (left to right) to the reading order (right to left) so I could use them in graphics. The order of Hebrew letters seemed to be a perpetual problem in many of the international graphics I got at the time. The program is dead simple, I even recently coded a Sinatra app that does it, but invaluable if you need it and need it quickly. Roupang, a programmer with the company, offered to write it for me. He said it would take him 5 minutes. And he was right. Yet after it got out he'd done that for me access to the programmers by production (and hence Quality Control) was restricted and the production team was 'reprimanded' for not seeking approval. Shortly there after Roupang's office was moved behind a glass wall into a different department. I was specifically denied entry to it. Yet a 5 minute program that was so useful that when I showed it to the graphics people, they immediately wanted to copy it and include it in their workflow. But again by giving them my little program, I increased quality but reduced my "job security." Because my "job security" was so tied to the specific tasks I'd been assigned to do not my contributions to the company.
This top-down segregated management is clearly not optimal, but stems from the view that people are like 'cogs' in that they are interchangeable and only do one job. And this view starts with the hiring process. What can you do (for us)? What can you do (to make us money)? What can you do (for me)? Seems to be the only filter that is used. A more valuable one is to hire great people and let them go where they can contribute the most. It's not only more human, I have a feeling it can also be largely more profitable. Other qualities like how much a person knows, how quickly they can learn, how good they are at benefiting the team, their creativity and passion then become valuable assets, not ones that need to subvert the system to survive.
While I worked in Quality Assurance, I wrote programs (with the help of tech guys like Jason Rowlands or Roupang Wang) that allowed my job to be more automated, even though this reduced my "job security" it increased quality by automating certain tasks. I'm thinking of dub-card creation, where me and Jason hashed out a Javascript script for Photoshop that allowed you to output as many perfectly formated cards as you wanted out of one source psd file. Together it took us around thirty minutes to hash out what I hadn't been able to do on my own. This actually made me more replaceable and I had to seek out Jason out of his normal work duties, and create space in my schedule to tackle the creation of the script. Yet it saved massive time and increased quality in the future. The sort of step that betters productivity and quality that businesses so crave. And it came from my own time. I basically had to slightly subvert the system to do it.
At another time I was teaching myself programming in the break-room. I had my massive book(I use it for weightlifting now) 'Learning to Program Python' plopped open while I ate. Roupang asked me why I was reading that. I told him I needed a simple program that would reverse letters in a line but not the line order to convert Hebrew and Arabic text from the input order (left to right) to the reading order (right to left) so I could use them in graphics. The order of Hebrew letters seemed to be a perpetual problem in many of the international graphics I got at the time. The program is dead simple, I even recently coded a Sinatra app that does it, but invaluable if you need it and need it quickly. Roupang, a programmer with the company, offered to write it for me. He said it would take him 5 minutes. And he was right. Yet after it got out he'd done that for me access to the programmers by production (and hence Quality Control) was restricted and the production team was 'reprimanded' for not seeking approval. Shortly there after Roupang's office was moved behind a glass wall into a different department. I was specifically denied entry to it. Yet a 5 minute program that was so useful that when I showed it to the graphics people, they immediately wanted to copy it and include it in their workflow. But again by giving them my little program, I increased quality but reduced my "job security." Because my "job security" was so tied to the specific tasks I'd been assigned to do not my contributions to the company.
This top-down segregated management is clearly not optimal, but stems from the view that people are like 'cogs' in that they are interchangeable and only do one job. And this view starts with the hiring process. What can you do (for us)? What can you do (to make us money)? What can you do (for me)? Seems to be the only filter that is used. A more valuable one is to hire great people and let them go where they can contribute the most. It's not only more human, I have a feeling it can also be largely more profitable. Other qualities like how much a person knows, how quickly they can learn, how good they are at benefiting the team, their creativity and passion then become valuable assets, not ones that need to subvert the system to survive.
Comments
Post a Comment