3 ms·
How do you know what to code or whether you should even be coding at all if you aren’t willing to spend the time knowing the business and the customers?
by scarface74 7y ago
How do you know what to code or whether you should even be coding at all if you aren’t willing to spend the time knowing the business and the customers?
- elif 7y agoBecause there are hard problems to solve. I'm into computer science not capitalism.
- scarface74 7y agoI bet the company you work for is “into capitalism”.
- elif 7y agoYou don't know anything about my particular situation, and happen to be very wrong. But assuming the role of your rhetorical me, there are thousands more companies just like "mine". they all provide paychecks.
- scarface74 7y agoAnd how do those companies earn money to give you a paycheck?
- elif 7y agoMost of these paychecks come from unprofitable companies. I'm not sure what point you are wanting me to walk into. Are we still talking about programmer management strategy?
- scarface74 7y agoAnd whether they are “unprofitable” or not doesn’t mean they don’t either seek to become profitable or get bought out - ie your livelihood is still based on capitalism.
- comicjk 7y agoPet peeve of mine: solving practical problems often involves capitalism in our system, but that doesn't mean practicality is capitalism. If you're writing ad code, you're working on capitalism only. If you're writing railway logistics code, you're doing practical work that any society needs, capitalist or not. And yes, this might involve understanding customers (or as a different system might call them, "people"). The important question would be "is this work useful," not "does this work involve capitalism."
- scarface74 7y agoSpeaking of railway logistics. Railroad car repair billing is esoteric and complicated. I wrote software dealing with this at one job. https://www.railinc.com/rportal/documents/18/260737/CRB_ProceduresManual.pdf https://www.railinc.com/rportal/documents/18/260737/CRB_Proc...
- lexicality 7y agoDo you not consider finding out what the person asking for things actually wants to be a hard problem to solve? I live to build things and continuously fend off attempts to slide me into management, but I find actually programming things to be only half the part of the puzzle. Humans are extremely bad at articulating what they want and will generally tell you the wrong thing up front because they haven't thought it through fully. A day spent with a stakeholder carefully picking their brains and properly crafting a plan can be just as satisfying as a day of straight coding - sometimes more so. (tbh this is probably also why I find UX work so interesting - getting a user to click on a button is considerably harder than making the button in the first place)
- matz1 7y agoIn the case of me >How do you know what to code Someone need something to be made or done, I help them by writing code. >whether you should even be coding I've been doing it for long time >if you aren’t willing to spend the time knowing the business and the customers? That's precisely the reason I prefer code. I'm not that interested or even willing to spend time knowing the business and the customers.
- thestepafter 7y agoI would argue that knowing the business and the customers makes you a much better programmer though. I would rather a programmer spend 75% of their time programming and 25% seeking to understand why they are programming instead of 90%+ heads down with little thought or attention to why they are programming.
- vlunkr 7y agoThis shouldn’t be a problem in a business with well defined roles. Boss/manager says customer wants x. I don’t care why they want that, and I may even think it’s dumb, but I can make it happen. After all, I get paid to solve problems with code, not interpret customer requests.
- scarface74 7y agoI was once interviewing for a job as a senior engineer/lead role. The CTO - my eventual manager - was asking me a lot of high level technical questions even though he was a very competent developer. Then, another developer who had been there for a decade asked me how I would go about writing “address validation routines” - It was a real world problem he was facing. I told him that I wouldn’t. That’s not the vertical the business was in. The best code is the code that you don’t have to write. Address validation software is a solved problem and there are plenty of third party CASS solutions and yes they cost money but being able to write address validation software is not the company’s competitive advantage or its differentiator in the market. It’s better to outsource it. That’s one less thing we would have to maintain and debug. The developer wasn’t impressed. The CTO was. I’ve spent my entire career asking questions and making sure we were building the right things. My first job out of college over 20 years ago I was the sole developer on a project to write a data entry system that was used by a new department with a dozen new employees. I had to actually talk to our potential client to gather specs myself.
- vlunkr 7y ago> I’ve spent my entire career asking questions and making sure we were building the right things. But this gets back to what OP was saying. Spending time asking questions == spending time in meetings. Not writing code. I’m not advocating for never asking questions or suggesting better ways, just saying that I prefer to work where someone else handles the bulk of that so I can focus on what I specialize in.
- magicalhippo 7y ago> After all, I get paid to solve problems with code, not interpret customer requests. But how to know what to solve... At least for my job, interpreting customer requests is key. The customer knows what they want, but they don't always realize what they want is not the best solution. Often I can steer the customer to a vastly superior solution to what they initially had in mind. Not seldom this superior solution can re-use existing code, sometimes entirely.