tmclaugh

This stuff is hard and we're all still figuring it out. Just recently we realized we built a monolith because we weren't paying enough attention to whether all the features should be a part of a single system. We're going to revisit that.

Don't look at your function as an app, it's an function and the best functions do a single thing. The definition of a single thing depends on the scale of view you're taking. But that may help you to realize your functions shouldn't be very complex and possibly small enough to test more easily.

However, that results in increased system complexity as you start adding SNS topics, SQS queues, DynamoDB tables, and all sorts of event triggers. Something I've adopted is using draw.io (others use Lucid Chart) to draw out the system logic. Maybe this helps?

https://www.draw.io/ https://www.lucidchart.com/

Maybe this makes the confusion worse?

I blog for work. I took a step back from product TechOps (we're a SaaS company) to write for us. We make a service that people like me, an ops engineer, would use. My rules for writing are I write about things that are interesting to me that fall within the general problem space and I don't sell product. I've struck on writing highly technical posts and including code or CLI examples of what I'm doing. If there's code I release GitHub repositories. A reader should be able to take what I wrote and instantly apply what they've learned. You can probably guess those posts outperform other posts anywhere from 2-5x depending on the popularity of the subject matter. If I strike gold we're looking at 20-30x performance.

The next key is promotion. I have a decent Twitter following (>1K) and have developed contacts with people in my profession. I schedule tweets for what I write that I think is really good. Hootsuite and Buffer can do that easily. If I want to strike gold I post to Reddit. I've posted here before but get minimal traction. I don't have enough time to invest in HN and just browse here casually. I spend time on Reddit regularly because of non-tech boards I love. (Hi /r/boston.) If I write something good for Reddit our traffic numbers see a significant spike. Those are the 20-30x performers.

All that said, ask yourself why you're writing. If you're writing for a specific audience then respect your audience. I've taken the talk up approach over talk down. I don't boil hard topics down to telling you what you should do without providing you a how. I'm also respectful of the communities where my audience congregates. That means I stay away from talking about our product. If my name is on something that discusses product I put those on LinkedIn and Twitter once or twice.

If you're writing for yourself, ask yourself again, are you writing for your enjoyment or your reputation? If you're writing for your enjoyment then whatever. Traffic numbers don't mean you have nothing of value to say. Just means you don't have an audience and that's going to happen today with the information saturation we all face. If you're writing for reputation, or because you'd like an audience to engage with what you've written for pure enjoyment, then promotion is key. I've used twitter because I'm on their regularly. My writing is only a small part of what I put onto twitter. I like posting pics of my cats, my food, and witty things I say to my girlfriend.

Wish I could share more and maybe be more articulate but need to work. FWIW, I have had a blog post become popular in the past week. Someone just filed a bug against once of the example repos. Knowing people are reading and inspecting my work to use it makes me happy and know that I'm doing something worthwhile. :)

I moved to product marketing. I have the benefit of working for a company that makes a product for ops engineers which was my previous role as well as exposure to marketing concepts from working at a marketing tech company previously. I understand the product audience in ways that traditional marketers do not and I understand our marketers in ways that most engineers do not.

It's fun and it's a nice change of pace. I've retained some engineering responsibilities. So what might be a passion project for someone on an engineering team, I develop something and then start writing blog content to help educate our audience around the subject. Our product has the ability to fire a JSON doc at an endpoint which gives you the ability to integrate us with other systems... To many ops engineers can't write a web service to bridge us with that other system. So, I'll write a small service and then after it's up on GitHub I'll write a tutorial and walkthrough of the service. If you just download what I wrote and run it, awesome. I've helped solve a customer issue. If you take my blog post and learn to write your own service, that's even better!

I'm an infrastructure engineer that's now working in the security space. What myself and others have found is security appears to be slightly behind ops in terms of adapting to the changing world of software deliver. I've highlighted a few key areas that I plan on expanding. Thoughts? Comments? Have I missed the marks?