deltasquare4

In HN context, a more suitable example is "fail fast" vs "persevere". Both are sound advices in the right context.

I'd say knowledge is not the only thing you need. It's one thing to know, and entirely another thing to be able to recall and apply that knowledge when it matters.

The application of knowledge gets trickier because there is almost always no "right answer". Everything is a decision, a choice. There may even be conflicting but contextually correct ways to process the same decision - i.e. Early bird gets the worm vs second mouse gets the cheese.

Well, we focus on building high-traffic/large-scale apps. The tech stack does not matter. Our skills are easier to demonstrate if customers have existing performance/design bottlenecks. Not something that other dev shops can claim to do easily.

It is. It creates two "problems" - our quality and price both are directly compared to the other shops, and the quality part is not always immediately visible.

As someone who runs a niche dev shop in India (not the kind that you describe though), I have found that new work has been hard to come by since last 6 months, especially for services like ours that are on the premium end. Even the general consensus around my network is the same.

The Indian dev "market" was pretty much commoditized during the pandemic, which prompted us to move our focus away to niche that could still pay premium for high skills and outcomes. It seems that rest of the market hasn't changed much even in this time. I can't speak for the job seekers, but the number of job applicants applying to us have risen sharply over last 2 months or so.

Well, I have found that it changes with numbers and age of the device. Generally speaking, you are right though.

They'd be adequate at running a bunch of services, but cannot bring the connectivity and reliability that AWS provides. I myself have been running a bunch of Chinese Mini-PCs to the same effect since last few years now, and found Reddit Homelab community to be a source of inspiration.

I have an ultra cheap VPS instance that I run wireguard on, and expose these servers to the internet through there. The Mini-PCs are like NUCs, so they hardly consume much power, and I have paid less than 6 months worth of comparable AWS costs to own and run them till now.

The two biggest issues I have are power backup - UPS works for only 3-4 hours, after which the servers shut down, and internet connections - I have 2 100Mbps fiber lines load balanced, but the reliability of consumer internet leaves things to be desired.

I spend roughly 2-3 hours every other month to maintain the whole thing, which is pretty much hands-off. I'd say it's been totally worth it for me, but I still use AWS for mostly S3 and SES.

We follow a model similar to concourse/governance. Yaml based files decide team and repo memberships. The PRs have to be approved by a group of people and it's dictated with CODEOWNERS file.

I have tried many of them on my blog, including MixPanel, Google Analytics, Clicky and Lijit. They all have their pros and cons, but they also require you to include a script on your pages, which can be slow at times.

I'm currently fairly happy with Google Analytics, but I'd move to a better server-based log analysis tool (AWStats on steroids, maybe?), if I found one.

Personally having encountered this issue, I do agree with Alex here. Most of the projects are left out unfinished, so collaboration sure does make sense here.

Sadly, I'm more productive around after midnight.

Great work, sir. You inspired me to try that for my own office :)

That would be awesome. Looking forward to it.