I think the 80/20 solution for reliable workflows is:

- Ensure the workflow is idempotent - if it stops or fails at any point, you should be able to start it from scratch and skip / happily redo various elements.

- Store the messages which trigger workflows.

- Track failures (if your log aggregation is good, even that's enough to start).

Then when the odd thing fails (or sometimes a bunch of things fail, because e.g. a core integration goes down) you can lookup the messages and have a little script or tool to go and re-queue them. This is an easy starting point that can keep you going for a long time until you really approach huge scale.

At my last company, I wrote a little tool for pretty printing our JSON logs. You pipe into it and it prints out. It's quite easy to write such a thing in Go, and useful for assigning preferred timezone conversions, and colors for your special common log items.

It's interesting to think that logging is now an undocumented API.

Google's Search and its AI result can alternatively help or hinder me, I find.

If I'm looking for a relatively straightforward and simple piece of information quickly (e.g. "wooden arch mirror stores linden") then the AI summary can be useful, because I don't need more than surface level info.

If I intend to analyze and understand something (e.g. "developer API issues Zoho Thrive"), then the AI summary and the general degradation in the quality of search results from Google really hinder me. I have to work to avoid a lazy and low value answer, whereas what I really want to do is go through various actual websites and reflect on them to gain insight.

I feel like the second scenario was never good with Google. You had to go through several search results describing other problems, and there was a high chance you still didn't find anything relating to yours; and then, it wasn't guaranteed that a years-old problem got fixed.

If I have an issue with my Mac, seeing an Apple Discussions result at the top readies me for disappointment.

Meanwhile, I've found a normal chat with current ChatGPT to be very helpful, as long as it isn't about itself or other OpenAI products.

My usecase would be investigating a potential integration - I want to go and see everyone's comments on the websites themselves. I'm not looking for an answer - I don't know the question - I'm looking for understanding.

I find it to be a fundamental hinderance as they use a cheap model. The cheap model gets confused based on conflicting reddit threads and then I get a wrong answer.

It's better to just ask codex to do the search for me, but this is much slower - but increasingly my go to. I wish there was a fast search api codex could hook into to answer internet questions faster.

Developing software is as much about the journey as the destination. I build a lot of my understanding of the actual problem in the pursuit of solving it.

There are many times when writing a feature that my spidey senses flare up and tell me that this thing is a lot more painful to code then I was expecting (and will be painful to maintain) and that a more elegant process may actually solve the problem, at which point I'll draw up an alternative option and talk to the product owner.

I've definitely started to see the consequences of the converse, which is large amounts of shite brittle code that solved the original spec narrowly, but is now an elephant on our back when we need to add other concerns to the system that cross over.

(BTW, this isn't against the use of coding agents entirely, its more against high-level agentic usage. I tend to use Claude Code to do little well defined tasks whilst I reflect on it).

As a millenial, I first got into something-like-programming by playing around with Game Maker. A few of my colleagues have said similar. I'm curious to see where others in my cohort started...

The show has its moments - mature, intelligent, human moments found little elsewhere - but its an intelligence that struggles to escape very typical network sitcom trappings. One wishes it had gone a little more in the direction of M.A.S.H. and ditched the pretense of having to make jokes every minute.

I have never seen M.A.S.H. but I agree the human moments were unique to this show. I didn't mind the jokes- I thought they kept it light, almost a reminder of how it didn't take itself too seriously.

I still have my jailbroken PSP, back from my school days. Great fun for playing emulated PS1 games. I imagine joysticks and buttons on modern handhelds are considerably better though.

CHRONO Trigger on snes emulator.

Can't wait to see the rooting hacks resulting from mousing over a strawberry and text saying "count the r's".

Another way to read it is that they are shifting blame away from their backend - which is so shitty that it experiences notable service disruptions when some of their existing users send unexpected headers - onto the client software.