masto

I was working on a bit of code to display results from a PCR test for veterinary use. Claude started randomly failing in the middle of conversations. Then I noticed the error code “bio”. Apparently it can’t tell the difference between bio weapons research and a web site to look at your fish’s lab results.

There needs to be a pithy name for this. I have been calling it context leak.

So management can cancel all of last week’s projects when they told us all we had to be using skills because the CEO read about them in the in flight magazine. Routines are the future, baby. DevOps already made a big announcement that they’re centralizing the Routines Hub. If you can’t keep up, we’ll get someone who says they can.

First I went to the Claude web site, because we've all been trained not to click on links promising things, right? Couldn't find anything there so I checked the domain carefully and clicked on the link, which takes me to my settings page with no discernible discount. I am no longer surprised by anything not working, perhaps Claude itself vibe-sent this mass email.

When I was in my 20s I was an insufferable know-it-all who found fault with everything.

I still spot problems and “push back”, but I have the experience now to know how to get people to listen and not just write me off as an annoying prima donna.

“Cloud to butt” was popular in the early cloud days. It went around Google internally, and caused some… interesting issues.

I’ll express my personal preference for VS Code with Cline. I don’t know exactly why, but its workflow feels right even though they’re all almost the same interface. I do like the huge choice of models and payment options. For something I use burstily, it makes sense for me to pay as I go.

I enjoy reading about other people’s approaches to motivation and creativity.

But I very much dislike when they phase it as “you need to” or “this is how it works”. Thinking everyone else’s brain operates the way yours does seems to be a frequent bias among bloggers. And managers.

I encourage those who write about their experiences to keep it in the first person.

I found a tiny bug in a library. A single, trivial, “the docs say this utility function does X, but it actually does Y”. I’m not even allowed to file a bug report. It took me some time to figure out how to even ask for permission, and they referred it to some committee where it’s in limbo.

On rare occasions, I use it for the virtual display; it's actually usable to sit outside and work with a giant display on the deck, or to dial myself onto the beach. But it's not exactly comfortable for extended use, and most of the time I'd rather sit at my nice desktop with multiple monitors etc.

I also have a Quest 3 and if I could only own one device, I'd take the Q3 hands-down. The games are fun, they get you up and moving, and although I'm not going to argue that the quality of the screens is the same or anything, it's more than good enough. I'll happily give up the virtual laptop screen in exchange for the library of VR games on the Quest.

I'm not much for consuming media so that aspect is lost on me. Unfortunately, that seems to be the primary use case Apple has focused on, if you can call the anemic dribble of content they've put out focus.

I've been using Clerk and it seems fine. I'm sure there's some drama, because everything comes with drama, but I just want to get on with building stuff.

Sometimes I spend half an hour writing a prompt and realize that I’ve basically rubber-ducked the problem to the point where I know exactly what I want, so I just write the code myself.

I have been doing my best to give these tools a fair shake, because I want to have an informed opinion (and certainly some fear of being left behind). I find that their utility in a given area is inversely proportional to my skill level. I have rewritten or fixed most of the backend business logic that AI spits out. Even if it’s mostly ok on a first pass, I’ve been doing this gig for decades now and I am pretty good at spotting future technical debt.

On the other hand, I’m consistently impressed by its ability to save me time with UI code. Or maybe it’s not that it saves me time, but it gets me to do more ambitious things. I’d typically just throw stuff on the page with the excuse that I’m not a designer, and hope that eventually I can bring in someone else to make it look better. Now I can tell the robot I want to have drag and drop here and autocomplete there, and a share to flooberflop button, and it’ll do enough of the implementation that even if I have to fix it up, I’m not as intimidated to start.

I don’t know if this counts as experienced, but I’ve spent about a year exploring Next.js, the last 6 months of which transitioning from exploring to building a serious project.

I understand the author’s frustrations. I have had similar ones when it comes to middleware and other parts of Next.js. I’ve also had those kinds of frustrations with every piece of software and framework I’ve ever used. A lot of times they stem from trying to shove a square peg into a round hole, and it only gets better when I finally develop the right mental model for how the thing works.

As a web developer going back to CGI scripts in the 90s, all this server and client side rendering, edge runtimes, etc., is quite foreign. But when I find myself screaming “why won’t it let me do this?”, the answer is often “because it doesn’t make sense to do that”. Auth is one of the places where that happened, and going through the process of “but why can’t I look the user up in my database from middleware” was a big part of wrapping my head around the parts of Next.js that I had been ignoring.

As far as being married to Vercel for hosting, not at all, if you don’t choose to use all their stuff. The sample Docker build they have works just fine to deploy anywhere, in my experience.

Maybe I’m speaking mostly out of ignorance of not having tried dozens of other modern TS web frameworks (I was on a big tech island for a decade and not in touch with what the cool kids were up to), but I rather like Next.js. I may feel differently when I want to start adding native mobile apps and realize I was lulled into omitting a clean API layer, but for now, adding features has been pretty smooth.