motoroco

just relating my personal/professional setup, I run Pocket ID for myself and made an agent account for claude. I have it set up with limited access policies to pre-prod Cloudflare resources and Tailscale servers for occasional tasks. I really like the control I have over access, grant it just when it’s needed and only exactly as much as needed. I wish SSO didn’t require biz/enterprise plans on so many services so I could use my OIDC in more ways

I got blocked by Microsoft for trying to advertise on DDG. I signed up for a new account (no prior M$ affiliation to start from), imported my Google Ads campaign, paid them a hundred bucks and was almost ready to spend. Then I uploaded my WA state business license and immediately got force logged out without any reason. An email came by to say I had been blocked. Support was utterly useless, there was no way to appeal that I could find, and I had to file a chargeback with my bank to get the $100 out. Don't waste your time & money, folks

I wanted to try out your service but I don't have any accounts with Google or Microsoft properties (incl Github), so I can't sign up

it's like costco memberships. the more you pay, the more value you feel you're getting

Yeah, Apple has a rule against simple wrappers. Check out bubblewrap if you haven’t already. It’s made by Google specifically for wrapping PWAs for the Play Store. There’s a tiny bit of work to make a keystore and manifest file, otherwise it just pulls from your PWA config automatically

Hello fellow map maker! I feel like I’m in the same boat. PWAs work great, I wish Apple would treat them as first class apps. I tinkered with launching a TWA for my app on the Play Store and it works pretty well but I haven’t published it yet. Probably a harder market to monetize than iOS but it seems like good advertising just to have the listing up

reading that tweet it came across to me that he set the agent up for failure. garbage in, garbage out! maybe a more fair comparison would have been a prompt to design & implement the renderer with a clean start. throwing it all out and starting from scratch is something these models, and people, are explicitly trained against doing

other than that, I don't think any serious person would trust a ralph-loop to accomplish something truly substantial. it's good for ensuring rote tasks get completed without the agent stopping too soon, but not much more

thanks for the tips about CarPlay development. I'll definitely target parity with the web app at first so hopefully it'll be something I can get access to eventually

regarding (1), yes, you keep the routes you made even after your subscription ends. the limits revert to the free tier so if it had more waypoints or was longer than that then you just can't add or move waypoints. removing waypoints to get back down below the limit works and you still export your trip to formats included in your tier (e.g. google/apple/waze)

you could absolutely pay for a month just to plan and export all your trips for the year, bookmarking them with google maps or saving the files for your GPS for instance -- not a problem. I am currently exploring alternatives to monthly & annual subscription schedules to further support seasonal usage, too

per your 2nd question, that's correct, only the creator needs to pay to have 2+ collaborators. you can try it out on the free tier with 1 person at first to see if it works for you. if you just want to share maps publicly you can do that without subscribing, too

thanks for trying it out! I actually happen to have a trip planned for next month in Shenandoah and Blue Ridge Pkwy, looks like that loop is a little out of reach for us this trip but I want to go back to hit Tail of the Dragon and others

I will make a native iOS app, but probably only after this summer roadtrip season passes. as far as CarPlay goes, I'll have to find a way to test that first (all my cars/bikes have "dumb" head units). I have the site added to my homescreen on my iPhone and it works for me at least, same thing on desktop I installed the PWA with Chrome and have it in my dock

you're right that other providers don't understand through-waypoints. it's more problematic for groups than individuals, but I haven't wanted to take on live navigation yet because it's so hard to do right and to do real world testing. I could just trust Mapbox to be good enough with their Directions API, but then I'm not adding that much value and so I'm still going with planning/collaborating as my primary focus and encouraging folks to export to google/apple maps or their GPS for now. thanks for all the suggestions!

I had the same issue planning roadtrips where the journey is the destination. I’ve paid for apps like InRoute despite poor UX, inability to access my routes from desktop, and zero collaboration support

I got increasingly frustrated looking for a replacement that I made https://plotalong.app in my spare time so that my wife and I could plan and save our motorcycle trips together. We still have to add the occasional “via” waypoint but it’s leagues better than Google maps at sticking to the scenic route

I need to update the marketing site and get new users into the map faster, but if you want a quick way in here’s an actual trip I took last year and spruced up to serve as a demo https://plotalong.app/maps/5c735b43-cb84-5a7b-8976-2d4e0c383...

New features and improvements come out regularly so if you have any suggestions send me a message

I'll have to try that. how does it compare to using playwright? it seems like one advantage of playwright is being able to use it with multiple browsers, at least as far as web development goes. if these browser-specific tools end up being faster or more token efficient then it's probably the better pick for task automation and claws

is the LLM not acting on your stated intent? maybe you can find a middle ground, where you can plan and act in small enough chunks that it doesn't start getting its "own" ideas about what to do, or how to do it

a chainsaw is a coarse tool and I liken it to vibe coding. you maintain at least some level of control, but the edges are rough and you might slice off more (or less) than you meant to. I want to model my usage more like a table saw, a precision instrument that can make the exact cut just as I planned it

It does not include shipping. I paid an extra $30 for shipping a 40mm cube. It’s the only reason I didn’t comment initially in support of an otherwise cool service

yep! it's really helpful. I have different accounts on my gitea server for various bots and agents with restricted access scopes. it's been great for making branch protections like mandatory reviews actually enforceable

gitea runs well on a low end server in my experience. self hosting on hetzner and it's somehow the holy trinity of cheap, fast and reliable. I previously (years ago) self hosted gitlab but I remember it being very slow which was the reason I moved on

I first heard something similar taking motion design classes in art school: every frame should look good. Transitions and animations that have bad in-betweens look bad overall

There’s no free lunch, it takes time and effort still. And expertise if you need it to be robust.

In terms of velocity, let me offer some numbers. In 6 months I generated >150k lines of code and merged 10k PRs to ship and iterate on https://plotalong.app

I follow best practices and isolate agents to continuously deployed dev environments, semi-manually review PRs and gate the release process between multiple protected envs. The project is getting close to 500 end-to-end tests in Playwright.

That’s just working nights and weekends. Before AI, it took my team at the office 4 years to produce this much work. There are some qualitative differences but the speed and results are real

I read it as a sort of prisoner's dilemma. If I'm being tracked and monitored everywhere I go, then it's in my best interests to do the same

Thanks for the feedback. As it is right now, only one person needs to subscribe in order to collaborate. I’ll definitely make that clearer in the marketing material!

I like your emoji suggestion. I realized little while ago I need to distinguish between different types of waypoints so this is great validation

I think you’re right about getting people into the actual app faster, before signup. I’ll have to prioritize that sooner than later

No, and while that does sound like a pretty obvious feature I should clarify the app is about planning, not turn-by-turn directions. You can open Plot Along routes in Google Maps, Apple Maps, multiple GPS formats and more

The idea is everyone opens the same route for coordinating and there’s just one source of truth for the group. And then when you’re all about to hit the road, everyone can use the nav app they’re already familiar with (or that’s built into their vehicle)

I will tackle the navigation aspect at some point if I do keep up on feature dev, though!

I wanted a faster, easier mapping app to plan motorcycle rides for myself and with groups, so I finally bit the bullet and started building my own at the beginning of this year.

I got to the MVP state which was useful for my personal use case in about a month. I took it further than that as a learning exercise and as a means to share it with others. Some features that came later are live cursors (like Figma), elevation chart and grade overlay, and QR-code enabled collaboration links to make in-person sharing simple.

Check it out! https://plotalong.app

Figuring out the exact UI/UX I wanted was the hardest part. I did the branding myself, handdrawn on paper, traced in Procreate, and vectored in Sketch. Fast iterations and a good test suite made it possible to try lots of different approaches and refine the one I liked the most. There are roughly 4000 unit tests and over 300 e2e tests that run on multiple environments with fully automated CI/CD.

I’m using Mapbox for the frontend and the whole app is basically just a monolithic Cloudflare Worker. Claude pretty much implemented the entire thing. I got a lot of mileage out of self hosting a Gitea project and recording all my planning sessions as Milestones and Issues. Claude has his own account without admin privileges. The process of managing a team of agents to build this practically autonomously was a bit jaw dropping and eye opening to be honest.

I would love to hear from other pleasure & sport drivers about the features they use or want the most in a routing app. I have an Android app in Play Store review, if you’d like to be an early access tester shoot me an email at my handle @plotalong.app

you can tap the Share button and use Add to Home Screen on any website in iOS Safari, including PWAs. it even works on hn

I've worked with some people who only seem to care about 2. as in, they don't try the feature in any way, but come back with comments about "this isn't tested enough" even though it has higher coverage than the codebase's average, and refuse to approve even though they'll never meaningfully review the content. it does seem to be mostly just theater in my experience

I have to agree with OP, in my experience it is usually more productive to start over than to try correcting output early on. deeper into a project and it gets a bit harder to pull off a switch. I sometimes fork my chats before attempting to make a correction so that I can resume the original just in case (yes, I know you can double-tap Esc but the restoration has failed for me a few times in the past and now I generally avoid it)

this was addressed directly in TFA with a link to https://journals.asm.org/doi/10.1128/mbio.03824-24

if you live in the PNW and would like to see some world class bonsai in person, the Pacific Bonsai Museum in Federal Way is a great destination. this time of year is also good for visiting the neighboring rhododendron and azalea garden

thanks for sharing your project. how was your process different compared to pre-claude design? would you use it again?

this is not true in my experience. prefab kits of all sizes (from sheds to houses to barns, like were once possible to order from a Sears catalog) have worse tolerances than a carpenter working on site. you can measure 3 times and cut perfectly, and still end up with a few mm gap (or sometimes worse) after tiny errors accumulate as you assemble piece after piece. it _requires_ measuring as you go and cutting on site to handle this small amount of drift and to really produce something of high quality. it doesn't come in a box

Darktable has really improved over the last couple years. It used to have some pretty confusing workflows and lots of overlapping modules, but somehow it's been getting cleaned up and polished into something of an intuitive app. It is still different but not so overwhelmed with features that you can't figure it out