Comment by _russross | Hacker News Reader

Comment by _russross | Hacker News Reader

I started contributing to OSM recently after a bike trail was built near me. The arial imagery typically takes years to refresh around here so I walked it a few times to capture GPX tracks and then drew the new path. It was exciting to see my contribution percolate out to various apps that depend on OSM, and meanwhile Google and Apple have both ignored my edit suggestions and deny the path exists.

So I have a fairly fresh newcomer's perspective and here's what I suggest to get started: keep it simple and start local. As others have noted, skip the complex mapping app and go straight to the website for feature edits, or download a simple mobile app like Every Door and focus on businesses and landmarks near you. Go on a walk around your neighborhood and look for old/incorrect/missing businesses. There may be errors near you about places you actually care about, and it is really gratifying to fix those.

There were restaurants near me that were missing, shops that had changed hands, etc., that were easy fixes but years overdue. I captured some basics (names, posted hours, the phone number posted on the door, etc.) with Every Door, and then later logged into the main OSM website and fine-tuned the entries. I got into a pattern of fixing one or two entries every evening when I went out for a walk and pretty soon my walking route was all up-to-date. Then I started paying attention to things like stop signs and cross walks and found a whole new kind of little, incremental edits that nobody else was doing.

It's been fun, I've made useful contributions, and my laptop is still Java free.

> I started contributing to OSM recently after a bike trail was built near me

I started after one long day of cycling, I ran out of water before I could get home. It was bad. After that, I have been updating OSM with all the water sources I run across. Drinking water in parks, bottle refill stations at stores, water fountains that I run across.

It's not much, but it's honest work.

Documenting public water sources is an incredible public good. I thank you for your efforts.

Thanks so much! I used it multiple times when traveling far away (and also contributed back what I spotted missing).

Every Door is definitely one of the easier ones and I like it but dang I wish the dev could get more funding to make it a nicer UX. Between the glitchy raster tiles and need to click load data all the time I feel it gets tiring fast.

I've considered building a separate app that's similar but oriented around the idea of using voice of pictures to update things (with AI extracting and formatting the data for you to review). But alas making yet another tool isnt ideal.

Isn’t that a door (haha) opened by our supreme new AI coding overlords: that now it’s possible to take the Every Door source code (it’s on Github), and throw it into a coding agent with the instruction to „improve <ux shortcoming of your choice> without breaking functionality“, and with some iterations and human validation that yes, the UX is now indeed better, play the changes back with a PR?

I can improve the UI to my taste but if the maintainer doesn't have time/money to maintain a big ol vibe coded PR for some very opinionated change isn't going to do much.

That would appear to leave me with the choice of just using it for myself (lame) or I then have a public fork to maintain :(

If there's basically no shot of it getting up streamed I think I'd rather make something from scratch to suit my needs without the extras.

Anyway, I hope the AI overlords help us fund more maintainers soon!

FYI LLMs are also very good at maintaining a fork ;)

Do you have any examples of forks maintained by LLMs and how they do it?

I keep my neovim config maintained by a LLM which is a fork of Kickstart.

My example might be an outlier on the easy side because they have a custom/plugins directory that will not have merge conflicts ever so I try to never modify the main init.lua and keep all of my changes separate in my own files. This prevents merge conflicts.

I can then ask my LLM to add a plugin or customize existing plugins by writing files in this separate directory to add/overwrite plugins and configurations. This works great for me.

If I were working in the main file creating merge conflicts, I could imagine it could get messy fast.

UX is one of the things AI isn’t good at. You can ask it for specific changes but just “make it nicer” is unlikely to work.

No but if you have specific asks like “make this resizable” “I should be able to drag and drop to combine” or “add a button on the top right to close” it’s quite good at directed changes

“Make it nicer” will lead to unpredictable and controversial changes, likely to clash with the project owner’s vision. The owner would be better off using the LLM as a collaborator, if anything, to decide where to take the design.

However, UX can also be improved by some simple heuristics that could be seen as mechanistic by the standards of modern LLMs.

Performance and responsiveness, for example, contribute to UX even when the UI remains fixed. I’d wager Fable-level LLMs could improve perf on most open-source GUIs by benchmarking typical use cases and progressively applying standard tricks like caching, buffering, batching network calls.

That's real work and requires real skill. The parent comment example prompt just doesn't work.

Claude Design can help. Fable is pretty good at UI when you narrow down to a specific page or—better—something like a modal, table, etc.

If someone is a user, they'll have specific complaints. Claude Design is more than capable, which can then be fed directly into Claude Code to implement, test, iterate, and ship.

I agree that it lags behind coding and imo the highest leverage area for Anthropic to invest in. Hopefully they do so; UI is consistently one of the main (severe) differentiators between OSS and paid options.

I had a brief chat with the dev at State of the Map (the OSM conference) a few weeks ago. I think he'd appreciate some help, maybe even a co-maintainer.

Any more you can share? Is he actively doing a search? I didn't see anything about it in the repo.

I make https://mapcomplete.org which also serves as easy-to-use data consumer and editor. I sidestep most of the complexity by having a lot of thematic maps, each focussing on just a single topic (e.g. mapcomplete.org/shops, https://mapcomplete.org/toilets, ...)

In a lot of areas the sidewalks and crosswalks aren't mapped even though they show up clearly on overhead imagery. I try to fill them in when I can, although this should probably be an AI task now.

It doesn't help that there are multiple ways to map sidewalks in OSM (as a property of the roads themselves, or as standalone paths parallel to the roads).

I got pretty into OSM for a while and that's definitely the thing that eventually put me off... So much stuff has multiple ways to be mapped, and it gets confusing/overwhelming. I understand that OSM has a very malleable data model, and that definitely has many benefits, especially for niche/edge cases. But I do wish the community had arrived at a settled convention for at least the most basic stuff, like the aforementioned sidewalks.

There's usually a reasonably clear consensus within a region or country.

That's less true for niche stuff (vs stuff where two different ideas are in wide use), but there probably isn't all that much of it either.

One way to approach it is to not worry too much about it, consistency can come later.

That's true, but the very feeling that I might be inputting things in a way that might have to be updated later kinda drained my motivation.

I do understand that even if does get replaced eventually, it still was useful for a while, and probably served as the basis for whatever came after.

But still. For whatever reason, in my brain, it just diminished whatever psychological reward I got from adding information to OSM, which is a shame.

At least in urban areas I've seen it's almost always better to define the sidewalks separately. The routing for pedestrians is often significantly different than for vehicles. This matters when the data is fed into Strava or Garmin for running routes.

I highly suspect Apple and Google maps are auto tracing satellite images. Apple maps has perfect coverage of every road marking on their maps that exactly lines up to the satellite images.

I believe most consumers of OSM just discard data like buildings and crosswalks and layer on top auto traced data which has 100% coverage.

One of the weaknesses of OSM is not requiring consumers to contribute back data like that.

> I highly suspect Apple and Google maps are auto tracing satellite images.

https://www.justinobeirne.com/google-maps-moat

they have been doing it for years.

Every Door is pretty cool but for iOS I prefer GoMaps more.

StreetComplete as well !

Reply on news.ycombinator.com