I tried two fairly basic searches and both had pretty fundamental flaws that make it very hard to consider using OM as a main navigation app.
First: it doesn’t seem to be able to use common informal names for places. If I search “Maccas” on GM, it correctly shows a number of McDonald’s restaurants, as well as Maccas competitors like Hungry Jacks. In OM I get a few places literally named “Macca’s”, followed by a bunch of much more tenuous results. No actual Maccas shows up.
Second, and probably more serious: it doesn’t tell me where something is. I typed in my street name, and understandably a number of results with that name came up. It tells me the city and state they’re in, but doesn’t get more fine-grained than that. With multiple of the same street in a given LGA, it is basically impossible to search by street address if the search result doesn’t narrow it down to suburb. This seems like an incredibly small fix to make, but also a mind boggling error to not already have.
TIL the nickname ‘maccas’ https://en.m.wikipedia.org/wiki/List_of_nicknames_for_McDonald’s
Yeah I find that a lot of FOS software is missing some features compared to commercial software. But many times after the initial learning curve it’s really not missing as many features as I thought. For example OM has a super easy way to record a track, which GM doesn’t have, and I now use it often for new hikes so I can find them again.
FOSS represents a world I would much rather live in, I think it’s a more equal world, so some minor inconveniences and even donating small bits of time and money to help the project succeed is a feature that is important to me.
I don’t think developers read this. Head over to https://github.com/organicmaps/organicmaps/issues
Your number one concern is that if you search for a word it doesn’t give you results for another word? Should they be training the search index with urban dictionary just to be sure?
My number one concern is that two out of the very first two things I tried doing worked spectacularly badly. It failed to turn up the expected results. If you read the post you’d know that of the two issues, the other one was the one I considered the worse of the two though.
Should they be training the search index with urban dictionary just to be sure?
If that’s what it takes? It’s not really my concern how it works, only that searching for X should turn up X, even if the way X was searched for doesn’t work with the sort of naive text match search that might work with a simple regex. Whether that’s typo detection or recognising ubiquitous alternative names.
Doesn’t seem very deep
What do you mean by this?
You say it seems deeply flawed, but these are not very deep issues to fix?
You know what, you’re right. They’re extremely big issues in terms of how big their impact on it experience is, which is what I intended to convey, but at a technical level they’re likely very surface issues that could be fixed quite easily. Especially the second one, which also happens to be the bigger issue for UX. I could have phrased myself better.
