Harry Wood has commented on the following diary entries

Post When Comment
Mapillarising Brazil almost 2 years ago

Thanks. Yeah I've just about managed to learn how to pronounce it (badly), but spelling Guarulhos is too much.

Mapillarising Brazil almost 2 years ago

Aye. She was bored back then too :-(

OSM Analytics launched! almost 2 years ago

Sorry I missed the missing maps event. Good to be able to catch up with a video!

So I was just trying out the tool and bit more, while at the same time making a wiki page for 'OSM Analytics' ...and the screenshot is "Image of the week".

School Edit Tracker almost 2 years ago

Yeah. A tool like this definitely creates an impression that is a competition, which can be a problem. I think that's something to bear in mind when choosing when to deploy it. There was something about the school project which was almost competition-like. We had a lot of isolated school cases to look through and edit in a fairly clear way which lots of people could get involved in.

Of course there's always lots of ways of gaming the system to score more points in a way which might not be all that helpful, and sometimes in ways which are helpful. For instance with these schools, I noticed somebody could go on a rampage just adding the isced:level tag based on the name of the school. A helpful contribution I suppose, but also an easy way to touch a lot of school objects and win the competition!

My previous uses of the tracker tool were more basic aerial imagery armchair mapping ideas. The wimbledon tennis courts project of the week and my Big Baseball Project. The baseball project had the advantage of being naturally US-focused and so we were only messing up crappy US data (and in 2011 it was mostly crappy data).

For UK projects, armchair mapping concerns are more of a problem. In fact I personally didn't get stuck into editing lots of schools, because whenever I looked at one I ended up thinking "you know I can't actually be sure where the outline of this school is. It really needs to be mapped by somebody who knows it"

What I'd really like is some way of making a competition out of recruiting more mappers. Maybe we could do something like "affiliate tracking links" where if somebody signs up to OpenStreetMap using a link you sent them, then you win! Of course that might encourage people to launch over-zealous spamming campaigns :-/

A missing zoo! almost 2 years ago

Ha! Doesn't surprise me at all that this was needed. Thanks!

A missing zoo! almost 2 years ago

oh yeah. Just a Gibbon so far in Perth zoo :-)

Statistical data of the Dutch OSM mappers. about 2 years ago

'central way to register' or the welcome messages could just be public. Hence my suggestion to welcome people on their first changeset (use public changeset discussions)

Statistical data of the Dutch OSM mappers. about 2 years ago

There's that "Long tail" shape again. I did a whole talk about this at SOTM2014

The welcome program is a great idea. Maybe the effect on the stats would be to "help users up the curve" thereby making the long tail and the elite spike less spiky. Or maybe not. It might also encourage more users to come along, making it longer. Either way, giving new users a good welcome will help build an OpenStreetMap community. It's only logical.

There was some talk of a having an OSMF working group dedicated to the idea, but I don't think it was established in the end.

I wonder whether the new changeset discussions feature might be useful for welcoming people. This is a strange idea really, because just sending an OSM message is more appropriate, but... changeset discussions are public, which changes the dynamic somewhat. What if we establish a convention of welcoming somebody on their 1st changeset? Then we can all collaborate on making sure new users get welcomed in a friendly way without accidentally duplicating messages and effort

10 years about 2 years ago

Happy 10year OSMversary Maning! Ooh. I'm just noticing mine is coming up next month. It's been a fun 10 years hey?

Should we teach JOSM to first-time mapathon attendees? about 2 years ago

We actually used to introduce people to JOSM from the get-go a lot more than we do now. Certainly at humanitarian events I've organised before Missing Maps started. In fact there was one back in 2010 where I had no choice, because we needed WMS-supplied imagery. And on that occasion it was bit of a nightmare. Thankfully things have moved on a lot since those days (most of the JOSM UI issues I mentioned there have been fixed since, and we're better at finding ways of not using WMS imagery) Even at the first few Missing Maps events, we followed the approach of presenting the choice, but then demoing the install of JOSM, or asking people to come with JOSM installed.

Why did we move to a more iD-ish approach? I think we used to have a very high proportion of absolute beginners, and some early experiences of mapathons, put us off JOSM a bit. It felt very chaotic as >50 people started downloading JOSM for the first time (the download itself can immediately create an internet bandwidth issue for a start) and then asking questions about how to install on different operating systems and versions of java!

But yeah I've always said, new users are actually better off going straight to JOSM, for a better more fluid editing experience. BUT.... certain types of new users will find it easier. As Pete says, "people who deal with any / many types of software professionally", which is probably at least 50% of the new folks coming to these events actually. So the thing is, depending on this proportion, and the number of experienced people you have on hand to help... getting lots of beginners started on iD can be a lot easier for event organisers. With iD you can write out instructions which will work for everyone, and these will fit on one piece of paper.

Via Monte Napoleone about 2 years ago

I like it! In fact I've just made it OpenStreetMap "image of the week" this week!

I get why so many people just don't bother to even try to vote for the OSMF when things like this happen about 2 years ago

Yeah. You do need to vote before the (very clearly stated) voting deadline. Sorry :-(

OpenStreetMap and Humanitarian OpenStreetMap Team Together about 2 years ago

It's long been observed that different people get different things out of OpenStreetMap, and have different ways of expressing their motivations. Of course some of us are intent on using OpenStreetMap just for a for a sat-nav application in Los Angeles, while others are using OpenStreetMap just for an immunisation programme in Lubumbashi. You say "OSM is about the process of collecting data and sharing knowledge". But knowledge of what? In what form? Perhaps a good way of describing it would be... it's knowledge in the form a "free and open licensed map of the world" :-)

I agree though, that it would be wrong to say we are only succeeding in our goals if we are managing to map the whole world. Wide coverage, reaching into every corner of the globe isn't necessarily the primary focus of the project. On an individual level we encourage people to focus on mapping their neighbourhood. On a project level, achieving wide coverage... I would say it is an important success metric (and one we are succeeding at!) but yeah... not the only success metric.

Certainly we are building a "free and open map of the world" if we only take it to mean the OpenStreetMap system of edibility, data representation, and community building can work the same worldwide.

Unknown Pleasures (of humanitarian mapping) over 2 years ago

It's "image of the week this" week!

Wanted: Simple amenity describer and foursquare style Q&A process over 2 years ago

The Kort game is designed to be asking various simple questions. Maybe it could do with some more interesting questions related restaurants and pubs (making it more foursquare-like) For me it's just asking relatively uninteresting questions about name language settings on things and names of service roads, so some questions about extra POI tagging might liven it up a bit.

Of course WheelMap is everybody's favourite example of an nice narrow focussed tagging app. Essentially it's asking one question... "how's the wheelchair access?".

Likewise coffeedex just asking "how much does coffee cost?"

onosm is for adding new things to OpenStreetMap, so not really the foursquare-like thing you're describing, but a simple tool with a constrained set of fields.

The Flash Map Mob over 2 years ago

It's like a mapping party but with restrictions on how you input the data.

Quite an interesting idea though. One nice thing about everyone putting their data in as they walk around, is that you'll get very rich meta-data about how the mapping party happened (timestamped edits corresponding to when the person was actually there) I imagine Derick's mapping party videos might look different. Otherwise his videos like this one: have distinct before, during, and after phases, and the 'after' phase has to last several weeks to give everyone chance to get around to inputting their data.

In general I guess it's better as a short satisfying feedback loop for newbies to see their data and everyone else's appearing all at once on the day. Trouble is, I would find it dissatisfying poking my smartphone to type in tags when I could be surveying three times the area with my normal techniques (which leave more work to do later)

Update JOSM to fix the Java Logjam almost 3 years ago

The servers were patched on Sunday evening with a change which means java6 applications won't be able to make SSL connections. Maybe you were using an old JOSM v7000 on java 6? Quite a few mac users ended up on this old version because upgrading from java6->java7 can be a bit of a pain on certain versions of mac. See mac install notes.

When I read this, I was worried we'd hit quite a few problems at our next mapathon event, getting people up and running with JOSM, because we have encountered quite a few mac users who can get up and running easily on JOSM v7000.

However I believe this issue should only be a problem if you also set JOSM to connect over https rather than http on the API connection URL settings.

Obviously using all these old versions and http is not desirable, so I'd like that wiki page to offer more advice on how to upgrade your java version, however it is a different process on various (old) versions of MacOS. Some versions might be a simple click-through to upgrade... but other versions have java 6 welded in place. Basically between oracle and apple they've conspired to make the whole thing a massive pain in the ass ...with a nice bit of toolbar sprinkled on top for good measure.

Police Scotland almost 3 years ago

It is!

Connecting Communities With Improved OpenStreetMap Credits on Mapbox Maps almost 3 years ago

In your final links "Collapsed (small) " I don't know if this gist has been changed since you wrote the diary entry, but it's back to showing all one link for the text "© OpenStreetMap © Mapbox", and no direct link to OpenStreetMap.

On these improvements have either not yet been rolled out, or you're not rolling them out there for some reason. Pages like this for example I know these are sometimes presented in an iFrame, and perhaps not intended as shareable URLs, but in any case, is the OpenStreetMap credit approach just not been updated there yet?

In general I've seen quite a lot of different inconsistent approaches to giving OpenStreetMap credit, even on Mapbox's own blog. e.g. Blog posts with images/screencasts like yesterday you'll tend to not bother linking OpenStreetMap at all. But here we're talking about embedded maps. There's been quite a few of these on the mapbox blog over the years. Some use your 'mapbox examples' site e.g. this one which has credit without links. here's one with the all-one-link style. Would be good to see these linking OpenStreetMap directly as you've described in this diary entry.

It's not because you have accurate data that you have to upload all of them in OSM almost 3 years ago

Interesting discussion. We have some brief guidelines on simplifying for imports, but this is mostly for larger scale imports. We've had problems with unsimplified shapefile imports where tremendous number of nodes were plonked in, despite relative inaccuracy of the data. In this case it clearly is very accurate survey-based data. In any case though, to put "too much" data in, is to screw up the data for anyone making normal use/normal edits. But how much is too much.

If we were to run an analysis of the golf bunkers worldwide, look at the nodes-per-metre, draw a histogram of that. I wonder what detail level mappers typically go up to. It's the kind of thing people do map in quite a lot of detail. And they're always curvy shapes of course. I guess finding a "high end" 90th percentile value from the existing database would have been a good way to decide on what to simplify this data down to (too late now, but could be fun thing to calculate)

I suppose the principle I'm suggesting here, is that we should encourage a detail level which is high, but not so high that other dedicated mappers are not attaining the same level elsewhere in the world using normal OpenStreetMap contributing techniques.