Blog What I Made

So I’ve left Multimap and I’m no longer a Microsoft employee, I have relinquished my blue badge and handed in my laptop. I’ve really enjoyed working at Multimap. The people I’ve worked with have all been friendly, incredibly intelligent and fun to be around. Not only has working with them been great, but the many times we’ve managed to meet up outside work has also been fun, including my wedding and stag do and lots of other occasions too.

Three years ago Fake Ed Parsons came to me in a dream and told me the secret to draggable maps. I knocked something up and then sent an email through to Multimap. Apparently the email was passed on to Sean Phelan, the founder of Multimap, who then passed an email around the senior management asking which of the following options they should use:

1. Tell me to stop
2. Sue me
3. or offer me a job

I’m obviously quite happy that they decided to go with the 3rd option! I worked for Multimap as a contractor for 6 months and then started working there full time at the end of September 2005. Since then I’ve worked on the API full time. Though I’ve written nearly all of the JavaScript in the API it wouldn’t be much use without the work put in by others in the team to build the many great web services that it uses. “Tiling the world” to make sure we had map tiles and imagery in as many places around the world as possible was also a huge undertaking that went really well. I’ve spent the last month or so knowledge sharing and I’m confident that the people who will be taking on the API work from now on will have no problem. I’ll also be doing contract work for Multimap in the near future too so I won’t be completely hands off on it.

The photo above shows my blue badge before I handed it back but I thought it would be amusing to also show an old business card of mine with it. Before I started working for Multimap I actually worked for a Manchester web agency - Fast Web Media. When I started working there they were actually a subsidiary of “FAST Search and Transfer” and were working on a “soccer search engine”. That project was eventually shelved but the company continued doing it’s flagship Premiership football scores and news website and started taking on more client work. Though they were eventually spun out from FAST to be a separate entity, FAST continued to hold the largest single share in the company. Fast forward to now and we find that Microsoft has bought FAST. The sale appears to have gone through so now it seems that Microsoft will now own the largest share of my old company. This is a fact I’m thinking I should highlight to any potential companies that I might work with, every company I work for gets bought by Microsoft!

A few weeks ago I handed my notice in to Multimap. My last day will be Friday 18th July. While I’ve really enjoyed my time here I’ve decided that I want to branch out and try some new things. I’ve had the privilege of designing and building one of the most important products that Multimap has delivered over the past 3 years - the Multimap JavaScript API. Specialising in this way has been a great experience but I’m intending to broaden my horizons to cover more back-end technologies, which is where most of my previous experience has been.

My intention is to start doing freelance work once I have completed my notice period with Multimap. I have arranged work with some good friends that should keep me pretty busy for the first few months after leaving but a large part of what I’m looking for in the future is variety so I will be very interested to hear from other friends who have interesting projects that they might like my help with.

Due to contractual obligations I may be looking to minimise the amount of geowanking I do for the first few months. Fortunately I do have five years experience in PHP, three years experience of intensive JavaScript coding and various bits of experience in Ruby on Rails and even a little Python. I also have good experience in setting up Linux-based web and database servers. I’ve been using and contributing to the interweb for something like 14 years, so yes, I do remember Netscape 1.0 (background images!), the launch of Yahoo! and the BBC’s wonderful “list of interesting web pages”. I was also putting music online when Shawn Fanning was probably sleeping peacefully in his dorm room, and I was doing it legally (I think)!

If you are interested in hiring me then you might like to take a look at my CV which you can find here (yes I know I need to replace the dodgy matrix). If you want to get in touch then you could try contacting me on my linkedin.com page, or just email me at my first name @ my surname dot net. If you’re having trouble getting me then drop a comment on this page and I’ll get in touch with you. I’m not looking to hear about a thousand unsuitable posts from agencies but if someone from an agency has something flexible to offer that’s relevant and can fit in with my other obligations then that will be fine.

I will still be attending the State of the Map conference this weekend. I’m really looking forward to it and it promises to be a lot of fun. I will also be representing Multimap for the last time while giving a presentation on “Using Crowd Sourced Data in a Commercial API” which I’m hoping will be interesting enough for those that attend. If the conference is anything like last year though there will be plenty of good presentations to contend with.

Technorati tags: resignation, multimap, microsoft

So I’ve now been up basically all night (5:14am, no intention to go to bed). Unfortunately I haven’t spent the whole night hacking, in fact the hacking task I was working on - getting the search API into the ruby library - was completed hours ago. I’ve actually spent the last 5 hours or so playing Rock Band.

We started playing cooperatively with two guitars and a drum set until around 3am when we got complaints about the noise of the drums (even though they’re not real, whacking some bits of plastic with real drum sticks makes quite a lot of noise!) Funnily enough though, the complaints came not from the sleepers but from the people playing Werewolf in a big group!

But back to the search API. I’ve added three new classes to the mmruby library: MMSearch, MMSearchFilter and MMSearchRequester. To perform a search you need to create a MMSearch object and then pass it to the MMSearchRequester class’s static search method:

s = MMSearch.new( { :data_source => 'mm.poi.global.general.atm', :address => MMAddress.new( { :qs => "L19 4UD" } ) } )
r = MMSearchRequester.search(s)

I have realised that there’s actually the routing API that also needs doing. It’s unlikely that I’ll get around to doing that today but I’m sure I’ll do it before long. I’ll post again once that’s done.

Again, not thoroughly tested just hacked together, any comments appreciated. Download mmruby-0.2 here.

Technorati tags: ruby, mashed08, multimap, web, service, REST

I’m spending this weekend at Alexandra Palace at the Mashed event that’s being arranged by the BBC. I went to the similar “Hack Day” event that they ran last year (and blogged about it) and had a great time. This time I’ll actually be there in a vaguely official capacity as Multimap and Microsoft are supporting the event. Multimap have already blogged about it but I thought I’d write a few lines too and release some software that might come in handy.

One thing that we’re really proud of at Multimap is our RESTful web service APIs. These give you access to geocoding and (faux) reverse geocoding services as well as access to our great search APIs.

As part of a project I’ve been working on in my spare time I’ve put together some ruby wrappers for these APIs so I thought I’d get these released in case they come in handy for anyone during Mashed. I’m not an experienced Ruby programmer so I can’t give any assurances of the quality but hopefully they’ll come in handy. You’ll need to have the ‘json’ library installed as well as Net::HTTP. I’ve only used it in a rails setting by placing the .rb files in the lib directory so your mileage may vary. I’d really appreciate comments, bug fixes and anything more you’d like to add. The libraries give you access to geocoding, national grid conversion together with a few handy geo-related classes (lat/lon, bounding box class).

Unfortunately I haven’t got around to wrapping the search API but it shouldn’t be too hard to do, I might even do it as a hack during Mashed. I might see if anyone’s added any comments before deciding whether to do it though

So download mmruby 0.1 here and tell me what you think of it below!

Technorati tags: ruby, mashed08, multimap, web, service, REST

David Troy

Twittervision got a lot of attention, sorta explained twitter. A way to show the world in a new way. Similarly flickrvision got a lot of attention. Shown in the museum of modern art. Spinvision.tv takes youtube videos and puts onto a flash player.

Twittervision local shows more local versions, also flickrvision local coming together.

Got me thinking a lot about all this data I’ve collected, what can I do with it? Photos and tweets in a particular area. Thought about a personal tricorder to see what’s around you, scan your local area. Frankly though it was a large amount of data it wasn’t large enough. Was thinking about what else out there could help this. Too may walled gardens of people approaching this, maybe there’s a way to link this? Came up with openlocation.org.

Suppose you land in Chicago, as you’re taxiing you take out your cellphone to find out what’s happening in Chicago, are friends available, events, places you’ve been recommended to go? Only solution is to scour bunch of websites and you won’t be able to do it quick enough. You need to be able to just say “what’s up in Chicago”. Find your friends, find the bar, meet your friends, you can do all that on the way to the gate.

This is a really hard problem, I liken it to being in jail, you get one call, you need to find the information straight away. This is case where you don’t care about maps, you care about proximity.

Problem is there’s a huge amount of information. Google’s geosearch will return results based on a basic page ranking but we don’t know how that works. Maybe we need to do this based on your social graph. All this enhances quality of life, if you can take out mobile and get complete picture of your surroundings.

Maybe it’s not a business, maybe it’s a technology or a protocol. If twitter was invented in 1994 it would’ve been given a port number. So maybe this needs to be a technology or protocol. We need to look at this as not simply a business.

We’re announcing openlocation.org. A lot of developers don’t get geo. Developers will give you many different approaches. We need to start a conversation on how to approach problems and agree on toolsets. Don’t want people to try to lock up data to become the facebook of LBS.

Need to think a lot about this, how do users interact with this. Make sure it’s simple unlike OpenID(?) More like Maker Faire, less like IETF.

Need to agree on goals. Openlocation.org is a community to bridge geo community with the web. Psychology. Sometimes maps are not great. Location relationship is important. This is a hard problem that we need to iterate on.

“Wrestling with Angels!”

GeoHash, lat lon converted to hash that removes accuracy as you remove digits from the end.

Got lots of important people involved. Get on the google group and get involved.

Openlocation.org: Location Services for Web Developers

Technorati tags: openlocation, open, location, where, where2.0, where2008

I’ve managed to knock something up that will let you track my journey from Liverpool in the UK over to San Francisco in the states. I do know I won’t be traveling the furthest but I still thought it would be fun ;-).

I’m currently (Friday lunchtime) sat in my house in Liverpool. I went for a run this morning so there is a little data in there already. I’ll be traveling down to London tonight and then going over to Heathrow and flying to San Francisco around lunchtime tomorrow. I’ll try to log my location as often as possible using the various means at my disposal. I have quite a few ways to do it so I’m hoping I’ll manage quite a good trace. I doubt I’ll get much if anything while I’m in the plane but I’ll give it a good go! Also I’ll be updating my position while I’m in California so it should end up with a full log of my trip over there. Finally, the page will automatically update itself every 5 minutes to see if my location has changed.

http://johnmckerrell.com/map/?show=john

Note the times at the bottom, “Last checked” is the last time that the page checked for an update and “last updated” is the time that I last sent it my position. So take a look and let me know any comments you have, especially if you find a bug! I may enhance it if I get time so you might want to refresh it as well from time to time.

One other thing to mention, even when I’m accurately updating my location with my GPS I’ll still only be updating once every minute, also that page isn’t capable of updating more than once every five minutes. I’m caching the location rather than using a database lookup to make it run a bit speedier. Oh yes, I do know the arrows don’t work in IE, unfortunately VML is not capable of drawing them.

Technorati tags: javascript, where, where2.0, where2008, map, tracking, wherecamp

Wow, before starting this blog post I thought I’d check the date on my previous post. I haven’t written anything in nearly three months! That’s pretty terrible so I’ll try to do a few more posts sometime. I’ve been pretty busy with a number of other things recently but that’s no excuse, I know interesting things have happened that I should have found the time to mention.

I guess the biggest of the things that have been restricting my time recently was the recent redesign of the Multimap.com site and the availability of Microsoft Virtual Earth content within the Multimap API. Oddly enough, this was the third time that I’d worked on a project to pull VE content into our API. The first two times we had done it by wrapping VE’s API into our own, passing function calls to our API onto VE’s where necessary in a similar fashion to OpenLayers Base Layers. Though this gave us access to VE content and allowed us to resell the content to the UK Yell.com site, it was never going to be the best method. The finished product required loading the entirety of both APIS (nearly 300KB just for JS), and because we only had access to exposed functionality resulted in some odd behaviour, such as when our API and VE’s were both trying to smoothly pan a map to a new location.

In December of course we were bought out by Microsoft and as a result were able to get direct access to their imagery. Making use of VE’s maps and aerial data was fairly simple but including the Bird’s eye imagery was a little more difficult. Fortunately with the help of a few guys at MS who sent through some source code, and with the existing good “custom map type” functionality in the Multimap API, I had a good working model pretty soon. We then spent a long time trying to come up with a really good way of communicating “Bird’s Eye” mode to users. Though we had trials and tribulations along the way I think we did pretty well with the solution we came up with, it’s live on multimap.com if you want to take a look (or see the screenshots above and below).

Another thing that I’ve been trying to get sorted out is the server that hosts my blog. I’ve been using the same dedicated server for a good 4-5 years now. That server was hosted at Sago Networks and it’s been through a lot, including Florida hurricanes, but has recently started hanging on a regular basis. A replacement NIC a few weeks ago gave promise of a reprieve but it has again crashed since then. Though I’ve had really good service from sago who have been happy to manually fsck it every time it went down recently, I decided I had to go with price and have now switched to Hetzner. Their prices were too good to miss, and though the fact that I don’t speak a word of German has caused a few issues along the way everything seems to be going well with it now.

I’ve even decided to be all modern and am hosting my site on a VMWare Server virtual machine. I’m hoping that this will allow me more control over that machine, making it safer to upgrade and reboot as I’ll always be able to get access to the console through VMWare. Hopefully it’ll also make things much easier when I eventually decide to move to a new server again (I’ve had this new server for a month now and have only just found the time over the long weekend to move everything across!) I’m also hoping it will lend me a little more security by allowing me to segregate important sites that I need to keep secure away from older less reliable code.

One other thing to mention, I’m going to San Francisco next week! Where 2.0 is the second most important conference for location based services providers (the most important, of course, being OSM’s State of the Map) but in past years I haven’t been able to attend. Fortunately this year some budget has turned up and I together with four of my colleagues will be there. Though the main event is the O’Reilley conference, I’ll also be going along to WhereCamp 2008 the following weekend, and as many other events as I can cram in on the Thursday and Friday between. If you’re going and I don’t already know it then get in touch with me on twitter or friend me on the WhereCamp site.

I’m also intending to map my journey there as much as possible, I have to get from Liverpool to London on Friday night, over to Heathrow on Saturday morning and then fly to San Francisco International airport at lunchtime (leaving 10am, arriving 1pm, still freaks me out ;-). Obviously it’s highly unlikely that I’ll be able to log anything while I’m flying but I’ll do my best, and I shouldn’t have too much problem on the train down. Check back later this week when I’ve figured out how I’m going to share my location with you

The other day, a friend sent through a link to an interesting “new” technology that Apple have applied for a patent for - podmaps. Reading through the article it seemed oddly familiar, perhaps because I came up with the same idea two years ago. As most good ideas do, this one came about over a few beers when someone suggested that Multimap should do a podcast. Of course the idea of a Multimap podcast was perhaps a little odd, but it did get me thinking about what we could do with podcasts.

After a little investigation and playing around with Apple’s “ChapterTool” I knocked up the Multimap Travel Directions Podcast. You should be able to try it by downloading the file and launching it in either iTunes or Quicktime. It takes advantage of all the great features available to podcast files. The file is split into “chapters” with each chapter being a step of the route. Each chapter has audio with a computer voice talking you through the directions for that step, an image attached to it which shows a map of that step, the text is included as the “title” (and is readable on your iPod’s screen) and there’s even a link to the route on multimap.com for when you’re viewing the podcast on your computer. (Note that the podcast above was made 2 years ago so uses old maps and probably links to the old multimap.com, keep reading to hear about the improvements though…)

So this is all very nice, but this thing took me a few hours to make by hand, surely there’s a better way? Well with all the great APIs that Multimap provides, yes there is. Though I was busy with a few other things I played around with this idea over the following 15 months or so. Every few months I’d write a few lines of code, come across a problem, get bored and put it down again. Finally though I got past all the problems (how to tell the length, in seconds, of an audio file, how to concatenate audio snippets, how to convert to AAC, and a few more) and managed to knock up a ruby script that could take a source and a destination and give you a podcast containing the directions. That was 9 months ago though and I’ve been sitting on it since. Seeing the news on Apple’s patent application has spurred me on to releasing it.

I spent a few hours last night updating the script so that it now works with our recently launched Static Maps API meaning that you’ll see a vast improvement in the map quality. The script requires OS X to work because it uses Apple’s ChapterTool and “say” command. It also needs SoX (for various sound conversions), the Perl CPAN module Audio::Wav and the FAAC library. I’ve packaged the script into a zipfile and put a README in there with some information on using it, I mention how to get and install the 3rd party packages in there too.

There’s obviously lots of improvements that could be made to the podcast, and I’m sure Apple will make many if they actually do release “podmaps”. All of the podcasts I’ve generated here have been made on my OS X Tiger installation, if you have Leopard you will be able to use the new speech synthesis voice that came with it to make better sounding podcasts. All the routes generated are currently driving directions but it would be a simple tweak to make walking directions. Extra functionality could also be added using more Multimap API functionality, such as highlighting places to eat as you’re passing them and things like that. For now I just thought I’d release it as-is and see what people think.

So, please download the script, play around with it, and let me know in the comments what you think of it. It might be fiddly to get it working but rather than get too technical in this post I thought I’d put that sort of thing into the README files. If you haven’t already, you’ll need to sign up for the Multimap Open API to get a key to use. Here’s a few more routes to give you an idea of what it can do too:

• From Euston station to Multimap’s office on Fleet Street.
• From Multimap’s office to Covent Garden.
• From Boston, MA to Cambridge, MA

• Download the route podcast generating script here.

Technorati tags: apple, ipod, podmap, routes, map, multimap, maps

OSMify

I’ve put together a new version of my OSMify bookmarklet. The main reason for doing this is because the old version is linking to the wrong set of Osmarender tiles; it’s linking to the ones on dev.openstreetmap.org rather than the newer ones on tah.openstreetmap.org. Rather than force you to regularly visit this site and setup a new bookmark every time there’s an update, I’ve decided to host the JavaScript for this version of OSMify on my site.

In the previous version, all of the JavaScript to make the new map types appear was contained within the one line bookmark. This time I’ve put all of the important JavaScript into a separate file hosted at johnmckerrell.com/files/osmify.js. The bookmark now only contains some simple JavaScript required to load that file in.

As well as allowing me to make improvements to the code this also makes the code a lot more readable, allowing you to read through it and understand what it’s doing. If you have any improvements that you’d like to suggest then either drop them as comments on here or email them through to me.

One improvement I did want to attempt was to make OSMify work on other sites than Multimap’s. The Multimap API’s custom map type support works in a very similar way to the Google API so it would be easy to support. Microsoft Virtual Earth/Live Maps also have a method for putting custom tiles into their API. Unfortunately with Google, they don’t expose the full API on their public site (i.e. maps.google.com) so it isn’t possible to add custom map types there. The current version of OSMify will work with third-party implementations, though they will need to expose a global map variable that is of type GMap or GMap2. I have also had to tell the code to add a new GMapTypeControl after adding the new map types as for some reason the Google API’s map type control doesn’t update when you add new map types (something the Multimap API does do). I haven’t bothered with Virtual Earth/Live Maps support either yet as I would need to code up my own widget/control for switching between these map types (or tile layers as they call them).

So, the link is at the top of this post, have a go with it and let me know what you think. I’d especially be interested in some feedback on using this with Google maps implementations, and any thoughts you have on getting this to work with more APIs.

Technorati tags: javascript, bookmarklet, osm, openstreetmap, map, liverpool, multimap, maps, favelet

I was invited to attend a private “hack day” at the offices of the Ordnance Survey on Friday last week for the launch of their new OpenSpace API. A number of other people from the Open Street Map community were also invited (leading to talk of attempts to extinguish the OSM project by doing away with us all!)

In actuality the guys at the OS were really friendly and had a day of activity planned for us. After the initial introductions we were split into two groups and went off for a mini-tour of how they collect their map data. My group first got to head outside and talk to one of the few hundred on-the-ground surveyors that the OS have. He showed us his collection of equipment, from the low-tech popeye to the 2 metre GPS mast, back-pack full of GPS equipment and the tablet PC for entering new data directly. It was really interesting to listen to his experiences of surveying and to compare it to my own. I’m not sure that he realised that most of those present would consider themselves to be “surveyors”, I’m sure he’d laugh if he heard it though considering the accuracy and quantity of the data that he would collect in an area would be far more than that taken by most OSM mappers.

After this we were taken off to the Photogrammetry department to find out about how OS are using aerial imagery for mapping. OS run two planes, one flying with a traditional film camera and one with a digital camera, to collect their own imagery of the country. They make sure that the imagery they collect overlaps and then use all sorts of advanced processes to convert this into a 3D model - processes such as wearing funny glasses while staring at a screen and clicking the imagery when features overlap. We saw some incredibly detailed imagery - yes, better than google maps - but unfortunately I didn’t get to see their imagery over my house and find my car.

After this mini-tour we were then given a longer introduction to the OpenSpace project itself, their reasons for launching it (a combination of “we want to” and “we had to”) and the terms and conditions of use, which unfortunately turned into an hour long discussion that nearly scuppered lunch! The terms that OS are using have various provisions for deriving data from the maps (which suggested a npemap.org.uk type postcode collector might be allowed) but also don’t allow for commercial usage. After we did finally have lunch we all signed up for the API and started hacking. Here’s what I ended up with, a low resolution plot of my route from London to Southampton:

Click for bigger

The API itself is built on top of the existing OpenLayers API. I think this was a really good decision by the OpenSpace team as the world really doesn’t need yet another JS slippy map implementation, and to have a large company developing with OpenLayers, can only help to improve that project. Rather than going with a WGS84 based system though, OpenSpace centres around the use of BNG (British National Grid). Though in theory this is an obvious choice for OS, considering that BNG is “their” system and is the system in which all of their raster based mapping is available (I think), the fact that they require coordinates to be entered using BNG by default is definitely going to lower the number of people that bother to use it.

Though a conversion system is provided by the API, I was interested to find that they are using the algorithmic method to convert BNG eastings and northings to WGS84 latitudes and longitudes, rather than the full National Grid Transformation OSTN02. I’ve never been too sure of the difference in accuracy between these and looking at the page on wikipedia about BNG I find it’s about 7m accuracy for the Helmert datum transformation which I assume they’re using. I guess that’s fine in most cases but Multimap does have a free web service available for doing the more accurate translation, get in touch with me if you want more information

When trying to put together the small implementation you can see above I also came across a few problems that also seemed to arise from the use of BNG. To give you a quick pseudocode run-down, my app basically does the following:

• Setup data
• On page load
  • Create new map object
  • Create GridProjection object for converting eastings/northings to lat/lon
  • Create Vector layer for drawing the line between points
  • Loop over the points for my route from London to Southampton
    • Convert position to BNG
    • Create marker at the point
    • Add the point to an array, for drawing the vector later
    • Extend a bounds object to contain the complete bounding box of the journey
  • Draw vector
  • Get centre of the journey’s bounding box, and the zoom factor to fully display it
  • Display the map of the journey.

I came across a number of problems when I tried passing my converted points into the OpenLayers.Bounds object and an array of the points into my vector. After playing around with it for a while I’d managed to get my markers to display, but not the vector, and the map wasn’t properly autoscaled to my journey. After looking at the vector example though I noticed that they were passing OpenLayers.Geometry.Point objects into the vector, and OpenSpace.MapPoint objects into the setCenter method. This is quite possibly a problem with OpenLayers rather than with the OpenSpace implementation, though I can’t be sure either way. After a little playing around though I found that adding the following line of code, then using the point variable with the bounds and vector objects fixed my problems:

point = new OpenLayers.Geometry.Point( mppoint.lon, mppoint.lat );

So, conclusions: I have to say that I don’t think that this API is ever going to change the world. There are so many mapping APIs out there that are freely available anyway that this one is just not going to have that effect (especially with Multimap offering the same landranger 1:50k data, in a WGS84 environment). That said though, I can see definite uses for it in the UK, and I know that many UK based organisations will be happy to use the BNG system. It also means that Ordnance Survey are going to be meeting their mandate to promote the use of their data and noone can say anymore that there’s no way to get access to Ordnance Survey data free of charge (or at least when the API is available outside the current closed release anyway). There may be lots of restrictions on use at the moment, but at least it’s something and hopefully this small step will lead onto a lot more open-ness at the OS. Also I definitely don’t want to put down the guys that have been working on this, what they’ve achieved is great, they just need to keep working on it, and working on the people above them to allow more data and more freedom of use for the data.