My Thoughts

Today, while looking into the new Google Maps Docs, I ran across the Styled Maps. They are quite powerful! While Cloudmade offers a similar service for a while, I didn’t expect Google to offer this, and for free.
A bit of technical background: these mapping services work by splitting the world into tiles. The whole world takes millions of tiles (depending on zoom level), and thus hundreds of gigabytes to render and store. While rendering and storing the world once isn’t that much effort for Google, rendering it in as many styles as it’s customers come up with does take quite a bit of processing power (For more background on the subject: See my Thesis on LBS)

While I have run a tile render server in the past, this was always an expensive, and not that scalable operation. So I’m suprised (but happy!) that Google now offers this to the world, for free! In just a few minutes, I was able to create the demo Waterworld shown above.

More examples can be seen on the Google Maps blog

Last week VGZ op Reis went live. It’s a mobile application for people on vacation, build by me.

It’s the first application I’ve worked on that simultaneously is released on iPhone and Android. Below you can find a (dutch) screencast which explains the functions of the application. It features several phone numbers that can be usefull on vacation, you can save your own phone numbers to it, you can share your location by email, and the iPhone version also allows you to create your own iPhone Wallpaper.

The design of the application was done by the brilliant guys at BUROPONY.

It’s always cool to work on new mobile platforms. And since I’m a big believer of HTML as a future cross-platform development environment, the Vodafone widget platform shows a nice promise. I had done some experiments with the platform, but now the first widget went live.

Together with The Saints I developed a widget for Sensation for the Vodafone platform. It shows the latest video’s from Sensation’s Youtube channel. It also shows the news around their world-wide events, and the latest tweets from the official Sensation channel.

For those running the Vodafone widgets, or owning the H1 or M1: It’s available in the Vodafone widget market. http://widget.vodafone.com/dev/widgets/sensation_16528

Since most of the content of Sensation is on public platforms (Youtube, Twitter), which all have an API, the widget could be developped fairly easy without further technical support by the organisation.

I regularly work at the NOS, the Dutch Public News broadcasters. Today the iGoogle widget I built for them went live. It shows the latest video’s from the website, in your own iGoogle page, or other iGoogle widget or Opensocial container. The video’s can directly be watched in the widget itself.

You can install it by clicking this link or read about it on the NOS New media blog (Dutch).

Apple seems to be a big proponent of HTML5: They use it for their new iAds, as Steve proudly announced during his last Keynote. So when I wanted to maken an Ipad webapp, I didn’t expect any issues. For a client of mine, I planned to work on an iPad webapp this week. It’s even the reason I bought an iPad.

Let’s start with the only good news: Setting an homescreen image, with

<link rel="apple-touch-icon" href="/custom_icon.png"/>

does work, and if you make the image 72 pixels wide, it looks great on both iPhone and iPad (the previous advice was 57 pixels, but this looks blurry on the iPad as it’s scaled up).

On the iPhone, there are some other properties you can set on a webpage to make it work offline: Besides the homescreen icon, you can set the status bar color, startup image, and viewport size, and make files work when not online. The result looks and works like an app instead of a webpage: No URL field, navigation buttons, a startscreen, etc. Apple invented these, and has extended their support in their OS updates. See for a good description: How to Make an HTML5 iPhone App and a great HTML5 example app is the PieGuy game. They work great, and can make it easier to build and distribute an application, without sending it to Apple for approval (the current appstore mess is a whole different blogpost).

Since Apple supports this in their iPhone OS, and the iPad runs iPhone OS 3.2, I expected this all to work. Unfortunately, it doesn’t.
On the iPad, the startup image and viewport size are ignored when run in webapp-mode. The viewport size is only used when shown in a normal browser window, and the startup image is never shown .

On an iPhone with OS 3.1, both do work. This makes the result, on an iPad, work more like a (broken) webpage, than an application.

Why did apple take working stuff out, when upgrading the iPhone OS from 3.1 to 3.2 for the iPad? Is this a simple omission which will be fixed, or is apple actively moving away from HTML5 webapps for the iPad, further pushing their appstore? Only time will tell.

Update:

I looked into this some more, and have more results:

If  you use a image with the right size, (1024*768), it does (sometimes) work on my iPad as startup image. My previous tests used a iphone-sized image, which is totally ignored on the iPad. However, it often takes a few tries before it shows the loading image, and sometimes shows a screenshot of the page, before showing the default loading screen. (so the flow is then screenshot->default.png->webpage).

I still can’t get the viewport tag to work in webapp mode: If I set the viewport to a predefined with, like:
<meta name=”viewport” content=”user-scalable=no, width=1024″/>
this does work in Safari, but is ignored in Webapp mode. If I set a scale:
<meta name=”viewport” content=”initial-scale=1.0, user-scalable=no”/>
and test in Safari, this does work on first load. However, if I rotate my ipad, and rotate back, the scale is set differently.

At the Mobile Monday Amsterdam, which I co-organise, I did a short talk on how to build your own Internet of Things Thing, in a day. The technical process is explained in my blogpost from last year: called The MoMoMeter.
Since we videotape all our talks, it’s included here:

The slides:

More on the IOBridge I used can be found on their blog
Photo’s by Filip Bunkens.

Last week I gave a talk at Social Strategy Talk on the crowdsourcing aspects of OneframeOfFame. Both the crowdsourcing of the frames, and the crowdsourced moderation are described. The slides can be seen above.

One computer to rule them all: why the iPad will be big.

The iPad is seen by lots of computer-savy people as too slow, limited, etc. However, after spending a few hours getting the PC of my mom up to speed, a few things became clear to me. Especially that her needs from a computer are way different from mine. She doesn’t need a filesystem,right clicking, context menu’s, and doesn’t want virus scanners and backups (but then I guess nobody wants those).

This article rang quite true for me: Why are computers for power users the same boxes as computers for experts? When you think about it, it’s quite crazy my mom has a computer (actually: an OS, it’s not a hardware thing) that can run a compiler. She’ll never compile anything, and doesn’t want to. She probably doesn’t need tabs in her browser (I usually have 100+ open), or even run the browser and the email client at the same time.
Currently, the only difference between a computer for a power user, and a novice is specs and price. Both will (in most cases, linux excluded) run the same OS, with the same options. The software might be different, but even there I run the same mail client, browser, and picture management software as my mom (although on a different OS). That’s strange, as I spend about 20 times as much time behind the screen as her.

Abstractions like directories are quite a hard system to understand, and I’m more and more convinced that the iPhone OS list of pictures (and, a similar list of docs) would be a way easier mechanism. Maybe less powerful, and not workable for the code of a website, but good enough for my mom.

Current computers are incarnation of the old typewriter: Even the form factor has only changed slightly (A box on a desk, with a keyboard). But the way we use computers has changed a lot: The first users were mostly producing content (letters, spreadsheets, etc). Today a lot of computing is the consumption of content (Youtube videos, browsing social networks, etc). Still we use a device with the same form-factor. Maybe we need a different device to consume media, which will suffice for 90% of the population?

I’m starting to feel the iPad is just that computer. Sure, it’s limited, but the things it does it probably does well. And it won’t have viruses, different browsers, or even right clicks. I have the idea that’s the right model(and it will make my life better as well!). I don’t know if the current version or form-factor will be perfect, but I can definately see computing as a whole split: An iPhone like OS for simple tasks, and a powerfull OS for the geeks, who need to run compilers, photoshop, etc. And maybe I like to use such a simplified system as well…

(Actually, the above is the same gripes I have with my MCE PVR: I hate to reinstall my VCR every few months, but I guess that’s a similar but different blogpost).

There have been some great responses to OneFrameOfFame.com on twitter. Here an except:
read more…

While Ford invented the Assembly line around 1910, I didn’t expect to employ a large group of people to work for me, on repetitive tasks. However, this week, I had 1052 people working for me. As part of the video clip project One Frame Of Fame we have thousands of users submit a frame for our video, with their webcam. These are all volunteered.
But to make sure all frames are actually people participating in the project, and to rate how well the users do, we use Amazon’s Mechanical Turk, via Crowdflower.
All images are rated by at least 3 people (but the average number currently is around 8 votes/image). Last night I looked into the statistics, and sofar 1052 people have worked for us, doing 61.000 ratings of our images. The top workers have rated thousands of pictures, while the people at the end of the longtail of workers have only participated in 18 frames (we offer images in groups of 18, so that’s the minimal amount of work that can be done).

Strangest thing is that, while these people work for us, I have never met them (and I probably can’t even get in touch). It’s a too large group to have a beer with, or even talk to. They might be on the other end of the world, yet they are clicking away on our project.

Without this world wide web thingy, this worldwide collaboration of thousands of people would never have been possible.