Who made the net? A new HTML5 microformat with answers.
If you watch a film, listen to music (if you have the inlay), go to the theatre or read a book, it’s generally easy to find out who was involved in making that thing happen. Websites creators however are practically anonymous and are more similar to commercials or posters in that you need to do some searching to find out.
This is strange because websites don’t have the same limitations that posters or commercials do and are actually far more suited to containing information about the people responsible for making them than film, theatre or music. So doesn’t it seem odd that there isn’t a clear way to find out who made a website?
Well, there actually are a few ways, but none of them seemed like a good solution to me – so I decided to come up with one by myself.
As part of a 24 hour hackathon a few days ago, I came up with a solution to this problem and with the help of fresh-out-of-uni-break-dancing-html-css-js-rising-star Jeff Lau we built two websites. The stats fans among you will be able to work out that that’s an average of one website every 12 hours.
You may have guessed, but the websites are http://www.whomadethe.net and www.wemadethe.net. Yes, the first thing I learned from this experience is that .com names are almost impossible to come by these days. Even the most obscure name you can think appears to be taken and irritatingly will almost always never be used by an actual website.
Anyway, back to the point of this article…
Some of you will probably be saying “Oh yeah, that’s just like that humans.txt thing isn’t it? I don’t know why I wasting my time reading this when I could be reading Reddit”. Before you go on Reddit and start arguing about whatever it is people argue about on Reddit, let me talk about humans.txt quickly.
humans.txt is a great idea by some clever Spaniards who seem to share the same frustrations that I do. The idea is to add a file to your webserver called (unsuprisingly) humans.txt which is a free form text file containing information about who build the site. It seems to have taken off to some extent with a template for humans.txt included in the massively popular HTML5 boilerplate and even Google are having fun with it.
However, although this is a great idea and it’s an excellent name (it contrasts with robots.txt which is a file used to tell search engines about how they should index your site) for me it has one major flaw. The data is free form – you can structure it any way you want. This means that it’s very difficult for other people to use that information. Also, it’s a text file. It doesn’t look good in your browser. It’s got no personality. What would a James Bond film be without their awesome intro credits?
So it seems to me like humans.txt is missing a couple of tricks. Fortunately, another load of clever people have already come up with most of the solution.
Schema.org and Microformats
Microformats and schemas are ways that you can markup your HTML to help other people and machines to work out what you’re talking about. Search engines already use microformats to help them display appropriate information when you search for things or use Google Shopping to compare prices of products.
The microformats and schemas are related but have different implementations and as far I understand the two groups have united in some way to try and create a standardised approach.
So I thought: “Why don’t we use microformats to label HTML to indicate who built a website?” It solves the problem of humans.txt being hard to use by others and also means you can do whatever you want on your credits page as long as it has the microformat information in it. Two birds, one stone.
Helpfully there’s already a schema for “WebApplications” which is closely related to a MobileApplication schema which you can see in use on the Google Play store. I’ve had a look around and can’t find any examples of this WebApplication schema actually being used… until now!
Alright then, how does this all work?
That’s why we built www.wemadethe.net. It explains how to create a credits page and how to label the HTML appropriately. We’ve had to adapt the WebApplication schema a bit to make it work how we think it needs to. We’ve also taken a leaf out of humans.txt’s book in places too.
To let everyone know where your credits are, we thought a link with a rel of “credits” would make it clear. Humans.txt uses a rel of “author” but we felt that “credits” implies a different thing. The great thing about this is that a user doesn’t have to download the credit information, but other people know where to find it.
<link rel="credits" href="credits.html" />
Then, you create your HTML and add some attributes to let people know what you’re talking about. (See wemadethe.net for examples).
We were keen to create a way to uniquely identify people involved in a project. The immediately obvious way is to use their email address, but I don’t think many people will want to put their email address somewhere spambots could easily get to them. So, for moment we added a “twitterHandle” to the schema. Ideally an OpenID type solution would be better.
The current WebApplication schema also talks about authors, editors, creators and contributors as people or organisations that could be involved in creating a WebApplication. We didn’t think those were all appropriate to web projects and that other types of people and organisations would be involved. So instead we’ve put down a “credit” type which has a “role” which can be defined as people think best.
<li itemprop="credit" itemscope itemtype="http://www.schema.org/Person"> <span itemprop="name">The Emperor</span>: <span itemprop="role">Project Manager</span> - <span itemprop="twitterHandle">TheEmperor</span> </li>
Now you’re taking the piss!
I know it might seem presumptive to start changing schema that people have spent time working on, but the Schema people do say that you can extend it as you feel is best and that they’ll look at how people end up using them.
Either way, this isn’t about competing with existing ideas – it’s about continuing a conversation that has already started.
That’s all well and good, but what’s the point in all this again?
Funny you should ask, because to explain that very point we created whomadethe.net. This is a simple site that shows you how the new microformat could be used (by anyone).
All you need to do is add the URL of your site to it and it’ll do the rest. It’ll look for a link with the “credit” rel attribute to see if you have any credits on your site. Then once it’s found them it’ll pull all the data out and add it to it’s own database.
Hey presto! Now we’ve got a searchable database of who made that website. Using the unique id provided by the Twitter handle we can now easily cross reference people and organisations. So in 10 years time, you could theoretically come to this and see everything you’ve worked on in those 10 years.
Validated information – unlike CVs and LinkedIn
One of the great things about this approach is that the credit information can only be put onto the site by people with access to the site’s files – i.e. the people involved in making it. That means that the information provided in the HTML is going to be more truthful than the information you’ll find on people’s CVs and on LinkedIn.
A feature we want to add to whomadethe.net is where you can approve any credits associated with your unique ID (i.e. your Twitter handle at the moment). This would allow people to give their own information against their name if they felt they had been misrepresented.
There is a danger that this could lead to a lot of negotiation over credits and “fake” credits could appear, but other industries are able to cope so I don’t see why ours wouldn’t.
Ideas for the future
Although you have to add your site manually at the moment, a site like this could start crawling the web itself and start pulling in the data it wants. In that way it starts to cross over with the Internet Archive as a way we could maintain a historical record of who made what on the web. It’s a bit like IMDB for people who make things on the web.
The microformat could also be used to associate credits with Mobile Applications. By sticking a link to a website on the App Store or Google Play store page for the app (or whatever store), you could search for credits on that URL and associate them with the app.
whomadethe.net could also cross reference sites against awards websites. So a website giving an award to another website could put a link with an “award” microformat attached to it. whomadethe.net could then use that information to show that the site had won the award.
The sites were hacked together in the AngelHack London hackathon, so they’re a bit rough around the edges. We’re working on getting them a bit more polished and making it easier to people to generate credit information and add it to the whomadethe.net directory.
The main thing is that it’d be great to hear what people think about this idea, so we’ll need to start getting feedback to see if this thing will sink or swim.
So, let us know what you think in the comments below or by Twitter or by sticking a a flaming poo through my letterbox if you really hate the idea.
PS – if you’re wondering, the photo that accompanies this article is a slightly sinister photo of me representing the “who?” part of the question. The champagne glass suggests celebration, as if I’m celebrating making a website. These are the kinds of loose associations you learn to make to justify your work when you do an Arts degree at University.