urlHosted

home

What is this?

urlHosted is an experimental web app that misuses the part after the "#" of a URL to store and read data.

The app is unhosted. See this definition from unhosted.org:

Also known as "serverless", "client-side", or "static" web apps, unhosted web apps do not send your user data to their server. Either you connect your own server at runtime, or your data stays within the browser.

This means this app neither stores nor sends any of your data to any server. Instead, the link that loads the site must bring it's own data. That data—if it can be read—will be displayed.

This is FLOSS—Free Libre Open Source Software under the GPL v3. The repository is at GitHub. The Software is hosted for free directly from GitHub Pages and anyone can do this!

How does it work?

Simple. Whenever you visit the site with payload data in the URL, the app renders that data as an article. When you visit the site without payload data in the URL you are directly presented with the editor interface.

At any time you can toggle the editor interface using the Edit button at the top of the app. Regardless if you came with a payload link or not. To make things easier, the text input for the main article text supports Markdown and also a subset of HTML.

Press the Share! button whenever you want to share your article or just store it in your bookmarks list. All data from the form, plus the current date and time is serialized and a new link is created. A box opens and presents the link to be copied. Note that this box closes as soon as you modify the data, because the generated link is outdated then. You can also close the share box at any time by hitting the Share! button again.

Why?

Well, it's an experiment. I am eager to see if someone will make something useful out of this. Either by being inspired or by using it directly as a blogging platform.

This tool has already a useful predecessor that I made shortly before. It is a tool to create web-font test cases instantly in a font development project. These test cases can then be stored in the according issue thread in the bug-tracker, just like here.

It will also be interesting to see where the technical limits are. URLs are designed to be short and to point to resources, not to be the resource by themselves. That said, I like the idea of saving these URLs at URL-shortener providers. Take twitter for example, you can instantly tweet pretty long texts, without having to host them anywhere. t.co, goo.gl or bit.ly will be your host.

Who?

My name is Lasse Fister. My contact data is here.

Caveats

I don't know about the technical limits like maximal URL length in different scenarios or how save/persistent your data is when stored at a URL-shortener. I also don't know how the legal situation is for hosting a service like this. If there is trouble I will probably take it down when indicated and see how to resolve that.

What else?

Here is a kitten
cat
CC-BY-SA-3.0 by Wikimedia user 1qaz2 (has no profile)