Some Backstory

Almost a year ago, one of the greatest - or at least, what I considered to be one the greatest - websites went down. Its name was 4scrape, and it allowed you to search for wallpapers in a way that's hard to describe, but easy to fall in love with. Not being able to live with this, I set out to make a website similar to 4scrape.

What is Wallchan?

"Wallchan" is the name I decided to give my project. Its name comes from "wallpaper" + "channel" put together ("chan" can also mean the kind of websites it crawls, such as "4chan", "7chan" or "2chan".). While similar to 4scrape in many ways, with the progress I have already made on it, I can safely say that it's also much better than 4scrape ever was. For example, I have a working implementation of generating histograms for each image, allowing users to search by colour. What this means is that if a user wanted a wallpaper that was predominantly red, they would be able to search for this, and find all red wallpapers. Searching for more than one colour is supported as well.

Features

- Automatic tag generation
- Automatic histogram generation per wallpaper
- Never downloads the same image twice
- "Desktop" images are filtered out, leaving real wallpapers
- Everything possible about a wallpaper is stored in a database
- Some of this info is size, ratio, colours, tags, etc
- Static generation of wallpaper pages to reduce strain on server, and increase speed for users
- The scraping is done on a different server than the one serving content to users
- Much more!

Where is it?

Excellent question! Wallchan will be released to the public before the end of 2010. I'm aiming for August, but you never know how long the kinks take to iron out.

Any demos?

Nope, but I plan on showcasing some of the technology pretty soon. When I do, I'll add videos here. For now, you'll have to take my word that it's pretty cool watching databases and folders populating with wallpapers and data. :)