The following screen shot shows what happens when you select all sources as the data source for your graph. Sales rankings from Amazon.com and Barnes and Noble are shown on the same graph. When the site is released, selecting all sources will actually show data from Amazon.com, Amazon.co.uk, and Barnes and Noble.

A lot of new features were checked in today. Since most were relegated to the back end, I’ll focus on a few changes to the signup screen. A previous post showed the new signup screen. The slightly updated version is below.

What’s new? The line under the logo has highlights behind the key words in the sentence. I also added a note out to the right of the email field letting new users know I won’t be abusing their email address. A block linking to the terms of service was also added to the form, and it must be checked for a new account to validate.

Some things you can’t see in the picture. The form implements tab indices correctly so that you can tab to subsequent fields. I encounter sites all the time that don’t do this and it’s frustrating. Also, any errors you have on the form will be highlighted when you submit the form. You’ll also get explanations of the errors so that you can fix them. The fields that were correct will remain (I’m endlessly annoyed by forms that clear all the values I typed in when a mistake is made).

Apparently the datacenter where Rankforest is hosted went down for a few hours last night. If you came to the site between 2 am EST and 9am EST and it was down, it was due to this. The power outages at the Dreamhost facility affected thousands of sites. Hourly rankings within that period are unavailable and the charts for today will be a little behind due to missing those points.

Currently, the comparison chart states that anyone with a free account can add a limited number of books. The current limit is actually 10. In preparation for the new version of the site, which is still a way off, I’ll be moving to limit the number of free additions to 1 new book (not currently tracked) and a few existing books (books that are already being tracked and that require no additional overhead). I’ll make the changes sometime after this Friday.

If you have a free account and have already added books, you won’t be affected.

If you’re new to the site, you can still add a book that you’re personally interested in. Data will pour in over time for the title and you can get a feel for the various charts. Additionally, you can add existing items to your collection to see how it feels to track multiple titles. As a benefit, these existing books will likely have much more data behind them so you can see longer term data now instead of waiting.

The main reason for doing this isn’t to limit or punish those with free accounts, or to goad them into paying for a subscription. It’s really about serving those with premium accounts better.

The new version of the site will track books from 3 sources and does more work behind the scenes. Spending time tracking free books takes away from spending more CPU cycles on the books premium subscribers are tracking. As the site grows, the number of books that can be processed in an hour grows close to a hard limit, so something has to give (one day I’d like that to be the point at which I buy another server, but for now this makes more sense).

While premium accounts will only ever be scaled upward in terms of functionality, the free accounts are a little more variable.

Thanks for your continued support.

I rolled out a new feature last night that enables archiving the hourly data for members with Pro accounts. The feature won’t work for those with free promotional Pro accounts from the 2006 launch; you need to have a real subscription.

To access archived data, click the hourly data link next to each item’s recent rank in your collection. You should see a page listing available dates, and clicking a date will generate a simple Excel CSV file. You can open it in a text editor if you don’t have, or don’t like, Excel.

I haven’t decided how long the data will be archived. The current implementation will be somewhat short-lived as the new site will debut in the coming months. It should, however, provide avid trackers with some peace as they will no longer have to wait until the wee hours to save off data points.

I’ve been spending time working on the backend processing system, trying to ensure that a book’s details are more accurate. Currently, some titles in Rankforest are truncated; some titles have multiple authors and yet only one is shown. Others have no author but they do have an editor. Additionally, in some cases the publisher gets listed as the author due to the organization of the Amazon XML responses.

Just a small update for the curious. Stay tuned for bigger announcements as release nears.