Amazon.com Platinum Visa Card

Why this site exists

The idea behind this was to provide a specialized directory of the Web which would serve as a knowledge base for all matters typographic on the internet. To do this we needed to do two things that directories like Yahoo! do not:

1. Deep link: Rather than just linking to the top of a web site, we link to as many internal pages as seems appropriate. If you're looking for information on Bruce Rogers' Centaur, you should be able to go straight to it, rather than having to guess that it's on the Serif web site.
2. Provide content: Not all information lives on the web. As a result, we're putting up pages with information that might otherwise never be presented as well as linking everything that we can find.

And of course we're really all about selling the magazine.

Notes about indexing

No more often than once every 15 days, each link in the directory is checked to see if it is still active by sending an HTTP "HEAD" command to the server. Any page which fails this test on two consecutive attempts is marked as inactive and removed from the directory. This helps us to eliminate dead links.

This does not, however, detect links where the page content has been changed or has a redirect page. If a link undergoes such a change and you encounter it, please let us know.

Privacy policy

We will never sell, rent, give away or otherwise distribute e-mail addresses collected by this web site. We hate spam even more than you do. This will not change. Ever.

Any demographic information collected is used strictly to create a composite profile of registered visitors to this web site. We may, at some point allow advertisers to restrict ads to particular subsets of the profile, but we don't have that technology yet.

We track the number of times that links in the directory are clicked. The IP addresses of the last few clickers are recorded to prevent someone from boosting their site's rating by repeat-clicking on their link, but this information is not used to connect clicks to users.

There's a bunch of stuff that gets dumped into the logs which we may look at occasionally, but we don't connect log data with individual users. We could, but it's a pain and we really don't care that much.

How this site is created

This site is a mix of multiple technologies. The directory tree of the site is used as its primary organization. Each link in the directory is represented by a file which contains basic information about the page being indexed (name, description, listing date, etc.). Any additional information specific to the page is contained in a series of XML files in that directory.

If a directory has any content pages, it's converted from XML into HTML and a directory entry is created for it.

A perl script, then moves through the tree on a nightly basis and creates an XML file for that section which is then converted into a PHP file for the final display. The PHP contains code to handle determining whether to add the and icons to the directory entry, adding the banner ad at the top of the page and handling all the personalization issues.

To-do list

1. page submission back-end
2. advertising link
3. Alerts to webmasters on annual(?) basis reminding them of their listing and checking for new pages--this will also notify for a new listing
4. Login/signup pages activated/created
5. Everything above completed and we can start the launch
6. Internal search
7. Do a nightly refresh of a "top 10%" page
8. Link to goto.com for global web search
9. Spidering for page search
10. Better front-end for editing entries in the directory
11. Directory updating should be run automatically in the middle of the night... maybe nightly, maybe not. This can wait until we're officially released.
12. User reviews of pages/sites
13. Pages pages pages