Some of the Dcomments.com features:

Automatic extraction of site's favicon.ico and converting it to .gif for displaying next to site's news. 
This makes it easier to identify and to spot a familiar blog just by noticing its favicon.

If blog entry (article) has images, they are extracted and thumbnailed.

If blog entry does not have any images, then a screenshot of the web page is made and thumbnailed

Number of links and number of images in an article is calculated and displayed under the excerpt.

Readability score of the article is calculated.

Feed XML is validated by a feed validator.

Important keywords 'tags' are extracted from an article. These are used to create 'tag clouds' per site, per-category and per-blog. Tag clouds use the values from the past 7-day period and changing every 24 hours, so the tag clouds are always 'fresh'

If the feed provides a <category> tag, its value is also recorded. This way the entries from the same category can easily be found.

A text-only version of the item is created and can be used by users who prefer to read the news in text format (very useful for small devices like cell phone browsers and for users on slow connections)

There are also some features that take place on the back-end like analyzing articles, feeds, generating statistics.

While this project is not an 'Open Source', we will be publishing some script samples for some of the site's features

Check back on this page soon.

For now this is a short list of what we use for development:

Linux, Apache 2.2.X, MySQL 5.0.X php 5.2.X, memcache, lots of pear classes, perl with several classes from CPAN

For backend a cgi version of php, perl