Sometime last year I conducted an interview with Greg Gershman, the founder of the RSS search engine BlogDigger. Here is the transcript of that interview:
SY: Hi Greg, can please tell us a bit about yourself? What?s your background and how did you get to start Blogdigger?
GG: My name is Greg Gershman, I live in Baltimore, MD. I?m a software engineer by day, and, lately, by night as well. I?ve been developing scalable web applications and middleware frameworks for the last six years, working for anything from government contractors to startups. Much of my work has been in a research and development capacity, either in prototyping new systems or examining future technologies. I?ve worked primarily in Java over the last 4 years, and have found it to be my language of choice.
I started Blogdigger in March 2003 while in graduate school as an experiment in using RSS feeds to improve search across blog content. I continued the experiment, eventually rewriting the original version to increase performance, moved from a shared hosting server and onto an individual box and later, multiple servers. It?s been quite an interesting ride thus far!
On the personal side, I?m married to Penny for almost seven years; we have
three kids, Ezra (5), Gila (3) and Eden (6 months).
SY: One of your most convenient services I personally think is Blogdigger Groups. I was thinking that day if there is actually such a service and believe it or not, I was actually talking to a few people if they?d be interested in taking this project alongside me. This idea of course came off once we checked out Blogdigger Groups. How has the impact of such a service been and what has been the response to it?
GG: Thanks! The response to groups has been very positive. Groups is quite a
unique application. There are many centralized RSS aggregators out there, but Groups is the only one that allows you to create a subscription list in a collaborative fashion. I?ve always liked the idea of combining the editable nature of wikis with the content nature of blogs, and Groups aims at that, giving you the option of allowing others to collaborate in the creation and editing of a Group by adding and deleting member blogs/feeds. The idea was to create an open, self-organizing directory of feed groups, so people looking for blogs or feeds on a specific topic could come, search through our directory of Groups, and get a ?starter pack? of blogs on a particular topic. As an example, we?ve got a comprehensive list of PR blogs (http://groups.blogdigger.com/prblogs), as well as a growing list of Women in Podcasting (http://groups.blogdigger.com/WomenInPodcasting).
As people began to use the service, they came up with ways of using it that were unanticipated. For example, since every Group is also available as an RSS feed of the latest posts from group members, people use Blogdigger Groups to create a single feed of their own content; blog posts/comments, Flickr photos, WebJay songs, del.icio.us links. Others use the Groups homepage itself as the homepage of a community site for bloggers of similar interests.
As for Groups? future, there is an update on the way that should improve the application, as well as add on some long-awaited features. This update represents not only user interface changes, but underlying changes to how a Group is created and edited, which should make lots of new things possible.
SY: What are some significant changes that we might expect to see over the next year with Blogdigger?
GG: In terms of our search engine, we?re going to continue to create new ways to search syndicated content. We already provide search by keyword, author, subject/tag, link and link type, as well as our Media search for audio, video and images. Others will be released soon, still others are in development. As an example, currently you can search Blogdigger in any character set (try this search for Shalom in Hebrew – http://blogdigger.com/search.jsp?q=%D7%A9%D7%9C%D7%95%D7%9D), and get back all the blog posts that mention any word in any language. I?d like to add in variables that allow explicit language selection.
As a business, I?m hoping that we can make some significant changes as well. We finally got incorporated after a few false starts, and are ready to look for partners and clients who can use our services.
SY: The search engine industry is getting quite big, and frankly, searching the web nowadays isn?t enough. You can search for all sorts of things, from Videos to Images to BitTorrent files. I noticed that a recently launched service by Blogdigger (http://blogdigger.com/media/) adds to it. Can you please tell us a bit about that service and maybe let us in to what the response has been so far?
GG: Blogdigger Media has been a big hit. In March 2004, we launched the first set of Media feeds of the latest links to MP3, WindowsMedia and BitTorrent files. These feeds were some of the first constantly-updated RSS 2.0 enclosure feeds out there. Over the next few months, we added other media types, such as QuickTime, Shockwave Flash and Images (GIF, JPEG). Finally, in December 2004, we launched the Media Search application, allowing keyword search and subscription for Media content.
The response to the feeds and the search has been tremendous. Many people ask me what the major difference between our Media search is from the various other audio/video/image search engines out there. The key difference is that, since we get our information primarily from blogs, and blogs by nature are tuned into what is current, Blogdigger Media gets current media content very quickly. As an example, we were one of the first to have video of the tsunami, while the other engines were delayed for a week or longer. Some Super Bowl ads made their way into our Media feeds before the game was over!
SY: Finally, what do you think is the future for RSS/Blog search engines (in specific). With things such as podcasting becoming so popular, searching for text blogs may no more be enough in a couple of years. What is your take on this?
GG: One of the main advantages of RSS-based search as opposed to traditional search is the clear distinction between content and metadata, allowing for a richer search experience The time-sensitive nature of blog content lends itself extremely well to subscription-based search services such as Blogdigger, where every search is also an RSS feed. That being said, there are ways to improve the current methods of searching time-sensitive syndicated content. Text is not the only content worth considering; for appropriately marked content such as podcasts and video blogs, RSS aggregation proves to be a quick and reliable way to index and search. In the future, RSS will fit well when used in conjunction with other new search technologies.