Drawn

Actually, that’s a really great cover you chose too. The good old days of the Hulk and artists like Marie and John Severin.

Actually, that’s a really great cover you chose too. The good old days of the Hulk and artists like Marie and John Severin.

Cover Browser is such a great site. It would be such a phenomenal reference tool if they made all of the art searchable by tags.

Cover Browser is such a great site. It would be such a phenomenal reference tool if they made all of the art searchable by tags.

Gerren, thanks for the kind words… please note that the site actually does have a search engine! You can find it right on the homepage ( http://www.coverbrowser.com ) or on every particular comic book series.

So far, the search is hooked up “only” to the automatically crawled keywords. Those are 220,000 keywords already that will find you a lot of things, you can search artists, writers, things, comics, comics + number, and so on. However, those automated keywords are also often “noisy” simply because the crawler algorithm isn’t perfect ( http://blog.outer-court.com/archive/2006-10-09-n22.html ).
Basically, the automated crawler will analyze… “what did people talk about in web pages that mention ‘amazing spider-man 294′? which words did they use?” This works good on many instance but leaves room for improvement; for example, people rarely talk a lot about “horses” just because a horse happens to be on the cover. It just doesn’t seem that noteworthy. But what if you want to find a cover illustration containing a horse? And that’s where I hope the community tagging can really shine, next to actually being a huge improvement to the tags quality in general (e.g. you can often see the artist’s name on the cover, and this way the tagger game allows us to add this as keyword).

So far, it was really a fresh “labs” project, but now that it was featured on Drawn, which I’m really grateful for, I can get going to start hooking up the human-helped keywords database to the live search engine (18,685 unique tags at the time of writing, which means much more submissions — the more often the same tag is submitted, the more “weight” it gets in rankings), and also, add a couple of new comic book series to expand the tagger game!

Gerren, thanks for the kind words… please note that the site actually does have a search engine! You can find it right on the homepage ( http://www.coverbrowser.com ) or on every particular comic book series.

So far, the search is hooked up “only” to the automatically crawled keywords. Those are 220,000 keywords already that will find you a lot of things, you can search artists, writers, things, comics, comics + number, and so on. However, those automated keywords are also often “noisy” simply because the crawler algorithm isn’t perfect ( http://blog.outer-court.com/archive/2006-10-09-n22.html ).
Basically, the automated crawler will analyze… “what did people talk about in web pages that mention ‘amazing spider-man 294′? which words did they use?” This works good on many instance but leaves room for improvement; for example, people rarely talk a lot about “horses” just because a horse happens to be on the cover. It just doesn’t seem that noteworthy. But what if you want to find a cover illustration containing a horse? And that’s where I hope the community tagging can really shine, next to actually being a huge improvement to the tags quality in general (e.g. you can often see the artist’s name on the cover, and this way the tagger game allows us to add this as keyword).

So far, it was really a fresh “labs” project, but now that it was featured on Drawn, which I’m really grateful for, I can get going to start hooking up the human-helped keywords database to the live search engine (18,685 unique tags at the time of writing, which means much more submissions — the more often the same tag is submitted, the more “weight” it gets in rankings), and also, add a couple of new comic book series to expand the tagger game!

sweet cover

sweet cover