Sphere Blog Search
Sphere is a new blog search engine founded with $3.75 million in capital from Trident Capital, Hearst Publishing and Scott Kurnit. That’s a lot of money–so what do they have to show for it, so far?
Blog Search
This is the core business, of course. Indexing blogs, tags, and letting people find them. For this, they seem to do a decent job, indexing things within the hour, and are probably constantly backfilling websites for more data. As a brief search confirms that they’ve hit my site and indexed some of it, a trip to the server logs shows that there’s a new spider identifying itself as “Sphere Scout&v4.0 (beta) - scout at sphere dot com” with ip address 64.40.115.32, the same as sphere.com resolves to.
There are the usual popular queries options on their homepage, showing top queries per hour and per week:
They also let you get an RSS feed for a given search term, say Chernobyl. You can adjust the language parameters, the sort type, and buffer by dates.
Custom Date Ranges
The feature that Sphere is currently hyping is the ability to quickly choose a custom date range off a histogram. This is their only unique feature so far. For example, this query for mohammed cartoons gives us the following slider:
This is not compelling for several reasons. First, it’s hard to get to. I have to select the “custom date range” option from a dropdown, wait for their slow page to reload, and then finally I have the custom date range slider. Second, it’s unclear what dates I’m selecting, as there is little to no captioning. Finally, on their partner page, they admit they were donated most of the graphics code for this component:
Adaptive Path/Measure Map - nothing but love for the lovely people at Adaptive Path. They designed our site, saved us from ourselves and then gave us the source code for our histogram feature - talk about a culture of help!
So the design of the site and their interface is all given to them by Adaptive Path. And, while it provides a solid base on which to grow new features and ideas, right now there’s nothing innovative or useful here.
Blog Profiles?
Sphere also has blog profiles; here’s mine. They’re rather sparse right now. For example, I just get to see my last three posts (one of which is duplicated strangely), that I’ve made 8 posts in the last 7 days, 14 posts/week (average), that 788 blogs link to this blog, and that I use 1244 words/post (average). These are things that I already know. If you want to see a good profile system, take a look at my Kinja Card.
Conclusion
Visit Sphere every few months and see if it gets better. Right now I’d stick with Technorati.
Related Links
This entry was posted on Tuesday, May 2nd, 2006 at 1:50 pm and is tagged with hearst publishing, trident capital, mohammed cartoons, server logs, range option, sort type, decent job, histogram, core business, slider, sphere, search term, scout, source code, queries, parameters, beta, interface, search engine, nothing but love. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback.
on May 2nd, 2006 at 8:55 pm
very cool, i will have to try that sometime.
i like this blog alot
whee
on May 2nd, 2006 at 11:38 pm
Sphere Blog Search Engine Needs To Train Old Dogs…
Sphere, a blog search engine, finally launched today (May 2, 2006) after much anticipation with a new look and logo since first hearing about Sphere in October 2005. My first impressions was that it presents a very simple interface…
on May 18th, 2006 at 12:43 am
[…] I tracked it down to these guys, but found better information from this blog. Short version is that it is a blog search service, in beta. Not very good yet. And they have all of 2 of my posts so far, even though they have indexed, well, a log of them. Of those 21 hits, 11 were for specific posts, 4 were for “/” (the main page) and 6 were for “robots.txt” (didn’t they figure out the first 5 times what an error 404 is?) I’m thinking I’m just going to block their IPs, as they are hitting me all night (so I hear the drive spin up all the time.) And geeks have a good reason I should not block them? […]
on November 17th, 2006 at 2:42 pm
Almost four million for a bot that shows that kind of spidering behaviour?
64.40.115.62 - - [17/Nov/2006:11:31:36 -0700] “GET /robots.txt HTTP/1.0″ 200 117 “-” “Sphere Scout&v4.0 (beta) - scout at sphere dot com”
64.40.115.62 - - [17/Nov/2006:11:31:37 -0700] “GET /feed/rdf/ HTTP/1.0″ 200 23347 “-” “Sphere Scout&v4.0 (beta) - scout at sphere dot com”
64.40.115.62 - - [17/Nov/2006:11:31:40 -0700] “GET / HTTP/1.0″ 200 45603 “-” “SET USER AGENT”
64.40.115.62 - - [17/Nov/2006:11:31:40 -0700] “GET / HTTP/1.0″ 200 45603 “-” “SET USER AGENT”
64.40.115.62 - - [17/Nov/2006:11:31:51 -0700] “GET /robots.txt HTTP/1.0″ 200 117 “-” “SET USER AGENT”
64.40.115.62 - - [17/Nov/2006:11:31:52 -0700] “GET /feed/ HTTP/1.0″ 200 24898 “-” “SET USER AGENT”
64.40.115.62 - - [17/Nov/2006:11:31:53 -0700] “GET /feed/ HTTP/1.0″ 200 24898 “-” “SET USER AGENT”
64.40.115.62 - - [17/Nov/2006:11:31:53 -0700] “GET / HTTP/1.0″ 200 45603 “-” “Sphere Scout&v4.0 (beta) - scout at sphere dot com”
Gosh - where do I sign up for my 4 million? I have a ten line perl script that does a better job than the thing above …
Sigh