A noval visual search engine sleagle.com, Beta version, is published. Visualizing, managing, overlooking, clustering, and browsing search results on sleagle.com. Any feedback is appreciated.
sorry i don't like it, it's a bit weird an i don't really get it. i see screenshots and images which are constantly flashing and changing and that bugs me. also the logo kinda sucks too.
Thank you for your feedback. We'll try our best to improve it according to your response. Sleagle is in fact completed by two students. It's a hard work to design a new interface which is both feature-rich and very easy to operate. It needs users have a little bit patience to know how to use it, such as maximizing a group in the right hand side map, filtering out improper results, and so on. The whole interface is divided into two parts: 1) the left hand side ‘text area’, and 2) the right hand side ‘map area’. These two parts are consistent with each other. If some results are filtered out from the ‘map area’, the corresponding documents are removed automatically in the ‘text area’. If the mouse is hovering on the title of any result in the ‘text area’, the corresponding block and attribute button will be highlighted. Preview images will be shown in the left area if mouse hovering on any block in the ‘map area’. The ‘text area’ can be switched from the ‘preview mode‘ to the original ‘text list mode’ by clicking the ‘magnifier icon’ or ‘reset button’. Any suggestions is highly appreciated.
It takes a bit of getting used to but is interesting as it divides the results up in a way that is useful for some searches. Does it use Google results as a base?
I'd like a more visible explanation of what the graphics on the right-hand side are. It's not obvious to me instantly, which if you want this to catch on, it should be.
I took me a little while to get use to but it grew on me as I started to use the navigation on the right. Easy to see which sites are relevant when hoovering since it shows the targeted page. I think a lot of the displays like format, size, domain etc a typical user wouldn't care but we know it's beta.
I'm not sure if it's just my internet connection, but the load time for the internet is around 1-5 seconds. That's a slow speed when comparing to a major search engine like Google. It could be your servers. Nice logo as well, lol. It looks just like Google's, nice though. All in all, I like your search engine.
You are right that a typical user wouldn't care format, size, domain etc. But maybe it can be useful for some special cases. For example, for a stranger’s name Vijay V Raghavan, most of the results of query ‘Vijay V Raghavan’ can be found in the domain ‘edu’. And there is a cluster named ‘publications’. So we can guess that this guy must be a professor working in a university (edu). One of my friends has used sleagle to search some mathematical papers. He only browsed results of the pdf format with size greater than 20Ks (documents of smaller size will be only abstracts but not the whole articles) and filtered out all other documents. For this purpose, he said, the size and format legends were useful. There must be some others uses that can be found by users. But most users will not care about it for sure.