^Cursor transforms the currently widespread methods of search
engines to lower the environmental footprint of such
technologies and to address the scalability issues of the
exponentially growing Internet. PCs, laptops and smart-phones can be
turned into a web server with a single click that further expedites
the publicly available amount of data. ^Cursor also boosts the
flow of information by letting users update news feeds
immediately
For CaretCursor architecture diagram click here
Web crawling goes green
Search engine providers utilize a vast amount of centralized computers to carry out the task of retrieving information from the Global Computer Network. ^Cursor greatly reduces the need for clusterized resources and lowers overall power consumption by utilizing the resources of its volunteers.
Community Web processing
Would you like to be part of the community that builds a
web search engine?
HydraByte is currently seeking to spawn the collaborative
contribution of everyday Internet users in creating a pioneering
semantic search engine, called C^ret
(Caret, like the name of “^” symbol). The
^Cursor (CaretCursor) application is the
web crawler component of C^ret in which volunteer users can
opt to run a customizable semi-crawler on their own
Internet-enabled devices as a background process. This
enhances the retrieving and processing of
websites that will eventually assemble search queries.
How it works?
Unlike traditional web crawlers, ^Cursor does not access or expose your personal information. Instead, the application repeats the following simple process:
- requests a package from a server through a secure channel,
- the package contains a limited number of webpage addresses preselected by a semantic unit,
- your ^Cursor application downloads the text source of webpages set by these addresses but does not follow any anchors present in the source,
- the downloaded data is compressed and then stored only temporarily in memory,
- after the download phase, the data is gathered in a package and sent to a processing unit through a secure channel,
- the processing unit extracts webpage addresses from the sources of gathered webpages that serves as input to ^Cursor,
- you have the option to directly submit a webpage you find interesting,
- you can also choose to make the websites found by your ^Cursor be accredited to your username, where this accreditation changes on each turn of the month, designating the username of the top ^Cursor, creating a racing condition,
- you can watch the progress of your ^Cursor and set the speed at which it crawls, although its resource footprint is minimal (low memory / disk / network usage). This footprint virtually creates no additional load to the computer that runs it (only textual information is downloaded by the Cursor).