Searching the CS websites

We actually have several publicly viewable websites here at CS. There is the primary site and in addition several projects have their own servers. To further muddle things, we have a content management system, conventional web pages and cgi pages.


(OK, big but not very friendly, I'll work on that.)

To help you out, there are several ways to search our site(s). The content management system has a search of the pages it manages. That is the search box at the top of a page with a sidebar. Changes to these pages are promptly reflected in the search results but only those pages in the content management system at itself are indexed included.

We also have an Extended Site Search which is a swish-e1) index and includes all pages on our CS web site and several research project sites or specialized sites CS folks are involved with. This search can be reached by either the Extended Site Search link in the sidebar, or by selecting the Extended Site Search graphic in the footer of many of our pages. Currently sites included in this search are:

The content of all html, xml, pdf, txt, java, and ppt pages on the above web sites are searchable with the extended site search. You can also seach for file and path names if you are looking for something in particular.

For most searching you can just type in a few words that you want to find. Swish-e will return pages containing all the words you enter. The search form itself has buttons for several common options. Additional options are also available.

To look for a phrase you need enclose it in escaped quotes(\"), for example:

\"wearable computer\"

Which will return pages that contain the phrase wearable computer2)

Leave one and two letter words out of your search as these are ignored by the indexer, so if want only pages with the phrase will your next computer be wearable the closest you can get is to search for:

\"will your next computer wearable\" 

Which will get you your desired match while

\"will your next computer be wearable\" 

Will get no matches.

To look for a file by filename or path you can specify the part you know with @@swishdocpath@@, for example:


The above search will show you pages which have cs353 in their path or filename. All the course pages at are in directories which begin with cs followed by at least the first three digits of the course number. So this is a good way to limit your search to a particular course, courses or course level.3)

Wildcard search is available with *, and of course you can specify several terms, so for our last example:

Prerequisite swishdocpath=cs3*

Will return all pages with the word Prerequisite on files for 300 level CS classes

Now, go look around and see what you can find!

Greg Priest-Dorman 2008/05/02 11:20

swish-e stands for Simple Web Indexing System for Humans - Enhanced you can decide for yourself if they met that goal for your use. It is a hard thing to blend advanced search capability and ease of use. Swish-e does do quite a good job of it for us. The prof of course will be in how long it takes you to find what you came here for…
as well as wearable computer with any 1 or 2 letter word between them as short words are ignored by the indexer
Prof. Ellman's courses are not listed at but on the campus vspace server.  While his pages are indexed here, they do not follow this convention and as a result would not be included in searches using swishdocpath=cs*. As Grace Hopper said, “the wonderful thing about standards is that there are so many of them to choose from.”