Search Tips

Here are some handy tips for using the VCN's search engine to find information hosted by our system. Our search engine uses a program called Glimpse, and you can get more technical information about it from there if you're interested.

Case Sensitivity.

Normally the search engine treats uppercase and lowercase letters as the same. This is called a "case insensitive" search. For example, if you were to search for "belch", the search engine would return any references to Sir Toby Belch, a character from Shakespeare's Twelfth Night, even though the "B" is uppercase.

If you want the search to be case sensitive uncheck off the checkbox labelled "Upper and lowercase text are considered the same." on the search page. Normally it starts out checked off for you already.

Finding Partial Words.

Normally the search engine will match whole words. But you can also have it search for parts of words as well.

For example, normally if you search for "info" the search engine will not find "information" for you. But if you selected "Search for partial words" then searching for "info" would reveal "information", "informed" - or even "spinfoam" if there was such a word.

Normally, separating words by spaces will have the search engine looking for both those words together. Thus, if you were to search for "Truth Beauty" the search engine would not be able to find "Truth and Beauty", because the word "and" there gets in the way.

The solution? Put in a semicolon and search instead for "Truth;Beauty". That will reveal any line with both "Truth" and "Beauty" in the same line.

One problem with computer searches is that they're usually so finicky. Which can be very problematic if you can't remember how to spell something. Our search engine lets you perform approximate searches, which is very handy.

Let's say you can't remember if the word you're looking for is spelt "deceive" or "decieve". Well, you could look it up in a dictionary, where you'd find that the first spelling is correct, or you could use approximate searching. Since there are two letters that differ between the two spellings select "2" from the "Number of misspellings allowed:" item and search for "deceive". The search engine will then find "deceive", "decieve", "decoove" and so on.

Another useful example might be if you can't remember if there's an apostrophe in a name or not. (is it People's? Or Peoples? Or Peoples'?) Just type in "peoples" and set the number of misspellings to 1.

Try to be fairly specific when you're searching for stuff. If you use a really general term you'll end up with zillions of references and it won't be very helpful.

So searching for words like "and" or "but" won't get you very far. Neither will using commonly used words on the Vancouver CommunityNet like "Vancouver" or "Community". Try to search for more specific terms.

Normally the search engine will display up to 100 files that match your search terms. You can change this to whatever you want.

Normally the search engine will display up to 30 lines from a document that match your search terms. You can change this to whatever you want.

That can happen. The search engine compiles a static (unchanging) index which it uses to find your information. If someone moves or deletes a Web page between the time the indexing happens and the time you decide to look for the information, you'll end up with a false hit.

Right now the search engine is set to index the entire CommunityNet once a day, at 3:30 AM Pacific time, during which time the search feature will not be available. It takes an hour or two for the index to be rebuilt.

As noted above the search engine does not update files continuously. If someone makes a change to a file then the search index won't reflect that change until sometime past 3:30 AM, when the indexing program is run.

Also, if the name of a file begins with the letters "ni-" then the contents of that file will not be indexed. This is intentional. Sometimes you don't want a file to be searchable, and so the no-index flag indicates that.

The search engine indexes files in their entirety. The problem with this is that most of the files on our system are HTML (Web) files and not plain text files. HTML files contain hidden information, called "tags," that define the appearance of the page. If you search for text that happens to appear in one of these hidden tags then the search engine will display a blank line as one of your hits. Ignore it.