 |
|
|
|
|
Concordance
The listing of words from a text, along with every
contextual occurence of the words. Concordance is very valuable
for research, and very rare to see in search engines, because it
requires efficiency to do well. PicoSearch offers concordance, and
we call this feature "results in context".
|
 |
|
Entry Points
Essential site design determines that your website should
have a
limited number of well-crafted entry points, where visitors
will
find the introductions to your services and links to pages of
increasing
detail. For most sites it makes sense to have just one entry
point,
called the homepage, where a navigation bar and/or site-map
can
begin. For umbrella services however, there may be several
entry
points, including those for partner sites and/or special
offers.
PicoSearch can build your search engine starting from more
than
one entry-point: up to 3 for Free Accounts, and an unlimited
number for Professional and Premium Accounts. Using this ability you can
control the packaging of related websites,
even if they are in different domain and server spaces. See
also Fully Qualified URL
|
|
|
Fully-Qualified URL
The fully-qualified URL is what someone needs to
type into a browser to get to your website document (also
often called the file or web page), and the PicoSearch Indexer
will need it to enter your site's entry points. It's okay to
use
just a referential URL in your HTML code if the location is
already
implied within the web page, but the fully-qualified URL must
specify
the resource name plus all subdirectories, the domain name,
and
the http://. For example, if you had
a web homepage called index.html and
your own domain called mysite.com,
then the full URL for that entry point would be http://www.mysite.com/index.html.
If you were being hosted by a web hosting service like GeoCities
or Xoom, you could have more directories in between. Maybe then
the full URL would look like http://www.webhoster.com/bigplace/portal/99/index.html.
And if you were a server owner with several related businesses hanging
off one root domain, then you might have several entry points, with
one full URL looking like http://branchoffice.myempire.com/index.html.
See also Entry Points.
Fuzzy-Match
We weight all of the words that are searched for
so that your users get the best match on the information that they
are looking for.
|
|
|
HTML
stands for Hyper Text Markup Language, and is the code that web
pages are written in.
HTML Source
The HTML source of your web pages is the actual text
of your web page, with all of the special tags that turn it into
the web page that you see on your screen. Most web page editors
will let you view the "HTML Source" or "Document Source." Another
way to get the source of your HTML is to open it in a program like Notepad in Windows (look under Accessories), or to load
it up in your favorite web browser and then view the "Document Source"
or "Page Source."
|
|
|
Indexers, Indexing
A well-organized index is the secret to every search
engine. Sure it's fun to imagine that search engines are eagerly
racing across the internet, like intrepid super-explorers, just
to fetch the answers to every question we ask... but that's not
really how a search engine works. What happens is this: an index
is made in advance, just like the one in the back of any good book.
And just like the index in a book, the quality of an index on-line
can vary a lot. People either will or won't find what they're looking
for depending on how good the index really is, which has been determined
in advance. So all these fancy search engines are really just look-up
engines, see? Don't be fooled by the technology. What you need to
know is that behind your search engine is a superior indexer - and
yes, the indexer in PicoSearch is very very good. So good in fact
that we can cross-index every word found to original sentence excerpts
in a flash, so that your users can see exactly where they are with
every search result. We call this feature "results in context",
and you won't find it in any of the traditional services.
|
|
|
Kilobyte
a kilobyte is 1024 bytes. Each character (letter, space, comma,
etc.) on your web pages is one byte. Average web pages are 2-3 kilobytes,
but size may vary depending on content. PicoSearch indexes by the page, so
kilobytes aren't a concern unless your account is really huge behind the
scenes, and then we might require that you have a paying account for
further support.
|
|
|
|
Relevance
Fuzzy-Match searching system ranks the documents, putting the best
(or most relevant) documents at the top of the screen. Relevancy
is determined by a number of factors to give you the best documents
first.
|
|
|
Search Engine
search engine is a set of programs that catalog information and
allow you to search through that information, like the index in
the back of a book. See also indexing and spidering.
Search Result Pages
web pages shown to your users after they enter a query for your
search engine. We display links to your web pages for what they
are looking for, along with original sentence excerpts to help your
visitors see on the spot how good the match really is. This helpful
display system is called "results in context".
Spidering
for those of us who have arachnophobia (fear of spiders), this web
analogy has been carried a bit too far, with people everywhere talking
about 'crawlers' and 'spiders'. Yes, PicoSearch can do spidering,
which really just means that the indexer in your search engine can
follow the links on your website to any public web pages, within
the parameters you specify. No information is duplicated, (secret
stuff is ignored), so the most complete possible index is made for
your visitors to search upon. See also indexing.
|
|
|
Template Control
Templating is available to PicoSearch Premium and Professional accounts. Templates enable full layout control of the Search Result
Pages, so that they may be fully integrated into your company's
website appearance. Templating even allows you to even sell your
own advertising space!
|
|
|
Web Color
Web-safe colors are the 216 colors which are guaranteed across
all browsers today, as opposed to the millions of colors which
may become a little distorted due to the spattered approximation effect
known as dithering. Web-safe is no longer as important as it
used to be, since most monitors display millions of colors, but web-safe
is a good place to start when choosing colors. Many browsers will take
simple color
names like "black", "gray", "blue", "red", or "green" - for a complete
list, here's a good link at w3schools.com color names. But if you really
want to be clear in web design, you must specify colors that are
made up of 6 hex values each for Red, Green, and Blue. This
means triples of 00, 33, 66, 99, CC, or FF, going from dark to
bright. Thus, pure red is #000066, purples would have red and
blue like #CC00CC (lighter) and #330033 (darker), and pure blue
is #0000FF. The '#' sign is a convention too, for HTML. Black
is #000000, gray is #666666, and white is #FFFFFF. You can see some great color
tables at design guru Lynda Weinman's site, sorted for value (brightness) and hue (color).
|
|
|
If you have a term that we didn't define to your satisfaction here
in our glossary, please let us know so that we can add it for you and everyone else who might be wondering too!
|
|
|
|
|
