June 17, 2002-- Newsletter #185
Full Text Search: The Key to Better Natural Language Queries for NoSQL in Node.js
Goodies to Go (tm)
June 17, 2002--Newsletter #185
This newsletter is part of the internet.com network.
Goodies Thoughts - Search Engines
We've all seen them. We've all used them. But how do they work and how many people actually use them?
Most surfers find search engines a very valuable tool in finding their way around the internet. But why are there so many search engines and why is it so competitive?
You would think it's a pretty simple process to construct a search engine. Create a database to catalog sites and a search function that allows users to access that information. Simple.
I'm sure Google and Yahoo! wish it was that simple. The information that search engines have in their databases is astounding. Some of the leading search engines like Google and AlltheWeb.com have over 2 billion pages cataloged in their databases. Yes, that's billion. And they use state of the art web servers to create some of the most powerful web farms on the planet ... all to just help the average surfer find what he is looking for.
Sound like a great deal of time and expense? It is. In order for a search engine to build its catalog of pages and sites, search engines usually use two basic approaches.
First is the web crawler. This little guy spends 24 hours a day, 7 days a week cruising around the web cataloging everything it finds. It does this by actually "reading" the content of each page, taking a look at the meta data, and noting any links to other pages and sites.
So, how does a web crawler know where to look? That's just it, it doesn't. It obviously knows all of the places it has visited before but in order to find new places to go, the crawlers usually use a very straight-forward approach; one letter at a time. For example, a brand new web crawler may start off with a very simple sequence like visiting a.com first. If a.com exists then it is cataloged and it's on to b.com, etc.
The second method of building a database is the registration process. This is made available to anyone that wants to register a site. Once a site is registered it's up to a more specialized version of the web crawler. Think of this piece of software as a sort of robot that only goes where it is told. While it does essentially the exact same cataloging as a web crawler, this particular piece of software can't be classified as a crawler because it doesn't actually crawl, it fetches.
The hardware necessary to run the more popular search engines would bankrupt many small countries. That doesn't include the people needed to keep the engine up and running or the developers necessary to create and continually refine the the engine itself.
So, how do search engines make any money? Some search engines make money through banner ads, however, that is one of the lesser sources of revenue. Some use links to their own specialized portals such as a computer or gardening portal. Sites wishing to be listed or featured in these specialized portals would pay a fee. Then there is the most popular revenue generator, ranking enhancement. Sites that want to be ranked in the top 10, 20, whatever, will pay a fee for that privilege. Often times there is a fee involved with every keyword that a site owner would want to appear under.
So, what makes one search engine more popular than another? That's anyone's guess. Over the last several years we have seen the popularity of many major search engines peak and fade. The only thing that seems to remain consistent is that the most popular engine never remains the most popular for long. Surfers have a tendency to always be looking for the next big thing. That is why many search engines put a lot of time and effort into being the most comprehensive and easy to use search engine on the web. By drawing more and more surfers to their engine they can begin to charge a higher premium to sites wishing to advertise or enhance their rankings.
Obviously, every search engine's approach and method of income is different. In the next issue we'll take a look at how you can make it easier for search engines to catalog your sites. We'll also take a look at how many search engines rank the sites and pages they visit.
Thanks for reading!
This week's question is hotspots. Do you know what they are and how they work?
Read answer below.
Q & A Goodies
Questions are taken from submissions to our Community Mentors. You can ask a Mentor a question by going to http://www.htmlgoodies.com/mentors/.
*** This question was submitted to our Mentor Community. The answer was provided by Eric Ferguson, one of our HTML Mentors.
Q. Can a form have two actions? I have a shopping cart from Red Hat interchange. They don't offer PayPal processing, which I have to use. I don't want to mess with the CGI process in the checkout. I LIKE what it does. But I'd also like to send the information to PayPal. Can this be done by simply having two actions for the same form?
A. To answer your
question plainly...No. A form must have exactly one action. However, the
server-side (e.g., CGI) program that processes your form submissions can perform
any number of tasks (e.g., updating a database, sending email, logging a
transaction) in response to a single form submission. In other words, the form
sends the information to another page or program and on that page is the code
for the tasks that you need performed. In the case of a shopping cart, the
information is probably sent to the database, the client and the merchant in the
form of an email. So, it looks like you would have to mess with the CGI code. I
did some checking on the Red Hat site and it seems that someone did come up with
a fix for using PayPal, but you do have to alter the code.
*** This question was submitted to our Mentor Community. The answer was provided
by Bob Conley, one of our Web Design Mentors.
Q. Is there any
way that I can have my pages pull the source of another file into them? For
example, I have a couple hundred lines of code for my menu system and logo that
I want to appear at the top of every page.
Is there a way I can have that source in one file and have all the other pages use it?
A. You can accomplish this by using an "include" file in your pages. Take a look at this tutorial: http://www.HTMLGoodies.com/beyond/ssi.html
*** This question was submitted to our Mentor Community. The answer was provided by Peter Szaro, one of our Web Design Mentors.
Do you use an Instant Messenger
service? Well, you're not alone. There are over 41 million other people out
there with you.
Interested in starting your own web
design business? This article may help get you started.
Microsoft is issuing a new release of
Internet Explorer for the Mac specifically designed to fix some of the problems
Hotspots are a very cool little trick. They allow you to designate one or more parts of an image as a link.
Let's say for example you made a graphic that had pictures from your birthday party, Father's Day and your vacation. You wanted to link each of the pictures in the graphic to a photo gallery page of that event. It would be quite a pain to have to separate out each picture in a table. The best solution is to use hotspots.
Here's an example:
<AREA HREF="page1.html" SHAPE="rect"
<AREA HREF="page2.html" SHAPE="rect"
<IMG BORDER="0" SRC="pictures.jpg" USEMAP="#Map1"
In the case above we made two hotspots on our image. Each one is defined in our image map. Just for grins we named the map "Map1". Clever, huh. Each link is defined by the AREA tag. Within each AREA tag is the HREF, SHAPE and COORDS attributes.
HREF - is the page or file that you would like to link to.
SHAPE - is the shape of your hotspot. In our case we are using rectangles but you also have the option of circles and polygons.
COORDS - defines the coordinates of your hotspot in relation to your image. In the first AREA listed we have coordinates of "73,13,241,67". The first 2 coordinates refer to the top left corner of the rectangle, 73 pixels from the left and 13 pixels from the top. The second 2 coordinates refer to the bottom right corner of the rectangle, 241 pixels from the left and 67 pixels from the top.
To use the MAP we created we have to tweak the IMG tag a bit. We have to tell the browser where to find the MAP we created by giving it the name of the MAP in the USEMAP attribute.
Hotspots can take some time to get just right when creating them by hand. If you have a web development package available you will probably find creating and implementing hotspots much easier.
And Remember This . . .
Happy Father's Day!
As a new father myself, I was interested in knowing how long Father's Day has been around. I was surprised by the answer.
In June of 1910 during the midst of the push for an official Mother's Day celebration, Mrs. Sonora Smart Dodd proposed a Father's Day celebration in her hometown of Spokane. She had heard of the movement for a Mother's Day. However, she had lost her mother when she was a child and was raised by her father. Her father had made many great sacrifices for her and her siblings and she wanted to show her appreciation.
As with Mother's Day, the idea was well received. Unlike Mother's Day, though, it would be some time before Father's Day became an official holiday in the United States. Even though many people and politicians thought it to be a good idea, politicians were reluctant to make Father's Day an official holiday for fear of looking too self-congratulatory.
In 1924 President Calvin Coolidge recommended to the state governors that they create a Father's Day observance. He left it up to each state to implement the observance if they so chose.
It wasn't until 1972 that Richard Nixon established Father's Day as a national holiday, sixty-two years after it was first proposed.
Thanks for reading Goodies to Go!