Recommended Web Tools Archives

User Assisted Search at Zeedex

Zeedex is a new breed of search that is attempting to get more user input into the search process.

Zeedex provides search results along with suggested search suggestions provided by users.

Zeedex user input works on the basis of lists. Users can create a list of other suggested search terms that are displayed based on a selected search term.

Lists consist of a title, keywords and search terms. The title simply names the collection of suggested search terms. For example: “Popular Dog Breeds”.

Keywords are the words used to determine when the list appears. Using our above example, we would want our list to appear for keywords like “dogs, dog food, dog shows”.

Search Terms are the terms we (as a user) are suggesting to the searcher to either refine or change their search. In our case, we can add “German Shepherd,Poodle, Golden Retriever, etc.” We can also add a short description of each term. In many ways, this list provides a mini-search result set that is related to the topic searched.

Other users can edit or add onto your original list.

Each list is also able to be sponsored. Sponsored lists include a text link ad and description. Currently, being a sponsor is free.

The goal of using these lists is to generate better relevant results.

Will this concept work?

I have thought about what Zeedex is trying to accomplish and I think they are close to having something very workable and usable. There are a few items which I think they need to consider as they move forward.

The search results themselves are not generated based on user input. They are in fact provided by Yahoo. Eye studies have shown that users will generally be looking at the left portion of the screen when looking at search results. Users of Zeedex (or any search engine) will be focused on the actually results not anything in the sidebar.

Clicking on user search suggestions adds the terms to the words already chosen in the search box. The searcher should be given an option either to add or replace.

The user lists are prone to being spammed, manipulated and rendered useless unless there are excellent controls placed on changing the list.

I also wonder if there is too much work involved for users. Do I really want to take the time to generate a list of suggested search terms for other people? It has to benefit the individual user as well. Tagging, for instance, provides an organizational mechanism for both the community and the individual. The lists in Zeedex serve the community but provide little value to the individual.

I think Internet users are interested in some kind of social search engine. The trick for Zeedex is to incorporate user input into the actual search results.

I wish I had the programming skills and the guts to produce a tool like this. I think the folks over at Zeedex would welcome more feedback. Take Zeedex for a spin and contact them with your thoughts. Be constructive and helpful. You never know, one or two tweaks and Zeedex could be the next coolest thing.
[tags]search, zeedex[/tags]

Live.com Search Engine Review

Microsoft rolled Live.com out of beta. Live.com is Microsoft’s new search engine that will replace the old engine at MSN.com. MSN will still exist, but all search will be driven by the Live.com technology.

At first blush, Live.com looks like Pastel Google. The colors are softer more subdued. For example, a search result consists of a Title, Description and a link. The title is a much softer blue than the bold blue of Google. The description is actually a dark grey not a true black like Google. And the link is a soft green, not a bold green of Google. In terms of readability, I find Google much easier to read. The softer subdued colors of Live.com make my eyes work just a tad harder.

How relevant are the results? It is hard to say without a lot of comparison but having done a few searches it is clear that the results are very different from Google in most cases. In general, some terms seemed more relevant other terms were less relevant when compared to Google. The verdict here is that search engines in general still have a way to go in pinpointing relevancy.

I think Live.com nails the interface for image search. It works like usual: type in a term, a bunch of images appear, etc. Clicking on the image will open up the image’s site of residence within the Live.com window. The other images Live.com found are moved to the left nav bar where they can be scrolled. Clicking on other images will bring up the original webpage within the main content window, all without leaving Live.com. Slick.

Live.com News search is sorta, well, blah. Nothing to exciting here.

However, they bounce right back with a pretty snappy local search. I typed in
“coffee shops”. Live.com uses your IP address to guess where you live. In this case, it was accurate and brought up my city. It populated the map with little numbered flags. Each number corresponded to a name and address in the left nav bar. I verified the list consisted of all coffee shops and no chinese buffets or McDonalds were thrown in. I only wish it search beyond the boundaries of the actual city. In metropolitan areas, it would be helpful to see the wider scope of possibilities.

Other items are still in BETA: feed search, QnA (Questions and Answers), Video, Academics and Products.

Overall, it still seems too Google like. I would have preferred a different approach to the User Interface like Snap. I guess a search engine is a search engine and a search result is a search result, but it seems there is room to provide more creative possibilities then a title, a description and a link.

[tags]search, engines, search+engines, google, live.com, microsoft, msn[/tags]

When did the Googlebot last visit my site?

The answer to this question is in the web server log files.

For beginners, there may be a more user friendly way.

• Go to Google.com
• type: www.yoursitename.com into the search box (where yoursitename is the name of YOUR website)
• When the results return, your site should be the top one listed.
• The bottom line of the search result is the URL for the website printed in green. Next to that URL is a link called Cached.
• Click on Cached
• The page that opens displays your web page as Google has it cached in their system. At the top is a header. The first line whould read:
  

  This is Google’s cache of http://www.yoursitename.com/ as retrieved on Sep 1, 2006 09:23:34 GMT

• The time reflected in this line is not only the last time the page was cached, but according to the Google Webmaster Central blog, it is the last time the Googlebot visited your website.

[tags]google, googlebot[/tags]

Snap Redone

It pays to pay attention.

I wrote a few months back that Snap was a search engine with great potential but little seems to have been accomplished. I had been a subscriber to the Snap RSS feed, but they went a long time without posting (Note: it may help to post at least once a month to let visitors know your company is still alive). So I unsubcribed and forgot about Snap.

I was on the Alexa site last week and saw Snap was one of the movers and shakers (up 1400% or something like that). What was going on with Snap?

Upon re-visiting Snap, I saw that much had been accomplished. Their blog mentions a relaunch on May 15th 2006. Things are indeed different.

Snap takes a different approach to search. Of course, the input is the same: a string of words of some sort. The difference appears in the results.

First, when keywords are being typed into the search box, Snap provides a popup list of recommended/possible searches. One might think this is similar to Google Suggest. However, it is different. Google Suggest provides only linear (or some might say anchor left) results. Type in “recommended” and get a long list of things following “recommended:”

• recommended daily allowance
• recommended calorie intake
• recommended books
• etc

Snap employs whole string querying and alternative form results. Type in “recommended” and the suggested search terms include items like “letters of recommendation”. While this approach to “suggest” technology is better, it still has to be improved. However, Snap learns from what users enter. The more users search, the better the search suggestion list becomes.

Second, Snap search results are not solely text based. A preview window is provided that displays the webpage of the text based result. This is similar to functionality provided by the French search engine Exalead. However, the preview is much larger (at least 10x) and easier to read than the preview provided by Exalead. Plus, with Snap, users can click on the preview and a “live” version of the website is loaded in the Snap preview window.

Do visually based search results help the searcher?

Yes. They prevent clicking to numerous non-relavent websites. The preview lets searchers know immediately if the site is relevant or not.

I am going to make Snap my default search engine for a while and see how it goes. Snap does provide an exciting shift in search engine practice. It will be interesting to see where things go from here.

[tags]search, snap[/tags]

Which are the most popular search engines?

A search engines main purpose is to provide answers to the questions its users ask.

Of course we know that Google, Yahoo and MSN are the big three search engines. I was curious about all the other places that people visit to get their questions answered. How popular are they? So I went to Alexa and starting typing away and creating little charts. I understand all the issues surrounding Alexa data, so the list that follows is just for fun and is not scientific in any way. It is just a way to recognize some trends. There are sites on this list that are not true search engines. but are included because they do provide “answers”. I have grouped the sites in tiers. The tiers are highly subjective and do not really relate to anything other than an easy form of presentation. The charts represent Alexa data over the last year.

Tier 1 (Top 10): Yahoo, MSN and Google reside at 1, 2 and 3. MSN and Google seem to switch positions every so often. Keep in mind this is not number of searches performed or anything like that. Yahoo and MSN are used for so many things other than searching.

Tier 2 (Top 100): search.aol.com (23), About (55). True, AOL Search is simply Google repackaged. Just interesting to note that AOL still maintains quite an audience. About is not a search engine per se, but is a place to go to find answers. People have a love/hate relationship with About. For what it is, it is a handy resource.

Tier 3 (Top 200): Netscape (147), Ask (154), Altavista (176), Lycos (187) are all grouped close together. Ask is probably considered the “fourth” search engine solely because it does not rely as heavily upon Google to provide its results as the others do. It only receives paid results from Google.

Tier 4 (Top 500): Answers (253), IWon (373) I was surprised to see these two up this high on the list. Answers pulls in a lot of information from other sources, primarily Wikipedia. It is not a true search engine, but it does pull in web based results from Google. It’s clean format and easy to find “answers” is a key reason why it probably has grown in popularity. I use it myself when I just need to get to a straight answer rather than flipping through pages of sites in Google. I have never used Iwon. I assume it is popular because it is connected to games and prizes. Everyone wants to win something.

Tier 5 (Top 1000): Technorati (598), Alltheweb (662), Dogpile (731) and A9 (911). Technorati is primarily for blogs but I felt it should be included here. I am glad to see that Dogpile has increased in popularity. It currently is my search engine of choice.

Tier 6 (Top 5000): Search (1643), Hotbot (2914), Clusty(2964), Icerocket (3536), Mamma (3596). Hotbot used to be my favorite search engine back in 1999 until I came across Google. I am surprised it still sees the amount of traffic that it does. I am glad to see Icerocket making waves, though primarily a blog search engine. Clusty is probably the most underrated search engine. It is an excellent resource for finding “clusters” of topics. It is helpful in doing keyword research. I have never used Search nor Mamma and will have to mosey over there and test them out.

Tier 7 (Top 6000): dmoz (4057), Teoma (4272), Webcrawler (4852), Gigablast (5076), Metacrawler (5278), Entireweb (5698). There are a few surprises here. While I have enjoyed searching using a few of these sites, I often wondered if anyone really used them. These numbers confirm that they are still alive. DMOZ, though a directory and not a search engine, was included because of its importance in feeding results to numerous other engines. According to the chart it is seeing a resurgence in popularity after experiencing much user decline. Webcrawler was the first search engine I every used back in…back in who knows when. Gigablast is fast and I love gigabits. I had to leave Entireweb off the chart since these Alexa charts only handle five at a time.

Tier 8 and beyond: I didnt chart any of the search sites beyond the 6000 rank. But here are the numbers exalead (13702), snap (13934), inktomi (20348), fybersearch (31838) northernlight (46187), myriadsearch (92359), golexa (129052), Zniff (151970). I love the features in Exalead. The biggest dissappointments in this whole post are Snap and Northernlight. Snap seems to have had spurts over the years but then just fades away. Northernlight burst onto the scene in the late 90′s with much fanfare. It is now almost useless. There are several specialized engines included here. I don’t imagine they will make any great gains in popularity but they are fun nonetheless.

The one trend of note for the majority of these search engines is the downward progession of popularity. Over the next five years it will be interesting to see if they even still exist.

FyberSearch Search Engine Review

FyberSearch is a small search engine that offers two unique pieces of functionality.

I had originally hoped that FyberSearch would help me search for feeds. While it has this functionality, the results returned are web page related not feed related. The search for a quality feed search will continue…

HOWEVER! I did come across two tiny tidbits that could be helpful.

First, Tutorial Search. Type in a topic and FyberSearch will return webpages that are tutorial related. Very handy indeed. It is not perfect and it seems to be more helpful with technology related items than other topics. For example, a search for “drywall” returned nothing. Whereas a search for “HTML” resulted in multiple pages.

Second, Gheto Search. Horrible choice for a name, but the functionality is fun. Here is the twist:

The last web page you see in the normal search results is the first web page you see in GhetoSearch.

It essentially turns searching upside down so that the unpopular items are shown first. But isn’t the bottom of the barrel the least relevant? Yes maybe it is. However, I have often found some gems on the 10th, 15th, 20th page of Google. So with FyberSearch, they do the work of turning the barrel upside down for me.

Maybe FyberSearch won’t become your favorite search engine, but at least there are two items that savvy searchers can tuck away in their back pockets.

Understanding Alexa’s Reach Per Million

I often wondered and could not figure out Alexa’s way of charting website traffic as ‘reach per million”. In a post on the Alexa blog the other day this explanation was given:

As long as I am on the topic, I might as well explain this reach per million business. Alexa does not calculate the actual number of users visiting Web sites… what we do instead is release raw data about how many people in a sample population (Alexa Toolbar users, among others) visit a site. The data is normalized to a sample size of 1 million users so that the reach doesn’t fluctuate as our sample base grows or shrinks.

Are you with me so far? So, for example, if we say a site has a Reach per million of 6000, it means that 6000 people in our sample of 1 million Internet users visited the site. Or, to put it more plainly, .6% of the population visited the site.

Now you know!

[tags]alexa, web statistics[/tags]

SEOMoz article rates factors that affect ranking

SEOMoz has an excellent article that outlines a whole plethora of potential factors that affect search engine ranking. Focus on the top 9 to let a website begin to make a difference.

GoLexa Search Engine Review

GoLexa is a cross between Google, Alexa and a ton of other metadata. Data freaks will love this tool. It takes Google search results and adds Alexa and other metadata to the result. For any given result returned, GoLexa provides links to numerous metadata site about that webpage/website. A searcher can get access to following:

• Thumbnail preview of website
• Google PageRank
• Alexa Rank
• Yahoo Rank
• Title of Website/webpage with link to actual site
• Short description
• URL of site
• size of webpage
• My Yahoo bookmark link
• MyWeb bookmark link
• delicious bookmark link
• Y!Q Contextual Search
• Answers.com reference search
• Hi-lite page with keywords highlighted
• Cached version from Google
• Text version from Google
• Similar/related sites at Google
• Pages indexed at Google
• View Source Code
• Spider Simulation
• What search engines list this page
• Email URL to a friend
• Monitor page
• Translate page
• Link Popularity
• Keywords on that page
• WHOIS
• Alexa Traffic Charts
• Wayback Machine History
• Site Report
• Loading speed and size of page
• Ping
• IP Address
• MSN rank
• Anchor text used with this page URL
• Adsense ads that would appear on this page
• Spell Check
• Trust Gauge
• URL Trends
• Google Banklink Alerter
• Yahoo Site Explorer
• TinyURL
• Extract Content
• Google versus Yahoo ranking

Phew!

Is it overkill? Absolutely.
Is it fun? Definitely.

Main use for GoLexa? Learn more about your competition.

Happy Golexing! (Doesnt have the same ring to it as Googling…)

[tags]Google, Alexa, GoLexa, search[/tags]

Dogpile Search Engine Review

Dogpile is a Searcher’s Search Engine. Searcher’s Search Engines are different than Marketer’s Search Engines. A marketers search engine is Google. The goal of most websites is to try and get noticed within Google. It has the most traffic, the most use, the most potential customers. However, engines like Dogpile are different. Dogpile compiles results from all four top search engines in order to give the searcher better results and greater access to those results.

The top search engines all produced different results. With Dogpile, all those are aggregated to give the searcher the best of the best. Dogpile offers a search engine comparison tool to demonstrate this. Enter search terms and watch the graphic as Dogpile demonstrates how the results from the GYM are unique and how Dogpile tries to pull them together.

After conducting a search, Dogpile allows users to see how Dogpile results vary from the majors. These results can be listed side by side. The user can then easily compare results. In actuality it allows the user to see MORE results and thus find exactly what they are looking for.

In terms of actually being able to track something down, Dogpile is a far more friendlier tool to the searcher than other search engines. Searchers ask questions, search engines provide answers. Dogpile helps searchers in two ways. By providing more relevance and by exposing searchers to a broader base of relevant results. As a searcher, consider the advantages of using Dogpile:

• Aggregated results from Google, Yahoo, MSN and Ask Jeeves provide the best of the best
• The ability to see results from the GYM alongside Dogpile results means MORE results. More results means a greater opportunity to find what you are looking for.
• Dogpile highlights results that are unique to each search engine. This simply translates into more resources for searchers to find their answer.
• Searchers are getting savvier. Dogpile matches their savviness.
• Dogpile saves time. Access to all the major search engines on one page keeps searchers from having to run searches on multiple search engines simultaneously.

[tags]dogpile, search engines, search, metasearch[/tags]

Understanding Search Engines

Here are two great articles about search engines.

The first, How Search Engines WorK (and Sometimes Don’t) explains the basic nature of a search engine and provides some detail on problems search engines sometimes face when crawling a page.

The second, How Search Engines Deliver Result Pages expands upon the first article providing a little more indepth understanding into the concepts of information retrieval.

Search Google, Yahoo, MSN and Jeeves concurrently

With the help of Myriad Search, search four top search engines (Google, Yahoo, MSN, and Ask Jeeves) concurrently and return weighted results.

Why weighted results? How is that useful?

Myriad Search takes the results from the aformentioned search engines and places a weight value to each. It then computes its own results list based on that weighting (More information of the weighting formula).

By combining the results from all four search engines, Myriad Search saves a lot of work and offers very valuable information:

1) More relevant search results
2) Find a more complete list of competitors
3) Find a more complete list of potential link partners
4) Another means to find authority sites

Why weight the results?

Each of the four search engines is flawed. None of them truly provide relevant results all the time. By weighting the results from each, a better sense of relevancy is gained (though still imperfect).

It is easy to use Myriad Search. Enter a keyword phrase in the search box and click ‘search’.

Myriad returns the results under 5 tabbed headings. The 2nd through 5th tab are the results for each individual search engine. Want to see the results just for Google? Click the Google tab.

The ‘All’ tab displays the weighted results. The description under each link is color coded. Each color represents a search engine (the tabs for the individual search engine results are colored accordingly). For each result, the user can then see what description of the site is returned by each engine. Often it is the same, but sometimes it is not. This is especially helpful in discovering how your own website’s description is displayed.

Under each results is a set of numbers:
#1 -80% – Google (4) yahoo(1) msn(1)

These numbers are fairly easy to understand:
#X – the rank within the weighted results
XX% – this is a percent authority/relevancy for a site
search engine (X) – this is the rank for the given search engine

So to interpret the line above: The result is ranked first (it is the most relevant). It’s weighted authority score is 80%. It ranked 4th on Google but first on both Yahoo and MSN. It did not appear in the top ten listings (I had set ’10′ as an Results Returned option) for Ask Jeeves.

Myriad Search Options:

There are several different options that can be used when using Myriad Search.

Results: How many results per engine? Choosing 10 will focus soley on the top ten results from each engine. Searching deeper, say 40 or 50, will potentially dig up more relevant sites that are worth looking at but are often missed.

Duplicates: By limiting duplicates, the results become more ‘pure’.

Google Key: Myriad Search uses its own Google Key. However, if at the end of the day it has used all 1000 of its searches on Google, it will cease to function. Entering a new Google API key (ie your own personal Google API) will let it continue to work.

Alexa Rank: Checking this box will throw the Alexa Rank for each site into the weighting formula. This also slows down the time it takes for results to return. Personally, I am wary of Alexa Rank since I don’t feel it is an adequate reflection of web usage.

Bias: Choose a number to add or remove the weight of a particular engine. Entering 5 adds the most weight. ‘Exclude’ removes highly irrelvant results from the mix. ‘Require’ ensures that the most relevant searches are included.

I personally have left all the options at their defaults, expect for number of results returned. It takes practice to see how each option affects the weighted results. It is best to use the tool in default mode until all the options are understood.

Google PageRank Update

One of the handiest websites to keep track of future Google PR Updates has been SEO Company’s Page Rank Export List History. There is also a blog for more specific comments about the update.

Google Blog Search

By now you have heard about Google’s Blog Search release. Is there anything to write home about here?

On the plus side, a blog search engine is able to handle infinitely more blogs than a blog directory. It is open to the whole world wide web not just those that have been submitted to a directory. Google enters this arena of search well after others (see IceRocket) have been at it for awhile.

Is the search tool helpful? Yes and no. I have a problem with it. I suppose it is a problem of searching in general. I am interested in finding blogs ABOUT ecommerce. I entered “ecommerce” into the blog search engine. The blog search engine searched blog content (ie posts). 14 of the first 20 results were all from the same blog. The blog it found the most was indeed a blog about ecommerce. It had the term “ecommerce” in the blog name. However, since this one blog had “ecommerce” in its name, posts from that blog came up the most EVEN IF THE POST HAD NOTHING TO DO ABOUT ECOMMERCE! Such results are not the most helpful.

It would be helpful to break the search into two parts. Search blog content and search blogs titles. The former would be like a traditional search engine (which is how Google’s Blog Search currently acts (sort of)), the latter would return blogs that are more or less ABOUT the topic entered. I want to enter “ecommerce” and have each page be an unduplicated list of blogs that are ABOUT “ecommerce”. I guess that is why blog directories are more helpful in that regard.

Use MyYahoo Tip with Google and MSN

Do you know the tip about getting a website crawled by Yahoo? In brief, by adding a website’s RSS feed to MyYahoo, Yahoo will crawl the parent site. It is how this site got quickly crawled by Yahoo.

Now that Google and MSN have their own News/Feed Aggregators, maybe the same will hold true for them. It is easy enough to test out.

Google calls their aggregator Personalized Home. A Google account is required to access. Once signed up, click the “Add Content” link in the upper left corner. Click “Create a Section”. Add the URL to the feed in the text box. Click Go. The feed will now be added. Whether you continue to use Google as your News Aggregator is up to you.

MSN’s News Aggregator is called Start. No account is needed to sign up. Clicking the Start logo drops down a list of options. Click My Feeds. Click Add Feed. Enter the URL to the feed in the Add Feed By Url text box. Click Add.

Sidenote: Start is cookie based. Once a browser cache is cleared the “personalization” and all entered feeds are lost to the user. This calls into question whether it should be used for what it is intended.

I would be interested in hearing if new sites that are not currently crawled by Google or MSN get crawled more quickly once their feed is added.

via Robin Good

Zniff Review

Zniff is billed as the “human search engine”. Essentially, Zniff is a search engine for the websites bookmarked via Spurl (I have reviewed Spurl elsewhere) . Websites that get bookmarked the most appear higher in the search results. The idea is that bookmarked sites are favored choices among actual users. Thus, a search of those results will produce a list of the most quality websites.

Results can be returned based on relevance or time. Searching by relevance produces results generally based on the number spurled. For example, I searched for “fishing”. The number one result is American Fly Fishing Trade Association (it had been spurled 4 times). Doing a relevance search for “dogs” providing an interesting list. A site spurled 18 times (and about all breeds of dogs) was listed second, while a site spurled once (and solely about schnauzers) was listed first. It seems the algorithm needs a little work.

Searching for “fishing” again, but by “time”, produced an article on psychics as the number one result. The top result will be the most recent one added. However, what does a psychic have to do with fishing? It could be that page got tagged that way. A weird result nonetheless.

For more complete results, Zniff can pull from the Yahoo search result set as well.

Remember, this is a limited audience of users. This search engine represents only Spurl users. Since Spurl does integrate with del.icio.us, I suspect that the full released version of Zniff may include results from that social bookmarking tool as well.

There is a Firefox Plugin for Zniff that includes it on the Firefox Search Engine drop down list.

Zniff is useful to see what others deem important. I did searches on “poverty”, “homeschool”, and “london”. All the top sites were generally useful. If anything, Zniff can be used in conjunction with traditional search engines to do research, especially if you want to know what websites other people have found to be helpful.

Google Toolbar for Firefox

In case you have been wondering when it was going to be released…here is the link to the Google Toolbar for Firefox

Answers.com – Online Dictionary, Encyclopedia and much more

Answers.com has become a favorite to search for items of a more encyclopedic nature.

Tired of wading through tons of sites and pages to find a simple answer to historical questions? or finding a basic definition for a word? Or finding a helpful article about a topic? Try Answers.com first.

Easy, one stop searching. I am finding Answers.com hard to beat in that department.

Depending on the nature of the term, Answers will potenitally return articles from the following sources:

Wikipedia
Who2
Houghton Mifflin – Dictionary and Thesaurus
Columbia Univeristy Press – Encyclopedia
Biogs
Best of the Web
WordNet
Medicine Net
Wizcom – Translations

and much much more.

Seriously, stop frustrating yourself trying to find basic answers to simple questions using the standard search engines. Next time you want to know the birthplace of Tony Blair or find out what a permalink is, try Answers.com

About.com Review

I has taken me a long time to warm up to About.com. I don’t know why. Maybe it was the orange color scheme or the sometimes disorganized pages or the ads right in the middle of articles (that DOES annoy me). As I was developing my own site, I actually began to read the content on About. I discovered there was a lot of excellent resources. Of course, it all depends on the Guide for the topic.

About uses “guides” who create content for the site. About is built on the premise

that people are the best Guides to the Internet

How can About.com help the beginning web-it-yourselfer?

There are a ton of helpful articles and resource that the beginner can access.

Let me point you to five categories that would be of particularly helpful to the beginner:

Internet for Beginners
Personal Web Pages
Web Design/HTML
Web Search
Graphic Software

Each category has it’s own RSS feed. Use the feed to keep up to date on the latest content that has been added.

Clusty Review

Clusty is another alternative search engine that can be used by keyword hunters.

Clusty provides traditional search results along with “clustered” results. These clusters combine the search results around a theme.

Clustering is best explained by Clusty itself:

It uses our award-winning Clustering Engine to organize search results into folders grouping similar items together. Thus a search for ‘pearl’ organizes the top 250-500 results into subject folders such as Jewelry, Pearl Harbor, Pearl Jam, Steinbeck Novel and Daniel Pearl. Clusty allows users to focus on the area of interest without all the chaff.

Not only is this a boon for searchers, but also keyword hunters. Keyword analysts can use Clusty to narrow the focus of more generic keywords.

For a more comprehensive keyword analyis approach see my article on keyword suggestion tools.

Snap Review

Snap has been an ever evolving search engine for the last several years. Snap recently changed their look once again. While the search results themselves offer nothing spectacular, there are two great features of Snap that users will like.

With every search, Snap produces a list of related search terms in the left navigation bar. They also provide the number of times that term has been searched. This is a great place to pickup a new idea for a keyword phrase or two. Unfortunately, there is no specific time frame given to the number. Is it number of searches in the last day? week? month? I could not find an answer to this. The only statement I could find was the following:

Search Count refers to searches by a panel of U.S. Internet users. This data is provided to Snap.com by an independent research firm.

Snap also provides additional reporting capabilities on keyword usage. Search a particular time frame (great for seasonal keywords) or search a greater list of related keywords.

Keep in mind that the audience of Snap is small. Don’t make big plans for keywords until other tools have been employed to measure the results (see my keyword suggestion tools article).

Submitting a website to Snap is quite easy. Just fill out this form.

Keyword hunters now have another tool in their pocket.

Librarian’s Index to the Internet

The Librarians Index to the Internet is an excellent information resource. The site can be broken down into two parts: human edited directory and an integrated search engine page.

Human edited directory

The LII human edited directory is not like DMOZ. It is not striving to be the largest directory of information but rather a collection of recommended resources arranged by topic presented by the librarians of California and Washington. It is actively added to and culled. The directory works like any other web directory. Click a topic and sub-topic and a list a recommended web resources is presented to you.

Why is this helpful versus doing a Google or Yahoo search? One never knows what Yahoo or Google may return. A directory such as this, which is human edited, will most definitely provide a high level of relevancy. Looking for resources on mental health? Click Health, Mental Health and there are twenty four recommended websites listed.

Do you know a site that you feel should be included? Use their Suggest a Site form and submit it.

Integrated Search Engine Page

LII provides a very helpful search engine page that includes the ability to search from 15 different search engines. This is extremely helpful for beginners who may be unsure about search engines and how they work. For each search engine listed there is a link to their help file and a link to their advanced search page. The search engines listed are:

• Librarians Index to the Internet – search the LII directory
• Google
• Yahoo
• Ask Jeeves
• Teoma
• Vivisimo
• Ixquick
• Google News
• Yahoo News
• Daypop
• Topix
• Infomine
• Bubl
• Internet Public Library
• Kidsclick

Not only is this site great for beginners of all ages, but also for seasoned Internet searchers who need a fresh look at new search engines and data.

Search Engines as Partners

This article makes the argument that webmasters should see search engines as partners instead of as adversaries. I agree.

Webmaster Guidelines for the Beginner

If you are just beginning to develop your own websites, it is helpful to check out the guidelines published by Google and Yahoo for webmasters. These guidelines offer basic instructions for webmasters concerning web development and search engine optimization. The guidelines are a summary of “what to do” and “what not to do”. They are not only a good place to start for the beginner, but oldie and moldies can relearn a thing or two from a review of these documents.

Basic Webmaster FAQ on Google
Google Webmaster Guidelines
Facts and Fiction about Google
Search Engine Optimization
Understand the Googlebot
Yahoo Indexing
Yahoo Search Results Ranking
Yahoo’s Web Crawler
Yahoo and Search Engine Spam

Happy Reading. (Seriously. Read this stuff. You just might learn something.)

Exalead Review

Exalead is a beta search engine. The company behind Exalead is based in France. Don’t hold that against them, Exalead is quite slick in many ways. Google’s excellence resides in their search algorithm. Exalead’s excellence resides in their user interface.

Let me walk you through all the features that make searching with Exalead a more powerful and helpful experience than most search engines.

After a search entry has been made, the results screen is divided up into 3 main areas: Main Tool Bar, Left Nav Bar and Center Content.

Center Content

This area displays the main search results. I start my description here because all the other features have to do with how these results are displayed. Much of what is displayed is typical of search engine results: Name of Site that is also a link to the website and a two to three sentence description.

A few items are added below the description that are value-added features. First there is the name of the web page that the results belong to. Clicking on this produces one of two results. If the link is the main web page of the site (ie www.mysite.com) then Exalead returns all pages that have been indexed from that site. This is similar to doing a ‘site:www.mysite.com’ search. If the page is a subpage (ie www.mysite.com/dir/thispage.html, then the bottom half of the screen opens with a a preview of that site! More on the preview in a moment.

Second, there is a list of categories that this page belongs too. The category feature is essentially a web directory. Clicking on a category brings up all the other pages listed inside that category. The page from which the directory click orginated from is always listed first. Having both regualr search results and a directory is not new. However, including the directory results right along with the main search results is definitely an improvement.

The default view of results also includes a screenshot of the web page. This is displayed to the right of the search results.

Clicking anywhere inside the search results box (when the mouse cursor hovers over the search results, a dashed box appears around it), other than on the links mentioned above, opens up the preview window at the bottom.

The web page preview is a cool feature. It rivals IceRockets preview feature. Once the preview is open, a user can choose to view the original, which opens a new browser window. Users can also add the page as a bookmark. On the right side of the preview screen is a Terms Found widget. Terms Found lets the user know how many times the keyword(s) searched for occur on this particular page. This widget also has directional arrows. Clicking the arrows leads the user through each occurence of the keywords(s) which are highlighted in yellow. Again, this is very helpful. It reduces the use of using the Find in this Page function on most browsers. Makes sense to include it right in the search engine itself.