# All Search Engines Are Not Created Equal? **URL:** \ **Category:** Factual Questions\ **Created:** [May 28, 2000, 5:46am UTC](https://boards.straightdope.com/t/all-search-engines-are-not-created-equal/20725 "2000-05-28T05:46:15Z")\ **Posts on this page:** 17\ **Page:** 1
**Author:** ![Jois](https://avatars.discourse-cdn.com/v4/letter/j/ea666f/32.png) [@Jois](https://boards.straightdope.com/u/Jois)\ **Post date:** [May 28, 2000, 5:46am UTC](https://boards.straightdope.com/t/all-search-engines-are-not-created-equal/20725/1 "2000-05-28T05:46:15Z")
Please explain why one search engine is better than another? Somebody mumbled that Google is now THE BEST. How? ---
**Author:** ![bibliophage](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/bibliophage/32/7615_2.png) [@bibliophage](https://boards.straightdope.com/u/bibliophage)\ **Post date:** [May 28, 2000, 5:59am UTC](https://boards.straightdope.com/t/all-search-engines-are-not-created-equal/20725/2 "2000-05-28T05:59:28Z")
We were just discussing this in the BBQ Pit ([Does nobody know how to use a search engine](http://boards.straightdope.com/sdmb/showthread.php?threadid=25418)). My favorites are [SavvySearch](http://www.savvysearch.com) and WebFerret. The former is a metasearch, meaning it searches with several engines at once. The latter also searches several engines, but it requires a download of software. I find it generally works better and faster than SavvySearch, but SavvySearch is better for phrase searches. I have not been impressed with Google, though some people swear by it. ---
**Author:** ![Chronos](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/chronos/32/134_2.png) [@Chronos](https://boards.straightdope.com/u/Chronos)\ **Post date:** [May 28, 2000, 6:33am UTC](https://boards.straightdope.com/t/all-search-engines-are-not-created-equal/20725/3 "2000-05-28T06:33:02Z")
There’s a few relevant factors: Size of database, speed, and relevance (that is, does it put the useful pages before the 99% junk) come to mind offhand. Note that I do not know how any search engines rank specifically on any of these points. I usually use AltaVista, but that’s just because when I started using the web a few years ago, it was the best, and I got used to it. It’s quite probable that some other engine has since passed it up. ---
**Author:** ![psycat90](https://avatars.discourse-cdn.com/v4/letter/p/7ea924/32.png) [@psycat90](https://boards.straightdope.com/u/psycat90)\ **Post date:** [May 28, 2000, 6:47am UTC](https://boards.straightdope.com/t/all-search-engines-are-not-created-equal/20725/4 "2000-05-28T06:47:52Z")
My personal favorite. [http://www.dogpile.com](http://www.dogpile.com) ---
**Author:** ![Ringo](https://avatars.discourse-cdn.com/v4/letter/r/779978/32.png) [@Ringo](https://boards.straightdope.com/u/Ringo)\ **Post date:** [May 28, 2000, 6:55am UTC](https://boards.straightdope.com/t/all-search-engines-are-not-created-equal/20725/5 "2000-05-28T06:55:06Z")
I use a metasearch site, [The Big Hub](http://www.thebighub.com) most often, and will sometimes use Dogpile (linked above) or others. One thing to remember is that they are not actively searching the Internet when the respond to your search request, they are searching their own database. These databases are updated continuously based on the criteria they tell their bots to use. Some databases are reviewed by humans. ---
**Author:** ![bibliophage](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/bibliophage/32/7615_2.png) [@bibliophage](https://boards.straightdope.com/u/bibliophage)\ **Post date:** [May 28, 2000, 6:57am UTC](https://boards.straightdope.com/t/all-search-engines-are-not-created-equal/20725/6 "2000-05-28T06:57:28Z")
This [site](http://www.wiz.co.uk/search.asp) among others gives top honors to Google. IIRC, a couple of years ago, the _Wall Street Journal_ gave top honors to [HotBot](http://www.hotbot.com). But [SearchIQ](http://www.searchiq.com/reviews/) gives top honors to [ixquick](http://www.ixquick.com/) which I never heard of before. The same rating ranks google well ahead of SavvySearch. I tried ixquick and I find it gives good results, but it’s slower than WebFerret. “ix” it may be, but “quick” it ain’t. At least not tonight. It does seem to give better results than SavvySearch, though, except for phrase searches. Ratings are subjective. They depend on many factors, including speed, comprehensiveness of the database, relevence of hits, etc. There’s no one best engine. It depends on what sort of searches you do, and how important speed is to you. WebFerret can be downloaded for free from [FerretSoft](http://www.ferretsoft.com/netferret/index.html). ---
**Author:** ![JonF](https://avatars.discourse-cdn.com/v4/letter/j/cab0a1/32.png) [@JonF](https://boards.straightdope.com/u/JonF)\ **Post date:** [May 28, 2000, 12:27pm UTC](https://boards.straightdope.com/t/all-search-engines-are-not-created-equal/20725/7 "2000-05-28T12:27:28Z")
Thre are many different ways that search engines can work. Sorry, I can’t often match engines with the way they work. Some possible methods: 1. “Quantity is all”. Send out a spider that hits as many pages as possible and index them . I believe [All the Web](http://www.alltheweb.com) uses essentially this principle. 2. “Humans know”. Only post links that have been reviewed by humans, [Yahoo](http://www.yahoo.com). Tends to have fewer links available 'cause human review takes time, but those links tend to be highly categorized and relevant. Often misses some categories completely. 3. “Web page designers know”. Rank links depending on how web pages link _to_ them. Artifacts tend to show up; [netscape.com](http://netscape.com) and [microsoft.com](http://microsoft.com) rank incredibly high in such engines. 4. “Searchers know”. Rank links depending on how many people clicked those links when viewing the results of searches. ---
**Author:** ![xekul](https://avatars.discourse-cdn.com/v4/letter/x/838e76/32.png) [@xekul](https://boards.straightdope.com/u/xekul)\ **Post date:** [May 28, 2000, 3:25pm UTC](https://boards.straightdope.com/t/all-search-engines-are-not-created-equal/20725/8 "2000-05-28T15:25:28Z")
Although Google’s index isn’t as large as some other search engines, I like it because its results are _relevant_. It determines the relevance based on how many other pages link to a given page. So, until recently, if you searched for `more evil than Satan himself,'' the first match was http://www.microsoft.com/ :) This is because a whole bunch of people use the word `evil’’ or ``Satan’’ to link to Microsoft. And the cache is great for pages that are gone or servers that are temporarily down. ---
**Author:** ![Arjuna34](https://avatars.discourse-cdn.com/v4/letter/a/eb8c5e/32.png) [@Arjuna34](https://boards.straightdope.com/u/Arjuna34)\ **Post date:** [May 28, 2000, 3:58pm UTC](https://boards.straightdope.com/t/all-search-engines-are-not-created-equal/20725/9 "2000-05-28T15:58:41Z")
It also depends on what you’re searching for. If I’m looking for some highly technical info (such as "what’s the dielectric constant of polyethelyne? or what are the different integrated circuits out there with a VGA controller on them?) I’ll use alta-vista, and be prepared to possible wade through some random hits (although the “near” keyword in alta-vista helps a lot). If I’m looking for a particular web page that I suspect is out there (for example, Virginia’s Division of Motor Vehicles page, or the home page for some company name or a city), I’ll use yahoo- since it’s created by humans, it tends to put home pages, etc. at the top of the list. Arjuna34 ---
**Author:** ![RealityChuck](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/realitychuck/32/195_2.png) [@RealityChuck](https://boards.straightdope.com/u/RealityChuck)\ **Post date:** [May 28, 2000, 7:52pm UTC](https://boards.straightdope.com/t/all-search-engines-are-not-created-equal/20725/10 "2000-05-28T19:52:16Z")
Google is currently the best because its system of ranking pages by number of links to it usually brings the link you want up very quickly. It’s also very quick, and the cache feature is great if the page has been taken down. Yahoo is awfully good if you’re looking for websites on a particular subject. Most search engines will find sites where your subject is mentioned; Yahoo finds sites devoted to your subject. For a metasearch, Dogpile is pretty good, as is Copernic (which is metasearch software). I also like Northern Lights, which reports results grouped by general subject, and [http://www.alltheweb.com](http://www.alltheweb.com), which is better at finding non-U.S. sites. ---
**Author:** ![omni-not](https://avatars.discourse-cdn.com/v4/letter/o/8dc957/32.png) [@omni-not](https://boards.straightdope.com/u/omni-not)\ **Post date:** [May 28, 2000, 8:36pm UTC](https://boards.straightdope.com/t/all-search-engines-are-not-created-equal/20725/11 "2000-05-28T20:36:40Z")
From Forbes issue “Best of the Web”: Favourite: Hotbot / [http://www.hotbot.com](http://www.hotbot.com) Best (feature): Highly customizable Worst: Search partners like Dealtime pop up no matter how irrelevant their services may be Others (in alphabetical order): Altavista / [http://www.altavista.com](http://www.altavista.com) Best: Enormous breadth and depth of information available Worst: It takes some effort to grasp all the site has to offer Direct Hit / [http://www.directhit.com](http://www.directhit.com) Best: Best for commonly searched items Worst: Obscure topics provide meagre results Google / [http://www.google.com](http://www.google.com) Best: You can search in a dozen different languages Worst: Harder to narrow your search Northern Light / [http://www.northernlight.com](http://www.northernlight.com) Best: Get alerts if a new Web site appears on a specific topic you have identified Worst: No customization features Savvy Search / [http://www.savvysearch.com](http://www.savvysearch.com) Best: Create and name your own personal search engine Worst: Antiquated site design ---
**Author:** ![Jois](https://avatars.discourse-cdn.com/v4/letter/j/ea666f/32.png) [@Jois](https://boards.straightdope.com/u/Jois)\ **Post date:** [May 29, 2000, 5:20am UTC](https://boards.straightdope.com/t/all-search-engines-are-not-created-equal/20725/12 "2000-05-29T05:20:28Z")
Thank you, now maybe I can narrow my searches down to 3-4 engines and save some time! Jois ---
**Author:** ![HorseloverFat](https://avatars.discourse-cdn.com/v4/letter/h/8e8cbc/32.png) [@HorseloverFat](https://boards.straightdope.com/u/HorseloverFat)\ **Post date:** [May 29, 2000, 10:42am UTC](https://boards.straightdope.com/t/all-search-engines-are-not-created-equal/20725/13 "2000-05-29T10:42:57Z")
> [@](#): > > \*\* > Thank you, now maybe I can narrow my searches down to 3-4 engines and save some time! Jois > \*\* Don’t bother, use a meta-search (a search of search engines) like [askjeeves.com](http://askjeeves.com). But [google.com](http://google.com) is really well done. Between yahoo’s index by subject and google’s content search its pretty easy to find what I need. ---
**Author:** ![psycat90](https://avatars.discourse-cdn.com/v4/letter/p/7ea924/32.png) [@psycat90](https://boards.straightdope.com/u/psycat90)\ **Post date:** [May 29, 2000, 2:52pm UTC](https://boards.straightdope.com/t/all-search-engines-are-not-created-equal/20725/14 "2000-05-29T14:52:01Z")
I’ll second a vote for Webferret. It’s pretty cool. ---
**Author:** ![London\_Calling](https://avatars.discourse-cdn.com/v4/letter/l/49beb7/32.png) [@London\_Calling](https://boards.straightdope.com/u/London_Calling)\ **Post date:** [May 29, 2000, 5:14pm UTC](https://boards.straightdope.com/t/all-search-engines-are-not-created-equal/20725/15 "2000-05-29T17:14:49Z")
Why is one engine better than another ? Kind of like cars – depends on the spec. What suits you may not suit another, some others might be plain inferior. Why is this ? Because each search engine (that doesn’t include meta engines like dogpile or Directories like Yahoo) has in place it’s own methodology for determining which sites in it’s index are relevant to a searcher’s request. After the engine’s done that, it has to determine the order of relevance - putting what it deems to be the most relevant at number one and working down. What can make an engine good is the way it makes those two determinations. It’s done by using an algorithm – a complex means of analysing the composition of every web page in it’s data base (and in the last few months also by analysing the web pages that link to other web pages). Once it does that analysis, it’s produces the results – takes about 2 seconds max. Pretty awesome stuff. So, the effectiveness of an engine is determined by (1) How tech advanced it is, and (2) How many people have worked out enough of it’s algorithm to manipulate it to their advantage and not yours. Cracking an algo is not unlike cracking a safe. I know of one guy who grossed $4 million last year as an affiliate selling Viagra. But he is the very, very best. Problem with the results from meta engines is that they can be easily manipulated by a web site’s meta tags. This can lead to irrelevant results for you but more clicks for the site that misled you into thinking it was on topic. Directories provide more reliable results as each web site is assessed by a human before being included in the index, the engines all use automatic submission systems. Aside from Yahoo, don’t forget the largest but as yet least well known Directory [http://www.dmoz](http://www.dmoz) - compiled by volunteer editors and twice the size of Yahoo. Google is probably the best engine right now for both speed and accuracy but its index isn’t as large as some others. Excite is still ok. In about three months the best engine – together with Google – will probably be one you can’t access directly. You get to it by using Hotbot or the ‘web pages’ tab that comes up AFTER Yahoo has returned your result. It’s called Inktomi but not many know of it because it’s in the background providing results for very many of the ‘named’ engines. It has a new algo it’s trying out and it’ll take a while for it to settle down. ---
**Author:** ![funneefarmer](https://avatars.discourse-cdn.com/v4/letter/f/9fc348/32.png) [@funneefarmer](https://boards.straightdope.com/u/funneefarmer)\ **Post date:** [May 29, 2000, 6:03pm UTC](https://boards.straightdope.com/t/all-search-engines-are-not-created-equal/20725/16 "2000-05-29T18:03:58Z")
I think most of it is how you narrow your search. I get good results from google and alltheweb simply because I’m familiar with them. That said if you want to get down to the nitty-gritty try the metasearches although frankly I think askjeeves is a load of crap, maybe I’m just not familiar with it. I keep a directory of search engines, [Netstrider](http://www.netstrider.com/search/directory.html), for looking up stuff as well. With something like Netstrider handy you can choose a search engine that fits your needs or style of searching. ---
**Author:** ![handy](https://avatars.discourse-cdn.com/v4/letter/h/b5a626/32.png) [@handy](https://boards.straightdope.com/u/handy)\ **Post date:** [May 29, 2000, 10:52pm UTC](https://boards.straightdope.com/t/all-search-engines-are-not-created-equal/20725/17 "2000-05-29T22:52:18Z")
To narrow your search, just put quotes around it, e.g. “monarch butterflies” I like [ask.com](http://ask.com) but [snap.com](http://snap.com) best cuz its got great new software & pretty ads on tv.