how can i start my own search engine?

Take a look at a few of the smaller engines out there. Mamma, etc. While Mamma was pretty big at one time it dropped into oblivion because it couldn't keep up. But it still requires a tremendous revenue to keep running.

I think that if you have to ask, you probably don't have the skill or capital to get it above a hobby site if you try a spider search engine.

A search engine that uses results from the big boys, would be relatively easy. But you would need something unique so that people would use you instead of what is already out there. So, study other meta search engines and try to come up with an idea that will make you better. If you can then deside if you want to spend the time and money to do it and if you could ever get your investment back out.

 

lol funny thread. I like the idea of starting a search engine too. What stops me is i have only one concept in my head that would make it unique but lack the funds to do so

maybe I'l go ask Sergei for advice

 

I count 3 steps, not just a couple. That third step is the tricky one.

 

Well, i am asking myself the same question, I am using google coop to get better results and really getting tired of it , what if want to index 80 000 pages, would datasearch would do? Someone mentioned metasearch engine, any good script existing?

For the one who want to start a country, there are islands in international waters near sweden, you can establish there, you won t have dsl connection though, I ve heard that you can buy sea platforms from exxon or total, fresh fish everyday and quite a big swimming pool .

 

Market is saturated I think, money will be high

 

Hmmm, OK. But what language should I use?

Can we assume the original poster was trolling since he hasn't commented on the thread in about 2 days.

 

You can buy one that is already developed with a search script and go from there configuring it how you want.
PM me, I have 2 for sale.
Or hire someone to build one. It may cost you around $50k to start, and then you can develop further over time.
Good Luck.

 

buy what? meta search engine? If so, show me!

 

Language isn't as much the issue as Database. Although one of the higher end languages would be better in the long run. If you really plan on creating a large database to be accessed by a large number of people that can handle many concurrent complex queries quickly, you will need to have a strong database. MS SQL (yes I'm prepared for the flame ), Oracle or one of the other expensive ones.

 

Believe it or not, there are search engines that have been developed, that don't have the same recognition as the ones frequently mentioned, or have basically run out of operating funds, that are on the market.
You can buy, or build anything if you have the funds. (Of course not proprietary stuff, but there are programmers that can get you damn close)

 

ok I misunderstood you. I thought you got one running. reading this thread, i refine my quest . I am looking for a script that do highly configurable metadata search and display graphically descent results .

 

Google was a graduate thesis project. It is one of many "a couple of guys in a garage" internet success story. Its clean, ad free start attracted people and it has rolled forward from there -- layering the advertising revenue potential.

I agree with some of the comments here. It is better to start a niche engine and build out from there. You will at least be able to build a core following and, based on their needs, expand the scope of the site.

It is worth noting that the cost of such projects is declining, at least from a hardware perspective, and it may be possible to grow with revenue. This is not the path for the impatient. But, success only looks fast from the outside.

 

You have to make a robot to surf the web from C++ or similar languages.

I had one in VB, that surfed a web page then saved all the content of it.
then the content and url was uploaded to a mysql database.

The data was then found via PHP

It's not a simple thing to do.

 

That was my point, when I see what I get from my hoster for a small price, I thought I could give a try with starting my own niche engine, now, what would be the best tool to start with?

 

A niche engine might be doable, especially if it's a small list of sites, but if you're talking anything remotely like what Google or Yahoo has, you're looking at pouring barrels of cash into and having a ton of server storage and processing power. It CAN be done in something like PHP/MySQL, but if you want something outside of a small scale, couple thousand sites, you're going to need to write a LOT of custom stuff and hire some really good guys. If I had a few months I could sit and write a nice search engine, but honestly unless you want to spend a lot of cash to do it right (and why waste the time and money if you're not going to do it the right way), then it's far from worth it.

 

is 1000 site a small scale? Barrels of cash, that s what I need

 

If I had barrels of cash, my day's of taking chances would be over,and I certainly wouldn't want to be on the hook for a few mill of someone elses.

 

I think that starting a search engine is a major decision; you need to seriously think about this as it will cost you time and money. People are so used to automatically typing www.google.com as their initial search engine, to compete with engines such as yahoo and google or even aspire to compete, is defiantly an extremely ambiguous notion. I think you missed the catch 10 years ago mate.

 

I am gonna be more specific, I have web activity in cinema, some big empty sites with no contents but actor s names, film s names, comments and ads crowd google results, they are, in my point of view, useless, what I want is take a selection of website, that I will select, and spider them, It can go up to a thousand sites and I could include the results of my cinema directory that hold 500 sites .

If I can t get my spider, I want a meta search engine script that could dismiss any of the sites I mark as irrelevant

I don' t really know where it goes but I know I want to give it a try, so I am kind of ready to test both solutions. Whatever if I make money out of it one day. It will be on a dedicated server that only host low traffic websites (I move them when they get too big )

I don t want to google the world , I want to ungoogle myself and people who are looking for good information

 

It sounds like you're doing something really low scale with less than a few thousand sites, that's nothing really, go for it. I've done larger searches with systems that have like 10,000 results. If someone wants to make an actual search engine that has more than probably 5k - 10k results, it would be a good idea to rethink it, and if that's what they really want to do then invest some money and have it done right, but with what you're doing, it's small enough php/mysql can handle that just fine.