Forget Cuil, check this out: one guy and a bunch of pcs | Hacker News Reader

Forget Cuil, check this out: one guy and a bunch of pcs

gigablast.com

45 pointsjacquesm40 comments

From the about page: "Now Gigablast has grown to several, highly-talented employees" So not quite one guy.

That said, after having read some of the comments on the page I think we should cut the guy (or guys) some slack. As jacquesm's link points out 55% of users in a double blind study found gigablast better than Google. We're talking about a couple of guys matching, and maybe beating, one of the worlds biggest software companies at their own game. A couple of guys matching Google's thousands of employees and thousands of servers.

In a word I think they're absolutely fucking awesome and deserve total and absolute respect.

They're not really competing with Google. Their big business model seems to be setting up a Gigablast cluster for customers to index a custom collection of info, a la the Google Search Appliance, but more vanilla. Further, they're not doing mail, or news, or reader, or wave, or groups, or chat, or android, or chrome, or GWT, or adsense, or youtube, or iGoogle, or docs, or app engine. So yeah, they have fewer employees, but they're not exactly doing the same set of things either - they're really focusing on performance in one area: search. Google may be about search, but they do a lot more than that.

The 10K makes it pretty clear: Google's in the advertising business, period. They hardly sell any search appliances.

If they are better than google and nobody cares, that is a frightening thought.

No, that's part brand recognition, part comfort and part trusting etablished companies.

Are they better than google at search (which is what I think you meant)?

Can they scale what they have to serve the number of SERPs that Google serves? Can they do that cost effectively? Can their algorithms produce as useful results in the face of "optimisations" by 10's of thousands of SEO/SEMs across the globe?

I'm guessing no.

My wife makes the most awesome gingerbread men (and brownies) but that doesn't mean she can put Greggs out of business.

Still props too them, they're doing something awesome.

"blind test" doesn't work. Read "blink: The power of Thinking without Thinking" by Malcolm Gladwell.

"blind test" doesn't work.

Please explain.

Based on what I understand from this presentation:

http://www.ted.com/talks/malcolm_gladwell_on_spaghetti_sauce...

Gladwell means that people don't always know what they want (or wont be able to tell you).

Always is the operative word in that sentence. Most of the time they know quite well what they want, sometimes they don't.

I read Blink. Gladwell need to remember that the plural of anecdote is not data.

He certainly raises some questions about blind tests compared to expert opinion, but as far as repeatable, replicable, methodology goes, my trust is in the blind test.

I always thought that to beat google you would have to be better by a huge margin. So even if its true that this searchengine performs 5% better (whatever that measures) it's not enough to convince people to switch.

I interviewed there once. Smart people; demanding, too. By that, I mean that I'm sure all of you would find it to be a rewarding and challenging workplace.

They made it clear that there was no VC money. Classic bootstrap. Matt wrote most of the original code base himself. I seem to remember they had at least several employees.

Ah yeah, the old work-for-equity scam. Classic sign that the company is going nowhere.

What make you assumed that bootstrapping company does work-for-equity scam? There are always other ways of bringing in money without venture funds.

You man it's not a get-rich-quick scheme ?

There are many routes to success, having a dedicated team that is co-owner of the project is one way of achieving that, but just like any other route it is not a guarantee.

'going nowhere' and 'being a regular business' are not the same.

In terms of relevance there seems to be a tie between gigablast and cuil (according to my highly unscientific study). And google search is still better. And I like cuil's design more.

In '08 they did a double blind study comparing gigablast with google and two thirds of the users preferred gigablast.

It's a dated study of course, and I can't find a reference for it (which doesn't help) but it impressed me quite a bit.

edit: found a link: http://www.searchenginejournal.com/gigablast-better-than-goo...

I misremembered the percentage though, not 66 but 55.

Not to knock them, from the little I know I'm very impressed, but this study showing Gigablast to be better than Google was conducted by Gigablast.

I just thought I'd mention it to save other people the 5 minutes I spent researching it because it just seemed to good to be true.

This study has an inherent flaw: the queries used were selected from the most common queries. Answering common queries is (relatively) easy, but Google claims (and other independent studies have shown) that Google is much much better than most of its competition for returning good results for uncommon queries (the "long tail" if you will).

Edit: I just (briefly) tested this myself. Searching for for my username on Gigablast returned 3 results: 2 of my HN comments and a wordlist. Google on the other hand found 3k results, with the top results being my linkedin and facebook profiles (which I would deem the "best results").

Can't agree more with that. I searched for some Cocoa part which was a tie between Google and Gigablast; a torrent which Gigablast could find, but at its origin; and the AD&D Sorcerer which Gigablast couldn't find anything about.

Gigablast might have better results for common queries, but those matter hardly to the typical HN user.

They definitely deserve some merits.

IMHO, though, i think the normal search engine becomes more like a marketing game now. Same as "most people prefer something other than Coca-Cola in blind test", yet they are still the biggest seller by far.

Unless something is changing in the landscape that affect users behavior (like social web - led by FB, Twitter etc), i wouldn't expect anyone can win against Google in this game without MUCH higher value propositions.

I'm not so sure.

Search is funny that way, you know instantly if a search engine is 'good' or 'fluff'. Most engines out there today are very comparable wrt to the quality of the results, some are better in one domain, others are less susceptible to spammers but overall there is not that much difference, it's comparable to different tastes in icecream. EDIT: once you've got a user they're not likely to switch away from their favourite flavour, it's what they're used to.

But if some group came along that quietly and without fuss presented you with a search engine that had exactly what you were looking at on page one every time I think the current crop would fade faster than you could say 'altavista'.

That's what makes technology industry interesting. Things are quite unpredictable and like it or not, none of us knows exactly how things will pan out.

Google's strength is obviously in brand penetration when it comes to search engine. After all, most people outside us hardly even know about "Google Chrome" after making such a huge splash all over the place. (http://www.youtube.com/watch?v=o4MwTvtyrUQ).

So, i wouldn't expect them to type in www.cuil.com or WolframAlpha without something is changing in the landscape that alter user behavior.

Seems like the last bit of news was from 2005:

http://gigablast.com/press.html

I wonder what he's been doing for the last 4 years?

I haven't a clue, but it seems like the site has been a pretty constant performer, I have never seen it 'down', it looks like he has about 50K uniques every day.

You have to admire his accomplishment, but the presentation grates on me -- the fonts are pretty unreadable and it all seems very rough around the edges. Even if the standard were up to Google's, I wouldn't use it for that reason alone.

So everything you dislike about the site could be fixed by a one-time contract from a good graphic/UI designer? Sounds like a compliment to me.

Interestingly the fonts are actually more readable to me than google's (slightly bigger, slightly more kerning). Perhaps he's only testing on linux - and I'm also on linux.

The presentation also makes scanning incredible hard. The results look cramped. Ugh.

They should fix their frontpage if they want to impress international users. I was thoroughly unimpressed when I searched for my last name ('grytøyr') and saw that it became 'gryt?yr' and included search results for 'gryt'.

However, searching for the same from the result page works, so it appears to be just a charset bug with the frontpage (which I’ve notified them about).

> Gigablast databases are scalable to 200 billion pages with minimal hardware. (100,000 servers)

While that's obviously a ton of pages, I've just never heard someone call 100,000 servers a minimal amount of hardware. Different points of view I suppose.

In the 'rants' ( http://www.gigablast.com/rants.html ) there is a lot of background information, he says he originally released when he had only 8 servers.

Also, there is no way he has 100,000 servers, he never took in outside capital. I think that part should be read as 'it scales to 100,000 servers), probably wildly optimistic since I highly doubt there is a gigablast implementation anywhere that has that many servers.

> Gigablast plans to be the best and most popular search engine on the internet within the next year and if you are interested in being part of such a business then this is the place for you!

while clicking around this reminds me of early google, however after a few results - I can already strongly suggest ways to game the system, first it ranks to heavy off of keywords in the domain name etc..

I use it as an alternative when google does not give me what I'm looking for.

It's small enough that it doesn't pay off to game it, but I'm sure that if it would take off (pun intended) that the 'war' would be on pretty quickly.

I used gigablast for a couple of searches and it just didn't give me the results I was looking for. I think I will stick with google until I find something that is consistantly more reliable

You nailed it there, that is the key. And when such a thing comes around we'll see how strong that google brand really is.

Well, it passed the ego-search test with grace. I almost thought people had given up on a search. There's just so much more you can do than what Google does.

I liked the subnavigation "Giga Bits" link.

Reply on news.ycombinator.com