Ask HN: Is there room for another search engine? • Hacker News Board

Re: Ask HN: Is there room for another search engine?

#121

Post by cocktailpeanuts » Sun, Mar 26, 2017, 8:28 PM UTC

bshimmin wrote: post #105

cocktailpeanuts wrote: post #83

Earlier quoted context omitted.

> We can agree to disagree about Google's immortality Dude where did I say Google is immortal? I'm just saying it's a wrong approach to try to kill google this way. There are many good ways to do it and coming from "How do i disrupt Google" is the worst possible way. > Obviously Microsoft has no chance, they're worse off than Google re: ossification and Bing is a hedge at best. Nothing is obvious when we talk about d…

I wonder, is DuckDuckGo's name a problem for adoption? I have no idea what it's supposed to mean, but it sounds very silly and I don't much enjoy recommending it to people in real life, at least in part for that reason. ("Google" was a silly name too, but at least it was just one syllable of silliness.) Privacy is probably slowly growing into a big deal in the minds of regular people, but when the incumbent works, an…

Yeah like you pointed out, if your product stands out so much that nothing else matters, then you can name it whatever you want. But duckduckgo is just a glorified meta search engine and they are not doing anything completely unique. In this case branding matters and I think their name is really bad for branding. There are many examples of really bad naming, like cockroachdb and swagger. Swagger ended up changing their name to OpenAPI (Thank god) but I don't know what the cockroachdb guys are thinking. I don't want to use a DB named after a bug (pun intended).

Re: Ask HN: Is there room for another search engine?

#122

Post by bobosha » Sun, Mar 26, 2017, 8:33 PM UTC

kapauldo wrote: post #33

If you keep the search paradigm of entering text and returning 10 links, it's probably not likely to succeed. But come up with a new pardigm and you can definitely shake things up.

This ^. I would also posit that coming up with a completely new paradigm for a mobile form-factor (cf. the mini-me approach of current mobile search), has lots of potential.

Re: Ask HN: Is there room for another search engine?

#123

Post by Skylled » Sun, Mar 26, 2017, 8:34 PM UTC

I'd think there would have to be something fundamentally different. It would have to be hardly recognizable as a "search engine."

There's too many clones on the market right now. Some with good purpose, like DuckDuckGo which can be simplified to "Google but without privacy invasion." Others like Bing could be just "Google but clunkier." (my opinion)

If you've got an idea on your hands that can't be described as "Google but..." then there's definitely room for another.

Re: Ask HN: Is there room for another search engine?

#124

Post by moultano » Sun, Mar 26, 2017, 8:42 PM UTC

thesmallestcat wrote: post #47

cocktailpeanuts wrote: post #43

Earlier quoted context omitted.

Those are all things Google can already do but decide not to. Which means if any startup attempts and gets a moderate success, google will of course decide to do the same. Your startup will be nothing more than an R&D department Google uses for free. Disruption happens by doing something completely different that happens to cannibalize the incumbent's business in a way that they can't change. All these little feature…

Keep in mind that it's been 20 years since Google's differantiator (PageRank) was invented. It's been about 10 years since Google's last meaningful changes to its search UI (autocomplete and "instant" search). So, while I respect Google's omniscience and omnipotence, I disagree strongly that Google/Alphabet are in a position to respond to a serious upstart competitor. Their primary response, as you hinted would be to…

I'd argue that showing answers on the search result page is a bigger innovation than "instant" was. It also requires much more new technology.

Re: Ask HN: Is there room for another search engine?

#125

Post by matt4077 » Sun, Mar 26, 2017, 8:46 PM UTC

MichaelMoser123 wrote: post #111

There is the problem of relevance - you must order your search results by relevance, Now you can have one global model of 'relevance' or you can have several models (i think the web is so big that one model is not good enough). relevance can be relative to language/origin ; number of links away from a wikipedia article ; coolness (has it a link from a known twitter account/news aggregator) ; age group ; a link from H…

Google obviously already operates such profiles, at a rate of one profile per user - with AI in the background.

Exposing options is rarely a good idea, as it only reaches a single-digit percentage of users.

And they say profiles are oh-so-relevant, but as far as I can tell, Google's main product (search ads) is still tied almost exclusively to keyword, region, and language.

Re: Ask HN: Is there room for another search engine?

#126

Post by zeta0134 » Sun, Mar 26, 2017, 8:48 PM UTC

arekkas wrote: post #95

Animats wrote: post #24

That's a good question, and something I've spent much time on. Cuil (2008-2010) tried. I knew some of those people. It cost them about $30 million to launch a full scale search engine. They had no revenue model. In retrospect, they were hoping to be acquired by somebody. It was some ex-Google people, trying to replicate older Google technology. They had a great launch, but the system wasn't very good and traffic rapi…

I was very young when Google became popular. Why did it become so popular? Was it just the technological advance of Google-Matrix (Page Rank) and Map-Reduce?

My mother introduced me to Google back in the day. She and several of her friends agreed that it was amazing because it was fast, simple, and got out of your way. And back then, that's exactly what it was:

http://royal.pingdom.com/2008/09/05/googles-first-website-fr...

At the time, their main competitors were already expanding into other markets, and Google was refreshing for doing one thing and doing it well, compared to its competition. It didn't clutter, it didn't try to distract with other services that they offered. It simply gave you the results you queried as a plain list of links, and got out of your way so you could browse those links at your leisure.

Re: Ask HN: Is there room for another search engine?

#127

Post by jgalt212 » Sun, Mar 26, 2017, 8:49 PM UTC

Animats wrote: post #24

That's a good question, and something I've spent much time on. Cuil (2008-2010) tried. I knew some of those people. It cost them about $30 million to launch a full scale search engine. They had no revenue model. In retrospect, they were hoping to be acquired by somebody. It was some ex-Google people, trying to replicate older Google technology. They had a great launch, but the system wasn't very good and traffic rapi…

> Yahoo had a search engine from 1995 to 2009. Yahoo is now a Bing reseller.

If so, why is Yahoo Slurp still showing up in my apache logs?

https://help.yahoo.com/kb/SLN22600.html

Re: Ask HN: Is there room for another search engine?

#128

Post by jacquesm » Sun, Mar 26, 2017, 8:50 PM UTC

Animats wrote: post #24

That's a good question, and something I've spent much time on. Cuil (2008-2010) tried. I knew some of those people. It cost them about $30 million to launch a full scale search engine. They had no revenue model. In retrospect, they were hoping to be acquired by somebody. It was some ex-Google people, trying to replicate older Google technology. They had a great launch, but the system wasn't very good and traffic rapi…

I miss DuckDuckGo in your line-up.

Re: Ask HN: Is there room for another search engine?

#129

Post by jacquesm » Sun, Mar 26, 2017, 8:52 PM UTC

Animats wrote: post #81

cocktailpeanuts wrote: post #59

Earlier quoted context omitted.

There are already many "serious upstarts" that are doing just fine but will never reach the status of Google, like duckduckgo, bing, etc. Some even say bing has better search quality than google. But that's not what matters. No matter how good the product is, if you can't get distribution it doesn't matter. Also, even if you do have distribution, it may not be the right context so it won't be effective. If anyone is…

prepare to be the most patient person on earth, because it won't happen fast, and probably won't even happen Maybe not. The Cuil experience is significant. The PR for the launch was good, there was a huge initial traffic spike, and then users discovered the product sucked and traffic dropped. The system at launch was terrible. Six months later it wasn't so bad, but nobody cared by then. They launched too early. But t…

I think the bigger problem with Cuil is they over-hyped their offering before launch which could lead to nothing but disappointment.

Re: Ask HN: Is there room for another search engine?

#130

Post by nostrademons » Sun, Mar 26, 2017, 8:56 PM UTC

hashmal wrote: post #119

kapauldo wrote: post #32

Earlier quoted context omitted.

It's bizarre that algolia is a thing. Elasticsearch is good and free.

there are lots of things like this. it's bizarre that i pay for email when there are good and free email services (and my privacy/security needs are pretty damn simple.) people don't mind paying for things they like , or for something that provides just a little detail that others do not. Algolia provides things that elasticsearch do not, such as being hosted, being extremely simple to setup and having a default web…

There are plenty of companies providing hosted ElasticSearch these days, starting with Elastic.co, the corporate sponsors of the open-source project. Amazon hosted ES is also a lot better than it was just a few months ago; they were stuck on 1.5 for the longest time, but now they've already got support for 5.1.1.