Hacker News Reader

Comment by giancarlostoro | Hacker News Reader

I've tried switching from Google to reasonable alternatives, and I usually have to go back to Google to search twice because the results are not what I was expecting, nowhere near. I'm not sure if this is only my own experience, no offence to other engines (and their engineers), but if you can't find what I'm looking for I'm going somewhere else to find it.

View on HN

Getting a bit tangential: I've been trying Bing again recently out of curiosity, and at least for my usage I find it now pretty reasonable. A few years ago it was barely usable, but now I'm fairly happy with it. It also feels slightly snappier to me than Google on loading results. The main problem is that a small but significant number of sites have a robots.txt that excludes all non-Google crawlers, making it impossible for other search engines to display equally good results in cases where that site has something relevant (unless they blatantly ignore robots.txt).

So wait, who actually blocks Bing but not Google? That is just completely odd sounding to me... But Bing is definitely one engine I'm trying to switch to, it just isn't reliable when I'm looking up programming related queries sometimes. Then again I always know all the key words to throw at Google to find whatever I'm looking for. Googling is a (query) programming language by itself.

Why would someone do that?

If a site owner wants traffic on their site - and most do - they want to make their site as easily discoverable as possible. That means being in all the search engines. Blocking one engine (or all except one) flies in the face of most business models.

The only exception I can think of is Google properties. By blocking out competitors, they can try to worsen the experience on other engines. Like imagine a video search on Bing that didn't include YouTube? But even then, it seems like someone finding a Google property via Bing still furthers their interests.

Few deliberately block Bing specifically, it's just a sloppy attempt to block unwanted crawlers, like the billions of crawlers running on AWS and DigitalOcean boxes. A quick-but-not-great hack is to have an entry blocking * by default, and then specifically whitelist GoogleBot to avoid getting delisted from Google. You could also whitelist other search engines, but that requires you to know and care about them, and for a long period Google was the only engine people really cared about, since it drove >95% of organic search traffic. I do feel we're past the peak of that period, however, and site owners are now getting more cognizant of non-Google search engines having at least some value in driving traffic.

Related, "Please Don't Block Everything but Googlebot in robots.txt": http://danluu.com/googlebot-monopoly/

So it's more out of ignorance than maliciousness. That makes sense.

Odds are the more troublesome scrapers are probably ignoring robots.txt anyway, so I wonder how useful it is.

Bing's image and video search have been superior to Google's for over a year now, in my opinion.

I use Bing as my default search engine, but it falls short of Google's results for a lot of programming-related queries. I still end up having to use Google a few times per day when I'm working.

It's hard to say whether we're bubbled or if Google is really better. I tried duckduckgo for a while. There's a lot to love in ddg, but every time I needed something not obvious, I needed to check Google results too, just to be sure. And quite often, ddg results were, even if I'm not entirely fond of Google trying to understand meaning in my request, a bit too mechanical for my tastes.

Been using duckduckgo for a year or two now. If the results are not satisfactory i can always drop a !g into he query to have them send me to the Google results.

That said, DDG is damn good at giving me relevant stuff on clear queries. But i find that Google and DDG alike gets into a bind if i get a bit technical.

That's mostly my barrier for switching out of Google. Thanks for the tip, I always seem to forget that DDG can quickly push you back to Google when convenient.

The real fun part is when you set it as default engine for a browser. Then you can do things like !w to search just wikipedia's articles etc right in the url bar.

There is also startpage.com (!s(tartpage) in DDG) to give anonymized Google results. And also ixquick (!ix(quick), natch) for a meta-search (not sure exactly what engines they cover).

https://duckduckgo.com/bang

I use Waterfox, which is the 64bit version of Firefox. I always use the builtin search located in the upper-right corner of the browser which allows me to choose which search engine to use on the fly. I use all of the most popular search engines depending on what I'm looking for. I find when I'm looking for services and products, Google seems to work best. The default I use is the Storm search engine which seems to return results that avoid commercially driven items. I would have to recommend that everyone not fall in love with a single source for web searching.

No need to leave Google for that purpose - just use Google from incognito mode in Firefox.

Reply on news.ycombinator.com