Ask HN: Google Goes Popular? | Hacker News Reader

Ask HN: Google Goes Popular?

23 pointsbertm16 comments

Lately, I have noticed when I am doing a scientific search on Google,maybe an algorithm, resent research paper etc., I get less scientific sources than I normally would and more what I would consider "popular" sources. Has anyone else noticed this? Let's face it Google is pretty much my lens into the world. I am not near a major university library, so I do not have much choice.

I do not like this "pressure" to fit into Google's popular mold. Does anyone else have a better search engine for meaningful sources. I'm not saying Wikipedia, About, or Ehow are not creditable sources... hmm... yes yes, I am saying that. Any Help? Maybe something that indexes noncommercial sites only...

View on HN

Are you using Google Scholar for this research? http://scholar.google.com/

sadly, Google Scholar mostly gives me a lot of articles... that are all behind a paywall.

You should e-mail the corresponding author and ask them to send you a .pdf. It helps if you claim to be at a university from a poor, developing country and your library doesn't have any subscriptions. Some authors will also have the .pdf uploaded on their website.

Upvote, but not for the suggestion to lie.

It's a terribly inefficient system and all government grants should require any resulting publication be openly accessible. But until we get there, simply look for their website or send them a nice, short email with the publication in the subject line. If there are multiple authors, you could have more success with a middle author than the leading or trailing authors (first and last are often the lab head). An alternative is looking for an administrative contact on their website.

...which does not make them less relevant, does it?

That mostly varies by area. Humanities are super tough to get. Most sciences, especially physics, will have a pdf that scholar will point you to.

Of course, if you are at a university you can basically get anything.

Add filetype:pdf to your query.

Agreed, advanced search is your friend when looking for solid research. In addition to looking for pdf's I often get better results filtering by .edu sites as well.

That will miss many non-US institutions. For example, UK universities use .ac.uk

If you want to include them, boolean logic still works for advanced search tags. You can search for 'somethingorother site:.edu OR site:.ac.uk'

When I was a student I used CiteSeer all the time, but I guess it depends on the area of your research.

http://citeseerx.ist.psu.edu/

Google search has been almost useless for me since they dropped implicit AND for implicit OR.

As in, the query [random trie] is now (random OR trie). The query [+random +trie] returns much better results (random AND trie).

I've seriously been considering starting a search engine just to get back to a ~2002 era Google level of usability.

I agree, plus their word mangling/assumptions are out of control. I often have to use double quotes to make it clear that yes, what I typed in is what I'm searching for.

I'm kind of curious: hypothetically imagine Google had super crazy search personalization... and knew when you were interested in scientific sources in a field versus just the wikipedia article. Would you be happier because your searches are great, or freaked out because you feel your privacy is being invaded.

I wonder if this might have something to do with it:

http://news.ycombinator.com/item?id=1082209

Wikipedia is noncommercial.

Reply on news.ycombinator.com