I'd be interested to have an article like this for slightly less popular search-heavy sites like Amazon, Facebook, Craiglist, Ebay, etc.
Get better at Googling
291–297 of 297 posts
Re: Get better at Googling
#292Earlier quoted context omitted.
It is sadly very true. See my comment here or what a lit of others write: https://news.ycombinator.com/item?id=26926309 Also I'd like to know what I should do to get your experience.
Likely there's some a/b testing and google has decided that you're more profitable without access to those tools
It's been happening since before 2010, but sure.
Re: Get better at Googling
#293Earlier quoted context omitted.
> I'm sorry, this makes zero sense and even if it did, is patently false. This feature works and I use it all the time. Lucky you. I can confirm qyis observations: it started to ignore me sometime around 2008 and hasn't worked reliably since. We have tried to tell them since then. I know Google runs a lot of experiments all the time but this one has been morenor less consistently ignored for me.
Can you offer a concrete example?
Give me access to a machine that isn't blocked from Google, and I will.
Re: Get better at Googling
#294Earlier quoted context omitted.
> I'm sorry, this makes zero sense and even if it did, is patently false. This feature works and I use it all the time. Lucky you. I can confirm qyis observations: it started to ignore me sometime around 2008 and hasn't worked reliably since. We have tried to tell them since then. I know Google runs a lot of experiments all the time but this one has been morenor less consistently ignored for me.
Can you offer a concrete example?
Hi again leoh. Did you try searching for "goolge" like I suggested yesterday?
If so can you confirm my results?
If not, can I have your cookies (just joking).
If you happen to work at Google, can you tell them that this is another way that they drive even previously ardent fans to hope that Google will sooner or later be split up so they can start focusing on search again instead of plotting for world domination.
Re: Get better at Googling
#295Earlier quoted context omitted.
Not really. Try asking it "how long does water take to freeze 300m above sea level" or "how much time does water take to freeze when 300m above sea level" It's a simple enough question -- it's able to show me a gradient curve when i give it "water freeze", but it's not able to parse that data into a simple format, which is the entire point of using a search engine like that. It's been the same for a number of simple…
Water only freezes when it is cold, it isn't time dependent.
The time it takes to become cold enough to freeze is pressure-dependent. "Room temperature" can be assumed in the case, or hell, I could have told it it was 21°C. I only wanted a ballpark figure to base a rough plan around ("This will take around this long, so I have this much time before I can do X"), which was certainly within the power of Wolfram Alpha to give me.
The point that this example is trying to illustrate is that Wolfram Alpha is not a search engine ("Find this out for me"), it is a definition engine ("Given these definitions, what is X"), and while it advertises itself as the former with the examples given, doesn't especially succeed at it to any specific degree.
Re: Get better at Googling
#296Earlier quoted context omitted.
I've worked in a lot of places where there were no docs and nobody knew how any worked. So be thankful that you have some docs.
And I've worked with auto generated meta code in Java 1.3. In practice, difference between it and Oracle docs was negligible. Hell. Their auto-gen code was more self explanatory than Oracle Java Docs.
I personally enjoy the Javadocs that tell me nothing about how to use the damn thing. Spending hours poring over it, with no examples, just a generated Javaoc version of the static types my IDE would have told me, and through much trial and error, I eventually figure out that I need -
new Doer(UnrelatedFactory.create("foo", new WTFBar(), 5, null, null), RandomEnumFromADifferentPackage.MEANINGLESS_VALUE, true).do()
But I'm not really convinced such documents are helpful. I mean, technically they're better than nothing (as in, literally better than nothing), but the difference between nothing and those docs is so small it approaches nothing ('as the worthlessness of the documentation approaches infinity the value of reading them approaches nothing', as it were)