News for hackers. Stuff that matters.
Re:Shirt Without Stripes by alberth on Monday April 20, 2020 @05:23PM (#22928734)
The # 1 google result for “Shirt without stripes” is this very own HN post

https://www.google.com/search?q=%22Shirt+without+Stripes%22&

Re:Shirt Without Stripes by albertzeyer on Monday April 20, 2020 @06:20PM (#22929404)
I guess, even in the original results, the problem is not really that Google search was not understanding the meaning of the search term (which is possible with today's models). Rather, it was a bit confused about what you are really searching for here. Maybe a comparison of shirts with and without stripes? The search query was just unusual, and it is not too unreasonable to guess that the query was not meant literally. At least this is a valid possibility, that the query was not meant literally. So it is reasonable to just return some results which might be related to the query, which will also be shirts with stripes.

If you argue this is bad behavior: Maybe we need a web query which really only takes the query literally. Putting the query in quotes will not quite have this effect for Google. Maybe some other syntax?

Re:Shirt Without Stripes by tcgv on Monday April 20, 2020 @07:04PM (#22929833)
> the problem is not really that Google search was not understanding the meaning of the search term (...) Rather, it was a bit confused about what you are really searching for here. Maybe a comparison of shirts with and without stripes?

In this very specific case I don't buy it. Sure, it probably applies for other queries, but if you approach a salesperson and ask him for "shirts without stripes" it's pretty clear what you want, and he wont bring you any piece with stripes on it.

Re:Shirt Without Stripes by indecisive_user on Monday April 20, 2020 @07:56PM (#22930310)
Agreed, it's not that weird of a search. Other, similar queries like "shirt without buttons/collar/pocket" seem to work just fine.

The only difference is that those are all physical properties of a shirt while stripes is a type of pattern.

Re:Shirt Without Stripes by zaat on Monday April 20, 2020 @08:54PM (#22930696)
Shirt without paisley - fails

shirt without buttons - preety much fail.

shirt without red button - as already expected, shirts with red buttons

Re:Shirt Without Stripes by ghayes on Tuesday April 21, 2020 @02:31AM (#22932551)
To be fair, the first and last query would also confuse me if someone asked me for that item. "Shirt without paisley" feels a bit like "cereal without elephants." You don't usually explicitly exclude an element that is relatively uncommon.
Re:Shirt Without Stripes by thw0rted on Friday April 24, 2020 @06:25AM (#22966495)
It's not just the way this is phrased, it's that there is no English-language formulation that works at all. Swap "shirt" for "tie", because paisley is quite common on ties. Now, try:

* tie without paisley * tie not paisley * non-paisley ties * ties that aren't paisley * ties other than paisley

You guessed it, in each case, at least half of the results are paisley ties. The only way to actually get what you want -- the set described by X, minus the set described by Y -- is to use the exclusion operator in the search, "ties -paisley".

This is great, and makes intuitive sense to somebody with multiple computer science degrees. But not only is it hard to explain to an outsider, it's actually quite hard to get them to think in a way that accommodates this capability, that is, in terms of set theory.

Re:Shirt Without Stripes by drongoking on Tuesday April 21, 2020 @11:09AM (#22935570)
I don't agree. Paisley is a well-known pattern type and some people really dislike it. "Shirt without paisley" is not a common request but its meaning is clear (if you do an image search for paisley you get lots of fabric images, so it's not like search engines don't know it.) I'd say the same for "shirt without red buttons." In general the pattern <article of clothing> without <feature> shouldn't be that difficult for search engines---especially since many are tuned for consumers.
Re:Shirt Without Stripes by 32gbsd on Tuesday April 21, 2020 @09:58PM (#22941262)
observations; firstly one can assume that google search can do no wrong. second observation is google search is made for noobs. thirdly you broke it because you are not a noob. 4th this will not be fixed because of said non-noob status.
Re:Shirt Without Stripes by lonelappde on Monday April 20, 2020 @08:53PM (#22930687)
This is way overthinking it. The search engines aren't semantically analyzing the images. They are just matching nearby text.
Re:Shirt Without Stripes by wyattpeak on Monday April 20, 2020 @10:00PM (#22931125)
I agree the parent is overthinking it, but that's underthinking it. It's been a long time since search engines were mere text matchers.
Re:Shirt Without Stripes by uniqueid on Tuesday April 21, 2020 @04:30AM (#22933097)
I’m uncomfortable with your use of “mere.” It might connote that what Google Search does now is an improvement.
Re:Shirt Without Stripes by wyattpeak on Tuesday April 21, 2020 @01:07PM (#22936882)
And that's exactly what I mean to connote, text matches make for appalling search.

I have the same reservations about Google as anyone, but rewriting history is never the right move. Moving beyond text matching was what made search truly useful.

Re:Shirt Without Stripes by uniqueid on Tuesday April 21, 2020 @02:31PM (#22937875)

    > Moving beyond text matching was 
    > what made search truly useful. 
PageRank is what made Google more useful than its competition. They had it since the beginning, and I like it.

What I don't like is to search for the band "Chrisma" and get results for "fruitcake sale!" because Google corrected my spelling to "Christmas", decided to look for related concepts, and then boost whichever result is the most mercantile.

Re:Shirt Without Stripes by wyattpeak on Tuesday April 21, 2020 @09:46PM (#22941212)
> They had it since the beginning

Yes, they have. Read what I said again, I don't dispute this. What Google does now is an improvement over text matchers. I never claimed that what they do now is better than Google circa 2000 (though I don't care to register an opinion either way on that).

Whatever they've done since, their product remains better than text matchers. Mercantile search is better than terrible search.

Re:Shirt Without Stripes by liveoneggs on Saturday April 25, 2020 @05:18PM (#22981594)
now they are product matchers
Re:Shirt Without Stripes by animalCrax0rz on Monday April 20, 2020 @05:37PM (#22928881)
Google is in the business of producing quick results to sell ads while keeping the cost low. If anyone does better, they will likely do it using a costlier algorithm, and if there is more margin in reselling tangible products than ads then Amazon is incentivized to use costlier algorithms that are more accurate. But I think the margin in both cases does not justify the use of algorithms that may be more accurate in some small percentage of cases but a lot more costly.
Re:Shirt Without Stripes by larl on Monday April 20, 2020 @06:11PM (#22929289)
animalCrax0rx is in the business of producing quick pithy posts to earn more karma while keeping understanding low and making claims that are not evidence based...

See how that works? That's not really what's going on. Sure, G. is incentivized to include pages quickly, but they are also incentivized to produce them accurately, and as the above poster indicates, this is quite a hard problem to solve generally.

A is also incentivized to sell items.

In many cases different algorithms will lead to quantifiably different results. The algorithm changes that work better for the measurement set will be kept and those changes which dont will be discarded. And both A and G do that within different constraints.

Re:Shirt Without Stripes by dEnigma on Tuesday April 21, 2020 @02:51AM (#22932627)
My cursor was hovering over the downvote arrow while reading your first sentence, before I realized what was going on in your post. Thank you for pleasantly surprising me!
Re:Shirt Without Stripes by animalCrax0rz on Monday April 20, 2020 @07:26PM (#22930031)
Google is aware of this problem in their search approach. It's a business problem, not a technical one. You're saying the same thing in suggesting they base their decisions on some measurement set. If solving the problem adds complexity, which it certainly will, and there is not enough improvement in accuracy for the majority of cases in their measurement set, why bother? You sound like the kind of person that attacks people for their opinion. So weird, dude.
Re:Shirt Without Stripes by nailer on Monday April 20, 2020 @07:55PM (#22930303)
Your ad hominem on the other poster is unnecessary, violates the HN guidelines and not an apt comparison.

Pointing out the obvious: Google is an advertising company. If the cost of producing an accurate result outweighs the advertising income on a given term, there is no incentive for Google to produce better results.

Re:Shirt Without Stripes by jefftk on Monday April 20, 2020 @08:54PM (#22930692)
This would predict that a query that has no advertising income will return no results, which is clearly not the case.

Having a search engine that people go to whenever they want to search for things is incredibly valuable, because they will come to you when they want to buy things and you can sell ads. But unless you consistently give the best results for all queries, people will go whenever does. It's worth investing strongly in all queries, not just highly monetizable ones.

(Disclosure: I work for Google, speaking only for myself)

Re:Shirt Without Stripes by 6510 on Monday April 20, 2020 @10:06PM (#22931146)
> This would predict that a query that has no advertising income will return no results, which is clearly not the case.

I was about to say there are no such queries but then I remembered having to type a captcha for seemingly automated queries. The captcha page has no results on it obviously. This is because automated queries do not produce advertising revenue. You have to buy them.

I've typed an insane number of queries since the beginning. A decade ago I use to be able to find truly exotic articles, I could find every obscure blog posting on every blog with 3 readers and I was pretty sure google delivered all of it. The tiny communities that came with the supper niche topics rarely produced a link I didn't already find. If they did it was new and I didn't google for a while.

Today google feels like it is a pre-ordered list from which it removes the least matching articles. Only if the match is truly shit will it be moved slightly down the page. The most convincing in this is typing first name + last name queries in imagines and getting celeberties who only have the first or the last name.

People wont go, it has to get much worse before they do.

edit:

With humans an pets a good slap over the head or a firm NO! will usually do the trick.

Re:Shirt Without Stripes by jefftk on Monday April 20, 2020 @10:28PM (#22931225)
> I was about to say there are no such queries

There are very clearly many queries with no advertising revenue, because there are many queries that show no ads. Trying some searches off the top of my head that I expected wouldn't have ads, I don't get any ads on [cabbage], [who is the president], [3+5], or [why is the sky blue]. On the other hand, if I search for a highly commercial query like [mesothelioma] the first four results are ads.

> A decade ago I use to be able to find truly exotic articles, I could find every obscure blog posting on every blog with 3 readers

My model of what happened is that SEO got a lot better. When Google first came out it was amazing because Page Rank was able to identify implicit ranking information in pages. Once it's valuable to have lots of backlinks, though, this gets heavily gamed. Staying ahead of efforts to game the algorithm is really hard, and I think a lot of times people's experience of a better search engine comes from a time when SEO was much less sophisticated.

> The most convincing in this is typing first name + last name queries in imagines and getting celebrities who only have the first or the last name.

This hasn't been my experience, so I tried an image search for [tom cruise], curious if I would get other Toms. The first 45 responses were all of the celebrity, and image 46 was of Glen Powell in https://helenair.com/people/tom-cruise-helps-glen-powell-lea... which is a different kind of mistake. Do you remember what query you were seeing this on?

Re:Shirt Without Stripes by alister on Monday April 20, 2020 @11:38PM (#22931609)
> This hasn't been my experience, so I tried an image search for [tom cruise]

I believe what he means is that searching for first name + last name of someone who isn’t a celebrity gets you celebrities who match either the first name or last name.

Searching for Tim Neeson gets you a wall of photos of Liam Neeson: https://www.google.com/search?q=tim+neeson&tbm=isch

Searching for Tim Cruise blankets you with pictures of Tom Cruise, but it at least says “Showing results for tom cruise“ so you know it did an autocorrect. When I tried other first names + Cruise, the effect is less pronounced than with the Neeson example. Maybe it’s because cruise is a more common name as well as an English word.

Re:Shirt Without Stripes by spunker540 on Monday April 20, 2020 @11:52PM (#22931682)
Thanks for clearly articulating what many people on HN seem to fail to grasp. It’s not that Google got worse over the years at surfacing the obscure content they used to so easily find. That obscure content has gotten completely buried under the mountain of content being published every day, and the cat and mouse game of SEO has evolved so rapidly that the problem space of generalized search is so much harder these days than it was 10-15 years ago. Not to mention the broader user base that they have to serve as well.
Re:Shirt Without Stripes by 6510 on Tuesday April 21, 2020 @01:46AM (#22932347)
It is the same thing! The mistake is obvious. For-profit content is prioritized. Google is the driving force behind the for-profit internet but sadly for google: you cant do organic ranking on commerce. ahhhh The index is now a [very limited] snapshot of the glory days.

You don't have to bother creating anything new unless you have something to sell and are willing to invest (big).

Facebook is actually a pretty pathetic implementation where we can still find content created by normal people. If people made traditional websites in stead of facebook groups and facebook pages NO ONE would be able to find it.

We've witnessed the great obliteration of what was once a nice place and now we have to hear google was not to blame?? The death by a thousand cuts is actually well documented.

We tell you what your site must look like or we'll gut it:

https://en.wikipedia.org/wiki/Google_Penguin Google Penguin is a codename[1] for a Google algorithm update that was first announced on April 24, 2012. The update was aimed at decreasing search engine rankings of websites that violate Google's Webmaster Guidelines[2]

There, this is what the entire internet must look like. We went from indexing to engineering here.

https://support.google.com/webmasters/answer/96569?hl=en

  rel="ugc"  
We recommend marking user-generated content (UGC) links, such as comments and forum posts, as ugc.

If you want to recognize and reward trustworthy contributors, you might remove this attribute from links posted by members or users who have consistently made high-quality contributions over time. Read more about avoiding comment spam.

Before this those elaborately contributing got actual credit for it. Do you think you got a choice in it? Google clearly demands you ban credit for comments. OR ELSE!

  rel="nofollow"  
Use the nofollow value when other values don't apply, and you'd rather Google not associate your site with, or crawl the linked page from, your site. (For links within your own site, use robots.txt, as described below.)

Woah association! How did we go from linking-to to association? It was important enough for readers but be careful to hide it from google. Such little unimportant websites simply shouldn't exist in our index. We command you to help keep our index clean of such filth!

Then the magical: We wont actually tell you what is wrong with your website! Ferengi Rule of Acquisition 39 "Don't tell customers more than they need to know." Get a budget and hire someone to do SEO. Deal with it, we don't care. No, you don't have any feedback.

Re:Shirt Without Stripes by 6510 on Wednesday April 22, 2020 @01:55AM (#22942685)
> There are very clearly many queries with no advertising revenue, because there are many queries that show no ads.

Queries without ads do produce revenue. They are an essential part of the formula.

Think of people standing around in bars. We cant argue that just standing there doesn't produce revenue.

The flowers on the table in a restaurant produce revenue.

Free parking produces revenue.

If queries without adds didn't produce revenue they wouldn't exist. More often enough it doesn't even take an extra query, the adds will sit behind the links.

Re:Shirt Without Stripes by jefftk on Wednesday April 22, 2020 @08:13AM (#22944689)
I don't think we disagree? Above I wrote: "Having a search engine that people go to whenever they want to search for things is incredibly valuable, because they will come to you when they want to buy things and you can sell ads."
Re:Shirt Without Stripes by nailer on Tuesday April 21, 2020 @09:59AM (#22935028)
> This would predict that a query that has no advertising income will return no results, which is clearly not the case.

No, it would predict that a query that has no advertising income will poor results. You can determine on your own whether that is the case.