Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Search engines and SEO spam

twitter.com

Post reply on HN
Print view View original

541–550 of 555 posts

Re: Search engines and SEO spam

#541

Post by mrkramer »

denton-scratch wrote: post #518
mrkramer wrote: post #396

Earlier quoted context omitted.

I wasn't speaking about that kind of entrepreneurism but about making legal software and legal web services that solve problems and are useful. So many Russian hackers got arrested when they travelled somewhere outside Russia and now they are serving 10 or 20 year sentences in US jails.

> making legal software and legal web services SEO is perfectly legal. So is spamming, regrettably. I think Goo no longer cares about the quality of search results; they have other business priorities, so SEO works. Spam is another thing again; we still, after 30 years, don't have an agreed definition of spam. We still don't have a flawless spam filter - far from it. So it astonishes me how much email spam I get prom…

>SEO is perfectly legal. So is spamming, regrettably.

SEO is legal but spam is not at least not in US and many other jurisdictions.

>I think Goo no longer cares about the quality of search results; they have other business priorities, so SEO works.

Google cares about spam but there is so much data and information on the web that it is impossible to figure out what is spam and what is not. Another big problem is fake data and information that is also very hard to figure out. Generally Google prefers popularity over quality because it is easier to detect what is popular than what is of good quality.

>Spam is another thing again; we still, after 30 years, don't have an agreed definition of spam. We still don't have a flawless spam filter - far from it.

Definition of spam is unsolicited message. So if I get pharma emails in my email inbox that I didn't ask for it is considered spam. SMS messages are another example for example if I get SMS message promoting free coupons but I didn't ask for it then it is spam.

Considering how much spam Google saw in the last 20 years both on the web and on the Gmail they should have some decent machine learning/AI algorithms which could flag spam pretty efficiently.

Re: Search engines and SEO spam

#542

Post by jimkri »

nobbis wrote: post #222

10 years ago, the original engineer of Google's search engine told me what he now wanted was asynchronous, human-powered search with curated results, e.g. a Google-like interface, but queries cost $5 and take 15 minutes. Money's no object for him, so he wanted to outsource the filtering, ranking, and interpreting of results. Would be even more useful today (albeit a tiny TAM.)

I'm seeing something similar to this when it comes to Lead generation for sales teams. The market right now is full of tools where you search for companies/prospects that you want to target and you end up with a list that is either outdated or incorrect data.

One of the solutions is to hire people remotely that will build a list from searching and verifying that the information is correct. There are a ton of scaling issues with a person building the lists, and you will still run into errors, but it has been the best way to verify and have the most up-to-date data.

A problem I noticed is people are so use to getting tons of results when they use Google or another database site, so they expect the same with human-powered results. SO I think a lot of expectation setting and also just working with the TAM that understands the impact a curated list will provide over a outdated database. Both have their costs but for a higher impact the human curation will always win over automation. (I have an example of 3 different lead gen sites having a different city for 1 company, the AI on the websites mixed up the location, a Human would be able to catch that)

Re: Search engines and SEO spam

#543

Post by kyriakos »

wodenokoto wrote: post #243
frenchyatwork wrote: post #145

Earlier quoted context omitted.

I think one of the fundamental things that make search work well about 1-2 decades ago was that web sites would link to each other, and that those links could vaguely correlate with reputation. There were link spammers, but there was actually a some decent organic content as well. What's happened since then is that almost all the normal "people linking to things they like" has gone behind walled gardens (chiefly Face…

Why has blogs and articles stopped linking to things? I'm reading a restaurant review site, and they won't link to the restaurant. The chef name is a link to a list of all articles tagged with the chefs name, rather a wikipedia link or something useful that can tell me who that person is.

Noticed this with online newspapers too to the extend they are reporting about a website or product and don't include a URL to it.

Re: Search engines and SEO spam

#544

Post by GarlicToum »

jimkri wrote: post #539
GarlicToum wrote: post #406

Earlier quoted context omitted.

https://garlictoum.com/

This is great. I've been juicing a lot lately and every site you go to is full of adds and useless content and you have to click `Jump To Recipe`. 1 thing that would be helpful is to add in the ingredient amounts.

Yeah, that was part of the long term monetization plan. Paid for user accounts ($3ish/month) that allow people to clip recipes, view previews, discuss, modify, and share.

Never quite got around to it unfortunately, as I wanted more users before new features.

Re: Search engines and SEO spam

#545

Post by marginalia_nu »

ffhhj wrote: post #245
marginalia_nu wrote: post #113

Earlier quoted context omitted.

Another big factor in what I do is prioritizing the opinions of certain indieweb sites when ranking domains, basically a segment of the graph consisting of humans with a particular dislike for seo spam. This makes ranking manipulation much less effective.

Good to see you in this thread! I just added Marginalia to the recommended search engines in a new search tool I'm building, to get programming answers faster. The search assistant builds queries for specific sites with "site:targetsite.com , programming question" (that comma is not a typo). When doing a query like that I get no results but these warnings: /!\ The term "," contains characters that are not currently s…

I've made the changes to the syntax (as well as fixed a bug that broke site:-queries), although I think it's probably not a great search engine for searching big mainstream websites like SO, the crawler is designed to de-prioritize them over blogs and what have you. Disregard what I said in the other comment about the domain names, use the full domain name.

e.g. this works fine: https://search.marginalia.nu/search?query=site%3Amemex.margi...

If you just do a site:-query without any search terms, you'll get a dump outlining the crawling status of the site. Like this:

https://search.marginalia.nu/search?query=site%3Astackoverfl...

You can see it's crawled 80 pages but not gone much farther since SO ranked so low.

Re: Search engines and SEO spam

#546

Post by compiler-guy »

siva7 wrote: post #323
techdragon wrote: post #63

Earlier quoted context omitted.

I sure wish I had problems that were totally unsolvable, they are so easy to measure progress on. /sarcasm I think it’s more likely that because they are just building hundreds of tiny tweak experiments and it’s someone else who desides what to build and if it even worked. Search quality is such a meta-problem that it goes beyond any real hope of simply working on it in anything beyond piecemeal trial and error fashi…

What is a danish?

A breakfast pastry something like a donut.

https://en.wikipedia.org/wiki/Danish_pastry

Re: Search engines and SEO spam

#547

Post by GarlicToum »

ajmurmann wrote: post #487
GarlicToum wrote: post #406

Earlier quoted context omitted.

https://garlictoum.com/

This is really cool! Bookmarked! One piece of feedback: It seems to somewhat liberally fuzz the term, doesn't tell me about it and not toggle it in the UI. I searched for "Natto" and got some great results at the top (Spicy Kimchi Natto!). However, a few results down the recipes start including "Nattu" which seems to be a Indian chicken dish and then quickly have nothing to do with the term at all and give me stuff l…

Yeah, that's the typo detection in search. The key thing is that "Natto" is prioritized, so that part is working correctly!

Tweaking search can be a whole full time thing. I'll write it down and maybe look into if I can tweak the search so if there are exact matches to not show alternatives.

Re: Search engines and SEO spam

#548

Post by heavyset_go »

type0 wrote: post #532
heavyset_go wrote: post #479

Earlier quoted context omitted.

Google used to prioritize blogs and original content like forum posts in search results, but they don't anymore.

What do they prioritize now, "reputable" news organizations enrolled in Trusted News Initiative?

Blog spam and pages filled with AdSense ads.

Re: Search engines and SEO spam

#549

Post by bobthebuilders »

ceejayoz wrote: post #345
nradov wrote: post #311

Earlier quoted context omitted.

If companies don't send him review units then he just buys them retail. He has already done this for many products.

Yes, I'm aware. That's less money in his pocket, and less ability to have the review be available on or before the product launch. There's still some conflict of interest, even if it's lessened. Only purchasing review units at retail would remove this conflict.

Incorrect. He can then return them.

Re: Search engines and SEO spam

#550

Post by B-Con »

mrkramer wrote: post #461
B-Con wrote: post #444

Earlier quoted context omitted.

That would have allowed people to artificially chase rankings even faster and more efficiently. It makes the problem worse, not better.

How is transparency worse than smoke screen that we have today? For example healthy and good websites could rank according to good content, good optimization, variety of multimedia content, decent design and UI etc. You can't have too much of good things and qualities. That would be something like writing a too good book or making a too good product.

Because the rank algorithms are subjective heuristics, not absolute metrics. All rank algorithms always have been. It started with the link metrics, then people started gaming that. It's been a signal/noise war ever since.

It's also dangerous to ask for the exact criteria because they are ever changing. Google et al don't want to be prescriptive about what a good site is, they want to recognize what a good one is. You make a good one, they'll figure out how to recognize it.

They can't sit down and publish "The Definitive Guide to a Good Website". That's just not their role and it will be out of date before it's published.

Post reply on HN