This author is in his own little bubble and doesn't understand the vast amount of blog-repost spam that google has to deal with. The way their algorithm most likely deals with this is a mixture of domain rank + tenure... how long has this copy of this article existed on this domain, and can we be sure this is the original copy? The author says the article was removed in 2006 (" [...] posts, were not accessible anymor…
If the originally indexed copy no longer exists, Google shouldn't down-rank a reposted version!
IME, it's not blackhats anymore causing the problem. It's (legitimate, but shady) marketing agencies and startups handling thousands of customers and with deep pockets to do SEO research.
I count those agencies as a variety of blackhat.
blackhat is as a blackhat does, its no difference how or why you screw users in someway by failing to be forthright and of candor. if you do it constituatively thats blackhat.
I've had this problem recently. I can craft a search for something just slightly obscure and specific that should, nonetheless, have had plenty of hits on the "old web", let alone now on the many-times-larger web. But "no pages found". Loosen up the search and it's nothing but Google-friendly blogspam that isn't remotely related to what I'm trying to find. I call bullshit.
Loosen it up? You mean google didn't automatically remove your keywords for you?
Heh, oh yeah, tons of that, usually the ones most relevant to narrowing the search beyond "everything on the Web". Thanks, Google.
So then I do the quotes thing, especially quoting phrases that 100% for sure must exist on some web pages, along with all my other keywords and pretty soon I'm at "no pages found". Pull back just a little, and it's page after page of entirely unrelated-to-what-I-want blogspam.
Thankyou for the link to your search guide - this looks tremendous! Especially when searching for a specific answer to something. I wonder what you think about discovery when you are looking for something unknown within certain parameters... Like, say you are looking for an "interesting film blog" - a search term like that will often lead to pages of "Top 10 Movie Blogs" lists that are all largely clickbait or not th…
I'm no gwern, but here's how I do discovery. To find a page on the internet, you need to hand a search engine something that can be reasonably expected to be on that page. So if you want to find an interesting film blog, you should not use that as a search term, because you'll find lists of blogs, not the blogs themselves. Rather, you should probably use the titles of interesting movies. There are many ways you could…
This is good advice - thankyou! I have definitely used this kind of approach before, it requires some creativity. It feels like there are possibly dozens of ways of approaching this - and obviously unlimited kinds of 'seeds' (as you say) for the search. I'm definitely looking for a guide that might encompass this kind of strategy.
> Anecdotally and perhaps unrelated - has anyone else noticed a decrease in the accuracy and general quality of Google search over the past 2-4 months Yes. Not just over the past 2-4 months, but over the past five years or so. It's become so bad that Google is no longer the most useful search engine for me.
It all started going downhill since Google's "Hummingbird" switch to be honest. Interviewing for Google, I actually brought this up with an engineer in the search team during the lunch. He said they haven't noticed any regressions. I said I figured that would be the case but I can definitely feel the difference as a daily user.
I noticed the same. I've wondered for years why it happened and sometimes when I'm frustrated I try to think about it. But I am not entirely sure that the degradation in search results for me happened only in the past 5 years. Maybe, but I'm not sure.
This has been been happening for years and getting increasingly worse. But it's not just old websites. I think it's certain types of content. Here is one specific example out of dozens I've seen. There is a short satirical rant "published" on Pastebin called The Java Way. Posted in 2015. Unfindable on Google. It was indexed and findable around the time it was posted. https://www.google.com/search?client=firefox-b-1-d…
Amazingly, three hours after your post, this hacker news comment is now showing up in the top 5 for the Google query. Wow.
One of the changes that made Google forget the old web is favoring https sites. This is a big benefit to new and commercial web sites because setting up SSL is still a burden for non-comercial publishers.
I feel like LetsEncrypt is a 5-minute burden. Unless I’m being naive - I’ve only used it in personal projects. Thoughts?
I have an old Linode with 4-5 personal sites running off a single Apache server. Until recently, it was running Ubuntu 10.04 LTS.
It took me a few days to safely upgrade to a new Ubuntu version with a new enough Python to successfully run letsencrypt, without also breaking the weird custom apache configuration rules that had accreted over the years.
I agree with you. Maybe the problem is that PageRank (or whatever they call it these days) has run its course. I mean, it supposed to gauge "what humans think is good", but it's failing miserably. It's indeed time for a more curated, artisanal, web.
PageRank is predicated on an assumption that most pages (and thus, most links) are created/curated by humans. This was true when it was invented, but appears to be less likely now. What gives me pause here is all the anecdotes in this thread about other engines getting results right. If the real answer is "PageRank has been successfully flooded by bots", then everyone would have bad results. What I suspect, off nearl…
You can see some evidence that suggests it when you search for a specific software or ebook to download.
Piracy is gone, but you will find hundreds of automatically generated credit card phishing sites full of Google Ads, sometimes promising pirated versions but serving a trojan, sometimes showing a credit card form. Some of them are on the first page, sometimes before legitimate websites.
While it's become impossible to browse the wider Web with Google, it's getting a bit easier elsewhere. A few helpful search engines: * https://millionshort.com/ * https://wiby.me/ * https://pinboard.in/search/ A recent movement to build personal Yahoo!-style directories: * https://href.cool/ (my own project) * https://indieseek.xyz/ * https://districts.neocities.org/ * https://the.dailywebthing.com/ The above resourc…
One of the changes that made Google forget the old web is favoring https sites. This is a big benefit to new and commercial web sites because setting up SSL is still a burden for non-comercial publishers.
I feel like LetsEncrypt is a 5-minute burden. Unless I’m being naive - I’ve only used it in personal projects. Thoughts?
Plenty of websites out there being kept up by volunteers without hardware access, the original owner having died or otherwise gone MIA. It's not always possible to add https, particularly when the site owner died 10 years ago and someone has an 'agreement' with the hosting provider to keep a website up as a memorial. No, I don't have any specific examples right now, but anecdotally I occasionally come across a website that's been in such a read-only form for as long as a decade due to family members being willing to continue paying the $15/year hosting fee, but not having the technical knowledge, passwords, or interest to fix problems. Sometimes there's evidence of a partial upgrade (the search engine stopped working due to a php upgrade), or a forum that has been converted to a static site entirely (the login buttons don't work either). In any of these cases, getting LE working is almost certainly more trouble than it's worth for whoever is currently paying the fees.