I do zero analytics on any of my web traffic, so this is just a guess, but I should be surprised if many people come to my sites via Google.

I am just a bloke with a Raspberry Pi running a website and two blogs.

I suspect that, of whatever traffic I get, most comes from links I (and others) post here, and from RSS subscribers.

Similarly, I don't think that I found any of the blogs or sites that I read regularly via a search engine.

Word of mouth (webrings, people posting about stuff they've enjoyed / their own work etc.) ftw.

@neil I think i found you via a post on the Orange Site, threw the RSS feed into Newsboat

@kevin Yes, exactly that kind of thing!

(And I can only apologise...)

@neil Do you know volume of traffic in gigabytes, at least?
@autiomaa No. I don't do anything like that.
@neil If google is no longer a search engine people will use another search engine. Every wrong answer drives traffic away. I can't imagine anyone will want to use this.

@prism

> If google is no longer a search engine people will use another search engine.

I suspect that this could be true for people who actively want a search engine.

I don't know how many people just want an answer.

I agree though that, if the proffered answer is wrong too often, that it will drive (some) people away.

@neil If they just wanted "the answer" they would get it from the summary of the site without clicking onto the article anyway. Not great for you but I don't see how it makes the situation worse. The AI is just removing the pesky source of knowledge outside of the googlesphere. Since google does not own all knowledge, there's no way that the result they'll get will be more reliable than searching the web. Even the AI bros will use perplexity.

@neil @prism my wife complained about Google presenting an AI answer whenever she searches. Which is a potential problem because her university have rules about AI use. I said "so just use a different search engine" and got back "what do you mean? I thought Google was the only one".

She is not super into computers - she uses them as a tool, rather than as an interest, but she is not a luddite by any means and learnt how to code a few years ago.

So that, is an insight into "normal people".

@neil @prism IMHO, if you expect a large migration away from Google due to enshitification you are dreaming - the vast majority of people have probably never even thought about whether an alternative exists.

Much like new tech (EVs, heat pumps, etc.) people won't switch to an alternative unless they know people who already use it. And this is true even though the cost of trying an alternative search engine is basically zero.

@neil @prism I guess the potential of an "inferior" search engine missing results that you would have found, had you been using Google, might be considered a cost?

Much like other new tech, people only consider the cost of switching rather than the cost of not switching (e.g. drivers are quick to point out all the things that their ICE car can do which an EV can't, but ignore the things that their ICE can't do but an EV can)

@neil @prism there seems to be little consideration given to how the pervasiveness of AI is a real problem in some contexts. E.g. schools often need to allow people to use MS Office in exams, but having Copilot available to basically do the exam for the student is an obvious no-no. Careful whitelisting/blacklisting on the web filter is required (no help from MS or the exam boards on this one... And yes, this was a real world incident I am aware of!)
@steve @neil You're framing a success case as a failure. She identified that google's AI results are no longer fit for her purpose, so you suggested she switch search engines. She might not have actually decided to do it, but that's only because she can ignore the AI results and use the actual real information. If it's just AI results, that forces the issue.
@prism @neil my point was that although she identified an issue, she was unaware of any solution, so was stuck with Google. In this case, I was able to advise, but the vast vast majority of people do not know anyone more knowledgeable about this stuff than themselves.
@steve @neil The vast majority of people don't have a tech-literate friend/child/cousin/former roommate who can suggest to solve their google problem with not-Google? Maybe I'm giving people too much credit but that doesn't seem like a huge hurtle to me.
@prism @neil I think you are out of touch with reality
Jacob Aron (@jjaron.bsky.social)

Ok I've designed an LLM prompt to make Google AI just give you a list of links. It's longer than you might think because it kept ignoring the instructions. This is efficient. This is The Futureโ„ข

Bluesky Social
@steve @neil @prism My littlun was taught about DuckDuckGo at school - I've no way to know whether that's one teacher or the norm, but it does give me a little bit of hope for the future (the catch being, people of littluns age have to get to adulthood before that information is invalidated)

@steve FWIW, until sheโ€™s ready to ditch Google ๐Ÿคž at some point, you can set up a โ€œcustom search engineโ€ in her browser that invokes the param to show the web view for results rather than the view with the โ€œAIโ€ (๐Ÿ™„) summary โ€“ see my linked toot ๐Ÿ‘‡

https://infosec.exchange/@itgrrl/114495075173286748

it may have the unfortunate side-effect of delaying her abandonment (or possibly open her eyes to other possibilities), but it will at least somewhat dis-enshittify her use of Google in the meantime

itgrrl :donor: (@[email protected])

@paco @[email protected] @[email protected] a little OT but FWIW the magic search URL param to prevent the AI slop summary showing on the Google search result page is (currently) &udm=14 - it corresponds to โ€œshow me the web view instead of the default viewโ€ you can add that param to your custom โ€œsearch engine(s)โ€ defined in your browser settings & use a keyboard shortcut to invoke a slop-free^ search from the omnibox ๐Ÿ’โ€โ™€๏ธ       ^ well, โ€œslop-reducedโ€ - you still get the slop in the actual search results ๐Ÿ˜• #udm14

Infosec Exchange

@neil I've stayed away from social media for a long time and was in two minds about joining mastodon. I'm really very glad that I did, and not just for the networking for my website. I've met a lot of interesting people here.

Not quite the same thing, but It brings to mind the online gaming community I was a part of when I first moved to Germany. I think without them those first months would have been very lonely. Still meet up with some of them irl. And there are some here I'd like to meet.

@GoingDownWithSundial

Yes! Community is a key part of why I enjoy being online.

I love the fact that I get to chat with, learn from, and enjoy the writings and other creations of, so many people, wherever and whomever they might be.

@neil curiosity is such a beautiful thing.

@GoingDownWithSundial

The Internet, as a tool for letting me explore and expand my curiosity through interacting with others, played and continues to play a massive part in me being who I am today. Absolutely.

@neil and I agree about the word of mouth phenomenon :)

@neil
Same. There's a few blogs I've found from looking for lists of RSS feeds but most of the listed ones are corporate. I found more interesting blogs through word of mouth here...

...word of fingers? Although I suppose if you dictate to your device it would still be word of mouth...

@feff

"Word of fingers" is rather apt. I like it :)

@neil In the early days of blogging, perusing referral logs was a big deal. I was once โ€œsummonedโ€ to join a top-secret web forum by a few people clicking on a link to my blog, with the (correct) assumption that I would notice the sudden spike in traffic from a weird address.

I wouldnโ€™t even know where to look to find referral logs now, and I presume theyโ€™re useless anyway.

@nic

> I was once โ€œsummonedโ€ to join a top-secret web forum by a few people clicking on a link to my blog, with the (correct) assumption that I would notice the sudden spike in traffic from a weird address.

Nice!

@nic @neil ah, that takes me back โ€“ I used to occasionally send messages to website admins by adding it as a param to one of their site URLs in the hope they might see it in their logs ( e.g. http[s]://domain/page.html?comment=hi - love your site, keep up the good work! :-) )

it was a simpler time, with much lower volume of crawler & bot traffic in the site logsโ€ฆ ๐Ÿ™ƒ

@nic @neil oh, not useless (depending on your definition) โ€“ the SEO folks are still all over them in pursuit of โ€œdriving trafficโ€ and โ€œoptimising click-throughsโ€, etc. ๐Ÿคข
@neil If you have any interest in getting analytics but do not want to add trackers to your website (which I totally understand), you can do some web server log analysis, e.g. with Matomo Log Analytics: https://github.com/matomo-org/matomo-log-analytics
GitHub - matomo-org/matomo-log-analytics: Import any kind of server logs in Matomo for powerful log analytics. Universal log file parsing and reporting.

Import any kind of server logs in Matomo for powerful log analytics. Universal log file parsing and reporting. - matomo-org/matomo-log-analytics

GitHub
@neil This is the "modern" replacement of AWStats. Both are on the #Debian repositories.

@gjherbiet

> If you have any interest in getting analytics

None whatsoever (https://neilzone.co.uk/2024/01/why-i-dont-care-about-websiteblog-stats/), but thank you!

Why I don't care about website/blog stats

Every so often, someone says to me

@neil Anecdote: I arrived on one of your blog posts from a search engine (probably DuckDuckGo) about a week ago. I rember thinking "oh, it's Neil from Mastodon's blog". I was looking to see if it was possible to import events into a Radicale calendar (not directly, apparently). Since I already knew who you were, I had reasonable confidence that you'd done a decent investigation. Thanks for posting about it!
@semiprime Interesting - thanks!
@neil
We run different sorts of blogs but...
As at today, most of my traffic comes from SEs and most of that from Google.
Masto is the SM platform that I'm most active on but the traffic I get from here is now virtually nothing. I get more traffic from Bluesky where I have less than a 3rd of the followers I have here. Engagement with my toots here which promote my writing is dwindling. Elsewhere, there's been a great response to my piece that was published last week. Here, its felt like 'meh'.
@neil
I have been getting some quite exciting word of mouth referrals recently and invitations for collaborations but none of it from Masto.
Thats fine.
If I remain here long term, it wont be to drive traffic to the blog, though.

@Jaimieserotica @neil Mastodon doesn't send any referrer headers by default (it can be enabled after a recent release to enable it, but it is suggested only do it on bigger instances to help anonymise visitors)

https://shkspr.mobi/blog/2024/12/mastodon-now-sends-referer-headers-hurrah/

@ben @neil
I'm not a techie, but I can say that I've been getting stats on traffic from Mastodon since I joined here at the beginning of last September. A few months ago I was getting somewhat more fedi traffic than I am now. I don't know what any of that means in the light of your comment, though.
@Jaimieserotica ahh, ok, a fall off of previous traffic will not be related to this, but more that, some of the unattributed traffic may also be coming from smaller Mastodon instances.

@ben oh interesting, I didn't know about that. I suppose I'm not terribly bothered about where the traffic comes from on my website, since I hadn't even thought about this :)

Thank you for the heads-up. Might take a proper look at how it works later.

@Jaimieserotica

> Masto is the SM platform that I'm most active on but the traffic I get from here is now virtually nothing.

Ooh, interesting.

I'm sorry to hear that, because you are a lot of fun here.

And I think that, early/earlier on, you were getting a fair amount of fedi-originated traffic? So I wonder what (if anything) has changed.

But your writing is wonderful and deserves more attention!

@neil
Thanks Neil  
It's not that overall traffic is down (its massively up over last year), just fedi traffic... It is what it is, I guess.

@Jaimieserotica

I recall reading one of your posts, about your traffic stats (and, perhaps, goals in that direction), so I am glad that things are going in the right direction!

I could have some readers, loads of readers, or no readers, and I am blissful in my ignorance :)

@neil
You've got one, at least! 
@neil @Jaimieserotica itโ€™s how I discovered @Jaimieserotica and her wonderful writing. I hate the fact I have to maintain a Instagram account to keep in contact with models as more have moved away from photography and modelling forums.

@rbphotographic @Jaimieserotica

> more have moved away from photography and modelling forums.

I miss the days of individual forums (as opposed to centralised forums like "reddits") - they were a staple part of life online for me.

@rbphotographic @neil
Instagram... *shudders*
I don't think I've seen so many unsolicited dick-pics in my life as when I had an Insta account (both times)
It's one of the reasons I came to Mastodon in the first place. Folks are generally much more respectful here.

@Jaimieserotica

I have no insight to the topic other than that a well known blog author I know of tracked visitors to his blog using referer links and up until last summer Mastodon-instances never had that. Even now every Mastodon server admin has to manually set the option to send referer-traffic.

This makes Mastodon-traffic show up as "direct/organic" rather than as "Mastodon". However, if you use som other means of collecting this information this is of course irrelevant. I just wanted to add this tidbit of info because it's surprisingly unknown to many Mastodon admins that this setting's default mode makes us "invicible" to how site admins usually measure traffic sources.

@neil

@neil I watch my nginx logs because ai scrapers hit my raspi hard. They wanted to scrape my mastodon instance not any more.

@neil I do zero analytics on any of my web traffic, so this is just a guess, but I should be surprised if many people come to my sites.

Every so often I look at the server logs with awk,  but  it's hard to figure out what's noise and I don't really know what I'd do with the information if I did

@dan @neil I use awstats, which is really easy to set up and works for me; however, the project is now end of life, and the maintainers are recommending another project.

Web analytics for your own sites is more than just interesting: it can alert you that you are exposing resources you did not mean to expose.

https://github.com/eldy/awstats

GitHub - eldy/AWStats: AWStats Log Analyzer project (official sources)

AWStats Log Analyzer project (official sources). Contribute to eldy/AWStats development by creating an account on GitHub.

GitHub
@neil your web site popped up on a search I did yesterday - looking for info on why localsearch was hammering my cpu and disk since upgrading my (Linux) OS

@not_a_label Perhaps it comes up more than I realise then!

Lucky me :)

@neil it was a 'oooh... I know someone famous' moment when I saw it ๐Ÿ˜‚

@neil I have this, which just extracts from the nginx logs.

Apparently my top referrer is... Me. Google. Also me. Google. Me again. Fediverse radio webring, and @Edent

https://m0yng.uk/report/

Server Statistics