Archiveteam did a full site crawl[1] when Anandtech announced they were stopping. You can browse the warc.gz files like a regular web page using https://replayweb.page
Alternatively you could use solrwayback[2] to index and browse the warc files.
You can still bring Wikipedia offline on a laptop (and mobile phone, for some of the larger ones), it is just that you'd need around 100GB instead. There is even a library[0] you can use to do your own wikipedia viewer.
I really like having the mobile version for fast searches, often faster than online. Useful for example while hiking or other out
-of-network places. Even some big stores have zero signal inside and sometimes I want to look up things. You can also get almost any Stack Exchange site.
If you live in a low, but not zero, bandwidth environment... since the rise of LLMs it's now cheaper to have the models do your dirty work. Before, you might have to search through pages of results, load MBs of data and still not find the answer. Offloading that to a data center and getting a few hundred kB back is convenient. Coupled with Kiwix and you can do quite a lot with a lousy internet connection.
This is not great. Decades of hardware reviews all the way back to the first GPUs becoming less accessible. Why would anyone consider taking it down when there was so much content that could be hosted with little effort while still making some ad revenue? Anandtech articles were still at the top of many Google searches.
Static hosting is essentially free. Add ads and it's a cash machine, not the opposite. Either there's something that they didn't tell us or they're incompetent.
The only thing needed is a staticization of their website, which any CMS they had could very easily be set up to do. Look at the archives of NYT, they're barebone pages that preserve the content without any dynamic areas.
Server load at that scale is measured in single-digit dollars per month, and bandwidth _might_ require two pipes with enough images being loaded. Multiply by 20 for replication and latency issues, and you're still only talking $200/month.
As a ballpark, even bad ad impressions bring in $10/mille, maybe $1/mille for something unintrusive. Does Anandtech get more than 200k impressions per month? If we're talking about a "hundreds of thousands of requests per minute" scale then I would certainly hope so.
It's most definitely free - cloudflare won't even ask you for the business plan (as long as their CDN's are caching properly). I had a site serving 90tb/month on a free plan (99.87% of the traffic was cached).
Simultaneously, the fact that anyone with any supposed business experience gifted that priceless level of ranking would decide to shut down the business is insane.
Like, the fact that someone is making money off of MySpace.com right now and Anandtech couldn’t swing it makes zero logical sense. To me it feels like they tried nothing and were all out of ideas.
Any time private equity does something stupid or short-sighted, remember this:
Private equity firms - or, at least, the ones that people complain about - don't own their own capital. They have to rent it from somewhere else, and those people get paid first.
The PE firm only really gets paid for their expertise when they make their hurdle. Ergo, PE is incentivized to make terribly short-sighted business decisions, because those are the ones that will bring in the money to make their hurdle. They get caught in a loop of buying and gutting otherwise productive businesses.
This capital structure made sense back when PE was a tiny part of the economy that bought and modernized small businesses, but now PE is more akin to a failing empire; with an entitled aristocratic class that will shiv any leader that tries to change the structure to be more sustainable. They are spending $2000 on candles and the candles have knives.
They were already bought out. By Purch in 2014, and then Future [0] in 2018.
This reads like some shit-for-brains VP at the acquirer couldn’t figure out how to make it work, so they’re putting it on ice.
The most destructive part of acquisitions seems to be the acquirer assigning a low-talent leader to the new acquisition, who then by virtue of no experience runs it into the ground, then blames its failure on the company itself.
Sadly, I’ve had to resort to ChatGPT for stuff like this. Their internal archive will last longer.
Of course, now there’s less and less of a way to see if it’s hallucinating.
Going through this with an old hardware project where ChatGPT says _____ vulnerability exists in their early units, but zeeero references, even on archive.org
anandtech.com now redirects to the forums instead its front page of articles. Here is what the website previously tweeted about its future a year ago after winding down operations.[1]
"And while the AnandTech staff is riding off into the sunset, I am happy to report that the site itself won’t be going anywhere for a while. Our publisher, Future PLC, will be keeping the AnandTech website and its many articles live indefinitely. So that all of the content we’ve created over the years remains accessible and citable. Even without new articles to add to the collection, I expect that many of the things we’ve written over the past couple of decades will remain relevant for years to come – and remain accessible just as long."
> AnandTech will stay online so readers can continue to access articles from our archive, and the forums will remain active to serve our community. Our sister site Tom's Hardware, will also continue to publish all the latest news, reviews and more from the PC world. Thank you all
Future PLC seems to be gradually shutting down all of its activities, one publication or site at a time.
It's a shame because many of them had been publishing for decades. Were they really completely unsustainable? The ads in magazines like Computer Music and Future Music were actually interesting and relevant, unlike typical garbage web ads.
I think it's more they've shifted away from their original focus as a specialist computer publisher into a more general publisher. I realised a couple of months ago they publish the TV Times nowadays, and also stuff like Country Life and Home and Gardens. Tech stuff is just another line on the balance sheet now.
I doubt if magazines (and websites etc.) in general are doing great, but for obvious reasons the more techy stuff is probably going to be a bit more vulnerable, particularly in print.
The fact that Future Publishing - once the scrappy 90s flagship of new tech journalism - now owns Country Life is something I will always find hilarious.
Chips and Cheese seems like a basically fine replacement for Anandtech. Things change, and the internet has gotten worse since then, but specifically chip benchmarking doesn’t seem too bad.
I think Chips and Cheese is more like a fine replacement for realworldtech.com sans the toxic and highly educational and entertaining forums. Anandtech was much more accessible to the general tech public, but also more commercial and thus hit and miss on the content (no judgement intended, gotta eat).
> I expect that many of the things we’ve written over the past couple of decades will remain relevant for years to come – and remain accessible just as long.
I honestly don't understand why companies do this. There's still SO MUCH traffic from Google for these things.
I've routinely went to AnandTech EXCLUSIVELY from Google. This means that their "news" and new content is of little relevance to me, as it's usually something from a few years prior that I'm reading over there. Yet somehow, "have to publish every day"?!
Part of me thinks that this is related to the inefficiencies of all these CMS, where it costs too much to run the site compared to the revenue from the ads.
Or is there another reason?
Frankly, as an nginx "practitioner", all of these sites could basically be cached and served from a single $50/mo server from Hetzner, Online or OVH. Aren't they're getting far more in ad revenue than that? How does it make any sense to close the shop when you've got such a treasure trove that you could continue milking easily for at least like half a decade?
Those who are currently publishing will tell you otherwise; traffic from Google has dropped off a cliff. There’s a little bit of long tail, but nothing like the volumes the tech media built its businesses around.
If I look at another tech site that stopped publishing, but where the content going back to year 2000 is still online [1], there's very little human traffic, vast majority is bot traffic.
Archiveteam did a full site crawl[1] when Anandtech announced they were stopping. You can browse the warc.gz files like a regular web page using https://replayweb.page
Alternatively you could use solrwayback[2] to index and browse the warc files.
1: https://archive.fart.website/archivebot/viewer/job/202409012...
2: https://github.com/netarchivesuite/solrwayback
Also Kiwix[1] is an excellent app for browsing websites offline. You can use warc2zim[2] to convert the WARC files to ZIM files for use with Kiwix.
I was pleasantly surprised to find that the DWDS (digital dictionary of the German language) app is actually Kiwix!
[1]: https://www.kiwix.org/
[2]: https://github.com/openzim/warc2zim
> Kiwix
... I haven't heard this name in 15 years probably. Back then you could bring Wikipedia offline on a laptop, it was only around 20-25 GB.
You can still bring Wikipedia offline on a laptop (and mobile phone, for some of the larger ones), it is just that you'd need around 100GB instead. There is even a library[0] you can use to do your own wikipedia viewer.
[0] https://github.com/openzim/libzim
Yes and much less than 100 if you can do without images
I really like having the mobile version for fast searches, often faster than online. Useful for example while hiking or other out -of-network places. Even some big stores have zero signal inside and sometimes I want to look up things. You can also get almost any Stack Exchange site.
If you live in a low, but not zero, bandwidth environment... since the rise of LLMs it's now cheaper to have the models do your dirty work. Before, you might have to search through pages of results, load MBs of data and still not find the answer. Offloading that to a data center and getting a few hundred kB back is convenient. Coupled with Kiwix and you can do quite a lot with a lousy internet connection.
This is a bit tangential, but is there a good way to archive Discourse forums and turn them into regular websites? Anyone have experience to share?
This is not great. Decades of hardware reviews all the way back to the first GPUs becoming less accessible. Why would anyone consider taking it down when there was so much content that could be hosted with little effort while still making some ad revenue? Anandtech articles were still at the top of many Google searches.
Being near the top of Google means you get an incredible amount of bot and AI traffic which costs a surprisingly large amount of money.
Static hosting is essentially free. Add ads and it's a cash machine, not the opposite. Either there's something that they didn't tell us or they're incompetent.
The only thing needed is a staticization of their website, which any CMS they had could very easily be set up to do. Look at the archives of NYT, they're barebone pages that preserve the content without any dynamic areas.
It is certainly not free at "hundreds of thousands of requests per minute" scale.
It's not free, but it's not expensive.
Server load at that scale is measured in single-digit dollars per month, and bandwidth _might_ require two pipes with enough images being loaded. Multiply by 20 for replication and latency issues, and you're still only talking $200/month.
As a ballpark, even bad ad impressions bring in $10/mille, maybe $1/mille for something unintrusive. Does Anandtech get more than 200k impressions per month? If we're talking about a "hundreds of thousands of requests per minute" scale then I would certainly hope so.
That’s the irony: forum traffic has shit cpm, but that’s what they’re keeping alive (for now).
Users coming direct from organic search to an article tends to have the best: they’re more likely to be buyers and/or find the ads interesting.
Main downside with ads on tech sites is their users block ads a lot.
Why would anyone take down a site like this? The content is done, the site is page ranked to the top, it’s passive money.
Because their revenue department doesn’t talk to their content department.
Embarrassingly parallelisable around the globe and embarrassingly easy technically to point DNS to a CDN.
Even if you have a server doing something 100k rpm isn't insane amount of traffic to handle. It requires a relatively modest server.
It's most definitely free - cloudflare won't even ask you for the business plan (as long as their CDN's are caching properly). I had a site serving 90tb/month on a free plan (99.87% of the traffic was cached).
What if you care about your users privacy and security and don't use Cloudflare?
Wouldn’t Cloudflare’s free CDN handle that? That kind of traffic is dirt cheap.
Not free after a certain point then gets expensive fast as you move out to CDNs.
It’s only expensive if you’re paying the AWS, GCP tax. They’re stupid expensive for simple file hosting, and for no good reason.
That is only true if you not use their cdns
Simultaneously, the fact that anyone with any supposed business experience gifted that priceless level of ranking would decide to shut down the business is insane.
Like, the fact that someone is making money off of MySpace.com right now and Anandtech couldn’t swing it makes zero logical sense. To me it feels like they tried nothing and were all out of ideas.
But that’s private equity for you.
I’m still upset Rakuten shutdown fatwallet. Those forums were the best.
https://en.m.wikipedia.org/wiki/FatWallet
Any time private equity does something stupid or short-sighted, remember this:
Private equity firms - or, at least, the ones that people complain about - don't own their own capital. They have to rent it from somewhere else, and those people get paid first.
The PE firm only really gets paid for their expertise when they make their hurdle. Ergo, PE is incentivized to make terribly short-sighted business decisions, because those are the ones that will bring in the money to make their hurdle. They get caught in a loop of buying and gutting otherwise productive businesses.
This capital structure made sense back when PE was a tiny part of the economy that bought and modernized small businesses, but now PE is more akin to a failing empire; with an entitled aristocratic class that will shiv any leader that tries to change the structure to be more sustainable. They are spending $2000 on candles and the candles have knives.
I wonder if Anubis or something could have been a viable solution but I'm sure they thought of that
Maybe they age hoping for a buyout.
They were already bought out. By Purch in 2014, and then Future [0] in 2018.
This reads like some shit-for-brains VP at the acquirer couldn’t figure out how to make it work, so they’re putting it on ice.
The most destructive part of acquisitions seems to be the acquirer assigning a low-talent leader to the new acquisition, who then by virtue of no experience runs it into the ground, then blames its failure on the company itself.
[0] https://en.m.wikipedia.org/wiki/Future_plc
Sadly, I’ve had to resort to ChatGPT for stuff like this. Their internal archive will last longer.
Of course, now there’s less and less of a way to see if it’s hallucinating.
Going through this with an old hardware project where ChatGPT says _____ vulnerability exists in their early units, but zeeero references, even on archive.org
anandtech.com now redirects to the forums instead its front page of articles. Here is what the website previously tweeted about its future a year ago after winding down operations.[1]
Originally heard this via https://x.com/System360Cheese/status/1951501044875477254.
The latest indexed frontpage in the Internet Archive is from July 28: https://web.archive.org/web/20250728143805/https://www.anand....
The original farewell article, which is now only readable through the IA: https://web.archive.org/web/20250726035557/https://www.anand.... One paragraph reads:
"And while the AnandTech staff is riding off into the sunset, I am happy to report that the site itself won’t be going anywhere for a while. Our publisher, Future PLC, will be keeping the AnandTech website and its many articles live indefinitely. So that all of the content we’ve created over the years remains accessible and citable. Even without new articles to add to the collection, I expect that many of the things we’ve written over the past couple of decades will remain relevant for years to come – and remain accessible just as long."
[1]: https://x.com/anandtech/status/1829489697384706555
Future PLC seems to be gradually shutting down all of its activities, one publication or site at a time.
It's a shame because many of them had been publishing for decades. Were they really completely unsustainable? The ads in magazines like Computer Music and Future Music were actually interesting and relevant, unlike typical garbage web ads.
I think it's more they've shifted away from their original focus as a specialist computer publisher into a more general publisher. I realised a couple of months ago they publish the TV Times nowadays, and also stuff like Country Life and Home and Gardens. Tech stuff is just another line on the balance sheet now.
I doubt if magazines (and websites etc.) in general are doing great, but for obvious reasons the more techy stuff is probably going to be a bit more vulnerable, particularly in print.
The fact that Future Publishing - once the scrappy 90s flagship of new tech journalism - now owns Country Life is something I will always find hilarious.
So “indefinitely” turned out to be only 11 months.
Just like lifetime subscriptions and warranties.
"years to come"
> "the site itself won’t be going anywhere for a while"
Seems people focused too much on the phrase "I am happy to report".
When the actual key phrase people missed was "for a while".
In hindsight, that implies knowledge it would be shutdown at some future date.
And while many people understand "indefinitely" to mean "unlimited", it also has a secondary meaning of "unspecified period of time".
Interesting, usually it's the forums of websites that die first, instead of the static content, due to requiring active maintenance and moderation...
Related, I guess:
AnandTech Farewell - https://news.ycombinator.com/item?id=41399872 - Aug 2024 (598 comments)
https://web.archive.org/web/20250726001915/https://www.anand...
https://archive.ph/bewtN
A forum member says that an archive exists:
https://forums.anandtech.com/threads/anandtech-editorial-ann...
Already got bit by this, remembered an external SSD I opened had a review there and wanted to compare hardware, and wasn't able to get to the review
Just learned they've stopped publishing. Sad! The old web is really dying. Seems like a bug though? They said they'll keep the site up "indefinitely".
Old web died when youtube reviews became more profitable.
It’s sad because I miss printed content like tech magazines.
We’ve gone full circle now with video summarizers turning videos back into text.
Chips and Cheese seems like a basically fine replacement for Anandtech. Things change, and the internet has gotten worse since then, but specifically chip benchmarking doesn’t seem too bad.
I think Chips and Cheese is more like a fine replacement for realworldtech.com sans the toxic and highly educational and entertaining forums. Anandtech was much more accessible to the general tech public, but also more commercial and thus hit and miss on the content (no judgement intended, gotta eat).
Indefinitely in the sense of an unknown amount of time (Not infinitely)
From another comment here, they did say:
> I expect that many of the things we’ve written over the past couple of decades will remain relevant for years to come – and remain accessible just as long.
I honestly don't understand why companies do this. There's still SO MUCH traffic from Google for these things.
I've routinely went to AnandTech EXCLUSIVELY from Google. This means that their "news" and new content is of little relevance to me, as it's usually something from a few years prior that I'm reading over there. Yet somehow, "have to publish every day"?!
Part of me thinks that this is related to the inefficiencies of all these CMS, where it costs too much to run the site compared to the revenue from the ads.
Or is there another reason?
Frankly, as an nginx "practitioner", all of these sites could basically be cached and served from a single $50/mo server from Hetzner, Online or OVH. Aren't they're getting far more in ad revenue than that? How does it make any sense to close the shop when you've got such a treasure trove that you could continue milking easily for at least like half a decade?
Those who are currently publishing will tell you otherwise; traffic from Google has dropped off a cliff. There’s a little bit of long tail, but nothing like the volumes the tech media built its businesses around.
If I look at another tech site that stopped publishing, but where the content going back to year 2000 is still online [1], there's very little human traffic, vast majority is bot traffic.
[1] https://hexus.net/tech/reviews/cpu/132-amd-duron-850/
This sucks. Same with cnet, so many spec sheets for old crt monitors gone
Not even techy people can keep a list of simple URL redirects active?
Pretty sure the techy people have left the building
It was already over when its namesake left for Apple really.
God this is sad :(
did they not provide a copy of the CMS to the Internet Archive??!
People probably already archived it on there themselves.
Big loss for the web.