archive.today: On the trail of the mysterious guerrilla archivist of the Internet – Gyrovague
Posted by jpluimers on 2026/02/23
I pulled this post forward from the blog queue in light of the recent Archive Today controversy (which started because of the Gyrovague article mentioned below). Please note that in this controversy, the Wayback Machine of the Internet Archive plays no role: it is purely about the Archive Today behaviour. Apart from this remark above the line I left this blog post in the original form I wrote it in, as I liked it a lot and quite a few published and queued blog posts still depend on it.
From a while back, but still a historic relevant article: [Wayback/Archive] archive.today: On the trail of the mysterious guerrilla archivist of the Internet – Gyrovague
Via [Wayback/Archive] difference between archive today and wayback machine – Google Search
Related:
- [Wayback/Archive] Archive.today: on the trail of mysterious guerrilla archivists of the Internet | Hacker News
- [Wayback/Archive] Paypal Donate
- [Wayback/Archive] Buy me Coffee: Archive.Today via[Wayback/Archive] archive.today on X: “@stronghead_yo I’ve replaced the direct PayPal link with
buymeacoffee.com/archive.todayThere’s no guarantee it will work but one positive thing is that it has no country selection.” because of [Wayback/Archive] Pei on X: “welp @archiveis” - [Wayback/Archive] Archive.is blog: Brave donation model scam
- archive.today – Wikipedia
- Wayback Machine – Wikipedia
- [Wayback1/Wayback2/Archive1/Archive2] github.com/volth (now defunct; active from 2015-2020)
- [Wayback/Archive] What are the differences between archive.is, archive.org and webcitation.org? · Issue #13 · rahiel/archiveror
- [Wayback/Archive] Archive.is blog (keep scrolling down to get back to 2013)
- [Wayback/Archive] Archive.is blog: Archive
- [Wayback/Archive] How does archive.today work? How do they a paywall free version of web pages? | Hacker News
Other interesting blog posts and tweets sketching some of the history and technicalities at Archive Today (with a reminder to self: check if the Archive Today bookmarklet has trouble on Firefox archiving for instance twitter.com and web.archive.org and a reminder that there are APIs that will help me develop better bookmarklets supporting Archive Today and the Wayback Machine):
- [Wayback/Archive] archive.today on Twitter: “about the future of mirror domains:
blog.archive.today/post/188657795411/i-see-the-new-md-domain-archive-md-and-9-months“
[Wayback/Archive] Archive.is blog — I see the new .MD domain (archive md) and 9 months… (also explaining the
archive.ecowner change)I see the new .MD domain (archive md) and 9 months ago you said (archive is) would stop working but it’s still going strong. Any backstory on the domains, which ones we can expect to keep going, other aliases that we don’t know about, etc? Also I cannot type the domain name with a dot, your link checker says not to include links. Maybe that’s a tumblr issue.
AnonymousBased on the story I can only predict that there will be approximately one trouble with domains per year and each fifth trouble will result in domain loss.
The forecast is no one of the current domains (with the exception of .onion ones) will work in the next 20 years timeframe, but there will be other domains.
I see it as an imminent problem and can only accept that.
The domain system is managed by the people with the commercial mindset so they optimize they workflow either for more money or for spending less time for the fixed work. If one of the domains requires them to read many silly emails they are happy to delete the domain. There are many attackers who know and use it, for profit (with a strategy to take control over a deleted domain and sell it later as it happened with archive.ec) or just to deplatform a website which they do not like.
Another problem is the domains are used on state firewalls for geo-blocking, some of the domains are blocked in Russia, some other in Australia, etc, while the website is still accessible using other domains. So to say even the active domains are half-dead already.
Unfortunately, any of these two problems prevent implementing some nice features, such as embed codes.
- [Wayback/Archive] archive.today on Twitter: “@MaxBlake5Z archiveiya74codqgiixo33q62qlrqtkgmcitqx5u2oeqnmn5bpcbiyd·onion archivecaslytosk·onion archive·today archive·is archive·fo archive·li archive·vn archive·md archive·ph”
archiveiya74codqgiixo33q62qlrqtkgmcitqx5u2oeqnmn5bpcbiyd.onion
archivecaslytosk.onion
archive.today
archive.is
archive.fo
archive.li
archive.vn
archive.md
archive.ph
- [Wayback/Archive] archive.today on Twitter: “@MaxBlake5Z archiveiya74codqgiixo33q62qlrqtkgmcitqx5u2oeqnmn5bpcbiyd·onion archivecaslytosk·onion archive·today archive·is archive·fo archive·li archive·vn archive·md archive·ph”
- [Wayback/Archive] Archive.is blog — whats the domain name? the name everyone calls it…
whats the domain name? the name everyone calls it is archive. is, when i visit archive. is the logo says archive. today and when i click that it takes me to archive. vn. whats going on?
AnonymousIt is intentional.
No single domain is reliable and I have no means to enforce control on each domain.
* archive.today – threatened with confiscation http://blog.archive.today/post/116913927371/the-domain-registrar-gransy-s-r-o-aka, also a troll attack caused service interruption https://blog.archive.today/post/138982909006/domain-problems-again
* archive.is – threatened with confiscation https://twitter.com/archiveis/status/1081276424781287427, asked not to use “archive.IS” for branding (that’s why you see “archive.TODAY” in the top-left corner; although many people remembered it as “archive.IS” and refer it so)
* archive.fo – threatened with confiscation https://twitter.com/archiveis/status/1188222460598116353
* archive.li – attacked by trolls impersonating police, caused few days service interruption https://twitter.com/archiveis/status/956025540028268547
* archive.ec – attacked by trolls causing service interruption and finally lost https://twitter.com/archiveis/status/1093608363647291393
* archive.vn – ok so far
* archive.ph – ok so far
* archive.md – ok so far
* a nice domain unrelated to archive – one day whois started showing someone’s else information and the registrar did not response, the domain was lostIn order of the above links:
- [Wayback/Archive] Archive.is blog — The domain registrar «Gransy s.r.o» (AKA…
The domain registrar «Gransy s.r.o» (AKA «regtons.com» AKA «subreg.cz» AKA «regnames.ua») was threatening us with blocking «archive.today» domain if some webpage were not removed from «archive.today» website.
This page is not related to Wikileaks, nor to the Gamergate controversy, nor to the war in Donbass. Not even to the copyright. They wanted to remove the snapshot of a deactivated LGBT-clothes shop.
We can only guess why the «Gransy s.r.o» employees consider this case so important – against the background of all the hot controversies that use «archive.today» to preserve the proofs which many people want to disappear – to renege on registrar’s neutrality and arrogantly use such a dirty method as threatening with blocking the domain in their intention to censor the client’s website content.
I have moved «archive.today» to another registrar.
Also mirror on «archive.is» is still active and I registered «archive.li» where I am going to set up an additional mirror. - [Wayback/Archive] Archive.is blog — Domain problems again.
“archive.today” is attacked by fake DMCA and Child Porn requests and may work unstable.
Today OnlineNIC interrupted the service twice based only on private emails with fake DMCA requests and rude wording about how bad the website is.
Can anyone recommend a reliable registrar for a domain with .today extension which will not disrupt the service without a court order? Or it would be easier to drop it and never again have a deal with newTLD? (I talked to Donuts’ VP during the past incident with “Gransy” and got an idea that they allow their resellers (registrars of newTLD domains) to interrupt the service as their discretion. So to say “we do not like you” policy. Thus, an successful attacker must not provide any documents to support her claims, she only needs to create a negative impression and your newTLD domain will be disabled; perhaps few times a day).
- [Wayback/Archive] archive.today on Twitter: “Please do not use
archive.ISmirror for linking, use others mirrors [.TODAY .FO .LI .VN .MD .PH]. .IS might stop working soon.” - [Wayback/Archive] archive.today on Twitter: “Domain problem again:
archive.FOis shutting down. Other mirrors are safe.” - [Wayback/Archive] archive.today on Twitter: “#switchplus seems to be attacked by trolls impersonating police. Absolutely no hints what the “illegal content” might be.…”
- [Wayback/Archive] archive.today on Twitter: “only
archive.ecwas lost, the rest work:archive.foarchive.isarchive.liarchive.mdarchive.pharchive.vnarchive.todayarchivecaslytosk.onion…archiveiya74codqgiixo33q62qlrqtkgmcitqx5u2oeqnmn5bpcbiyd.onion“
- [Wayback/Archive] Archive.is blog — The domain registrar «Gransy s.r.o» (AKA…
- [Wayback/Archive] Archive.is blog — If sites have been saved from other archives,…
If sites have been saved from other archives, which archives have they been saved from?Anonymoushttp://archive.is/webcache.googleusercontent.com
http://archive.is/peeep.us
http://archive.is/hghltd.yandex.net
http://archive.is/www.youdao.com/cache*
http://archive.is/www.sogou.com/websnap*
http://archive.is/www.webcitation.org
http://archive.is/www.family-source.com/cache/*
http://archive.is/hl.mailru.su
http://archive.is/cache.baidu.com
http://archive.is/cache.yahoofs.jp
http://archive.is/nova.rambler.ru
http://archive.is/cc.bingj.com
http://archive.is/blekko-webcache.com
http://archive.is/wayback.archive.org
http://archive.is/megalodon.jp
http://archive.is/hiyo.jp
http://archive.is/arquivo.pt
…
The list is not complete, because archiving is initiated by user requests. And they can archive any URL they like, even if I am not aware of that website.Almost all of these snapshots are indexed using both URLs.
For example, this snapshot – http://archive.is/sa4Uo – taken first byarquivo.ptin 1997 and then re-archived in 2013 on archive.is, can be found by both the original URL: http://archive.is/http://www.lusa.pt/impreg/impreg/12MAR97/not0158.html and the URL or archiver: http://archive.is/http://arquivo.pt/wayback/wayback/id742788index5?pos=1&l=pt&sid=CFA9EE0F6FBB9DC227425C893349A213 - [Wayback/Archive] Archive.is blog — Why did you change the URL back from archive-today…
Why did you change the URL back from archive-today to archive-is? Archive-today is a better name and URL.AnonymousI see two problems with .today: it is longer and the URLs are not clickable sometimes:
If there would be many users who prefer
.today, I will change it back. - [Wayback/Archive] Archive.is blog — Thanks for the outstanding service. I’m building a…
Thanks for the outstanding service. I’m building a stand-alone archiving tool for “Web 2.0” pages which is heavily inspired by your approach. I know you mentioned repeatedly that the codebase is not useful out of context and I understand that, but I’d like to ask again just for the Javascript components, which are the ones that would be the most useful and portable, but also the trickiest to get right without iteration. Any chance I can reuse the website-specific plugins and/or the inliner?AnonymousIronicaly, the JS part is the most useless one and not portable, it requires PhantomJS with many patches applied.
- [Wayback/Archive] Archive.is blog — So I wanted to archive a page it asked me if the…
yes, old archives are not altered.
- [Wayback/Archive] Archive.is blog — GET request for archiving web sites would be nice….
GET request for archiving web sites would be nice. Web Archive has this feature: web archive org/save/{URL} works easily. (I had to change typing url, because of, tumblr doesn’t allow urls) I use archiveis because of it has more sophisticated features than Web Archive. Thanks for this nice tool.AnonymousGETs are disabled to prevent the crawlers (such as Googlebot) from making submits.
- [Wayback/Archive] Archive.is blog — Why the project isn’t open source?
The codebase is far from the point where the open/close-source difference may even have sense; it is not in the form of a re-deployable product; besides the html standards (and big list of exceptions) the code reflects specific hardware and network, which pages are popular, the behaviour of the SEO-bots and the users (in order to tune the caching strategies), etc.
It would be a big work to create an alienable archiver (whether open or closed source) which anyone could set up on their own premises. - [Wayback/Archive] Archive.is blog — Is there a way to link to the most recent archive…
Is there a way to link to the most recent archive of an article by including the URL in an archive. is link?Anonymous - [Wayback/Archive] Archive.is blog — You mentioned there’s no hot backup as of yet…. (oktober 2015)
5Tb is weekly growth :)
- [Wayback/Archive] Archive.is blog — If and when archive_dot_is can no longer support…
If and when archive_dot_is can no longer support its archived materials is there a way users can be notified of discontinued service. Like if something really bad happens financially and archive_dot_is can’t be supported anymore will it’s users be notified and will there be some plan like move everything over to the waybackmachine?AnonymousFinancially there should be no problem. Although, something bad could happen. So consider having a backup at another place.
- [Wayback/Archive] Archive.is blog — I have received several bug reports about…
I have received several bug reports about archive.is saving empty or 404 pages from Google Cache although there expected to be some content.
It seems that there is more than one Google Cache, and what you get depends not only on the URL but also on which one of the Google datacenters serves you request.Examples of pages saved via different proxies:
http://archive.is/https://webcache.googleusercontent.com/search?q=cache:_PVt8WPb4DEJ:*
http://archive.is/https://webcache.googleusercontent.com/search?q=cache:CO15sF9zSrQJ:*I think, the archive should perform few requests simultaneously and then save all successful versions.
- [Wayback/Archive] Archive.is blog — I looked in the FAQ, didnt find an answer, so I’ll…
I looked in the FAQ, didnt find an answer, so I’ll ask here. Have you considered trying to connect your database with the wayback machine’s database? So that when someone searches for a page, they are actually searching both archiving websites simultaneously.AnonymousThere is http://timetravel.mementoweb.org/ for that
- [Wayback/Archive] Archive.is blog — What about the CA “Let’s Encrypt”? They seem good….
What about the CA “Let’s Encrypt”? They seem good. Nonprofit. You still support HTTPS connections, just only when explicitly instructed or you have HTTPS Everywhere installed. HTTPS sites get better SEO as well. I fail to see how this site is any more controversial than similar platforms like Wayback Machine or Megalodon.AnonymousThe Archive’s ability to preserve short-living content of social media turned it to a lovely instrument of troll wars (Alt-Right vs. SJW, Ukraine vs. Russia, …)
and although the Archive tries to be neutral to those battles, it was often under the fire of technical and social attackers.
The pattern of attacks resulted in our infrastructure became similar to those of Wikileaks, SciHub, 8ch or DailyStormer – many mirror domains, fast-flux IPs for ingress and egress, etc.
If there are attacks which have already been made against one of the websites in this karass, the rest have to be prepared.
Revocation of SSL certificate as the result of some social attack is very likely, so I would even argue for using plain http in links to the Archive. - [Wayback/Archive] Archive.is blog — I have been told that “Higbee and Associates”…
I have been told that “Higbee and Associates” copyright trolls [1][2] (and probably their clones) have been used Archive.Today’s snapshots as “evidences” of “crimes” of their victims.
We are not associated with those guys and never heard about them before. Archive’s snapshots are not notarized nor use strong anti-forging technologies, so they cannot serve as legal evidences. Shortly, if you received a copyright claim with Archive’s snapshots as a proof – it is a scam.
There is the second floor: other guys (for example [3]) try to extort more money from the victims of aforementioned scammers for archive snapshots removal in order to protect them from further attacks.
It is a scam too, do not pay them.
3. http://archive.today/2020.02.17-183946/https://sumbit.nl/prijs.html
- [Wayback/Archive] Archive.is blog — You refer to the wonderful Wayback Machine. Have…
You refer to the wonderful Wayback Machine. Have you had any discussions with the Internet Archive? Would you be interested in integrating your cool service with theirs somehow?AnonymousNo, I hadn’t any discussion.
But I’ve heard some rumors about their plans to use real browsers to execute javascript, etc.
- [Wayback/Archive] Archive.is blog — Any plans to provide an API or open source the…
Any plans to provide an API or open source the code? This would be a really helpful addition to many of the bookmarking tools.The site supports OExchange (http://www.oexchange.org/), so the bookmarking tools can use it.
- [Wayback/Archive] Archive.is blog — Hello. I am new to this website. Does this website…
Hello. I am new to this website. Does this website act like the WayBack machine where pages are automatically saved by the website for users to see? Or does this website need a person to on purposely copy and paste a url into the archive to save its contents?
AnonymousIt needs a person to on purposely copy and paste a url into the archive to save its contents.
At the time Archive.Today was created, The Wayback Machine was not able to save pages on demand, while increasing share of dynamic and realtime content in the Internet (so to say, the transition from “writing“ to “speech“) was destroying the very idea of Hypertext (when every page could link to another and every reader will find there the same content) and it was frustrating. So, our idea was not to crawl and preserve the entire Internet, but merely to freeze the pages before linking to them. And this implies the presence of the person.
- [Wayback/Archive] Archive.is blog — what happened to the bookmarklets is is just…
what happened to the bookmarklets is is just replaced by the extension now?
AnonymousBookmarklet still works. There were few issues which makes the bookmaklet inferior to extensions so I stopped promoting it:
1. People do not understand where is “bookmark bar” and even what is it. The bar is not visible by default in modern browsers.
2. Firefox has blacklist of websites where all the bookmarklets do not work. The list includes “twitter.com” among others, to the bookmarklet did not allow to save tweets if you are using Firefox.
3. Since recently (somewhere between Chromium 80 and 85), `
document.URL` on Wayback Machine pages returns url of the original page, having “http://web.archive.org/web/YYYYMMDDhhmmss/” prefix stripped. This breaks bookmarklet too. - [Wayback/Archive] Archive.is blog — Are Wayback Machine links no longer allowed to be…
Are Wayback Machine links no longer allowed to be backed up in your archive? The archive process seems to keep rejecting them.
AnonymousThere is an issue with Wayback Machine snapshots which are just saved to Wayback Machine.
There seems to be some sort of eventually consistent storage, so if you just saved a link to Wayback Machine and immediately send the WM link to a friend (or feed in to Archive.Today), they might see an empty page on WM. In 10-30 minutes the WM page is visible to everyone
- [Wayback/Archive] Archive.is blog — Is there a way to programmatically check if there…
Is there a way to programmatically check if there is an archived version of
{url}atarchive(.)is/newest/{url}My usecase is when a source site is unavailable I want to show my users possible archived versions. With wayback machine I can check somejsonif there’s something there before showing my users a link, but onarchive(.)isthe best I can do is always send them the link. Then since often there is nothing archived here yet, users get trained to not click thearchive(.)islink.Anonymoushttps://archive.is/timemap/http://www.w3.org/ for plain text answer
It looks like there is JSON-answers in a newer standard (http://timetravel.mementoweb.org/guide/api/), but I was not aware of the newer version and no one has complained yet (since 2015) that it does not work.
- [Wayback/Archive] Archive.is blog — Can you please implement a feature similar to…
Can you please implement a feature similar to “Save outlinks” of Wayback Machine?
AnonymousNo. Saving pages here is much more heavier process than on Wayback Machine. Here it runs Chrome, executes JavaScript, scrolls page down and up, while on WB it is just a curl download.
- [Archive/Archive] Archive.is blog — Have you thought on contacting Internet Archive…
Have you thought on contacting Internet Archive about hosting a mirror copy of your whole collection of websites? To clarify, it would be stored as an item, not Wayback Machine pages. What you have is very important! Alternatively, have you thought on offering a torrent of all pages? Thanks for reading!AnonymousIt wouldn’t work this way. The storage form is very far from “items“ to upload on FTP or distribute via torrents. It requires computers to run the renderer code which is being changed almost every day (see a lot of “fixed” over here?)
- [Wayback/Archive] Archive.is blog — If you run out of money, will Wayback Machine take…
If you run out of money, will Wayback Machine take over the backed up pages? It would suck if they all just disappeared. Big fan of service BTW
AnonymousI can leave a will, but how can you be sure that they will be happy to receive it and that they will dispose of it in such a way that you will like it? - [Wayback/Archive] Archive.is blog — Are the Archive-Today going to have the same…
Are the Archive-Today going to have the same search features as the Wayback Machine of The Internet Archive has, such as search by date and sitemap?
AnonymousMaybe.Even full-text search is planned, but there are too many current tasks :(
--jeroen
The Wiert Corner - irregular stream of stuff 
Leave a comment