Google is not some opaque corporate entity (I mean, yes it IS, but...), it is made up of individuals, many of whom read this site.
Google should offer to send (the public content from) their DB directly to The Internet Archive. It results in LESS overhead for Google than this attempt to scrape it, and results in better information.
This is unbelievable. We're all used to Google abandonware by now but the resources required for storage of shortened URLs and any dev maintenance must be so miniscule to be a rounding error to a rounding error on any budget anywhere in the company.
What can possibly be the motivation for this? I see no rationale for it in the linked post.
Google already has severe trust issues with anything they launch. They hurt themselves because people expect whatever is being launched to be abandoned so why invest in it? Reputation matters.
This is a great summary + call to action. The ArchiveTeamWarrior tool from the Internet Archive is a pretty sweet piece of kit, it is fun and satisfying to watch it crunch away at URLs!
Well, to answer the post title question directly: far away from google services for years. It was obvious to me 10+ years ago that they only sustainable business products for them is search, yt, and marketing tracking/performance tools. Nothing else is stable, nor secured and there were examples of it before that shortener existed. G+ exposed how vulnerable google is regarding addon services and now they can’t even keep up with the archives (usenet). URL Shortener… please, that by definition looks like vaporware regardless of the vendor, anyway.
people definitely should donate resources to archive team but as someone who has been participating in this project it's my understanding that all of the URLs at risk have already been backed up
I'm against link rot and I hate how Google doesn't maintain old projects. But this is one shutdown I 100% agree with.
Having an official Google domain that anyone can hijack is dangerous, given that many people's main internet identity is GMail (aka their Google account). I know anyone can create an offshoot (goooogle.org, etc), but Google was using goo.gl too.
It was easy to redirect a goo.gl to a Google login page (which is on a real Google domain), and trick people into authorizing access to their account.
I consider myself savvy, and I got a pretty convincing one recently. The email looked legit, and the link was a goo.gl link that ultimately landed me on a legitimate Google login page. It didn't trick me, but it did take me a few minutes to figure out how it wasn't legit.
NOTE: This article is kinda misleading. They already stopped letting people add new links in 2019. And now, they're only removing "inactive" links, AKA links that had no activity since 2024. If you visit a link right now, it will be kept. Here's more info: https://blog.google/technology/developers/googl-link-shorten...
It always seemed strange to me that people would use a link shortener (that they didn't own!) for anything other than Twitter, or the small use-case of manually typing a URL into a different computer that you couldn't sync in some other way. In what way was this not the obvious outcome?
So if the issue is `goo.gl` is useful for phising, why not shut it down and leave a readonly copy on some other domain? Sure it still breaks every link, but for people who are in-the-know can use the alternate domain by manually translating it.
I've never quite understood URL shorteners. They seem like a way of opting into link rot and sending tracking data to third parties who aren't necessary for the client-server connection.
Is there a major benefit I'm missing? I could kind of see them if you have a character limit, want to hide the URL, or have to type a URL manually. But manual typing is rare, and even microblogging services are expanding character limits. Hiding the URL seems slightly sketchy, but you can achieve it without a shorter URL so maybe that's not a real benefit.
Anyway, I'm actually curious about this because people seem to love them.
(and this is aside from all the very valid issues and concerns people have with Google shutting down a widely used service).
URL shorteners significantly increased in popularity as Twitter did. There was originally a 140 character limit that became quite squeezed when adding most URLs, especially to blog posts where the title is part of the URL.
Later, adding things like analytics and tracking (eg: not just in social media, but also in email campaigns) became another reason to use them, especially for those less tech inclined.
Historically there were a lot of CMS's with very small character limits. I remember a war between my high school teachers trying to make their pages work, and the IT department trying to ban all URL shorteners categorically (which admittedly I never understood, given that the DNS interceptor should be able to catch the redirected domain)
Conspiracy: They're shutting it down so competitors' AI and search crawlers can't visit the links, but theirs can, since archiving efforts aside only they have the DB.
Given that it doesn't make money, it has zero value alive and at least epsilon value dead. Plus they don't need the links to work to collect metrics.
Might I suggest that instead of linking to the article at theatlantic.com you instead link to the archive.org archive of it? I couldn't visit it due to Cloudflare thinking IPv6 is the devil and then still couldn't read it due to the paywall. Here's the archive.org link that sorta works for me, until their JavaScript removes the majority of the page content shortly after loading:
I let the ArchiveTeam Warrior docker run in the background on my file server, configured to contribute to whichever project archiveteam picks. When I first set it up, that was archiving telegram channels, but I took a peak yesterday and it was backing up goo.gl links. Apparently I've uploaded 94gb's worth of links already.
My question is, why can Google themselves not just provide a dump to archive.org themselves? Having volunteer middlemen doing the work seems like an artificial crisis.
It seems like Google doesn't want a single machine visiting so many links in such a short time. I wonder what the "bad response" could be?
Google asks for a login, sleeping 20 minutes.
Server returned bad response. Sleeping 8 seconds.
635=302 https://images.google.com.pk/imgres?imgurl=http://www.pakiboutique.com/wp-content/gallery/zainab-chottani-pret-collection-2017/zainab-chottani-luxury-pret-2017-7.jpg&imgrefurl=http://www.pakiboutique.com/zainab-chottani-pret-collection-2017&docid=0gx_lzfAk8YInM&tbnid=qKdQuVyBbOHJfM:&w=948&h=1280&source=sh/x/im
Google asks for a login, sleeping 20 minutes.
Server returned bad response. Sleeping 11 seconds.
636=302 https://images.google.com.pk/imgres?imgurl=http://www.pakiboutique.com/wp-content/gallery/zainab-chottani-pret-collection-2017/zainab-chottani-luxury-pret-2017-7.jpg&imgrefurl=http://www.pakiboutique.com/zainab-chottani-pret-collection-2017&docid=0gx_lzfAk8YInM&tbnid=qKdQuVyBbOHJfM:&w=948&h=1280&source=sh/x/im
Google asks for a login, sleeping 20 minutes.
Server returned bad response. Sleeping 29 seconds.
637=302 https://images.google.com.pk/imgres?imgurl=http://www.pakiboutique.com/wp-content/gallery/zainab-chottani-pret-collection-2017/zainab-chottani-luxury-pret-2017-7.jpg&imgrefurl=http://www.pakiboutique.com/zainab-chottani-pret-collection-2017&docid=0gx_lzfAk8YInM&tbnid=qKdQuVyBbOHJfM:&w=948&h=1280&source=sh/x/im
34 comments
[ 0.29 ms ] story [ 56.7 ms ] threadGoogle is not some opaque corporate entity (I mean, yes it IS, but...), it is made up of individuals, many of whom read this site.
Google should offer to send (the public content from) their DB directly to The Internet Archive. It results in LESS overhead for Google than this attempt to scrape it, and results in better information.
What can possibly be the motivation for this? I see no rationale for it in the linked post.
Google already has severe trust issues with anything they launch. They hurt themselves because people expect whatever is being launched to be abandoned so why invest in it? Reputation matters.
Two weeks ago [1] they were doing 37k a minute with ETA just barely before end of the month. Now it's ~55k a minute, and ETA of just 5 more days.
[1] https://news.ycombinator.com/item?id=44688478
After stripping any past statistical data from each entry, it shouldn't be that much of data per URL...
Having an official Google domain that anyone can hijack is dangerous, given that many people's main internet identity is GMail (aka their Google account). I know anyone can create an offshoot (goooogle.org, etc), but Google was using goo.gl too.
It was easy to redirect a goo.gl to a Google login page (which is on a real Google domain), and trick people into authorizing access to their account.
I consider myself savvy, and I got a pretty convincing one recently. The email looked legit, and the link was a goo.gl link that ultimately landed me on a legitimate Google login page. It didn't trick me, but it did take me a few minutes to figure out how it wasn't legit.
NOTE: This article is kinda misleading. They already stopped letting people add new links in 2019. And now, they're only removing "inactive" links, AKA links that had no activity since 2024. If you visit a link right now, it will be kept. Here's more info: https://blog.google/technology/developers/googl-link-shorten...
No, only those that were deemed "active" in 2024 will be kept.
Step 2: shut it down
Step 3: make every link redirect to an advertisement
Step 4: profit
Is there a major benefit I'm missing? I could kind of see them if you have a character limit, want to hide the URL, or have to type a URL manually. But manual typing is rare, and even microblogging services are expanding character limits. Hiding the URL seems slightly sketchy, but you can achieve it without a shorter URL so maybe that's not a real benefit.
Anyway, I'm actually curious about this because people seem to love them.
(and this is aside from all the very valid issues and concerns people have with Google shutting down a widely used service).
Later, adding things like analytics and tracking (eg: not just in social media, but also in email campaigns) became another reason to use them, especially for those less tech inclined.
Given that it doesn't make money, it has zero value alive and at least epsilon value dead. Plus they don't need the links to work to collect metrics.
Google shifts goo.gl policy: Inactive links deactivated, active links preserved - https://news.ycombinator.com/item?id=44759918 - Aug 2025 (189 comments)
Google's shortened goo.gl links will stop working next month - https://news.ycombinator.com/item?id=44683481 - July 2025 (222 comments)
Google URL Shortener links will no longer be available - https://news.ycombinator.com/item?id=40998549 - July 2024 (49 comments)
Ask HN: Google is sunsetting goo.gl on 3/30. What will be your URL shortener? - https://news.ycombinator.com/item?id=19385433 - March 2019 (14 comments)
Tell HN: Goo.gl (Google link Shortener) is shutting down - https://news.ycombinator.com/item?id=16902752 - April 2018 (45 comments)
Google is shutting down its goo.gl URL shortening service - https://news.ycombinator.com/item?id=16722817 - March 2018 (56 comments)
Transitioning Google URL Shortener to Firebase Dynamic Links - https://news.ycombinator.com/item?id=16719272 - March 2018 (53 comments)
https://web.archive.org/web/20250527052607/https://www.theat...
* Microsoft has code that references a URL shorterner
* The URL shortener is in-house and points to a Microsoft property.
* However, the documentation team don’t keep this stuff up to date, resulting in you getting error messages containing a broken link.
You would have thought it would be more than possible for Microsoft to keep their own house in order.