Yeah, I also hate back-button hijacking. I suspect some websites do it to artificially force more page views for ad revenue. Try a long-press on the back button to view the history for that browser tab and click on the most recent page you think won't redirect.
Usually with this, it's like 20 entries, so pushes everything else off.
The ones where it's only a couple entries mostly seem to be the ones where there's multiple articles on a single page and it's at least might be attempting to be helpful?
Youtube does it, and it just continues to blast the wrong video you accidentally just auto-started because instead if fucking off, it shows other videos with the bad video getting just reduced.
I recently looked into this after it seemed like Facebook messed with my back button on a private mobile window:
Someone pointed out that it's nice to have, for example, your email provider know that you probably want to go back for a message to your inbox instead of going back to the previous page.
But what if browsers monitored which sites abused the feature and showed a pop-up when you click the back button, just like they offer to show you notifications? They could show you:
This site has been reported to hijack the back button. Would you like to go back to the last domain that you visited?
This could easily be fixed by the browsers but they don't. Sure wish these back button tricks would stop. Especially news sites try to keep you from getting back to your search and makes your page refresh over and over. I wonder if that behavior counts as hits to their advertisers.
I don't know about "easily." replaceState() is actually intended to make single-page apps easier to use, by allowing you to use your back button as expected even when you're staying on the same URL the entire time.
Likewise, single-page apps are intended to be faster and more efficient than downloading a new static page that's 99.9% identical to the old one every time you change something.
Fixing this bad experience would eliminate the legitimate uses of replaceState().
Now, what they could do is track your browser history "canonically" and fork it off whenever Javascript alters its state, and then allow you to use a keyboard shortcut (Alt + Back, perhaps?) to go to the "canonical" previous item in history instead of to the "forked" previous item.
Also: Algorithmic generated feeds where you try to click on one thing, but you click on the next thing in the list and when you click back, the feed looks completely different because it has new information on you. That thing you wanted to click on is gone and will never return.
I'd love that, my entire frontpage is the same 30 things over and over unless I deliberately sort for something then it's a DIFFERENT 30 things over and over
What's worse is that YouTube sometimes doesn't do that, i.e. when you hit back it shows the same list from the cache or something. It gives you hope and makes it worse on those occasions when it does fully refresh on back.
Though on desktop I've completely switched over to using FreeTube, and I've been loving it. The order of the videos in the feed does not change. It's great.
It's not for this, of course. It's because in the world of single page applications built in react and angular where there is no physical back, like no actual server page to go back to just JavaScript, you have to code in what the back button means. Even though there's no server calls to ask for a new page. New page. Most people still expect that forward and back will still go forward and back in standard navigation.
Sites like this it's pretty clear that they just overwrite that with the last 20 calls to their own page, but the alternative is that single page applications would not be able to have forward or back functionality
It's a very "dumb" implementation of a generally useful feature. Browsers don't keep track of how many times you're redirected to the same site or try to consolidate the back-button list accordingly, but they certainly could. Wouldn't be surprised if there was a plugin to this effect.
They actually do. To avoid infinite loops. If a URL redirects to the identical URL for more than ~5 times most browsers will refuse to load and show an error instead.
That's why sites like this will generate new URLs with the same content.
I think it's taking you away to a login page, logging you in, then bringing you back.
I can see the point if you were going to ask or answer a question, but 99% of the time you just want to see how somebody else didn't get their problem solved by some random Indian guy who people assume works for Microsoft, who think the solution to everything is running "sfc /scannow" which has replaced chkdsk as the command most likely to take a long time, do nothing, and make the question asker go away without a solution to their problem.
What helps with this is clicking links with mousewheeldown, I automatically opens in a new tab. Also MWD on the tab label will close it, so you don't have to aim for the 'x'.
A mouse with thumb buttons is really handy as they do foreward and back, double clicking that gets you out of the issue caused in op
Yes. Sometimes this is malice. Sometimes this is an attempt to drive impressions and page views.
This can also be caused by poorly configured web applications that update in real time. If, say, some sports website is giving you real-time data about the game as it progresses, a poorly configured web application might be creating a dynamic URL for every change. When you access the older page, it will be instructed to take you to the most recent data, so pressing back is taking you to old data on that page, and then immediately realizing that data is old so refreshing it with the most relevant data.
This is a super common misconfiguration in single page web applications. Domain.com will take you to an application that renders at domain.com/en-us/home. Pressing back takes you to domain.com, and guess what happens next?
This is basically 99.99% of these cases. I would say if its on some shitty news site with 1000 ads that somehow sneak by AdBlock and UBlok Origin, it's case 1. Otherwise, it's case 2 or 3.
This is one of the absolute greatest reasons to support opening most everything in a new tab (as long as you don't end up like my mom who at one point had over 100 tabs on her phone). Doesn't matter if it's a link from the same website, from a search engine, or whatever else there is. New tab.
Then on android Firefox you accidentally hit the back button and it closes the tab and you can't go forward and you already navigatedc away from the originating page on the other tab forcing you to open your history and try to figure out where the hell it is.
Edit 2: From the 3 dots menu INSIDE the tabs view you can access a list of recently closed tabs, not nearly as fast as a 3 key combo, but maybe better than looking for the tab in the history.
Also apparently there's an extension that may help.
I always just assumed it was a form of "dark pattern" meant to try to stop people from leaving their website once they've entered (e.g., coming from a different site, you can't just hit backspace or click back to immediately exit their site. You're stuck now).
I think that's right for a website where you accidentally clicked an ad and now it's trying to convince you you have a virus and you need to download their virus to remove it. Or maybe for an ad pop-up where annoying you might increase the chances that the content makes it into your brain.
But for a news website i have trouble seeing the logic.
You've reminded me of a similar frustration that I've never found the answer to - though it may be adblock related - in that whenever I open a link to eBay it completely wipes the history for that tab. Or possibly it opens a new tab and kills the parent. Either way I always forget about it until the next time and then it drives me mad all over again.
What makes me angry here is, I am 90% sure the browsers could code against this.
If the user clicks a control on a webpage one time, the stack can declare "One user click! You have earned yourself One (1) navigation." Then, the click activates some JavaScript that moves you to a new webpage. That new webpage has an auto-loader redirect that instead runs a 300ms timeout, and then takes you to some other page. The browser, meanwhile, has seen this, and establishes "We are still only operating off of that One (1) click. So, instead of adding a new page to the user history, we'll replace that first navigation."
I have yet to hear a satisfactory reason as to why that's not possible.
We just got vertical align last month. There's so many things they should be working on but are too busy trying to add more ads or monetization features.
I think the web is just too long in the tooth at this point but there's nothing we can do.
CSS features like vertical alignment would be defined by web standards. Those fall under the non-profit org W3C. They're pretty slow about things as to not break the fuck out of everything.
Browser behaviour like merging redirects falls on browsers tho, so yeah, we can blame Chrome or FF on that one.
Aren't they scamming their advertisers too? Because if you click the back button a bunch of times it's gonna reload a bunch of them on every click. At least if your internet is fast enough.
Impressions are usually deduped, meaning multiple impressions from the same user during the same session are just counted as one. The big ad networks are extremely careful to avoid miscounting of any sort and will generally err on the side of undercounting rather than overcounting (since telling advertisers they got more impressions or clicks than reported is way better than telling them the numbers were accidentally inflated). Of course, there's the occasional bug, but it mostly works as expected.
just click again, but fast enough to get the redirect, but not too fast to miss it and double click, and try not to do it a third time or you're going back a few ages.
yo honk honk am here to help. Right click the back button to bring up a menu of several previous pages select when it was the search engine or whatever you used before. For Firefox. If you're on chrome, you can cry. Honk honk, goose out.
Seems cool, but it's currently missing some pretty important languages (Hindi, Urdu, Thai, Bosnian/Croatian/Serbian, Swahili, etc). I'd put up with something limited like this if it was FOSS and/or selfhostable but it appears not to be
While it offers a concurrent alternative to Google translate, it still lacks some features, as @[email protected] mentioned, many languages are missing. In my case, I sometimes experiment with terms across various languages, sometimes Hindi ("O param Devi Kaali"), sometimes latin ("Vita mortem manducat, Mors manducat vitam" is a latin phrase I wrote myself, following Latin grammar rules), sometimes Hebrew (especially for Gematria calculation using numerical values from Hebrew letters (Aleph is 1, Bet is 2, Gimmel is 3, and so on) after translating/transliterating a word/name such as "לילית"). For these kinds of experimentation, DeepL can't really be of use, so I need either Google Translate or Bing Translate (both support the aforementioned languages).
This is why I have dozens, if not hundreds of tabs open. Usually I open links in a new tab so I can easily tab back to where I came from. Using a hierarchical tab manager makes this work better because when you're done with the topic, you close the whole branch... theoretically.
This tactic also seems targeted at mobile users where it's harder to break the loop.
Not sure about that site specifically, but others that's done it to me was easy to get around. Most of them are thwarted with basically double clicking the back button.
Well enjoying a game of footie is your first mistake. Oh wait, I got confused, too many euros around, sorry about that. Footie is what I call soccer, aka foreign football.
Enjoying a game of football is your first mistake.