This is not a solution for the majority of users. The problem with keyboard shortcuts is that most users won't know what happened. All they'll know is that suddenly they're not on the same page any more, and then won't realize they've navigated backwards. Therefore they won't think to navigate forwards again, and they won't get back to their form data, even if it is cached.
This happens to me all the time in Emacs. I'll press something by accident, make major unintended changes to my file, and have no idea what I did. Fortunately, Emacs has a way to look up key presses, but this is a relatively advanced feature that typically doesn't exist in other tools anyway.
Whenever you have a feature with zero discoverability (e.g. essentially all keyboard shortcuts), you have to be very careful to avoid the "what did I just do?" phenomenon, even if you provide ways to get back to a previous state. I personally think this change by the Chrome devs is extremely reasonable in that regard.
Edit: I realize that these changes are aggravating to users who actually use them on a regular basis. However, please also realize that they are not zero-cost, because they do impose a constant and non-trivial burden on everyone else (especially in cases like backspace which are relatively easy to hit on accident).
So make them an option and stop taking the Chrome route of stripping out everything that more advanced users use regularly.
Of course, this is why I run Firefox. I can do things like force OCSP enabled and get error pages when it fails - Chrome's solution was to call OCSP useless because their less technical users couldn't understand it.
Even Firefox has buried the option behind about:config and the scariest thing for me now is that we'll lose the control we get with current FF options and extensions as they "modernize" the browser.
Advanced users want this level of control.
Backspace-as-back-button is not an "advanced feature". It's a hotkey that essentially no other software uses for the same behavior and one which conflicts crappily with the fundamental behavior of that key. It will take you, generously, a day to relearn the new hotkey. Instead you'd rather the developers maintain indefinitely an option and corresponding logic to support your existing use-case.
If you can't tell, my level of empathy for this kind of griping is roughly zero. Adapt to the new hotkey and save the complaining for something that actually matters.
Except for Windows Explorer, but who uses that, eh?
Your level of empathy is immaterial to the conversation, but thanks for sharing.
In a not-at-all shocking turn of events, it turns out Alt-Left and Alt-Right are in fact the hotkeys for navigating backward and forward through history, so Chrome aligns perfectly with Windows Explorer in this regard.
I definitely believe removing backspace-to-go-back in the browser is the right choice. I did know about that one, solely because every so often I do it by mistake. The problem isn't that I lose data -- I don't. The problem is that this shortcut does nothing but cause me problems I don't want. I never want the functionality, and yet it keeps happening to me anyway because it's so easy to trigger unintentionally. There's no case for this shortcut continuing to exist.
[1] http://windows.microsoft.com/en-us/windows/keyboard-shortcut...
If I've got a folder structure
1 | |--1.1 | |--1.2
And I click into 1.1 and then into 1.2, if I then press backspace I end up on 1 (the parent), not 1.1 (the previous in the history).
Then click on "Folder3" to go directly to Folder3.
Then press BackSpace, you will go back to Folder5, not Folder2.
To go up the shortcut is ALT Arrow Up.
Now when there is nothing to go back to (you opened a sub folder directly), Backspace goes up instead of doing nothing.
Users who want it can install an addon for it.
Chrome on the other hand has such weak addon APIs that they are often unable to achieve what's needed. Firefox seems to be moving in the same direction unfortunately.
I think it's worth it for those developers to at least maintain full addon APIs, even though there's a trade-off here it greatly increases the flexibility of the product.
* alert if any form filled by user wasn't submitted (in case site doesn't do that already)
* let 99% of user who don't know what they did (they are in minority compared to number of times backspace is pressed to actually navigate back) retype their form if they don't navigate forward (which recovers their input, or should)
* add a third hand to human race so it doesn't take all the hands to navigate back from keyboard
* lure lenovo to restore the back / forward keys on thinkpads and make that implemented on most keyboards
The number of times I've accidentally hit the back-a-page button on my old X201, and wiped my in-progress form entry...
They smooshed those keys right around the arrow keys. The arrow keys I use to jumping around text.
That's worse than having a dual-use for backspace to me.
Backspace is far enough away that it's not hit by accident.
It really feels chromium team should install telemetry:
* how many unsubmitted forms edits lost by mistake
* how many backspace keystroke to do "history-back"
They should also consider alt-gr + left as "back-history", using two hands for this could be avoided this way.
It's absolutely amazing how many comments (500++ as I write this) that this "news" generated.
OCSP introduced a non-trivial delay (cited as ~300ms to a second) and there are also privacy concerns with it as well[1]. There's also the fact that it's set to soft-fail when the OCSP servers can't be contacted making it useless in that case[2].
Even if we ignore the above reasons for its exclusion from Chrome they do provide CRLSets which provide similar functionality but are pushed via browser updates (supposedly it's updated daily which makes it faster than OCSP). I'm not sure it's fair to say that they called it "useless because less technical users couldn't understand it". How else can CRLSets be explained then?
[1] https://www.imperialviolet.org/2012/02/05/crlsets.html [2] https://www.imperialviolet.org/2014/04/19/revchecking.html
Yes, notice that I called for OCSP force/required, not just OCSP soft-fail. I've been running OCSP required for years and never had any noticeable slowdown. Maybe there's one on the first connection of the day to a site or something?
Chrome's CRLSets are extremely limited and cannot begin to capture the full number of domains which have been revoked. They can get the most popular sites and that's about it.
OCSP stapling is the superior solution which will solve this in Chrome and other browsers, but for now OCSP is the only reasonable solution in a world where millions of SSL certificates have leaked and been revoked and you care about your privacy. CloudFlare alone revoked millions of certificates. These have indeed not all made it into the CRLsets.
An excellent summary of the problems is here: https://www.grc.com/revocation/crlsets.htm
Chrome's CRLsets contain less than 1% of current revoked certificates. I can't in good faith call that a valid solution to the problem.
[1] https://www.chromium.org/administrators/policy-list-3#Enable...
[2] https://www.chromium.org/administrators/policy-list-3#Requir...
That seems unstoppable for a long time now. There's no browser for power users in modern world.
Did you really use backspace and "move back" on regular basis? Why? You most likely aren't browsing the web with just your keyboard, so unless you are a leftie it requires you to either move your hand back to keyboard or move your hand across the whole keyboard and why would you do either of these when you can just move back with your mouses thumb button?
Because I'm far more likely to accidentally press the mouse thumb button, hence it's rebound.
I really hate the cargo-cult tradition of disabled menu items that don't tell you anything about why they're disabled, or what you can do to enable them. That's so frustrating, it's like the computer knows something you want to know to do what you want to do, but it won't tell you, because you have to guess.
Even better: Clicking on a disabled item should start a dialog to help you satisfy the prerequisites before performing disabled the action, instead of just ignoring you or telling you to go somewhere else.
I was able to trade alien pets to a highly advanced xenophilic civilization who wouldn't otherwise talk to me, in exchange for civilian access to their space! I felt like Harry Mudd sneaking through Klingon space with a space shuttle full of tribbles!
Mind you I'm enjoying it but many parts fall extremely flat. Like corvette balance and topped war scores.
I totally agree that Stellaris kind of falls flat atm IMO - the foundation is rock solid, the game looks gorgeous and the music just blows you away...but then you play on Hard and literally nothing happens for 6 hours. Eventually a war happens and instantly degenerates into a massive my-blob-chasing-your-blob-benny-hill-nonsense :/ And you can't even achieve anything because of 100 warscore limit! Apparently there are serious improvements coming soon though(Allegedly a lot of content wasn't polished enough before the deadline so minimal viable product happened)
In our attempts at multiplayer, we've already reduced ourselves to calling this "Yakety Saxxing".
You might be surprised by some actual wars.
EU4 has taken close to a decade to get where it is.
(I.e. "You just traveled to the end of a line and deleted a ". You could have done 'A[esc]dw' or whatever it is. Like I said, I'm bad at vim over here.)
Firefox does not have "Go back" accessible through the menu, although the "Backspace" keystroke works. Chrome also does not have a "Go back" accessible through the menu that I can find. Hovering over the Back button shows how to click to access history.
Does not work for me, FF 38. This feature used to throw me out of forms all the time, there has been a couple of years that it stopped.
Now the only FF keystroke that annoys me is the CTRL-q for closing the window, that conveniently is just by the side of CTRL-w for closing the tab, and does not save the window state.
Are you on Linux? Backspace to go back only works on Windows and OS X because this behavior was apparently un-Linux-like. You can change this behavior with the browser.backspace_action about:config pref:
http://kb.mozillazine.org/Browser.backspace_action
Here is a Firefox bug to make Backspace work on Linux Firefox the same as it does on Windows and OS X:
Do any of these options work to get your state back after accidentally closing the browser?(1) Is control-w significantly different than pressing the x on the Window? Just curious.
(1) https://support.mozilla.org/en-US/kb/how-do-i-restore-my-tab...
(Might not work if you delete history on close)
If this is true, it's a very convincing argument – I'd buy it for a dollar. In any event, my personal preference is to keep this backspace behavior, and my browser of choice (Safari) implements this caching perfectly.
At this point, not only am I ok with the accidental back navigation, but also I actively rely on things working this way. For instance, here on HN I routinely hit back and forward to see what people wrote, and Safari keeps my drafts perfectly well.
alternatively, just don't browse away when you have something filled in a form - just ask the user if they meant to do that
I do wish they had mad a user configurable flag rather than forcing people to use plugins to re-acquire functionality they had before.
if you have written a long paragraph it shouldn't just be lost because you accidentally pressed a key on your keyboard
I build complex web applications, which are stateful in many ways that don't involve forms (and involve things like websocket connections, etc). There's no way that the browser could possibly recreate that state by going forward.
So web apps of any complexity must necessarily intercept the delete key and break its functionality. And to the extent that the web increasingly consists of such apps, it means that "delete to go back" has long since being a standard feature that users can count on. That's just a fact on the ground. Time to remove the expectation entirely.
I incline to believe that users that were filling out long forms (and subsequently complaining about losing data because of hitting backspace) were filling out classic old-school static HTML forms, where automatic form restore by Chrome would certainly help.
That space is where complex, dependent-on-javascript-for-no-good-reason forms live. The kind often encountered when dealing with financial & public institutions. It's unlikely that automatic form restore is going to work for a lot of these (especially multi-step wizards & the like) so preventing an extremely common accident seems extremely prudent to me.
Of course there is—just don't delete anything immediately. All your state stays there like it's in its own (hidden) tab and gets lazily garbage collected "some time" later. If the user navigates back, there's nothing to rebuild as everything is still sitting there intact.
In any event, there are way too many replies to keep count of in this thread, but all I can say is that however Safari does it – it just works(tm). Accidental backspace has literally never been an issue for me since they implemented this. There are convincing arguments why backspace as a short cut is a bad idea to begin with, but old habits die hard, and I for one am very happy that Safari makes this work.
This all seems terribly expensive relative to removing a hotkey used by fewer than 1% of users.
Great work!!!
How so? Can you elaborate?
There is no rational defense for overloading a single button with two totally unrelated functions (navigation and text input). It's a design wart that people are defending on inertia alone.
In a better world, there would be much clearer defined intermediate interfaces between keyboard hardware, operating system, and final software (e.g. the native app, system shortcut, or browser javascript) performing the action for the user. Every interface would be clean and straight-forward, and easy for the user to adapt to her preferences.
In the world we live in, there are a confusing mishmash of 6–7 intermediate layers, which differ on each operating system and keyboard, some of which can be partially adapted by the user, but many of which can’t. These layers are poorly defined, not understood by users, and reach into each-other with various ugly hacks, and they often co-opt each-other’s keyboard shortcuts in arbitrary and confusing ways.
But along the same lines, I would like it if browsers waited 5 seconds before actually killing a tab that you closed. That would not only fix your problem (if you applied it to page history changes as well), but solve most accidental data loss problems, and most importantly for me, be the end to having to reload entire pages just because I held down Ctrl-W for too long.
EDIT: But if it were an option or an extension, I would gladly enable it to help for 99% of sites I use and either accept the occasional weirdness or blacklist it for Youtube et all
That's my main pain point (I really do think about frequently), but if others are having problems with AJAX-generated forms or whatever, it would help them too.
Another example would be losing your place in on an infinitely scrolling page (yet another web anti-pattern) because you accidentally closed the tab. I would appreciate having a second to correct my mistake before I having to do the"scroll wildly for 30 seconds until it catches up to where I was" song and dance.
But it is hard. No browser has succeeded in serializing full tab state: dom, forms, js, pending requests, etc. Chromium has been working on it and has large pieces of it but it is very very hard to get even close to fully correct.
You're letting perfect be the enemy of better.
IIRC Safari had some CMD+Backspace ("Clear history") that I kept on triggering the same way, as in a Cocoa text field it means "Clear from cursor to beginning of line"), but apparently they (thankfully) removed it too. The process, and resulting effect, was the same.
Basically in terms of usability this boils down to the key behaviour following an implicit mode (triggered by what's focused), making it hard to discover (or discovered the hard way) and unpredictable.
This is similar to Safari's braindead CMD+Z that does both "undo in a text{field,area}" and "reopen last tab".
† even though I myself know what happens I am taken aback every single time because it just breaks the flow, resulting in grievous annoyance. Therefore, props to the Chrome team.
i thought you were applying some subtle commentary about the change until i saw you agree with it.
Caching local content as drafts-in-process does in fact make a great deal of sense.
How to present that poses a new set of UI/UX considerations. Perhaps not grave, but significant.
I do hold strongly that the "backspace to navigate* motif is either profoundly broken or tied to unacceptably weak state stickiness of editing content itself. It's a callousness, capriciousness, and lack of consideration on the part of browser developers to users concerns (as opposed to advertisers) which I'm very hard pressed to defend on any basis.
What if my application needs to do something during data entry that can't be replicated by auto-populating the fields?
I believe this is taking power away from web developers and putting it into browser developers. That's a bad thing.
>Why not disable backspace only when some form has data entered? Why remove the feature in all cases when the reason only applies to some?
It still did not help them gain more than 4% marketshare.
Good thinking 99.
The more genuine way to phrase this would've been something like "Your solution: use different keys for moving away from a form and erasing data", which sounds a lot more reasonable
> The problem: Moving away from a form can result in data loss.
> Your solution: Make it harder to move away.
> The correct solution: Cache tab history completely and make it easy to move forward and backward in a tab's history by reusing cache and maintaining form contents.
(Source: https://bugs.chromium.org/p/chromium/issues/detail?id=608016...)
Safari does this, for instance, and yes it does happen that I accidentally the whole thing by hitting backspace, but it's OK because I know if I just go forward again, it'll all still be there – fully intact.