Slashdot Mirror
Opera Developer Comes With Address Bar Speculative Prerenderer Feature (opera.com)
Earlier this month, Opera announced a new interesting feature with Opera 43 developer that predicts the website you're about to go to. The company explains: There are two ways we can predict what page the user will soon load. When the current page tells us so, and when we can determine from the users actions that they are about to load something. Pages can use the tag, and for instance Google uses that for search results if they are pretty sure of what you will load next. When someone writes in the address bar they are humanly slow. Sometimes it is obvious what they will write after just 1-2 characters but they will just keep writing or arrowing through suggestions for millions or billions of wasted clock cycles. We expect this feature to results in an average of 1 second faster loads from the address bar. The company insists that this feature saves time and energy without compromising the security. What's your thought?
Wow, a whole second? Definitely should be implemented. Don't bother addressing the bug lists, only do new features. It is waaaaaay more fun.
So arrowing through the suggestion list is bad because it uses "millions or billions of wasted clock cycles", but downloading and rendering an entire website is perfectly OK?
I think they meant that it wastes the user's time, not that it wastes clock cycles.
Every failed guess is more bandwidth, server load, client load wasted.
Never, ever type "goats e..." in Opera.
Stop wasting my present time trying to predict what I'm going to do in the future. Nothing is more aggravating than waiting on my computer to finish whatever it predicted I would do, so that it can do what I was actually wanting to do. Because now, when I go to a different page than the one it predicted, it has to stop rendering the prediction and start rendering what I wanted. Just fuck off with this shit.
How do I turn it off?
That is all...
Once you are finished with typing in your url your connection is already saturated with loading the wrong page and will be for a few seconds more.
Is this the Opera development news feed now?
Why is Opera getting so much press lately? Did they hire a PR guy or something?
My eyes reflect the stars and a smile lights up my face.
Cynical me thinks this was done to game the browser usage statistics.
E.g. I type "o-p-e-r"...
Browser immediately opens "Download Google Chrome"* page.
*or other browser of your choice provided it doesn't have a mouth-frothingly insane "speculatively download and render potential malware" feature... because nobody ever left any security loopholes in any code ever.
Even the existing not-very-smart-bar feature in most browsers keeps wanting to google "http://mytestwebserver.local" or "192.168.1.254" instead of doing what I obviously want.
In a survey of 100 programmers, 111111 thought that duck-typing was a good idea.
people only knew about bookmarks. How often do I see someone type 'google' in the google search bar, click 'google' in the google search results and then type 'facebook' once on a new google page...
Our shared connection will be just fine with all the unnecessary page loads happening...
I'm thinking that the browser is constantly guessing what I'm going to try to load and is preemptively loading additional content in the background just in case it's right. It may eventually get the right answer, but more often than not it's probably going to be wrong; I'm just wondering how much additional bandwidth this feature is going to use.
Long signatures suck.
Or am i the only person who uses Google's advanced search page just to prevent the distracting mess that is typing into Google Search? Alternately, I just start with the word 'fuck' since that disables the feature, and when I'm done typing, I'll arrow back and remove 'fuck' and press enter.
I don't care in the least if my computer sits idle for a few seconds waiting for me, the user, to tell it what to do. I care very much if it arbitrarily decides to waste some of my all-too-limited monthly bandwidth incorrectly trying to second-guess my intent.
Dear Silicon Valley (or in this case, Oslo): Kindly fuck off and quit acting like the whole world has the same nice gigabit FTTP connections you've come to enjoy. Over half of the US (and more than half of the planet) doesn't have effectively unlimited high-speed broadband available. Please behave accordingly.
I don't want the browser second-guessing me, and by the way doing a ton of things which would be even difficult to explain to me!
What next: speculative execution of unloaded pages' Javascript? What could possibly go wrong?
FFS! Get first our current mess running smoothly and securely. Mirai anyone?
(I know, I know: Mirai isn't a browser problem. Still the mentality is the very same: "Lets fix tha... Oh, shiny!").
Shees.
Me? I just excised Javascript from my browser. I'm not using Opera (and given the above article I'll avoid it if I ever can), but if I had, I'd try to excise that kind of monstrosity too.
I still remeber when Opera was the best browser around - not that long ago, in fact. I was a loyal fan until they became yet another Chromium skin and the company ditched their entire codebase, spending ages to release a non-Windows binary in the process.
I now look forward for Vivaldi.
Their government logged and spied-on browsing history to be filled with unfortunate AI driven preload, when the algorithm land them into troubles.
Léa Gris
I am sick and tired of desktop systems trying to go the extra mile. Do NOT go the extra mile. Just do what I tell you to do. Do not try to predict, do not try to run the show. I run the show, you shut the fcuk up and do what I tell you to do, when I tell you to do it. You are not intelligent enough to know what I want or I need, so do not even try. Just stay in the background quietly, ready to do, as promptly as possible, what I tell you to do. Do not bother me with nonsense. Do not remind me of things, unless I explicitly tell you to do so. And, whatever you do, do NOT use oodles and oodles of resources for it. You are not the star of the show. You are just a servant.
Opera has this feature called Turbo which was basically a compressed web proxy - it really makes a difference on poor connections and was specially useful for me in the early days of mobile broadband data.
They really used to pioneer useful features back in the day...
This "feature" just screams UNINTENDED CONSEQUENCES.
If it's like the suggestions that Google puts up when I'm typing a search query, then it's useless and annoying to me, and I'd turn it off, just as I've turned Google's inept attempts at reading my mind. The problem is, the suggestions are almost always either wrong, or incomplete. It would be useful sometimes, if it allowed me to populate the search field with one of their suggestions and add text to the suggestion BEFORE they do the search.
In the case of Opera, what they consider 'obvious' has a good chance of NOT being what the user desires. People often type in a URL that points to something other than the main page of the site they plan to visit. Unless Opera can figure that out, (and that's pretty unlikely), then it will waste time and resources loading a page that the user may have no intention of visiting.
'The Economy' is a giant Ponzi scheme whose most pitiable suckers are the youngest among us and the yet-unborn.
They have a chance of 50% to get it right and if /. is already open, it gets to 100% that the next site is PornHub.
Don't fight for your country, if your country does not fight for you.
SLSIA.
What a great way to burn through all that extra bandwidth that you never use. *cough*
Just cruising through this digital world at 33 1/3 rpm...
FTA... We try hard to not load anything that would be bad to load.
Really? How are they assuring this? Is it phoning home to compare to a list? Downloading some huge list for comparison? This begs for explanation PRIOR TO any public experimentation.
This is a tech news site
The company insists that this feature saves time and energy without compromising the security. What's your thought?
Potentially wastes a lot of bandwidth and generates visit logs on sites I don't visit.
It must have been something you assimilated. . . .
thank you.
can just fuck off. Leave me alone.
Let me do shit MY WAY. I don't need your help, I reject your help.
Fuck your ads and your tracking and your deep mind AI bullshit.
Also all your "social" bullshit too. Fuck that shit in the ass with a bulldozer.
The more you push this shit the more aggressively I will push back.
This will waste time for me in two ways: in a small way, the flickering display of suggestions will distract me and instil doubt from the answer I already know I'm typing; in a large way, I expect it will act like autocorrect does in wordprocessors and webforms, and I will have to erase what was filled in (maybe a more than once) to get it to be what I intended to type.
What if the request modifies data? Sure, you're not supposed to do that with a GET request I guess, but it's possible that accessing a certain URL might trigger some large action (e.g. creating a zip of a server's content for development purposes). I wouldn't want that to happen every time I start typing an URL that looks like that one.