Catch up on stories from the past week (and beyond) at the Slashdot story archive

 


Forgot your password?
Close
typodupeerror
×
Chrome The Internet IT Technology

Chrome Should Get 'Extremely Fast' at Loading a Whole Lot of Web Pages (cnet.com) 203

Posted by msmash from the for-what-it-is-worth dept.
Chrome is going to get a big speed boost -- at least for web pages you've recently visited. CNET: With a feature called bfcache -- backward-forward cache -- Google's web browser will store a website's state as you navigate to a new page. If you then go back to that page, Chrome will reconstitute it rapidly instead of having to reconstruct it from scratch. Then, if you retrace your steps forward again, Chrome will likewise rapidly pull that web page out of its memory cache. The speed boost doesn't help when visiting new websites. But this kind of navigation is very common: Going back accounts for 19 percent of pages viewed on Chrome for Android and 10 percent on Chrome for personal computers, Google said. With bfcache, that becomes "extremely fast."
This discussion has been archived. No new comments can be posted.

Chrome Should Get 'Extremely Fast' at Loading a Whole Lot of Web Pages More

Chrome Should Get 'Extremely Fast' at Loading a Whole Lot of Web Pages

Comments Filter:

  • Used to do this... (Score:3)

    by WoodstockJeff ( 568111 ) on Wednesday February 27, 2019 @04:41PM (#58190702) Homepage

    before everyone became concerned with "expiring pages" and made it so you couldn't go back without re-posting data. Then they made it so you couldn't even view the page source without reposting data.

    So, we're going back to what we had 10 years ago?

    • No. Expiring pages were a necessity of a dynamically generated internet. What we're doing is incorporating the cached system from 10 years ago without breaking what we have now.

      • Re: (Score:2)

        by PJ6 ( 1151747 )

        No. Expiring pages were a necessity of a dynamically generated internet. What we're doing is incorporating the cached system from 10 years ago without breaking what we have now.

        "A dynamically generated internet" is just code for breaking the fundamental design of the web, by serving variable content for identical requests based on ephemeral server-side state.

        Every application I've reviewed that required disabling back navigation or expiring pages you just visited was seriously broken. They "worked", usually, but they weren't designed by someone who knew what they were doing.

        • "A dynamically generated internet" is just code for breaking the fundamental design of the web, by serving variable content for identical requests based on ephemeral server-side state.

          Not quite. Either

          a) You're under the impression that requests are identical. They aren't. Requests are stateful and depend on the context with which they were made.

          b) You're under the delusion that a completely static internet depending on the request is a good thing. That would be fundamentally broken design in todays world. The internet isn't the black and white text crap that was featured on Slashdot last week.

          Speaking of Slashdot if you now click on this link: https://tech.slashdot.org/stor... [slashdot.org] you may n

          • Re: (Score:2)

            by PJ6 ( 1151747 )

            "A dynamically generated internet" is just code for breaking the fundamental design of the web, by serving variable content for identical requests based on ephemeral server-side state.

            Not quite. Either

            a) You're under the impression that requests are identical. They aren't. Requests are stateful and depend on the context with which they were made.

            b) You're under the delusion that a completely static internet depending on the request is a good thing. That would be fundamentally broken design in todays world. The internet isn't the black and white text crap that was featured on Slashdot last week.

            Speaking of Slashdot if you now click on this link: https://tech.slashdot.org/stor... [slashdot.org] you may notice a few things. You may notice that it's your user name in the top right and not mine. You may notice that you made a post. You will also notice that since you last visited that page there is additional content replying to you and calling out the absurdity of your claim. That comment was brought to you through the power of the fundamentally broken web as is your ability to reply to it.

            You're presumably a developer, and yet you can't conceive of any way at all that stateless requests can work for a web application with logins and changing content? Instead of thinking about it for a second or two, you lash out and call this apparently foreign concept "delusional" and "absurd". Nice.

            And you're right, Slashdot breaks the pattern with its volatile paging. I suggested a fix to this broken design years ago, back when they asked for feedback. I'd settle for them fixing the editors first, thoug

            • you lash out and call this apparently foreign concept "delusional" and "absurd". Nice.

              You're the one who started with generalities replying to my comment on dynamic content. I really don't know what response you were expecting other than that it is absurd, and that you posting here is proof of why it is actually a good thing to have.

    • No everybody else have been doing it for decades, but Google was just behind, possible because it creates fewer ad views they can bill for.

  • More RAM waste for benchmark "wins" (Score:1)

    by Anonymous Coward

    As always, Chrome continues to be the fattest memory hog of all browsers, and it only "wins" in synthetic benchmarks, which mean very little in real-world usage cases.

    I keep wondering why people are so keen on giving up their internet secrets to Google, while their spyware slowly eats up all their RAM.

    • Re: (Score:1)

      by Anonymous Coward

      Why isn't this rated higher. Chrome does not need more speed. It needs to have its memory use decreased by a factor of 10 or more. Just did the following test over the weekend on a 16gb system with 2gb page file.
      Opera 12.18 - 65 tabs - 680mb used.
      Chrome 72 - opened 18 of the 65 tab that were in Opera. All 16gb ra and most of the 2gb page were in use. Opening the 19th page had the kernel kill chrome processes.

    • Why have empty memory?

      • Re: (Score:1)

        by Anonymous Coward

        Why not?

        Things using up all my RAM only brought me misery. Upgrade to the RAM ceiling of your motherboard or computer and you're still fucked because the browsers eat everything again. Well, ceilings of 32 or 64 gigs are common now but don't forget that RAM prices exploded and might soon come back to 2011 or 2012 level..

        Things not using all my RAM? They cost me nothing. I don't benefit by going out of my way to fill the RAM. Doing so is as meaningful as "offers" to spend $80 so as to save $5. If I don't eve

  • Would you pay for this feature? Nor me. (Score:3)

    by shanen ( 462549 ) on Wednesday February 27, 2019 @04:50PM (#58190746) Homepage Journal

    As usual, someone else is insisting on what the users REALLY want. The financial models don't really make it possible to do otherwise, eh?

    Let me put it this way: If you had the option to donate $10 toward the implementation of this feature, would you? What feature do you actually want instead? Wouldn't it be nice if someone cared enough to ask?

    ADSAuPR, atAJG.

    • Ask away but as long as there is gold in them thar hills via the ad revenue stream your voice will be ignored. It'a a design feature.
    • Every effort must be made to maximize the speed at which the consumer browses when they are shopping.

      • Due to your brevity, I'm not sure if you were deliberately being insightful or it was some sort of joke. However, it is true that the economic model is driving the behaviors, but only indirectly in this case through the google's invasive ad business. Even from that perspective there are other options for new features that make much more sense than this.

        If the user is dissatisfied with the speed, they can buy a faster Internet connection or a faster computer or both. Much more than $10 in such cases.

        Error in

        • Reusing data you already have can be much quicker than waiting for the speed of light (or at least the speed of internet) to reload content from remote servers.

          • Re: (Score:2)

            by shanen ( 462549 )

            Yes, I understand what a cache is. I even know why additional working memory can improve cache performance, and the new computer I mentioned will probably have more memory for caching. If that was your question, then it is answered. Politely, even though such a question could be regarded as rude.

            On the other hand, if you have nothing to say, then perhaps you should say nothing.

        • Due to your brevity, I'm not sure if you were deliberately being insightful or it was some sort of joke."

          I'll admit that I was going for the chuckle by being so brief but it's a statement of truth.

          The average client with a storefront is not going to tell their customers that their computer is outdated and internet connection too slow to purchase their products. It is the developer's duty to provide cost savings to the client by reducing server load and to improve page performance for the customer to aid retention and thereby increase sales. It is the developer's duty to use every available tool in the box t

          • Re: (Score:2)

            by shanen ( 462549 )

            Hmm... I feel like there is some confusion about the nature of advertising here. Probably too complicated a topic? One of the dimensions is the old struggle between substance and presentation, and another major dimension involves privacy versus relevance. The more technical dimension of how fast the ads are displayed seems relatively minor to me. There's also an element of propaganda involved, insofar as improved caching is part of it. I frankly believe the liars are more concerned with the number of repeti

            • Hmm... I feel like there is some confusion about the nature of advertising here.

              I was honestly thinking more about e-commerce sites than sites whose primary source of revenue is advertising. I don't see any disconnect in the logic, however. Page loads faster, consumer browses faster, server loads are lowered, consumer appreciates experience more, consumer uses web site more. Profit????

              • Re: (Score:2)

                by shanen ( 462549 )

                I'm not following your logic. As long as the competitors offer roughly comparable shopping experiences, then there is no competitive advantage if the browser makes all of the websites look better. Of course the ceteris paribus is never fully the case, and the e-commerce website that has the best programmers has the advantage, but once again without regard to the browser. Actually, improving one particular browser may even upset the competitive balance if certain customers prefer the "wrong" browser, for wha

    • Software is not a perfect democracy. Not Chrome, not open source projects, not those by mega corporations, not those by single developers.

      You want something, why don't *you* put the effort in. Hire a developer to code that thing you want and integrate it in Firefox. In the meantime just because you don't want something doesn't mean that Google hasn't achieved great market share understanding their users.

      • Re: (Score:2)

        by shanen ( 462549 )

        That is the big-donor model. The one that has worked so poorly for Ubuntu and various other examples. Depends on having much deeper pockets than mine, but even deep pockets can be negated due to bad decisions by or ulterior motivations of the wealthy donor.

        I could explain more, but your unprovoked and dickish rudeness merits nothing. Nor shall I hold my breath waiting for you to come up with a constructive or useful thought.

        • That is the big-donor model.

          There's no reason groups of small donors couldn't do it. Or even individuals who contribute larger amounts of time/expertise, rather than money.

          • Re: (Score:2)

            by shanen ( 462549 )

            That is precisely what I am advertising. The specific mechanism that I advocate can be described as a "charity share brokerage". There needs to be a coordinating mechanism of some sort, and the objectives of the CSB are to address some of the problems of existing crowd funding mechanisms by supporting more planning and accountability.

        • The one that has worked so poorly for Ubuntu and various other examples.

          ... How poorly has it worked for a Linux distribution that came from a Debian fork and had it's limelight as the most popular distro on the market? Or are you talking about a specific case not a general one?

          • Re: (Score:2)

            by shanen ( 462549 )

            Mostly I'm talking about the failure of Ubuntu to become a serious alternative for "the filthy masses". Linux remains a niche market. Actually, the most successful new OS is probably Android (and yes, I know it has some links to Linux), but that's actually another bid-donor model. Again, it's driven by the good (or bad) decisions of the donor (whose real and non-charitable objective is to obtain more advertising revenue).

            No, I obviously can't prove things should have gone differently, but... I'm still convi

            • Mostly I'm talking about the failure of Ubuntu to become a serious alternative for "the filthy masses". Linux remains a niche market.

              And what feature could you donate that makes it a "serious" alternative for the filthy masses? Ubuntu didn't fail from a development for a target audience point of view, it failed against a large well funded and vertically integrated monopoly with a lot of money to throw at contenders in terms of marketing.

              Linux could be the perfect and most ideal system in every way and offer free blowjobs with every download and it won't ever become a "serious" alternative given it's absence of advertising, expectation su

              • Re: (Score:2)

                by shanen ( 462549 )

                You obviously don't understand or don't agree with my description of the problem. Notwithstanding, I have put my constructive suggestion on the table. So what's yours?

                • Notwithstanding, I have put my constructive suggestion on the table. So what's yours?

                  No my suggestion of the big-donor model is still perfectly acceptable. Your example of Ubuntu as to why it doesn't work has no relevance as the project failed to achieve what you are looking for for completely different reasons.

                  My suggestion remains unchanged. The original idea you had was getting the features you want, not gaining mass market acceptance for a product, something which has failed for multiple distributions following multiple different development models.

                  • Re: (Score:2)

                    by shanen ( 462549 )

                    You don't want to get me started on all of the problems with Ubuntu. Many years since I've been able to recommend it to anyone. I'm not even saying that the big donor made any of the huge mistakes that have doomed so many such projects to early oblivion. It's more that his priorities are wrong from a real world perspective, at least for the parts of the real world that I live in. One of my theories is that he's been overly influenced by his programmers, who push for flashy new stuff (mostly because it's mor

                    • I agree Ubuntu has many problems. I think they've been pulled on all arms at once in different directions. Let's create a desktop OS for the tablet which gets deployed on the cloud and make sure nothing works.

                      Your second paragraph mostly indicates that you have quite limited understanding of my suggestion, but you either don't care enough to ask for clarification or don't know the questions.

                      We are having a discussion. I have given you a reply based on my interpretation. How am I supposed to ask for a clarification I am not aware I need. The only person here who is capable of knowing if I didn't understand something you wrote is you.

                      Insofar as this discussion is probably timing out (as Slashdot does things), I feel like this is a parting attempt to be clear.

                      We can keep this up for another week before that happens

                    • Re: (Score:2)

                      by shanen ( 462549 )

                      Okay, now I feel I'm being dragged back to square one... Of course much of the problem is that this is so old to me that my viewpoint is jaded (and I take it for granted, too). Much of this goes back before I ever heard of Kickstarter, though now I see the CSB (charity share brokerage) as a solution to the most glaring problems of crowdfunding. Underneath it's really a human freedom thing, per my sig, and the REAL problem is that we human beings aren't very bright. Therefore we need to figure out ways to ke

                    • Right I understand it now. That would be a workable system but you're still at the whim of the 3-5 suggested improvements. Ultimately this is still someone else suggesting what users want with the exception that if they get it wrong there's the potential not to get funding at all.

                    • Re: (Score:2)

                      by shanen ( 462549 )

                      The reason to limit it to 3 to 5 at one time is because too many choices becomes overwhelming. Once that happens the choices are rarely free, but usually influenced by irrelevant factors, or even manipulated. However, that is also addressed by replacing the funded projects with others (rather than letting the project collect the excess jackpot donations (like the ones that destroyed the Diaspora alternative to Facebook)).

                      If a project can't attract sufficient donors, then that's fine, too. Don't forget the s

    • As usual, someone else is insisting on what the users REALLY want. The financial models don't really make it possible to do otherwise, eh?

      Let me put it this way: If you had the option to donate $10 toward the implementation of this feature, would you? What feature do you actually want instead? Wouldn't it be nice if someone cared enough to ask?

      You don't have to ask. If there's something you want to see in Chrome, go add it to Chromium. Yes, this will be a fair amount of work, and you'll have to work with the Google engineers who act as gatekeepers, but it can totally be done. If you aren't a programmer, get a few thousand of your closest friends together and each donate $10 so that you can collectively hire a programmer to do it.

      With closed source you don't really have this option, of course -- and of course with closed source you'd probably

      • Re: (Score:2)

        by shanen ( 462549 )

        Your tone sounds like you disagree, though you seem to be advocating just how I think the CSB should work.

    • Re: (Score:2)

      by AmiMoJo ( 196126 )

      I'm sure users will be mightily disappointed that their browser got faster. That's the last thing they want.

      I think we have found the bottom of the barrel here. This is rock bottom for criticism of Chrome.

      • Re: (Score:2)

        by shanen ( 462549 )

        I am NOT saying that saving time is intrinsically bad. If I was saying anything along those lines it would be that I would rather save time by seeing fewer ads, but the decision for this feature is clearly driven by a desire to shove more ads into my face. I certainly would NOT pay for more ads if I had my druthers. If the development of Chrome (or any other browser) was more clearly driven by the desires of the users, then I think the objective would be fewer ads overall, with higher relevance for the surv

  • I'm a little confused here. WebKit has had a back-forward cache for as long as I can remember, and Chrome forked off of that. How is this not already part of Chrome?

    • Firefox too (Score:1)

      by Anonymous Coward

      Firefox added this long ago, so hitting the back button went from fast (reload assets from disk and ram caches) to instant. I think this didn't last for long because it's all too easily defeated by the presence of any HTML5 and/or Javascript garbage.
      Perhaps Google's new feature will be able to be disrupted by a single script or ad picture etc. that invalidates the page and force a full re-render, except google will hand optimize its own sites/applications and Google AMP.

      I don't care either way, I'm lucky if

      • I feel like I've still got it over here on Pale Moon. If I back up to a form here on Slashdot, or on those other rare sites that actually use HTML+CSS instead of doing everything with Javascript, whatever I put into the forms is still there...

    • I'm a little confused here. WebKit has had a back-forward cache for as long as I can remember, and Chrome forked off of that. How is this not already part of Chrome?

      The WebKit implementation was incompatible with their multi-process model, so they had to rewrite it. Though I am kind of shocked they didn't do this YEARS ago.

  • Very common. Really? (Score:5, Insightful)

    by fahrbot-bot ( 874524 ) on Wednesday February 27, 2019 @05:29PM (#58190970)

    But this kind of navigation is very common:

    Not for me. When I'm viewing a page that has multiple sub-pages of interest, I tend to open a new tab for those sub pages. For example, one tab for the /. main page and a new tab for each article I read -- similarly for actual news sites. :-) Don't really know why I would want to go back and forth within a single tab.

    • You must be new to the internet. Back in my day, there weren't no pic-tures or scripts and everything went back and forth lightning fast just like the creators intended. And we had to remember where we came from iffin we ever wanted to go back 'cause there weren't no tabs. And we liked it that way! NGOML

      • You must be new to the internet. Back in my day, ...

        Not really... I used Mosaic (and compiled it from scratch) when I worked at the NASA Langley Research Center as a sysadmin for their supercomputer network - many Sun workstations, a Cray-2, Cray YMP, and 3 Convex systems. I was actually at work there the day the Morris Worm [wikipedia.org] hit.

        • Mosaic was amazing for its time. Wasteful tool bar at top, but a pretty sweet graphic while loading.

        • Heck, I was probably thinking more of the good ol' days of Gopher. I recall first using Mosaic and thinking it was kind of cool, but no match for the speed of Gopher. I still believe that images ruined the internet, even more so than the Army Of Lamers.

    • I have a macro key next to my left shift key, which I've mapped to the "close window" command. My style of surfing the web is to shift-click to open each link in a new window, not a tab, and use the macro key to close pages. I use the OS taskbar to manage windows (the way a window manager is supposed to work), not whatever custom tab management each application devises.

      I absolutely love this arrangement, but apparently I'm the only person in the world who does it this way. It pisses me off how many web p

    • Don't really know why I would want to go back and forth within a single tab.

      The answer is obvious: it's because it doesn't make sense for your particular workflow.

      For my own workflow, even with three browser windows full screen on three different monitors, I soon end up with so many tabs open, the tabs shrink to where I can't read the page title, and I start to lose my mental map of how to get back to other contexts I've recently visited.

      This typically happens when I'm involved in adding a lot of new inform

  • Yes, I've been trying to do this since 1985 and my DeLorean is running out of dilithium crystals. My obnoxious son chides me with these words:

    'You are old, father William,' the young man said,
    'And your hair has become very white;
    And yet you incessantly stand on your head -
    Do you think, at your age, it is right?' . . .

  • Personally, I wish Chrome would compress multiple back arrow operations in your browser and jump straight back to the desired page in history. It would save a lot of reload time.

  • I...thought they did this already.

    I have a better proposal. If I am on a web site with a login, and I take so long to enter an awesome post that I am auto logged out behind the scene, preserve the blather from the vaporized long text box somewhere...anywhere...so I can recover it. Back + login = clean form thanks fer nuthin'.

    In short, if I had an human assistant and said, oops, see I was logged out, put that text back in thanks, then they would do it.

    • Re: (Score:1)

      by Anonymous Coward

      This would be a useful thing to do - you can extend this by including the text area for the post you'll lose because you clicked on a user name by accident.
      But well, to Google..

      - It's a liability, for security or discretion. You will also lose your post because you used an "Incognito" window this time and the text area logging feature is disabled when in "Incognito". People will get burned by inconsistent behavior like that.
      - You "should" be using gmail, gdoc or "Google Posts" etc. that will auto-save the p

  • Retracing your steps forward? (Score:5, Funny)

    by TeknoHog ( 164938 ) on Wednesday February 27, 2019 @06:37PM (#58191238) Homepage Journal
    Only Alt-Right people would do such a thing.

  • Cache the Tylenol (Score:3)

    by Tablizer ( 95088 ) on Wednesday February 27, 2019 @07:06PM (#58191356) Journal

    Browser caching has been a source of many headaches in our org with regard to CMS's and dynamic web applications. You can put header tags that allegedly turn off caching so users always see the latest, but they don't always work right on all content types (HTML, images, CSS, JavaScript, PDF's, etc.). All browser brands we've tested have at least one caching bug.

    If browser makers haven't perfected current caching, then this new fancy-ass caching will probably have even more bugs.

  • Now Chrome will take up 80% of my memory rather than 50%

  • Wasn't this in WebKit (and therefor Chrome) 10 years ago?

    https://webkit.org/blog/427/we... [webkit.org]

  • Opera had this years ago (Score:4, Informative)

    by sgunhouse ( 1050564 ) on Thursday February 28, 2019 @12:23AM (#58192382)
    Opera Presto (which is to say, versions 7 through 12) had this years ago, though the early versions didn't handle dynamic pages well. It was one of their stronger features, and when they did change it most people wanted the option to put it back the way it was. It doesn't do anything for javascript/HTML benchmarks as it only deals with pages you've seen before, but it helps immensely on workflow.

    Best example, you do a search and get a results page, then have to look at the pages in the results to see if they are what you're looking for. So you load one page, then go back to the results page, then load the next ,,, until you (hopefully) find what you want. With this RAM cache (as Opera called it), returning to the results page is as fast as if you'd just switched tabs, so you don't need to open the links in new tabs (and thus don't use as much memory and CPU).

  • This is a feature Opera had for at least a decade before they rewrote it to just be Chrome.

    No modern browser comes close to Opera circa 1998, and it's absolutely pathetic what people think are features a browser should have.

  • ...is a quite dynamic multiplayer online game like slither.io? Saving the states of these kinds of pages will likely introduce bugs.
  • Hog the memory, memory hog.

  • That's the bit that's interesting to me. How do they know that going back accounts for 19 or 10 percent of the traffic?

  • I browse 12 hours a day. Improvements are good, I need more though.

    I have 8 cores, 16 threads. 32GB.
    Let me tick a box "insane fast mode" I don't care if it uses 24GB memory, I want preemptive tab updating in the background of tabs they know I open 70 times a day.

    I also want, since I browse like a drunken master, to not open a tab I already have open. If I have one open already somewhere else just switch that tab in its place. So I'm never on duplicate tabs. (They do this now, poorly)

  • I see a pattern with Chrome.. the faster it gets in loading pages with every update slower it makes our computer. Where are they going with this ?

  • As if reliably disabling cache in Chrome for development purposes wasn't difficult enough...

Slashdot Top Deals

"...a most excellent barbarian ... Genghis Kahn!" -- _Bill And Ted's Excellent Adventure_

Close