Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
Trust the World's Fastest VPN with Your Internet Security & Freedom - A Lifetime Subscription of PureVPN at 88% off. Also, Slashdot's Facebook page has a chat bot now. Message it for stories and more. ×

Comment Re:A Painful But Necessary Transition (Score 1) 221

So please tell me what is the point of Firefox even existing at that point?

Because we need someone who isn't an OS vendor or an advertiser making an open source browser and to champion open standards. But that does us no good if it results in an inferior browser.

Apple is indifferent, Microsoft would rather we go back to IE6, and Google would just as well take over the whole web and track your every move (and then they'd pull an IE6 on us just to be extra evil). Firefox is the outsider, the rebel.

There exists a suitable balance between customization and performance somewhere. But right now Firefox is increasingly intolerable because if you use add-ons, one tab slows the whole thing down. NoScript, Ghostery, uBlock, Anti-Adblock Killer, etc are all great. But all that work they do comes at a cost of further bogging down the single process. e10k means multiple processes, and that means we can layer on these things and have them going on in multiple tabs without grinding away on a single core in the age of 8-core workstations.

Comment A Painful But Necessary Transition (Score 3, Insightful) 221

I know this must've been a hard decision to make at Mozilla but I feel it's not the right one.

You do a great job of outlining the pros and cons. That said, I do have to disagree that this isn't the right move. I would argue that it is in fact the right move; it's just that the right move is the most painful move.

Firefox is a wonderful browser. But I fear we're losing sight of just how limited its legacy core is. Legacy Firefox offers no threading, no privilege separation, and no meaningful isolation between tabs or windows.

The browser-as-an-OS concept is no longer a gag, but the actual reality of how browsers work. Browsers are expected to do everything from executing code (JS) to graphics (WebGL) to video (HTML5, etc). Furthermore they are being treated as a multitasking operating system - via multiple tabs - with those tabs all competing for resources. Worse, some of those tabs may be hostile to the system or to other tabs.

This is something Legacy Firefox is ill prepared for, and in doing so it's the odd man out among the major browsers. Legacy Firefox is the MacOS Classic of browsers; a time-tested piece of software with parts going back to the earliest days of the Web. But like OSes 15 to 20 years ago, the world has moved on; it's akin to MacOS Classic going up against MacOS X/WinXP/Linux. The lack of real, preemptive multitasking and security has become a major liability, and becomes downright embarrassing when you realize that Microsoft of all companies was doing things like putting their browser in a low-privilege context a decade ago. Similarly embarrassing is the fact that a single runaway tab can take out the whole browser!

But all is not lost. Firefox can and is being upgraded with electrolysis (e10k). e10k Firefox has taken far too long to be developed - Mozilla should have been working in earnest on this a decade ago - but at long last it's here. And it finally brings with it all of the threading and isolation features that will make the browser safer and more reliable. Or more to the point, it will make the browser competitive in these respects with Edge/Safari/Chrome.

However just like giving up MacOS Classic meant giving up the OS's legacy applications, there is a price to pay for giving up Legacy Firefox: XUL and legacy add-ons. XUL is incredibly powerful, but the Moz devs have laid out a very good case for why it (and the rest of the legacy add-on system) can't be used with e10k Firefox. There's no concept of threading or safety; it's an API that has an unsafe level of access to the browser and can't handle being split up among threads. Its power is why we power users love it so much, but that power is dangerous. Worse, maintaining that power ultimately gets in the way of operating the browser with a safer multi-threaded environment.

And I won't dance around the issue: losing XUL and the legacy add-on system is going to be painful. Just losing the Classic Theme Restorer alone is going to be complete and total hell for this crowd. Never mind the other add-ons that enhance privacy, block ads, and do so many other nifty things. And not all of those add-ons can be remade for e10k Firefox, since they rely on a level of power that will no longer exist.

But you know what? It has to happen. Just like with MacOS Classic, at some point we have to stop using an archaic, unsafe environment origially designed around unitasking in order to move on to something better that can actually fulfill our needs. Even if we were to explicitly design/limit Firefox to Slashdot-level power users - and I would argue that doing so would ultimately be the end of the browser - it's still not in our interest to be using a browser that, at the end of the day, relies on cooperative multitasking. It's a crappy (if not horrific) execution paradigm for the real world. And while I admire the Pale Moon devs for what they're doing, Pale Moon just prolongs the problem. We still have to face this demon some day, if not today.

Is it going to suck giving up Legacy Firefox? Hell yes. But what other option is there? To continue using a browser core that can't handle a single rogue tab? No. We're going to have to grin and bear it, and then after the transition to e10k we as a community are going to do what we do best: make it better. And we'll do so by developing new add-ons for e10k, leveraging the strength of open source software development, and ultimately pushing Mozilla to better serve our needs. Without this change Firefox has no future, and even with e10k it may still have no future. But with e10k at least there's a chance.

Which is not to say that the Mozilla devs are saints. Far from it in fact. We wouldn't need Classic Theme Restorer if they didn't screw with the UI in the first place. But despite their painful inability to see why cloning Chrome is the wrong way to go, they're not wrong in this case. We need e10k, and to have e10k XUL has got to go. After that's done, then we can get back to beating some sense into the UI team...

Comment Re:Compares to Old Unlimited Plan how? (Score 3, Insightful) 62

I have to wonder how this compares to the old unlimited data plan (which I'm still on)

It's exactly the same plan, with exactly the same limitations, at exactly the same price.

Which wouldn't be so bad, except that everyone else is cheaper, and everyone else offers some amount of tethering. Which is damned useful to have in a pinch.

Comment Good, It Just Wasn't the Same (Score 1) 258

Good, I'm glad to hear they're putting back the Playmates.

Even though it's primarily a men's magazine, I still pick up an issue on occasion. It just hasn't been the same over the last year; the faux high-culture style of the magazine lost something essential when it lost the nudes. They aren't the only reason to read the magazine, but it makes for this interesting mix of wit, beauty, and far too many ads for liquor.

Comment Unfortunate, But Necessary (Score 4, Insightful) 99

Given that outside of the major publishers, Steam is treated as the de-facto marketplace for PC games, at first I wasn't happy with this move. But after giving it some thought, I think this is going to be for the better.

Right now Steam is suffering from two major problems that, as a casual buyer, make the store unpleasant to use.

  • Straight up garbage games. These are games thrown together using stock or stolen assets, with no real development effort, all in the name of making a quick buck. It's the noise in the overall signal-to-noise ratio of the store.
  • An extreme case of overchoice/analysis paralysis. There's too many small cap games, exacerbated by the garbage game problem listed above. 38% of all Steam games were released in 2016 despite the fact that Steam has operated as an open storefront now for several years. The number of games being introduced each year is growing, and consumers are having a hard time keeping up.

To paraphrase from Ye' Olde Wikipedia: "Having too many approximately equally good options is mentally draining because each option must be weighed against alternatives to select the best one". Which really, is kind of a horrific concept because it implies that choice (and competition) is bad. But outside of AAA titles with large marketing budgets and immense brand recognition, most of the games in the Steam store are unknowns, so customers are coming in and facing too many choices without nearly enough information to choose between them. Which isn't a problem if you already know exactly what you want (Call of Duty) and are just coming to the store to buy it. But it is a problem if you only know what kind of thing you want (a first-person shooter) and want to see what's available.

Essentially requiring a deposit on sales is going to lock out a lot of low budget developers, which taken at face-value is anti-egalitarian. But from a consumer perspective it's going to improve the store by cutting down on the noise. Games from developers who were likely never going to become successful in the first place now won't be cluttering up the storefront. It may keep the next ARK from being discovered, but it will also prevent the next The District from clogging up the store's search results. Developers lose, but arguably it's a win for consumers.

Which really goes back to a central argument about Steam and app stores in general: what should they be, a free-for-all or a curated store? The former allows everyone to participate, while the latter allows for a more structured experience. And judging from the consumer discontent, it seems that people would rather have the latter. Which at least for the PC is fine; the PC is an open platform, so it doesn't limit choice. It just makes it harder for a no-name developer to get noticed.

On a side note, I hope this also helps to curtail Early Access shenanigans. There are too many games that are being sold badly incomplete, and of those Early Access games, too many of them will never get finished. There's a dirty secret that I think everyone in the industry has had to re-learn the hard way: publishers suck, but having a middle-man funding game development means that at least games are more-or-less done before they are sold to consumers.

Comment Re:Translation... (Score 2) 78

Intel needs a new microarchitecture to replace Core. Core was an exceptional design, especially considering what it replaced and how much the early performance gains were like if you bought an early Nehalem CPU. Hell, even Core itself traces its roots back to the P6 microarchitecture after Intel abandoned Prescott (which was sold as the Pentium 4 back in the wild days of the clock speed wars) which goes back decades. It's pretty clear that Core is tapped out in terms of what can be squeezed out of it and Intel needs to go back to the drawing board like AMD did and use all of the lessons they've learned to make a new architecture.

It's not for a lack of effort on Intel's part. Despite the misconception, Core isn't one microarchitecture. Intel has revved the architecture several times at this point, always incorporating some of the latest theories and designs on branch prediction and the like.

The issue is that Intel's on the bleeding edge of single-threaded performance, and it's increasingly looking like there's nowhere left to go. Your options are either to increase the clockspeed - something that's proving impractical due to power consumption going crazy past 4.5GHz or so, even with FinFETs and other adaptations - or you increase the IPC.

And on the latter point, Skylake/Kaby Lake is already a wide, deep out-of-order architecture with more execution resources per core than most threads can take advantage of. The linchpin to IPC is out-of-order execution to fill these complex cores, and on that front Intel is already well into the diminishing returns stage. The re-order buffer is now 224 entries, which is deep enough that you're most often blocked by instructions you can't re-order around well before you reach the full depth of that buffer. And Intel's branch prediction is one of the best in the industry, so there's not much room left to improve performance by eliminating bad branches. Fast radix division? Already done. Low latency caches and basic instructions? Done. Decoding x86 into swiss-cheese and reassembling it into highly tuned macro ops? Been doing that for years. Meanwhile the damn backend is already 8 ports wide.

At the end of the day Skylake is already taking virtually every opportunity to increase IPC, and for most tasks they're likely close to the theoretical maximum IPC (in an information science sense) for those tasks. So what's Intel to do? It's one thing to say that Intel "needs a new microarchitecture", but what would that microarchitecture do differently from Skylake?

All that's really left for Intel right now is low-level brute forcing. This means tweaking data paths to avoid the already small numbers of times where the processor is bottlenecked by those paths waiting for data, implementing larger L3/L4 caches, etc. Otherwise the only path not yet explored is extreme speculative execution in the form of eager execution, which is massively wasteful from an energy efficiency standpoint, and still not all that great because branch prediction is so good.

Comment Re:Multiple Screens (Score 1) 59

I remember it was (at least in the stores) hyped that separate screens (like the controller that comes with the Wii U) would be sold separately and we would be able to have 4 screens on it. Well that never happened.

There was never enough bandwidth for that to happen. Communication between the tablet and the console happened via 5GHz 802.11n; the system was designed with just enough bandwidth to support 480p60 at reasonable distances. Adding a second control would mean, at best, you're down to 30fps. And even then, that doesn't account for overhead from the console now having to split its time between multiple tablets.

4 tablets was right out, and as far as I know, no promises were made to that. Nintendo only ever commented on 2 tablets, and that was more of a "well, it's technically possible with some drawbacks..." kind of comment.

Comment Re:Why is everyone copying mobile? (Score 1) 78

Why is everyone copying mobile?

Because that's how consumer electronics manufacturers were able to reach the next billion-plus users. The PC market became saturated, and what built on top of that is the mobile market. Those billion customers wanted something cheap, simple, and secure.

The traditional PC is great for many reasons, but it's not for any of those reasons.

Comment Corrected Title (Score 4, Insightful) 168

The story's title isn't quite accurate, so I've gone ahead and corrected it.

"IMDb Is Shutting Down Its Long-Suffering, Vitriolic Message Boards After 16 Years "

The contents of comments sections and message boards are getting worse year-over-year, and IMDB's are no different. Through no direct fault of their own, mind you, it's just that as the number of users on the Internet continues to expand, those users are living up to the Greater Internet Fuckwad Theory.

Comment Re:We ALREADY HAD cable TV without the box! (Score 1) 108

Think long and hard about why that won't scale for a minute

Well they did it for a number of years, but you're not entirely wrong. Having to do a truck roll every time someone changed service plans is a dumb way to manage access in the 21st century. Remote management of a data network makes all the sense in the world, especially as we get farther and farther away from cable's traditional multicasting-style roots.

(Comparatively, I suspect people would be up in arms if you had to go to the local cell phone store every time they wanted to change phones or plans)

Slashdot Top Deals

In a consumer society there are inevitably two kinds of slaves: the prisoners of addiction and the prisoners of envy.

Working...