Actually, science works DESPITE many scientists not having integrity.

That's one of the reasons that I actually FAVOR creation science despite the fact that I think it is not a correct theory.

Because there are a ton of scientists out there who do fudge their data, or take their data and deliberately misinterpet it, beyond those who take their data and accidentally misinterpret it.

Then others go around and BELIEVE the studies, far in excess of the claims of the study. So then you have general academia all saying "well, science proves that lack of funding will destroy education." Or whatnot. Because they heard it, and want to believe it.

And there, to break through all the dogma, is the creation scientist who says, "uhhh.... a lot of your data looks okay, but you draw absolutely nonsense conclusions from it. Here, look at the limits of accuracy of your pb-pb dating, and consider...."

Now: Don't take my word for it... I'm making up numbers to give the gist of what I'm about to say, but 90 percent of the scientists will refuse to listen to what he said. 9% will attack him virulently. And 1% will say, "That's very interesting. I still think the theory is basically valid, but what would that imply, then?" And sometimes he discovers a fraud. Sometimes he discovers a new law.

So it is the creation scientists who battle the dogma, and help ensure that science moves forward.

So don't worry that science wouldn't move forward if scientists didn't have integrity. For it is very possible (and indeed, from what I have seen it is moderately often true) that don't, and it does.

You just have to figure out what to ignore.

More accurately, the "exciting" choice is the inexpensive choice, and inexpensive means more launches, or more money available for other programs.

If I'm already scheduled, that doesn't impact me unless it puts me on the schedule more often. If it only means OTHER astronauts get to go up too, well, screw them! :-)

Ironically, if you do text and drive, you are likely to become disabled.

Most likely not, actually. Driving is one of those activities where the risk is disproportionate - text/drink/etc and drive and those who suffer are more likely going to be third parties than yourself (who is protected by a myriad of safety devices and the car itself). Especially since a collision generally means you're going forward, and frontal impacts have been well studied and front crumple zones are big and absorb and distribute the energy over a long period of time. (Side impacts are the worst since there's very little crumple zone before intrusion into the passenger cabin happens).

What you describe is with the current situation, which is currently crashing all around us. Bus usership is skyrocketing in the city where I am (Hampton Roads). If that trend continues ... and I admit that I do NOT see national statistical evidence that bears me out, but if it continues, then the automobile may be less relevant.

Now, that said, I think there is something far more important than electric buses, and that would be electric metro trains that can dock at speed.

  First, because electric is best suited to short stretches without power, and buses and autos fail at that.

Second, because if you can use a system in which every trip is nonstop, one way, and moderately fast, it will clean out the short-hop airline industry AND some of your more extensive auto use.

Third, because if a reasonable public transit system is offered for medium- and long- trips, then short-hop electric becomes more viable.

While I had some problems with the stream for about the first 1/2 hour or so, they eventually got it stable.

  You do realize, of course, that that was likely hands-down, the most amount of streaming viewers of any single internet broadcast, right?

  But I am pretty sure there are meetings going on right now at Apple, and some streaming "experts" are seeing another side of Tim Cook...

It may be the most viewers (it certainly took down apple.com for a couple of minutes at a time when it just returned permission denied), but it's supposed to be flawless.

Apple isn't new to livestreams - this is probably the 4th or 5th they've done. It was started as a way to crash their new datacenter for iCloud - just have everyone with an iOS device or AppleTV go and watch it live and see how many bits they can push out.

And Apple can push the bits out, but their servers started collapsing under the load (which is unusual in an of itself). And the whole Chinese translation thing was completely odd.

And yes, it's something Apple typically prepares for because it's planned out way in advance. So I do find it inexcusable that the live stream had those flaws because Apple prides themselves for practicing, anticipating and handling practically every conceivable situation. They even have "audience member has emergency" as part of the contingency plan and a way to quietly remove them and give them the aid they need without disturbing or disrupting the whole event.

So no, sorry. The people behind iCloud and the livestream do deserve to be ripped a new one for these flaws. Because they are supposed to have everything down pat weeks in advance. About the worst I can excuse is if the stream stops and requires reloading the page to resume because the local CDN got overloaded. But that takes only 30 seconds tops.

No, they didn't. The total number of lanes declined. In essence, they went from having four lanes, with no dedicated turn lane, to three lanes for most of the street, expanding to four (with one being a dedicated turn lane) at every other intersection. So, for the bulk of the street, the number of lanes declined.

Four lanes with no dedicated left turn lane turns into three lanes when someone wants to turn left.

And add to that the chaos of having to do lane changes because people get stuck behind left turners (and the corresponding people who want to turn left but were in the other lanes to avoid left turners in the previous intersection) means traffic just gets all jumbled up.

Put in some proper traffic lights to help clear left turn lanes so people don't jam it when it fills up and spills into a straight through lane...

Basically all that happened was in order to build a bike lane, they had to reconfigure a bunch of intersections and in so doing also happened to improve traffic flow.

by Anrego (830717) * Alter Relationship on Tuesday September 09, 2014 @08:27AM (#47862303)

Is it ever chosen for new projects though? Would there ever be a reason to?

I get that for the type of programs originally written in COBOL it makes no sense to do a complete re-write. Things like accounting and payroll and inventory management are pretty static, and once you've got a working system, why change it.

But does COBOL offer any reason to start a new system with it?

Surprisingly, few new projects do it. But then they implement a whole programming layer on top of whatever the application is written in to do similar things.

So many crappy languages have been created over the years that really are meant to implement business logic without having to recompile the source code or have managers etc. learn C or Java or whatever because they want to implement a 10% discount on card members.

So you see lots of abstraction layers built up and maintained. And it's funny, since COBOL was designed for this - it's designed for managers and all those kind of people to understand. Accountants and the like love it (which is why the financial industry runs off of COBOL) because they can read it and comprehend what it does without being a programmer.

I think it plays heavily in to the decision, and the timing. Apple and Amazon are the two brand names that have a content and hardware eco-sphere, so they compete directly in a section of the market. Amazon is intent on taking more of that share, and the Fire phone is a part of the strategy.

The business cases for both are completely opposite, however.

Amazon makes devices to sell content and services. They sell Kindle hardware, Fire hardware, etc., at cost (practically). The goal here is to promote use of Amazon's music, movie, TV, e-book and app stores, which is why they're heavily locked down and discourage use of alternative sources of content. Sure you can sideload, but that's finicky and annoying enough that most users wouldn't bother.

Apple, though, sells content to push hardware. Apple doesn't really care about the content and is more than happy (generally) to let the content industry dictate prices. Sure Apple gets a cut, but that's just storage and bandwidth and handling the transaction and such. (It's not easy to be able to sell something and let users re-download it at will and keep transaction records. Hell, there's a lot of places that charge $5 to let you redownload your product up to a year later). The money for Apple is in hardware. The content merely helps move the hardware, so Apple puts a little effort into selling and promoting the content.

You're essentially saying "systems that rely on a key item are problematic. The attacker need only that key thing."
But all systems rely on a key thing. So you're not really saying anything at all.

Except that key thing is highly transient and changes frequently enough that if it's any length of time old, it needs to be verified.

It's why systems often verify email addresses once every year or so - not just to avoid spamming, but to make sure the person their sending stuff to is the same one.

Hell, you run into the same problems in coding today - you see people accidentally re-use file descriptors leading to all sorts of interesting bugs, or even people killing processes by PID only to realize the PID was reused!

My credit card company has just recently send new cards with the microchip.
Now I have seen the chip reader on 80% of the card readers I have seen.
And only Wal-mart has it implemented and working. Target has the new reader, but it isn't implemented.

It probably IS implemented, it's just waiting on the processor to actually flip the switch to enable the chip reader.

It's a bit more involved than just swapping out old hardware with new hardware - the whole operation of chip+pin is completely different. And it's different enough that POS systems that integrate in credit and debit processing software MUST change as well.

Presumably, Home Depot and Target are busy rewriting their charge card processing software to handle the new format - but in the mean time, they can certainly have the hardware down and ready for it. Then there's retraining as well - because when doing stuff like returns and such, often because the POS machine stores the card, it can reverse the charge without the user doing anything more than signing the slip. But chip+pin can't do that, so the system needs to ensure that it can send the refund request properly to the machine and the customer service folks need to be trained to tell the user to insert their card. (And no, it won't accept any card, it generally must be a card on the same account, so if you made a purchase specially on Visa and forget it, you can be stumped when the normal MasterCard you use fails).

The hardware's just the basic part. It's the whole POS integration's that the difficult part. Wal-Mart has it easier because they're always tweaking stuff and policies so they've probably live-beta'd the new changes all the time.

In fact, the retailers most likely to have the new machines working are the small ones, provided they upgrade (usually as part of equipment refreshing or broken replacements), since there's no integration.

Heck, my local comic store has an interesting charge machine - they had their old one replaced with the same model because it broke, then earlier this year, that got replaced with a brand new one with a bright shiny high-res LCD. That apparently is probably running the old software in an emulator because it provides all sorts of status information on the screen, but the actual credit card processing uses large pixelated letters that emulates the low-res old screen. No, it's not simply using a large font, it's actual pixelated text where you can see the individual dots rendered. (And you know it can go finer because the stuff around the emulated low-res LCD is in color and fine text).

I hope that whatever Romero is doing doesn't turn out to be Free-2-Play or co-op or with multiplayer focus.

The beauty of his best games was that they were single-player, with some very fun multiplayer as a bonus. The current gaming industry mode seems to be co-op or multiplayer primarily with maybe a very short single-player campaign thrown in.

I understand that this trend started primarily as a way to prevent some kid in Estonia from having a nickel in his pocket that didn't belong to the gaming industry, and I don't fault them because their nature is to be money-grubbing monsters who basically hate their customers. But somehow, the great single-player games managed to make a nice profit. Nice enough to finance a stinker like Daikatana.

Actually, multiplayer was EXTREMELY popular because it turns out people didn't really want single-player. It's been that way since DOOM was released in the early 90s. Remember the early builds that had a nasty habit of taking down office networks? Turns out a LOT of people wanted to play against other people, rather than bots. And it was extremely popular with people willing to schlep their desktop PCs, monitors and all the peripherals to other people's houses in order to do "LAN Parties". (And this was back in the CRT days, so hauling your 17" monitor was generally a big deal).

This multiplayer aspect is what drove DOOM, Quake, etc sales, not the single player campaign. In fact, there's very few first-person-anythings that have (half-) decent single player mode (the Half Life, Halo and Portal franchises come to mind), because most players don't do the campaigns, ever, they just get it and hit "Go online".

So since then, multiplayer competitive generally was the order of the day because tons of people played it.

It was maybe a little under a decade ago that people started taking cooperative modes seriously - seeing that while a good majority of people play multiplayer only, there's a sizable chunk who do play the single player mode and rarely do the multiplayer, or at least do it a lot less than the core group.

The vast majority of people you see who buy the latest shooter basically do it to play multiplayer. Be it against random people on the Internet, or with their friends

Never mentioned anything about ajax or responsive etc, only about support for SNI. Also, but of selective quoting on the part about loosing XP customers, you forgot to include the bit where I said "would rather loose XP based people that those who use the latest Chrome builds etc and won't connect because of security alerts". - in other words if one of those two sets has to be lost for some reason, I would select to loose the older XP set. Obviously it would be best to loose neither, but given a enforced choice then the XP users are toast (and they count for less than 0.5% of our users, so really not going to loose any sleep over that)

There should be no reason anyone's using XP SP2 as that's been obsoleted a LONG time ago in favor of XP SP3, which works just fine. It was only recently that XP SP3 fell out of extended support.

So you're not likely to lose too many people (unless you cater to people running ancient unsupported highly vulnerable versions of Windows). I would think the majority of XP users would be running SP3 for a while now.

IBM addressed this with POWER7 and newer in a fairly innovative way. They have an option called TurboCore mode which turns off half the cores. The ones still running can use the disabled core's caches, and because of the space available for heat dissipation, clock speed could be bumped up. The result was half the cores, but almost the same performance due to the faster clock and cache available.

Intel has it too - it's called "TurboBoost". Basically if the CPU is not under thermal restraint, and the other core is basically idle, then it will go into a practical shut-down state while the clock on the active core is boosted. This is designed less to save per-core licensing and more for single thread performance - if the workload consists of one thread, then rather than have the other core stay there idling for threads that aren't coming, it can be put in low power and the one running real code can be boosted to speed up

Takes into account a lot of PC workloads are still single-threaded but there are a few that can benefit from more cores.

I wonder how much they are duped. I repeatedly see spam-ads in forums for "moneyrelative.com" except that isn't what it is. It's money ... ummm... something else.

And the spam ads always look like somebody misplacing a comment from a completely different forum, "uhh, yeah, what kate said, it's amazing how you can earn yada yada dollars in only a month..."

But they never say what you do for the money.

Now, I *DO* earn yada yada dollars in a month, because I am drafting working drawings for real bridges that get built. I KNOW what I do is valuable. But if you are employed... no, let me adjust that.

There are LOTS of jobs out there for which you can be paid a reasonable looking wage for no particular service other than time spent. How is that? Because the purpose you are serving is reducing the risk to your "employer". So you can get a job flipping houses with no money invested, taking 10% of the profit, but ALL of the loss (that is, bankruptcy) if the market crashes. And that is how your employer steals money from banks. Or you can get a job running drugs: your risk is 20-to-life if caught and you DON'T rat out your employer; or death to you and your family if you do. He makes millions, you make a good , what, maybe $70k in a year. And so on. Now, this money-relative-dot-com (name changed to protect the probably guilty) , I half suspect is nothing more than recruiting US accomplices to the Nigerian gangs. But I don't know, and I'm not willing to go over to the site and risk my browser (in case it's actually recruiting spam bots).

But I rather suspect that most people know that it's illegitimate when they sign up for it. But they hope that it's legit, and even if not, they hope that somehow they'll slide. Because that reeeeeeaally need the money, and they've maxed out their credit cards, and a person shouldn't just give up when they can do SOMETHING to get money.... ... but a lot of times, they set themselves up for that, deliberately, before, when they chose to get the things they wanted, knowing that they'd have to go to the shady side to pay it off.

It's not an interesting quote, it's the truth.

People rag on Apple etc., when they only do dual-core, and yet, it's the reason WHY they do dual-core.

Quad-core (and "octacore") are a joke because the power consumption is high. Even at 100mW/MHz (=1W/GHz) which is what ARM does, the sheer speed of modern SoCs mean they dissipate a fair bit of power.

E.g., a 2.5GHz Snapdragon with 4 cores is 10GHz or 10W of power consumption. Of power-efficient ARM at that, which is 10W of heat in a small area that has very little way of being dissipated (there's memory chips over the SoC, and PCB underneath).

You're going to thermally limit quick. I've seen the calculations and it's roughly two cores going full bore 100%, while the third and fourth can only manage 50% to be at the most processing you can have without thermal shutdown. This was running basically right at the max junction temperature with room temperature cooling around it. Add in PoP packaging and enclosed packaging and your max performance is much lower.