A good explanation of setting speed limits at the 85th percentile. This is by a pro-motorist group, so you could claim bias. The other results on that google search are from government pages, both state and federal, and should be trusted.
For those too lazy to follow the links, countless studies have shown that the safest place to set a speed limit is the 85th percentile of vehicles on a given road. Going too slow has an increased chance of accident, and exceeding the 90th percentile also shows an increased chance of accident.
The only problem with that is that widening roads actually does not reduce traffic jams.
I suppose you could say I have one - actually I have three.
I started with a Garmin FR 405, got a FR60, and recently upgraded to an Garmin FR 220.
I am an avid runner, and they all track my workouts. The 405 and 220 are GPS watches. I have heart rate monitors (chest-strap, which I trust a thousand times more than a wrist-based solution at this point). The 405 was fairly large on my wrist, but the FR60 and 220 are actually reasonably sized.
They revolutionized my training when I started wearing them five years ago. I can get instantaneous feedback while I run, and I can track mileage and pace information over an entire season. I run faster now because of the Garmins, and my workouts are more intelligent.
Granted I only wear them while working out. I like not having to strap a phone to my body to get additional data, and I like that they are dedicated devices for the task. The FR60 goes months or years between battery changes, and the 220 can do a long weekend's worth of runs on a single charge. As just a watch the 220 can last weeks between charges.
The rest of them time I am content pulling my phone out of my pocket to check the time, see alerts, and so on. The Pebble is interesting (mainly because I see it as letting me know how important that last vibrate from my phone was), but I simply cannot justify it yet.
I am able to load https on Slashdot. You have to be a subscriber, but that is one perk. It costs me about $10 every few years, so I am willing to pay for a secure connection and no ads.
Based on some of the articles I have read, Google has thrown a lot of resources at the problem and now have hashes that are capable of identifying certain photos even if they have filename changes, resolution changes, and and so on.
It does not sound like too difficult of a problem - instead of relying on SHA5 file hash, run an it through a program that gives you an array of what the image would look like when displayed and then hash that. Use some margin of error to take into account compression, etc. and you could say with some confidence that one file is the image in question, even if the original JPEG is now a half-resolution GIF.
Of course having the resources to run that on every single image that goes through Gmail's servers is another issue entirely.
Agreed - every complaint about self-driving cars has been for the migration time when there are both autonomous and human-driven vehicles on the roads.
When you take human drivers out of the equation, and autonomous vehicles are the norm, utilizing things like mesh networks to keep other nearby vehicles informed, all of the complaints suddenly disappear.
Autonomous cars might wait at lights longer, and stop for more yellow lights, but imagine a line of vehicles stopped at a light all accelerating at the exact same moment and rate. Imagine vehicles re-routing around an accident with correct ratios going to alternate routes so no one alternate route gets slammed, leaving other routes empty.
I worked retail for a long time, including an Apple Store. I cannot remember the policies at Apple when I was working there, but most places will not take a verbal approval code.
If the person on the other end of the phone (generally you get to them by calling the 800 number on the back of the card) has the ability to run the transaction, they have the ability to clear whatever prevented the card from going through the first time. They would have to - they have to clear the hurdle before they can run the transaction themselves.
So policy at most places is that the telephone operator clears the issue (usually it is a daily spending limit that card issuers never mention) and then the store runs the card again. There was no procedure for manually entering a verbal approval code.
My memory of Apple Retail (this was '04-'06), however, is that they had almost every contingency covered. The POS machines all had USB modems attached so that in case the Internet went down at the store, credit cards could still be processed. We even had the old CH-CHUNK imprint devices when everything went pear-shaped. I do seem to remember having the ability to enter a manual authorization code for a credit card transaction. It is Apple Retail - there are supposed to be no hurdles keeping a Specialist from keeping a customer happy.
[Note: Written as someone who really liked the original series, but merely watched the prequels.]
So you can largely look at the prequels (Episodes 1-3) as "The Downfall of Anikan". Going into the movies you knew the outcome (Anikan was going to be a Jedi, only to be turned to the dark side). They had three movies to show this.
The problem? The real turn seems to happen somewhere between the end of episode 2 and the beginning of episode 3, and it really is not that believable. So you have three movies to show me the downfall of a man, and you choose to largely have that turn happen in the time between two movies?
From page 212:
Credit Cards With Intelligence? The Battelle Memorial Institute is studying the feasibility of a credit card with a built-in micro-processor. Such a card has already been developed in Europe, and will soon be tested. It is expected that intelligent credit cards will provide added security without requiring large computer networks.
Everyone who shopped at Target last fall saw how well that was implemented here in the U.S.