I'm on FF 30 and LinuxMint (downstream from Ubuntu), it seems to be working fine. My FF is whatever is in the repositories.
I remember that NYT graphic/article on the avalanche in Washington a couple years ago, I had to view it in both FF and Chrome to see all the graphic effects.
I just looked at it again, everything seems to work in Chromium. In FF, some of the graphics don't play automatically as they do in Ch, and some of the videos don't play at all. Chromium seems to play everything fine.
It seems odd that browsers on Ubuntu would be getting CORS errors — while the same browsers for everyone else would not. CORS, being a policy, should apply consistently, no? Perhaps you've changed a setting somewhere?
Either way, I've made a note to look into setting Access-Control-Allow-Origin headers on all the static assets we publish out to Graphics8...
OK, I've been digging into this a bit further. When I do a request from a remote machine, I do get an Access-Control-Allow-Origin: * header, but from my local machine, this header is missing.
I assume that the proxy on this network discards the header and thus breaks the site.
So it's not as easy as walking into your bank, sending a text message to the ATM and walking out with a lot of free money.
You actually need physical access to a USB port of the ATM which is hidden somewhere inside. So I guess the hack here is that the money inside an ATM is very well hidden and even protected by exploding paint barrels so it cannot normally be stolen even with physical access for a long period of time. But it is possible to get the Windows XP based software to dispense money from the safe by injecting a trojan through USB.
Unfortunately, the article does not make that very clear.
Also you have to get to an USB port, seems like attackers know for some ATMs where to cut a hole in for that. But the real problem is that the machine isn't isolated well enough through hardware design. Just imagine you could access and replace the HDD. While it may use an XP exploit, that's not the real problem.
Since the life time end of normal XP (not even the embedded version), those "XP will doom our money" news spawn everywhere for no good reason.
2. I'm running some realtime benchmarks on a number of hosted service providers. If I can get the geolocation it let's me add that to the benchmark results. Granted, a banner that provides info on that would be better than the bog-standard browser prompt.
3. Yeah, sorry about that. I'm interested to see where I'm sending traffic to so am capturing outbound clicks. It seems there's a bug on Chromium on Linux where the `event.metaKey` isn't set to `true` when it should be. See https://code.google.com/p/chromium/issues/detail?id=95874
From the rather vague text I can imagine they are doing matrix multiplication in JavaScript and assigning the end result to a CSS transform. Instead of doing multiple transforms in CSS which is slower. The end result is smoother HTML5 animations but I haven't yet figured out where they reinvented the web.
It's been a while since I looked at transition (and 3d transition) support from css -- but this sounds really counter intuitive -- shouldn't the browsers be able to make these changes themselves, in c++/c/hand-tuned machine code at that?
"A transformation is applied to the coordinate system an element renders in through the ‘transform’ property. This property contains a list of transform functions. The final transformation value for a coordinate system is obtained by converting each function in the list to its corresponding matrix like defined in Mathematical Description of Transform Functions, then multiplying the matrices."
On a side note: this looks like something that should work (transformation/transition) with javascript disabled -- we've (rather) recently gotten drop-down menus that don't require javascript -- seems like a bit of a step backwards to require javascript for (some uses) of basic transformations if that's being used as part of the essential ux for a page...
I'm sceptical. You can actually use the upper button on the left side of the app pictured in the article [1] to calibrate that app. So you could make it show whatever inaccuracies you wanted. And it's far from an "unfixable sensor problem" that "would put a dent in Apple's reputation".
Also, analog bubble levels aren't always perfectly accurate either.
According to users in the thread about this on macrumors, the error isn't linear, and if you calibrate it, you still get inaccurate readings once it's off-level.
> The issue is that it cannot be calibrated out. I've tried,
> it can't be done. You can zero it in one orientation but then
> the error is doubled in the opposite orientation.
Because the sensor inside every individual iPhone/iPod touch could be built slightly differently, to make the level more accurate for your device, you need to calibrate before use.
To calibrate:
1. Find a flat surface.
2. Hold your device upright (in portrait mode and home button at the bottom), put the bottom edge of the device on the flat surface, then press Calibrate button (the one with a target image on it) to calibrate the portrait mode.
3. Turn your device 90 degree clockwise (now it should be in landscape mode and home button on the left hand side), again keep it upright, and put the current bottom edge (long edge) on the flat surface, then press Calibrate button to calibrate the landscape mode.
4. Put the back of the device against the surface, then press Calibration button to calibrate the face-up mode
Apple’s built-in compass (which has not calibration) also shows many devices to be off. I’m not sure why that person chose a third-party app to show this off.
No to mention that it might not be an Apple problem, but an industry-wide problem. I've tried to replicate this on a brand new Sony Xperia Z1. It marks -2 degrees on a flat surface. Never occurred to me to test these things before.
Makes me think of the movie Mercury Rising[1], where a kid "wins" an NSA crypto challenge and is subsequently hunted by them because he cracked their code.
The developer console has a few CORS errors, though:
> XMLHttpRequest cannot load http://graphics8.nytimes.com/newsgraphics/2014/06/18/tigris-.... No 'Access-Control-Allow-Origin' header is present on the requested resource. Origin 'http://www.nytimes.com' is therefore not allowed access.
> XMLHttpRequest cannot load http://core2_euw1.fabrik.nytimes.com./info. No 'Access-Control-Allow-Origin' header is present on the requested resource. Origin 'http://www.nytimes.com' is therefore not allowed access.
I suppose accessing through nytimes.com would work, but it redirects to www.nytimes.com...