At that point, "quick and easy" math is largely irrelevant. If you can drive the MM x MM pixel display in the first place, you'll just apply an appropriate image filter to provide the best upscaled visuals -- which isn't necessarily an integer-multiple pixel expansion.

Observation: your web browser is already doing this constantly; a heck of a lot of web images are scaled these days (although generally downscaled rather than upscaled as we're talking about here).

Any attempt by the MPEG-LA to renege on this grant (a massive public announcement within this industry) would be blocked by Estoppel (at least in legal systems which have that concept). Plus MPEG-LA has the additional deterrent that the backlash would be exactly what they're trying to avoid, only worse: it would promote market fear of H.264 for web use, forcing one of the format's competitors to rise to the forefront.

Simply put: no. Software "averaging" may smooth out noise, but it will not add information that was not present in the first place. Missing dynamic range at the hardware is just not there to be recovered in software. In digital camera sensors, dynamic range is limited by saturation of the sensor's photosites. Once a photosite has collected enough photons, it registers maximum charge -- information about any further photons collected at that photosite during the exposure is lost. In fact, adding more photosites per unit area increases the per-photosite noise and chip areal overhead. Noise reduces dynamic range at the low end, and less charge capacity per photosite reduces dynamic range at the high end.

As another poster notes, you might change the effective exposure received by each photosite (perhaps by Bayer-array like neutral-density filtering). Or you can do what Fuji did with the S3 pro: make a matrix of photosites of different sizes/sensitivites to improve dynamic range. Fuji's sensor, while nice, has hardly taken over the digital imaging world.

On a more constructive note, Ctein wrote up a nice exposition on The Online Photographer about both near-term sensor technologies entering production and long-term avenues for advancement in digital imaging technology.

It's more than that: it's also for every case where the lookup logic is NOT handled by the database. Consider when queries are fielded by a separate service, such as a dedicated search engine (e.g. Solr/Lucene), leaving the database is relegated to just primary key lookup for full records/documents. At that time the benefits and tradeoffs offered by the various NoSQL solutions suddenly become a LOT more interesting, because that's what these tools specialize in.

This has already been done. The first I personally encountered such was in a then-new university building in the mid-90's. It had security panels at various points with individual illuminated LED display buttons. When not active, each button face was a rather enigmatic black. On the first press, the panels would "wake up", make (I kid you not) a sci-fi show warbling sound and scrambling animation on each keyface, then present a set of shuffled digits on the various keys. Each press reshuffled the displays.

This made perfect sense, since back in the 80's a sysadmin for the local university showed me the "breathe on the keypad" trick to see what keys are being pressed by users. Forget fancy Photoshop or IR imaging tricks...

Given how incongruous the randomizing keypads were at the time, part of me always suspected that they weren't actually security panels but part of a long-running installation art piece. The cameras wouldn't have been even a bit out of place. ;-)

Or perhaps more interestingly: is the Rb gene the newt's only defense against cancer? Specifically, have newts developed alternative cancer defenses that support Rb suppression during regeneration?

git cherry-pick newt/5f5c3c4f

Sniping is not a problem, it is a solution. You have to be a fairly naive eBay bidder to reveal your bid limit before bidding is essentially over. If you place a plain bid, you are vulnerable to 1) bidding wars from weak-willed folks (i.e. human beings) who don't set a personal bidding limit and 2) those who will "data mine" by incrementally upping their bid until they beat the top bidder, while at the same time trying to limit their upside risk. From a bidder's perspective, a snipe is ideal: it encourages you to do your research and set a firm bid limit up-front. I find this to be a vastly more relaxing way of bidding, since I've done my homework beforehand and avoid the temptation of stupid bidding wars. Likewise, it doesn't expose your bid (or even your intention to bid) until the last moment.

Frankly, I think sniping should be the standard bid mechanism for eBay auctions. I suspect that they'd never do it because it would reduce revenue by some amount.

[Pause for evil laughter omitted]

Note that MS under Gates' watch had successful (and ruthless) business practices to make sure that MS made heaping tons of money, even without being a major market innovator. It was often easier to let others innovate, then use a combination of financial might, second-mover advantage, and sometimes a bit 'o market leverage to move in and take over.

I'm frankly a bit shocked at how much this news item echoes Ballmer's earlier pathetic whinging about iPod and then iPhone. It's unacceptable that a major corporate CEO should sound like such a broken record when the message being repeated is "failure!"

How about "Rob Pike Decries Complexity of Java, C++" instead?

|Rob Pike| >> |Google Engineer|

To provide some concete figures here, 4x5 large format film (the smallest format considered "large") can be drum-scanned to produce images in the 200-300 megapixels equivalent range. Quadruple that for 8x10 film. Ultra-large format (ULF) formats are even larger, up to 20x24 inch film. Folks working with hybrid (analog/digital) processes with ULF mostly don't bother with full drum-scan resolution. Even for very large prints, there's just a stupidly large amount of information. And the real dynamic range and highlight behavior for black and white film blows all current digital sensors out of the water.

I had to laugh at that gigapixel photo of Obama's inaguration -- all sorts of artifacts along the stitch boundaries. A single wide-angle 8x10 (or hell, 8x20) image would have blown it away.

As someone who has shot film and digital side-by-side, yes. Film isn't just "disposable digital sensor rolls." Each kind of film has unique working characteristics. To quote Pascal Dangin from this New Yorker article:

I'll elaborate on that "ten dollars' worth of someone's ideas" bit: It's very loosely akin to being able to choose from a set of experienced digital post-processing artists, each with a distinct look. Film companies put a lot of money into tuning the characteristics of each line of film, whether color or black and white, for the desired results.

That's an amazingly ignorant statement. Computers and the software that runs on them are just tools. These tools are evaluated based on a collection of merits. Licensing concerns are just one of many factors that influence decisions to adopt a particular software system. Compatibility, up-front costs, ongoing costs, and suitability to task are some important others.

In many cases there exists exactly zero FOSS software systems that satisfy certain application needs. We're not talking about boring stuff like MS Office vs. Open Office, either (and even that can't pass muster in many organizations). Examples of verticals where FOSS systems are weak or nonexistent include: scientific software systems (very hit or miss; some outstanding FOSS projects in scientific verticals and some huge voids), machine control, color-managed print workflows, just off the top of my head. There's a world of other examples. In some cases, open source solutions exist but simply aren't up to the standards of the competition and the organization's needs.

In the end, it's never really a matter of "FOSS or die". It's always a positive choice to solve the problems that need solving, using the available tools. If FOSS tools aren't even available, then they aren't a choice. Even when they are available, they may not measure up as the best choice, at least to anyone who isn't playing FOSS zealot.

What on earth are you talking about? You're making up a random argument ("without copyright protection and enforcement") that has nothing to do with the summary or TFA. Those discuss the reprehensible and inefficient tactics of suing members of the general public for file-sharing, and warping of the law to suit the tastes of large rightsholders (e.g. the US' DMCA and similar). No mention is made of eliminating copyright or of not enforcing against corporations (who damn well should know better).

As far as revenue in the real world, many independent artists and small labels (often a single individual) have cropped up in recent years who are successfully selling non-DRM'ed downloadable music to the general public, either directly or via intermediaries (c.f. Amazon, Beatport, iTunes, etc., etc.). For the small artists, I expect they are likely doing vastly better than they ever would through a traditional recording company contract.

Note that the phrase in question will be both translated and summarized from the text of the law. I wouldn't read too much into it without a look at the original.

Also note: without some such clause, ISPs might be legally barred from useful and necessary activities such as addressing ongoing DDOS attacks.