We Need More Science

Tech Time Warp of the Week: Return to 1974, When a Computer Ordered a Pizza for the First Time

On December 4, 1974, a hapless pizza restaurant worker answered the phone and heard a strange, robotic voice. “I’d like to order a pizza,” the voice said. “A large pizza, please. Pepperoni and mushrooms.” The worker asked for the address, but then hung-up when the voice took too long to respond.

The caller on the other end was Donald Sherman. But it wasn’t his voice. He had a rare disorder called Möbius syndrome, which results in facial paralysis and makes speech difficult. Sherman was calling from the Michigan State University’s Artificial Intelligence Laboratory, where he was using one of the very first text-to-speech systems to try to order a pizza.

Sherman was using a system designed by John Eulenberg and J. J. Jackson and consisting of a CDC 6500 mainframe computer nicknamed “Alexander” and a device called the Votrax voice synthesizer.

The first few places Sherman called thought it was a prank, but at last someone took the call seriously. Alexander could speak clearly enough to place the order, and the pizza was delivered.

It was the first time anyone used a computer to order a pizza, and more importantly, it proved that text-to-speech systems could be used to effectively communicate in the real-world. Forty years before Siri.

Tech Time Warp of the Week: Return to 1974, When a Computer Ordered a Pizza for the First Time

Sherman was using a system designed by John Eulenberg and J. J. Jackson and consisting of a CDC 6500 mainframe computer nicknamed “Alexander” and a device called the Votrax voice synthesizer.

The first few places Sherman called thought it was a prank, but at last someone took the call seriously. Alexander could speak clearly enough to place the order, and the pizza was delivered.

Facebook Is Making News Feed Better By Asking Real People Direct Questions

Getty Images

It’s a well-known fact that Facebook’s flagship feature, News Feed, is run by algorithms.

Essentially, invisible computations are going on all the time that automatically optimize future items you see on your feed, depending on the actions you take now—what you click on, what you like, what you comment on. The goal, as CEO Mark Zuckerberg told WIRED in 2013, is “to build the perfect personalized newspaper for 1.1 billion people and counting.”

But Facebook knows that it can do better than relying solely on these cold computations.

As detailed in a new piece on Backchannel by former WIRED writer Steven Levy, Facebook is currently running a focus-group-like program that asks people direct questions about News Feed items in an effort to improve post relevance. According to Levy, the pilot program started last August, testing just 30 Facebook users in an office in Knoxville, Tennessee.

It has now expanded to 600 people around the country, who are paid by Facebook to work answering News Feed questions four hours a day from home. Eventually, Facebook could offer some kind of direct questioning to its entire population of users.

The project works like this: each of these 600 Facebook users is presented with 30 top News Feed stories in a random order. Then they go through each story one by one. They can comment, share, follow a link, or choose to ignore the story. After that they answer eight questions about each item, including how much they cared about the subject of the story, how welcome the story was in their News Feed, how entertaining it was, and how much the story connected them to friends and family. Finally, they are asked to write a few sentences describing their overall feelings about the News Feed story.

Facebook itself acknowledges there are problems with how News Feed is currently set up. It’s already very good at delivering personal news from close friends—things like marriages, childbirths and vacations—but it’s also overrun with items that are sugary sweet and designed to tug at your emotions, which Levy has dubbed the “Dozen Doughnuts problem.”

The donut-y content contrasts with a “vegetables” of real journalism and hard news. When so many of those donuts are presented to you at a time, you’re bound to click on at least one item. And that click sends a strong signal to Facebook: you want to see more of the same thing.

Facebook could interfere. But especially in the case of News Feed, it prefers not to be heavy-handed. “We really try to not express any editorial judgment,” Adam Mosseri, News Feed product director, tells Levy. “We might think that Ferguson is more important than the Ice Bucket Challenge but we don’t think we should be forcing people to eat their vegetables even though we may or may not think vegetables are healthy.”

Preliminary results have already emerged. As expected, news from close friends—especially tagged and photo stories—has been consistently rated as highly relevant. But other things, like the meaning of a “like,” has proven to be more ambiguous. It could mean anything from the approval of a story to validation of a user’s connection to the author.

Unfortunately, so far, it looks like users are less willing to engage with “meaningful” stories or news, preferring anything that triggers a strong emotional response. But Facebook is hopeful that when it begins asking users about sets of stories instead of individual items people will start to reward informative content.

Though some Facebook employees are quoted in Levy’s story as wanting to do the right thing by fixing the News Feed, the real reason why Facebook may have a vested interest in making News Feed the best product it can be is glossed over. Facebook made $2 billion in ad revenue last quarter, more than two-thirds of its total $3.59 billion in ad revenue for 2014.

And where do those ads live? In News Feed. If the social network can crack the problem of what users really want from News Feed, they can presumably apply those learnings to ads, too—and make those ads irresistible to its users in the process.

How Imgur’s New GIF-Maker Stacks Up Against Other Tools Out There

There’s a new kid on the GIF-creation scene, and it’s a good kid. This week, Imgur launched a new “Video to GIF” feature that lets you enter a URL, adjust the parameters of a clip, and get a great-looking GIF or GIFV file within a few seconds. It puts that animation on an Imgur page, so if you have an account, you can share it on Imgur and host comments from all the wonderful Internet people.

Of course, Imgur is not the first or the only free GIF-creation tool on the Web. We take a look at the newest GIF generator, and see how it stacks up against some other offerings we’ve used.

Imgur Video to GIF

What it does: Converts any video URL to a GIF or GIFV. The newest GIF-creator on the block is also one of the most versatile—at least if you want to convert a video. The new Video to GIF option at Imgur allows you to just pop in a video URL—YouTube, Vimeo, Funny Or Die, or pretty much any other video service. You select an entry point for the clip, pick a length from 0.5 to 15 seconds with the scrubber, and even add a text subtitle if you want. The tool spits out a GIF if the file is less than 10MB, or a GIFV if it’s bigger. The results show up in an Imgur template, but you can open the file alone in a new window or tab by right-clicking on it. You can also embed it in a web page using an iframe, like this:

GifYouTube

What it does: Converts any YouTube video to a GIF or WebM video. If you’re working with a YouTube clip, GIFYouTube is probably the fastest way to get what you need; you just add “gif” before the “youtube” in the address bar, and you’re off to the GIF-making races. You can turn clips into GIFs that are 1 to 15 seconds long, and GIFYouTube gives you the option of viewing it as a WebM video file or a GIF. Your results show up in a GIFYouTube template, but you can also view the GIF by its lonesome if you copy image URL. GIFYouTube doesn’t do text overlays, and the GIFs are grainier than Imgur’s output, but it’s a really quick way to get it done from a YouTube page.

EZGIF

What it does: Lets you edit existing GIFs. EZGIF is a go-to post-production tool if you’re really serious about your GIFs. You can use it to create GIFs from images or uploaded video (it doesn’t create them from YouTube links). But it’s the editing options that are killer. You can resize and crop existing GIFs by their URL, and you can add effects to existing GIFs. The “GIF Effects” tab lets you apply filters, flip the GIF horizontally or vertically, add a text overlay, or change its speed.

Imgflip

What it does: Converts video files, YouTube videos, and images to GIFs. Imgflip is certainly a versatile GIF-creation tool, but it requires registration to get the most of it. There are tons of options: You can fine-tune the speed, size, and image quality of the GIF, and you can add text, crop, rotate the GIF, and even reverse it. But in order to create a GIF from a YouTube video, you need to create a free account. You also need to create an account to “claim” your GIF, or else it will be deleted after an hour. Imgflip’s free GIFs are watermarked, too—you need a $10-per-month Pro account to get watermark-free images.

Picasion

What it does: Converts a series of static images to slideshow GIFs (see above). Picasion doesn’t offer the video-to-GIF abilities or crazy tweakability of Imgflip, but it does crank out clean GIFs without an annoying wrapper page. No registration is needed, either. The service lets you set the size of the GIF and the frame-by-frame playback speed.

GifMaker.me

What it does: Converts images to GIFs. This one’s another quick-and-easy service for turning a bunch of images into a slideshow-style GIF, but you can only view your work on GifMaker’s page template. There’s no clean GIF-only URL, which is a bummer. You can adjust the size, speed, and repeat settings for your GIF with this service.

Deal With It GIF Creator

What it does: Converts images to “Deal With It” montages. This site only does one thing, but it does it well: It makes a pair of sunglasses fall from the sky, stop where you want them to, and then displays the text “Deal With It” at the bottom of the GIF. You can edit the text to say what you want, change up the text color, resize the sunglasses, and even add several more pairs of shades to the mix. Assert your authority!

Facebook Is Making News Feed Better By Asking Real People Direct Questions

Getty Images

It’s a well-known fact that Facebook’s flagship feature, News Feed, is run by algorithms.

But Facebook knows that it can do better than relying solely on these cold computations.

Imgur’s New GIF Tool Is Awesome. Here Are Some Other Great Options

Of course, Imgur is not the first or the only free GIF-creation tool on the Web. We take a look at the newest GIF generator, and see how it stacks up against some other offerings we’ve used.

Imgur Video to GIF

GifYouTube

EZGIF

Imgflip

Picasion

GifMaker.me

Deal With It GIF Creator

Vinyl or CDs: Tech Doesn’t Do Sentimental — Listening Habits Show What’s Next for Cloud

incredibleguy/Flickr

I am old enough to be of the generation that grew up with vinyl. Unlike the romantics of today I remember the scratches, the arguments over borrowing my brother’s records, putting them on my worn and generally abused record player and the reality that after the first few plays, even with that brand new stylus, the quality degraded to that “warm” or, in reality, muffled sound that I remember. The revival we see today is a choice, largely based on the sentimental feelings that we tend to attach to music. Yes, I miss the artwork, the feel and smell, but do I miss the general faff? Not really. Even with a vinyl revival, we have moved light years away in terms of the volume of recordings available digitally today. There are very few who would want to go back to how it used to be. The choice of vinyl is from the world of irrationality not the rational.

The point is that technology doesn’t do sentimental. Compact Discs wiped the floor with vinyl records because they could hold more, were more robust and the sound was consistently good. Downloads have taken this further mainly driven not by sound quality but convenience of format, i.e. the iPod. We are driving down the all too familiar silicon integration cost convenience curve which is underpinned by Moore’s law; skip the box, skip the media and the album. I just want the song. In fact, I don’t have the patience to download, I’m going to stream it… and so on we go.

How do we get from this to thinking about the future of cloud? Well, for me the two are related. My first ever job was working for the American mixed signal chipmaker Analog Devices. Mixed signal simply means analog and digital signals; analog being the “real world” and digital being that of a computer. So their trade was and is in the conversion of the real world into the easier to manipulate, more robust and generally cheaper world of digital – and back again.

When I first started there, the division I worked in made an esoteric analog-digital converter which translated the movement of transducers found on airplanes, tanks, steel mills and missiles into digital so control systems could make the right decisions. There were two methods for making the same thing: the “Hybrid” and the “Monolithic”. At that time, 1987, the Hybrids ruled the day. In those days, the term “hybrid” was used to describe a component where we had to use two different types of silicon to get the thing to work. Hybrid literally means “different elements”. Conversely Monolithic means “(on) the same piece”. The key point was that the monolithic product was cheaper to make – a lot, lot cheaper. It didn’t require the hands of a surgeon to place components on ceramic substrates with gold interconnects. It relied instead on an automated semiconductor process. You know which production method eventually won out.

Let’s move the argument to the current state of cloud computing. Hybrid clouds are very much in vogue and characterize a status quo where the current internet-based cloud doesn’t meet the requirements of all the applications we have. So, we marry the public cloud with private cloud architectures, where we traditionally have foregone flexibility and elasticity for security and control.

As happened in the past, for those analog-digital semiconductors, the world of hybrid cloud is also just a transition, simply a stopping point, not a destination. We currently have pools of computing connected by a variety of communication methods, and we simply haven’t worked out the process to reach the “monolithic” stage. But we will. Current cloud computing is generally made of three things, elastic CPU, RAM and Disk. These can be reached via the internet or some fixed network. Just as semiconductors evolved architectures and designs so that they could become “monolithic”, the same will apply to cloud computing, but writ large across the globe.

The evolution of cloud from pools out on the internet will therefore evolve from today’s triple play to a “quad play” as the fourth element, the network, is integrated and automated for both private and public cloud types of environments – or to give network back its original name, Inter Process Communication, (tipping my hat here to a long line of luminaries who have made this point, from the father of Ethernet, Robert Metcalfe, to Professor John Day pioneering the next evolution of the internet).

Cloud is not a technology but a dynamic way of optimizing your consumption with the availability of resources – the same evolution that has taken us from vinyl to CD to download. The challenge is therefore to restore the original vision of the founding fathers of the core technology that underpins the digital and with it the global economy. The arrival at a global, monolithic “platform of computing” where the network is the computer isn’t a vision or choice. It’s simply the realization of a world set out and supported by a model which has remained faithfully predictable for the last 50 years.

Matthew Finnie is Group CTO and EVP of Cloud Services at Interoute.