new web technologies

Verizon's Planned App Store Is The Dumbest Carrier Idea In Some Time

Reporting from the world of terrible ideas, The Information claims Verizon Wireless is scheming to create a new Android app store to compete with Google's own Play Store. 

The carrier's plan is currently at an early stage, according to reporter Amir Efrati, who adds that there's no concrete launch timeframe. As a result, it's hard to know whether Verizon's plan, if it exists, is much more than a trial balloon at this point.

Still, let's hear it out. Verizon is supposedly wheeling and dealing behind the scenes, hoping to give its upcoming store a global reach by getting other wireless operators and tech companies on board. For now, its supposed plan is to use data—like user location, time of day and social data—to power app recommendations.

Cue the eye roll. 

Verizon Makes A Play ... For Your Wallet

Telecoms have long wanted to prove that they’re more than just dumb pipes channeling in data for other companies’ benefit. Unfortunately, that ambition usually—make that, "almost always"—translates into unwanted features and pricey services forced on consumers.

No one likes having their Internet service or real estate on their phones held for ransom. One big reason the iPhone took off the way it did was that Apple managed to wrest away control of the phone from the carriers, much to the benefit of users everywhere.

In that context, this reported plan looks like little more than a lame money grab. Verizon subscribers already have access to Google Play, and they can even pay for those apps from their wireless bills. So an alternative app store doesn’t seem to fill any need or solve any problems for people.

There’s also no reason to believe that Verizon, whose own stock apps are for the most part awful bloatware, really understands very much about app development. It would have to, if it wants build relationships with the developers who would populate its store.

Let's not forget that Verizon tried this before in 2010, with little to show for it but a big bucket of fail. That store, aimed at Android and BlackBerry users, limped along for three years before the carrier finally pulled the plug in January 2013. 

Follow The Money

If the Verizon plan is a reality, it seems to reflect a collective fear among carriers that they've failed to dip their buckets deeply enough into the app revenue stream. Sprint also just announced its own app store, called App Pass, a subscription service that assumes users will be happy paying $5 a month to use a collection of paid apps selected by Sprint. That might be a bargain if you believe Sprint has your best interests at heart—as opposed, say, to those of its corporate partners.

Unsurprisingly, money likewise seems to be at the heart of whatever Verizon thinks it's cooking up. The Information reports that its app-store strategy is a direct response to Google scaling back its app revenue-sharing arrangement with carriers and hardware manufacturers.

How unlike the mobile carriers we've come to know and love.

Lead photo courtesy of Shutterstock

UPDATE: Verizon denies that it's readying an app store, but The Information's Amir Efrati is sticking to his guns, tweeting, "Verizon spokeswoman says 'no plans' for app store. I and @theinformation stand by this report 100%..." 

Date: Wed, 20 Aug 2014 01:19:34 -0700
Author: :: Category: Mobile

back to top

Want A Job In Tech? Become A Linux Expert

Over the past 20 years, Linux has ascended from a Finnish student's hobby to the world's most dominant operating system—one that runs everything from high-performance computing to mobile. Yet companies still can't get the Linux talent they need, at least for the price they're willing to pay for it.

See also: Keep Learning Linux—It's The Future

To help meet this demand, the Linux Foundation today announced a new Linux certification program. But you have to wonder whether this is anything but a stopgap.

The Need For Linux

A quick glance at job postings suggests that demand for Linux jobs has declined after a peak in 2009:


But the truth might be otherwise.

After all, Linux servers now represent 29% of all server revenue, up five points when compared with the fourth quarter of 2012, according to IDC. And that's just paid Linux. Digging deeper into IDC's numbers, while Microsoft Windows still claims over 50% of the server market, Linux—paid and unpaid—servers are rapidly closing the gap.

Source: IDC

Some of that unpaid Linux adoption comes from the cloud providers like Amazon Web Services, which is built atop Linux. Ditto Google and every other major web company. That adds up to 30% of all Linux servers shipped in 2014 going to cloud services providers, all of which need Linux talent.

See also: Why Data Scientists Get Paid So Much

But there's another reason not all Linux jobs find their way into that graphic above: not all Linux jobs are labeled as "Linux."

Android, the world's most dominant mobile operating system, is Linux. So is Amazon's Kindle operating system, a variant of Android which, in turn, is a variant of Linux. The list goes on. Were we to update that chart with all the different names Linux assumes, we'd see dramatic, robust demand for Linux.

Who's Hiring Whom?

Or we could just ask those in charge of hiring budgets. According to the Linux Jobs Report, released earlier this year, employers are aggressively looking for Linux talent, but too often coming up short. Nine in 10 hiring managers said it’s “somewhat” or “very difficult” to find experienced Linux pros. 

See also: Linux's World Domination Is Complete

Among other findings:

• 77% of hiring managers have “hiring Linux talent” on their list of priorities for 2014, up from 70% in 2013;

• 93% of hiring managers plan to hire a Linux professional in the next six months; 

• 46% of hiring managers are scaling up their plans for recruiting Linux talent over the next six months, a 3-point jump from 2013; 

• Among the hottest areas of expertise are: systems administration (58% of hiring managers are looking for this expertise), Linux application development (45%) and systems architecture/engineering (45%);

• 86% percent of Linux professionals report that knowing Linux has given them more career opportunities, and 64% say they chose to work with Linux because of its pervasiveness in modern-day technology infrastructure.

Training Up A Generation On Linux

Which is what makes the Linux Foundation's news interesting. The Linux Foundation had already introduced new training programs earlier this year to increase Linux expertise. 

Now it's giving employers an industry-standard, neutral way to evaluate talent. Red Hat and others already offer their own certification programs, and these will continue to serve a useful purpose. But having a program delivered by the Linux Foundation that spans different distributions is a welcome addition.

See also: Why Uber, And Not HP, Is The Future Of Technology

The new certification program exams include designations for Linux Foundation Certified System Administrator (LFCS) and Linux Foundation Certified Engineer (LFCE). The cost is a reasonable $300. 

Even so, more is needed. All major trends—cloud, mobile, Big Data—are built on Linux, for the most part. While this certification program is a great start, what we need even more is widespread teaching of Linux as part of core curricula at the university and even K-12 levels. 

I'm not just talking about engineers, either. As I wrote recently, with software eating the world, any company not managed by people experienced with technology is handicapped. The same is true for all levels of the org chart, and Linux is central to much of what we build.

We need a generation learning Linux, because we already have a generation building the world on Linux.

Lead image by Michal Dočekal

Date: Wed, 20 Aug 2014 12:17:39 -0700
Author: :: Category: Work

back to top

Twitter Tightens Photo Policies Following High-Profile Deaths

Twitter is tightening its policies on harassment and graphic imagery following the deaths of Robin Williams and U.S. journalist James Foley, who was reportedly beheaded by jihadis in Syria. On Tuesday, the company stated it will now remove certain images of deceased individuals at the family's request.

Twitter announced its clarified guidelines on images of deceased Twitter users soon after graphic images reportedly depicted Foley's beheading by members of the Islamic State of Iraq and Syria began circulating on the social media site. Twitter quickly deleted the images as they circulated and suspended many of accounts that were sharing them. Some of the accounts however, were reportedly reinstated.

Deleting images at the request of families is a conservative change for Twitter, which has proved a crucial tool for documenting and distributing news and information happening around the world. People regularly share graphic content and images, and the company suggests users mark media containing sensitive content to prevent it from automatically displaying in tweets. 

Under the new guidelines, Twitter won't delete just any image, however, and it makes it very clear in the updated policy that images can be removed in certain circumstances like immediately before and after death, but if images are newsworthy, they may not be scrubbed from social network.

In order to respect the wishes of loved ones, Twitter will remove imagery of deceased individuals in certain circumstances. Immediate family members and other authorized individuals may request the removal of images or video of deceased individuals, from when critical injury occurs to the moments before or after death, by sending an e-mail to When reviewing such media removal requests, Twitter considers public interest factors such as the newsworthiness of the content and may not be able to honor every request.

The new policy also comes on the heels of Robin Williams' death on Aug. 11. His daughter, Zelda Williams, received disturbing images and harassing comments following her father's death that eventually led her to quit the social network.

In response, Twitter suspended the accounts that harassed her and Twitter promised to "evaluate its policies" around tragic events to prevent future behavior. 

Lead image by Anthony Quintano

Date: Wed, 20 Aug 2014 09:07:17 -0700
Author: :: Category: Social

back to top

Google Wants To Control Every Character You Type

Imagine a typeface that unites all the world’s languages. A publisher could print a book in Arabic, Cherokee, Egyptian Hieroglyphics and more—all without swapping out fonts.

That’s what Google is attempting to accomplish with Google Noto, a free font family that currently supports 96 languages, and aspires to support them all. Noto stands for “no tofu,” where tofu are what font professionals call those empty white boxes that appear when a character isn’t supported in a typeface.

Started in 2012, the Noto font family now spans 100,000 characters. This month, Google partnered with Adobe to release a new collection of fonts—Simplified Chinese, Traditional Chinese, Japanese, and Korean—which can be used separately or bundled together in one file so a writer can switch between languages without switching to another font.

This isn’t the first time a technology organization has made an effort to universalize the world’s fonts. In 1987, the Unicode Consortium began developing a way to make computer type compliant with global languages. The result was the Unicode Standard, a system of character codes designed to eventually represent every character in every language on Earth.

See also: Google's Gender Diversity Push Is Paying Off

Unicode Standard wasn’t really adopted by Web browsers until 2008, and still isn’t a complete solution for capturing all the nuances of global languages in a culturally sensitive way. It was designed with character universality in mind, not particular languages. So the playing field is still ripe for a new global typeface. Whether Google is the right team to take the field, however, is debatable. 

Pakistani-American writer Ali Eteraz told NPR that he isn’t sure a massive software company like Google is the right steward for the project: “I tend to go back and forth. Is it sort of a benign—possibly even helpful—universalism that Google is bringing to the table? Or is it something like technological imperialism?”

In other words, when Google is the only entity making decisions, critics fear that the actual language speakers left out of the process are the ones who suffer. Critics already have found issues with Noto’s handling of Urdu, which they say incorrectly adopts Arabic characters. 

See also: How And Why Google Is Open Sourcing Its Data Centers?

Google has already taken on an enormous effort, but one way it could attempt to improve cultural sensitivity toward global languages would be to support languages that even Unicode has overlooked. NPR used the example of Nastaʿlīq Urdu, a type of calligraphic script used in famous Urdu poetry. Right now, the only way to share it online is through image files.

Google Noto has already made strides toward not only supporting common modern languages, but minority and historical ones too. While supporting such languages will require extensive research and development, it’s the only way to truly achieve Noto’s ultimate goal of “visual harmonization across languages.”

Photo courtesy of NASA

Date: Wed, 20 Aug 2014 08:25:05 -0700
Author: :: Category: Web

back to top

Want To Start An Open Source Project? Here's How

You have a problem. You've weighed the pros and cons of open sourcing your code, and you know you need to start an open-source project for your software. But you have no idea how to do this.

Oh, sure. You may know how to set up a GitHub account and get started, but such mechanics are actually the easy part of open source. The hard part is making anyone care enough to use or contribute to your project.

Here are some principles to guide you in building and releasing code that others will care about.

First, The Basics

You may choose to open source code for a variety of reasons. Perhaps you're looking to engage a community to help write your code. Perhaps, like Known, you see "open source distribution ... as a multiplier for the small teams of developers writing the code in-house." 

Or maybe you just think it's the right thing to do, as the UK government believes

Regardless of the reason, this isn't about you. Not really. For open source to succeed, much of the planning has to be about those who will use the software. As I wrote in 2005, if you "want lots of people to contribute (bug fixes, extensions, etc.," then you need to "write good documentation, use an accessible programming language ... [and] have a modular framework." 

Oh, and you also need to be writing software that people care about. 

Think about the technology you depend on every day: operating systems, web application frameworks, databases, and so on. These are far more likely to generate outside interest and contributions than a niche technology for a particular industry like aviation. The broader the application of the technology, the more likely you are to find willing contributors and/or users.

In summary, any successful open-source project needs these things:

1. Optimal market timing (solving a real need in the market);

2. A strong, inclusive team of developers and non-developers; 

3. An architecture of participation (more on that below);

4. Modular code to make it easier for new contributors to find a discrete chunk of the program to work on, rather than forcing them to scale an Everest of monolithic code;

5. Code that is broadly applicable (or a way to reach the narrower population more niche-y code appeals to);

6. Great initial source code (if you put garbage into GitHub, you'll get garbage out);

7. A permissive license—I personally prefer Apache-style licensing as it introduces the lowest barriers to developer adoption, but many successful projects (like Linux and MySQL) have used GPL licensing to great effect.

Of the items above, it's sometimes hardest for projects to actively invite participation. That's usually because this is less about code and more about people.

"Open" Is More Than A License

One of the best things I've read in years on this subject comes from Vitorio Miliano (@vitor_io), a user experience and interaction designer from Austin, Texas. Miliano points out that anyone who doesn't already work on your project is a "layperson," in the sense that no matter their level of technical competence, they know little about your code.

So your job, he argues, is to make it easy to get involved in contributing to your code base. While he focuses on how to involve non-programmers in open-source projects, he identifies a few things project leads need to do to effectively involve anyone—technical or non-technical—in open source:

1. a way to understand the value of your project 

2. a way to understand the value they could provide to the project

3. a way to understand the value they could receive from contributing to the project

4. a way to understand the contribution process, end-to-end

5. a contribution mechanism suitable for their existing workflows

Too often, project leads want to focus on the fifth step without providing an easy path to understand items 1 through 4. "How" to contribute doesn't matter very much if would-be contributors don't appreciate the "why."

On that note, it's critical, Miliano writes, to establish the value of the project with a "jargon-free description" so as to "demonstrate your accessibility and inclusiveness by writing your descriptions to be useful to everyone at all times." This has the added benefit, he avers, of signaling that documentation and other code-related content will be similarly clear.

On the second item, programmers and non-programmers alike need to be able to see exactly what you'd like from them, and then they need to be recognized for their contributions. Sometimes, as MongoDB solution architect Henrik Ingo told me, "A smart person [may] come[] by with great code, but project members fail to understand it." That's not a terrible problem if the "in" group acknowledges the contribution and reaches out to understand. 

But that doesn't always happen.

Do You Really Want To Lead An Open Source Project?

Too many open-source project leads advertise inclusiveness but then are anything but inclusive. If you don't want people contributing code, don't pretend to be open source. 

Yes, this is sometimes a function of newbie fatigue. As one developer wrote recently on HackerNews, 

Small projects get lots of, well, basically useless people who need tons of handholding to get anything accomplished. I see the upside for them, but I don't see the upside for me: if I where[sic] to help them out, I'd spend my limited available time on handholding people who apparently managed to get ms degrees in cs without being able to code instead of doing what I enjoy. So I ignore them.

While that may be a good way to maintain sanity, the attitude doesn't bode well for a project if it's widely shared. 

And if you really couldn't care less about non-programmers contributing design input, or documentation, or whatever, then make that clear. Again, if this is the case, you really shouldn't be an open-source project. 

Of course, the perception of exclusion is not always reality. As ActiveState vice president Bernard Golden told me over IM, "many would-be developers are intimidated by the perception of an existing 'in-crowd' dev group, even though it may not really be true." 

Still, the more open source projects invest in making it easy to understand why developers should contribute, and make it inviting to do so, the how largely takes care of itself.

Lead image courtesy of Shutterstock

Date: Wed, 20 Aug 2014 06:02:00 -0700
Author: :: Category: Hack

back to top