GigaOM » web applications

Key technologies for the smart city

Wed, 28 Mar 2012 06:55:30 +0000

As more and more people move to urban areas there is an escalating need for the smart city, where technology, sustainability, citizen well-being and economic development integrate. Currently five key technology sectors are enabling the smart city: smart grids, smart transport, smart water and waste management, smart building systems, and the enabling ICT platforms for the smart city. This report examines each and provides recommendations to those key players — IT companies, telcos, utilities and real estate developers — that wish to benefit and harness those technologies.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

Submit once, sell everywhere? Mozilla to open mobile Web app store

Tom Krazit — Wed, 22 Feb 2012 18:57:36 +0000

In an app-centric world, those who are trying to embrace mobile Web development have to think in terms of stores and marketplaces. Mozilla announced plans Wednesday for its own take on a mobile app shopping experience, one built around the promise that Web applications will bridge the gaps between mobile devices.

There weren’t a lot of details revealed by Mozilla’s press release, but the company plans to talk more about the Mozilla Marketplace next week at Mobile World Congress and will invite developers to submit Web apps. The idea is to give Web developers a prominent place to hawk their HTML5 Web applications. App discovery is a huge problem in the native world, and even though Web applications don’t really need a distribution channel other than a Web site, without a big signpost directing mobile users to Web apps it could be a lot harder for smaller developers to get noticed.

Mobile Web apps offer the promise of being able to target everyone with a single development effort, as opposed to building separate iOS and Android versions of an app (not to mention tablet versions). But right now native apps offer hooks into the phone’s hardware that Web apps can’t quite mimic and most consumers are quite familiar with the concept of stores like Apple’s App Store and Google’s Android Market.

While Mozilla avoided any discussion of the details, it suggested two ways in which it might try to change that equation: standard APIs (application programming interfaces) that could help developers reach down into the phone, and “a new identity system for the Web that puts users in control of their content, tying apps to the user and not the device or platform.” Stick around for our MWC coverage next week to learn more.

A report in the San Jose Mercury News suggested Mozilla might also be ready to unveil a prototype mobile Web phone based on the Boot To Gecko project announced last year. Google has also tried to plant seeds for mobile Web computing with its Chrome OS hardware, but it’s pretty clear that has yet to make an impact.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

How Codecademy got so hot, so fast

Colleen Taylor — Tue, 24 Jan 2012 01:18:44 +0000

Codecademy is on fire right now. The startup, which teaches users how to program with an interactive and social web application, has garnered more than 1 million users (including bold-faced names such as New York City Mayor Michael Bloomberg) and made learning how to write computer code trendy with its “Code Year” program aimed at the New Year’s resolution crowd. And all this from a startup that’s only five months old, with just five full-time staffers.

I sat down with Codecademy co-founder and CEO Zach Sims to hear about how the company got to this point so quickly, and what’s on deck for the months ahead. Here are a few key takeaways:

Necessity breeds invention

The idea behind Codecademy emerged out of the founding duo’s frustrations with the status quo of learning how to program. Co-founder Ryan Bubinski was already an experienced programmer who spent his weekends and free time during college teaching other students how to build web applications; but Sims was not nearly as familiar with coding. When the two entered Y Combinator’s summer 2011 class together in the hopes of launching a web startup, Sims tried to learn how to code on his own so that he could be of more help on the technical side of whatever business they founded.

“I was watching videos and tutorials and reading books,” Sims said. “But I found I learn best by building things and breaking things, not by just reading something. I wanted something interactive where I could learn in bite-sized pieces, and actually practice what I learned along the way.” So Sims and Bubinski decided to use their time at Y Combinator to build exactly that — and Codecademy was born. The company is now backed with $2.5 million in venture capital from a handful of elite investors including Union Square Ventures, O’Reilly Ventures, SV Angel and Yuri Milner.

Timing is everything

Codecademy’s message — that knowing how to write computer code is becoming just as important as knowing how to read or write — could not have come at a better time. While many sectors of the economy are suffering from layoffs and underemployment, the tech industry is having a full-on hiring crunch. Nearly every tech industry executive I talk to is currently looking to hire as many good engineers as he or she can find. The only problem is that not enough people right now have the programming skills necessary for those jobs.

Screenshot of an introductory Codecademy lesson (click to enlarge)

“Programming is the new literacy,” Sims said. “We’re all walking around with these phones in our pockets, using all these apps, but no one understands how any of it works. There are just not enough engineers, and this is the job of the future.” That Codecademy launched just when this started to become apparent on a larger scale has been key to its early success.

Listening to users — online and off

When Codecademy’s users started getting together offline by scheduling real-life meetups, the company decided to follow them. Last week saw the launch of official Codecademy meetups, and there are now official meetup groups in 171 regional areas worldwide to let people get together in person to discuss their progress learning how to code. Also last week Codecademy launched a Q&A feature within its web product to let people talk to each other via online forums while doing the Codecademy lessons.

Getting bigger, but staying scrappy

For now, Codecademy does not make any revenue, but in the future it could start charging for more advanced lessons and premium services, Sims said. Any revenue generation plans are a bit farther out on the horizon, as right now the company’s focus is on growing its user base and adding new lessons to the core free service. One thing is certain, though: Codecademy has no plans to become an accredited learning institution that charges for a degree.

Sims, who studied Political Science at Columbia University but dropped out several credits shy of graduation, said that Codecademy is founded on the belief that skills are the most important factor in getting good work — not educational credentials. “If you look at a lot of Silicon valley companies, they don’t hire based on a degree. A lot of the best programmers in the industry never even went to college at all,” he said. “Your skills should speak for themselves.” That sounds like a worthwhile lesson in itself.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

Email is still hot: Why SendGrid got $21M in VC funds

Ryan Kim — Tue, 17 Jan 2012 14:00:34 +0000

All those email messages from web applications and services for notifications, updates and subscriptions are growing at a fast clip, and that’s helping to spawn new opportunities in so-called transactional email. SendGrid, a TechStars graduate, is a big beneficiary, becoming one of the largest email platforms in the world, with 40,000 customers and 2.6 billion emails sent a month.

Now, the Boulder, Colo.-based company says it has raised a $21 million Series B round led by Bessemer Venture Partners, with additional funding from existing investors Foundry Group, Highway 12 Ventures, SoftTechVC, 500 Startups and TechStars. This follows a $5 million Series A round raised in 2010. SendGrid will use the money to grow internationally, increase hiring and invest in product, developer relations and customer support.

Since launching in 2009, SendGrid’s cloud-based SMTP platform has been a valuable tool for many web application companies that stay in touch with their users through emails. It has delivered some 27 billion emails to date. Companies such as Pinterest, Hootsuite, Foursquare and Path all offload their transactional email to SendGrid. It’s a growing market, as companies rely on email to update users and keep their services sticky. SendGrid isn’t just handling a lot of emails; it’s doing some smart work to get past spam filters and blocked IP addresses to ensure emails reach their destination.

“Web application companies are under tremendous pressure to build fantastic services quickly. Offloading a core service like transactional email management and delivery is a smart and simple choice,” said Jim Franklin, CEO, SendGrid. “Today’s funding validates our original vision and supports our global expansion plans as we continue to build out our team.”

SendGrid also said it’s launching on Windows Azure, Microsoft’s cloud platform for web applications. Windows Azure customers who sign up for SendGrid will receive 25,000 free emails each month. SendGrid is showing the opportunity in infrastructure apps that help developers plug in needed services that keep their apps running.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

Migrating media applications to the private cloud: best practices for businesses

Mon, 05 Dec 2011 08:01:23 +0000

If your company has a cloud application with a predictable audience size or one that is costing you more than $25,000 a month to host, you may want to consider maintaining a private cloud. This paper provides an overview of the factors that decision makers who are developing a public-to-private cloud-migration strategy should consider, recognizing that public versus private cloud strategy is not an all-or-nothing proposition. It also details pitfalls that must be avoided along the way and provides a case study of Zynga, a company that has found a way to use both the private and public clouds to create a hybrid solution. Companies mentioned in this report include Akamai, Foursquare, Nimbula and ARM. For a full list of companies, and to read the full report, sign up for a free trial.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

The rise of the do-gooder developer

Katie Fehrenbacher — Mon, 11 Jul 2011 07:00:14 +0000

Apps will save the world — or at least there’s a trend around developers donating time to working on apps to make the world better. These projects give the developer community a chance to give back and work on rewarding projects, while at the same time providing some much-needed innovation around data sets for the public good.

Developers are more in-demand every year, particularly now that a seeming Internet and social media bubble is growing, and as young developers graduate college (or skip university altogether), they seem to be increasingly looking for do-gooder developer projects to work on. Perhaps it’s the young age of these new developers, the “free-for-all” mindset of the coder, or the relatively high pay for many developers, but donating time to hack projects seems to be de rigueur in many programmer circles.

Data that can be mashed up and blended, for free or at a low cost, to create these types of philanthropic apps is being unleashed at a rapid rate, with continuously easier access and open source tools. The rise of big data, in general, is leading to the need for more types of ways to organize and access this data.

At least three groups come to mind, that have helped pave the way for the rise of do-gooder app development. First, Google, which for years has touted its “do no evil” slogan, has encouraged employees to have their 20-percent time and also runs a philanthropic arm. Google commonly does things, like when Google.org donated 20 million CPU hours of Google Earth Engine to groups in the developing world to use.

Second, the rise of Y-Combinator, (and similar incubation groups) that encourage young entrepreneurial and developer types to come together and channel their coding creativity. Lastly, the popularity of business competitions like those from the X-Prize Foundation have helped show companies, investors and entrepreneurs that these innovation prizes actually work.

Do-gooder developers

Last month, Hack for Change, sponsored by Change.org, brought together around a hundred developers that spent 24 hours building do-gooder applications like winner Good Neighbor, which sends users a text message when a neighbor in need requires help. Another dozen philanthropic apps were coded and launched at the event.

Random Hacks of Kindness (RHOK) is a group that came together in 2009 — backed by Google, Yahoo, the World Bank, Microsoft, and NASA — to hold events where developers spend time working on applications for disaster relief and fighting climate change. Last year RHOK spawned I’mOK, a mobile app that helps people in disasters alert loved ones that they are OK, and Google also used RHOK to refine its PeopleFinder app, a similar app that connects disaster victims.

Grass-roots programs like Hack for Change and Random Hacks of Kindness can be a good places for small developer teams to flesh out ideas that can become larger projects, or startups even. Or, as in the case for Google, potentially projects that can be taken back to their companies.

The trend is also moving beyond the grass roots to the corporate and government worlds. Last week, the Environmental Protection Agency launched its Apps for the Environment contest, which calls for developers to use EPA data (and any mashup data) to produce apps that can help the environment. Winners of the EPA contest win a trip to Washington, D.C. to present the app, and the EPA will highlight the winning apps on its website. GE has also tapped into this type of contest with its smart grid challenge, though GE’s contest is more about creating innovation for GE around the smart grid sector than helping people or saving the planet.

The real challenge for these do-gooder developer projects is that donated developer time sometimes leads to apps that aren’t fully baked or all that helpful. But the good news is that since these projects are largely open-sourced, future developers can often connect with the project and help take it a step further.

Image courtesy of Guillherme Chapwieske.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

Is the End Nigh for Database Appliances?

Derrick Harris — Wed, 02 Feb 2011 00:40:31 +0000

Whenever a new appliance hits the market, the question is asked whether customers will actually be willing to pay big bucks for a product that locks them into a specific architecture at a specific price point that makes scaling an expensive proposition. If the tale of Schooner Information Technology is any indication, the answer to that question is no, at least when the target is web applications. Today, Schooner, which just under two years ago made a splash with its high-powered, $45,000 database appliance — including securing a partnership with IBM — made a hard left turn to selling software only. The implication, from my perspective, is that, with few exceptions, vendors pushing appliances need the right product for the right customer, or they might be in trouble.

The goal of the Schooner appliance was to significantly boost web-application performance by packing 512GB of flash memory into each appliance to serve MySQL or memcached data, as well as high-powered chips and networking gear. Schooner was optimistic there was a market of companies that would just want to buy an appliance and forget about having to worry about scaling their data layers or tuning for better performance. However, as my colleague Stacey Higginbotham pointed out in covering Schooner’s launch, “I’ve seen appliance efforts play out before. And I can’t help but think of that abandoned panini press buried in the back of one of my cabinets.” It appears that Stacey was right on this one, unless nobody was buying the appliances to begin with.

In an email response to my inquiries on Schooner’s new software-only approach, CEO Jerry Rudisin laid out the lessons learned that led to Schooner’s decision to put its physical appliances to rest: (1) customers loved the prospect of boosting performance, but wanted to reuse their existing servers; (2) customers who do large-volume purchasing can get hardware at price points far lower than Schooner could sell it to them; (3) customers have preferences of hardware vendors and don’t want to be forced to buy particular gear if they want the foundational software; and (4) the appliances only came in the 512GB capacity, which was too small for some users and too big for others. Customers always drool over the prospect of faster-performing applications, but, it seems, not at the expense of everything they believe in terms of IT budgeting and avoiding lock-in.

So, as of today, Schooner now sells only the software that made its appliances run, leaving customers to choose their own x86 server platforms on which to run it. The aptly named Schooner MySQL is the MySQL version, while the memcached-based data store is now called Membrain. Schooner says Membrain is ideal for running NoSQL databases, which generally target unstructured data not fit for SQL-based relational databases. However, Schooner’s scale-up-instead-of-scale-out value proposition remains in tact. Whereas scale-out approaches have become a popular option for boosting performance due to the prevalence of cheap boxes, Schooner’s software is still optimized for flash memory, which also is becoming less expensive by the day. According to Rudisin, customers can consolidate their database footprints by packing fewer servers full of more flash memory, including from former nemesis turned technology partner Fusion-io.

Maybe Schooner also learned a lesson from one-time competitor Gear6, which launched its own memcached-based appliance just a week after Schooner made its debut. Gear6 went belly-up in early 2010 and ended up selling its assets to flash leader Violin Memory. Gear6 had a longer history and a broader product line than does Schooner, but Violin’s use of the Gear6 IP is telling — it’s using Gear6′s caching software to underpin cache appliances that sit in front of network-attached storage. Yes, Violin’s appliance start at $40,000, but it’s easier to sell high-priced storage appliances than it is to sell high-priced web-database appliances, which Violin is not selling.

It’s worth noting, however, that not every vendor has trouble selling database appliances. Oracle claims it’s making a killing with its Exadata Database Machines, and Teradata and Netezza (now owned by IBM) are doing just fine, as well. Even EMC decided to get into the appliance game upon acquiring Greenplum and its database software. But those are large vendors selling data warehouses and analytic engines to large companies, which is a different game than simply trying to improve web application performance like Schooner is attempting to do.

Image courtesy of Eileen Henderson.

Related content from GigaOM Pro (sub req’d):

Mobile 2011: Trends Not to Expect

Colin Gibbs — Tue, 21 Dec 2010 23:00:28 +0000

Predicting what not to expect in mobile next year is a lot easier than predicting what will happen. You can be sure, for instance, that consumers won’t pay for on-the-go TV next year any more than they have in the past, and that carriers won’t embrace “openness” any more than they are forced to. And in areas like mobile advertising, tablet devices and mobile payments, there are several trends you shouldn’t look for in 2011 — despite all the hype to the contrary.

Are Web Apps Becoming Over-Reliant on One Another?

Simon Mackie — Thu, 11 Nov 2010 16:00:58 +0000

More and more, web apps are becoming interconnected with one another. In general, of course, this is a very good thing, but it nonetheless introduces dependencies between these apps. And as more apps come to rely on one another to provide critical services in our day-to-day work lives, the greater the consequences to both our data and our productivity if and when these apps fail.

Mozilla Wants Everyone to Have a Web App Store

Kevin C. Tofel — Tue, 19 Oct 2010 21:00:24 +0000

Mozilla today released technical documentation of a proposed open ecosystem for web apps that’s built on standards such as HTML, CSS and JavaScript. The storefront and apps offered through it will be usable across both desktop and mobile devices that use modern browsers. Although Mozilla is proposing the infrastructure and standards for a web app store, it says that it has no intention at this point to build such a store or distribute apps. Instead, the proposal is being offered to the open source community as a concept for developers to use.

The Open Web App Ecosystem, as Mozilla dubs it, will bring several benefits to both users and developers, the company argues. Paid transactions will be handled with authorization standards such as OpenID, for example, and purchases will be seamlessly transferable between browsers, so that a web app used on the desktop could later be used on smartphone by the same user. Developers can offer apps through their own web storefront and users gain a personalized app dashboard: Here’s the sample one Mozilla is providing today.

Watch this video for free on GigaOM

If Mozilla’s proposal sounds similar to what’s to be expected from the web app store in Google’s Chrome OS, it is and it isn’t at the same time. Both approaches will offer robust web apps, likely paid and free, within an easy to manage browser-based storefront. The key difference is that anyone can run their own web app store with Mozilla’s approach, instead of relying upon a centralized ecosystem. There’s merit in Mozilla wanting to give such power to developers, but by going independent, programmers will lose marketing advantages often associated with a centralized app store.

Ironically, in this age of mobile apps, it’s worth noting that Apple developed a web-app store for the original iPhone in 2007. Even with the iTunes App Store humming on all cylinders, Apple’s web storefront continues to exist for iOS devices. I suspect that with web apps becoming more robust due to evolving standards, Mozilla is preparing itself for the future. But it’s just as likely that Apple, and Google, for that matter, will be just as ready with their own improved centralized web app stores. When that future arrives, will there be room for independent web ecosystems like the one Mozilla is proposing?

Related content from GigaOM Pro (sub req’d):