The arrival of Stride, Atlassian’s Slack competitor, was probably the company’s biggest launch of 2017. While the company generally allows developers to easily integrate with its products, Stride’s API remained in closed beta for significantly longer than the product itself, which exited beta last September. Today, however, Atlassian is opening the Stride API to all developers.
As the company notes, this is the first API that sits on top of the new Atlassian API platform. Thanks to this, Stride developers will get access to a new app management console that makes it easier for them to manage their app’s credentials, for example. In addition, Atlassian is also making a new documentation interface available for Stride developers.
Developers can also include app cards inside conversations and create action buttons (or even build a pop up dialog for when they need to show multiple button actions, for example). These buttons can both act on Stride itself (to open the app sidebar or a dialog, for example) or call on a backend service (which could be any REST endpoint). The team also notes that Stride apps can upload files (think presentations, videos and pictures) into conversations.
Atlassian says about 1,000 developer signed up for early access to the API.
What’s maybe more important, though, is that the company also says that “tens of thousands of teams” now use Stride. That’s not exactly at the same level of Slack, which has more than 6 million active users, or Microsoft Teams, which is now in use by more than 125,000 teams, but it shows that there’s some momentum behind the platform. The modern workplace, after all, seems to have a need for an ever increasing number of tools that provide constant interruptions.
Amazon today is rolling out the biggest makeover for its Alexa Skills Kit Developer Console – the console where voice app developers create their skills – since its debut back in 2015. The new console has been redesigned with a focus on improving developer workflows, says Amazon. It now offers separate sections for “build,” “test,” “launch,” and “measure” – meaning developers can create skills in a visual interface, test their performance, push them live, then see how consumers respond in order to make adjustments.
While Amazon’s own developer console doesn’t address those exact same use cases, there is a bigger emphasis on visualizing the flow of the conversation with Alexa than in the previous version.
Now, developers will have a visual interface for creating skill components, like intents, slots, sample utterances, and the dialog model.
Afterwards, they can test their skill in the console using either voice or text – including multi-turn dialog – while seeing the conversation play out in the left-hand sidebar.
The “Launch” page will help developers address common validation errors, and offers support for publishing private skills for Alexa for Business. (Before, these business skills could only be published using the Alexa Skills Kit Command Line Interface.)
And in “Measure,” developers can track metrics around sessions, utterances, unique customers, and now, cohort analysis, too. This will help them better understand their customer retention and engagement over time.
Above: The old and new experience, side-by-side
“We’re constantly introducing new features for developers to enhance skill building and management, and one piece of feedback we received from developers was for a simplification of the workflow,” explains Amazon’s Paul Cutsinger, who led the redesign project. “We took developer feedback and used it to improve the site structure and information architecture. The new developer console makes navigation easier with an updated and integrated user interface,” he says. “Developers can create and manage their skill in one integrated environment with a built-in checklist and validation support at each step.”
Overall, the new skill builder does look like a more modern piece of software, is now branded with Alexa’s logo and blue color scheme, and appears less intimidating to newcomers to Alexa skill building as a result of the redesign.
However, the software is still in beta and Amazon warns that while it can handle most skills, there are some features that are unsupported for the time being.
Part of the battle for smart speaker market share is keeping the developer community happy. On that front, Amazon has a mixed track record. While it has a bigger end user base for smart speaker voice apps thanks to Echo’s popularity, the apps have so far been somewhat difficult to monetize, despite Amazon’s more recent support for in-app purchases and subscriptions.
Many developers are doing voice apps as more of a hobby, in the hopes that the ecosystem will develop over time. Other apps built by larger companies are seen more as the cost of doing business – if you’re Uber, for example, or a smart home device maker or a news publisher, you have to establish your presence on the platform.
HackerRank, the skills-based recruiting platform and online coding challenge community, today announced that it has raised a $30 million Series C funding round led by JMI Equity, a fund that specializes in helping software companies scale. Exiting investors Khosla Ventures, Battery Ventures, Randstad and Chartline Capital Partners also participated in this round.
As HackerRank co-founder and CEO Vivek Ravisankar told me, the company now has over 3.4 million developers in its community, which has grown organically since the company’s launch in 2012. What’s maybe just as important, though, is that the service has also brought on a wide variety of companies to its HackerRank for Work program that are looking to use the services platform (and community) to recruit developers. According to Ravisankar, these customers include five of the top eight commercial banks, for example, as well as auto manufacturers, retailers and others. Ever company is now a software company, after all, and they are all looking for talent. With these customers, HackerRank was actually cash flow-positive for a part of 2017 and expects to return to that in the near future.
Over the course of the last year, HackerRank also expanded beyond core programming skills and adding support for other technical roles, including DevOps positions, database specialists and others.
“To be frank, we didn’t think we needed $30 million,” Ravisankar told me, but JMI looked like a good match for HackerRank and this round, which brings the company’s total funding to $58.2 million to date, gives it a long runway to expand its product portfolio.
Specifically, Ravisankar is focussing on three areas: doubling down on customer acquisition and HackerRank’s go-to-market strategy, investing more in its community, and using machine learning and data science to better match employer and job seekers.
It’s this last part that’s probably the most interesting. HackerRank sits on a trove of data about what skills job seekers possess and which ones employers are looking for. “It’s very hard for a lot of companies to quantify what makes a great developer,” Ravisankar explained — and the same goes for developers who don’t always know what they are looking for. So at this point, the HackerRank team is trying to figure out how it can best use its data to say whether a developer is a good fit for a job.
HackerRank currently has about 150 employees but this new round will allow it to hire a few more, too. Chances are, it’ll use its own platform to do so.
GPUs on GKE (an acronym Google used to be quite fond of, but seems to be deemphasizing now) have been available in closed alpha for more than half a year. Now, however, this service is in beta and open to all developers who want to run machine learning applications or other workloads that could benefit from a GPU. As Google notes, the service offers access to both the Tesla P100 and K80 GPUs that are currently available on the Google Cloud Platform.
The advantages of the kontainer/GPU combo is that you can easily scale your workloads up and down as needed. Most GPU workloads probably aren’t all that spikey, but in case yours are, this looks like a good option to evaluate.
Overall, the Kubernetes Engine (which seems to be Google’s prefered nom du guerre for the GKE now) saw its core-hours grow 9x year over year in 2017. That’s no surprise, given the hype around containers and the fact that the GKE, as it was called at the time, only launched in 2016, but it does show that Google may just have a winner here.
A number of iOS app developers have been mystified by a new wave of app rejections related to their use of Apple’s emojis. They’ve suspected that a new App Store crackdown is underway. However, the company hasn’t changed its policy on Apple emoji usage in apps, nor its enforcement, according to sources familiar with the App Store review team’s processes. The policy does seem to be inconsistently enforced at times, though.
That’s led to previously approved apps receiving rejections, while other apps in breach of policy have been let in.
As Emojipedia had determined, we’ve confirmed that Apple will only allow apps using emojis in specific contexts, like in a text field.
Meanwhile, any other usage should be banned by App Review, including when emoji are used as elements in a game, as replacements for buttons or other parts of the app’s user interface, as sticker packs, in app logos or icons, or in promotional images, also as Emojipedia had suspected, based on the pattern of rejections.
While emojis exist as part of the Unicode standard, Apple’s implementation of that standard is copyrighted. That means the company is within its legal right to control the usage of their own emoji designs, especially in their own App Store.
However, Emojipedia founder Jeremy Burge takes issue with the fact that Apple should have such a policy around its emoji at all.
“It seems reasonable to me that Apple would want some level of control over emoji use in the App Store, but banning it outright from anything other user-inputted text feels a step too far in my opinion,” he says.
That said, Apple’s decision to reject apps based on their use of Apple emoji is not a new occurrence. If you go back far enough on Twitter, you’ll findmanyexamples of developerscomplaining about the same thing over the past couple of years.
iPhone X update for Reaction Match out soon 🎉. First attempt was rejected at app review for including emoji on the leaderboard 😂 – they have always been there… 😩
Adding to the more recent confusion, as Emojipedia also pointed out in its reporting, was the fact that Apple’s own app development course on coding using Swift offers an example of an app with emojis that seems to breach its policy.
The real issue here is that the App Review team has not consistently enforced the policies around Apple emoji use. In addition, Apple it doesn’t speak up to clear the air when it’s aware developers are confused.
That leads to a situation where developers will just try to sneak their app through, even though it seems to be in violation of a guidelines. (That sometimes works, too.)
But in the end, it wastes developers’ time because they later may get caught by App Review. They then have to go back and overhaul their app to address the problem at a much later stage of development.
Apple declined to comment about the emoji-related rejections.
Scheduling tech conferences is hard, especially in May, when seemingly every company wants to hold a major event, including Google and Microsoft. Typically, Google I/O and Microsoft Build, the flagship developer conferences for both companies, happen within a week or two of each other in May. But not this year.
Microsoft today announced that its Build conference in Seattle will run from May 7 to 9. Google I/O is scheduled to run from May 8 to May 10. That’s not ideal.
Google probably put Microsoft in a tough spot given that it announced its dates first. Both companies use these events to make major announcements that affect both their respective developer communities and their users. Last year, for example, Microsoft used Build to announce both new developer and cloud tools, as well as the latest version of Windows 10.
By scheduling Build right before I/O, Microsoft clearly hopes to steal some thunder from Google.
Most attendees probably only attend one of these shows, so that shouldn’t be an issue for most. Microsoft typically schedules two days of keynotes for Build, though, so May 8 will be an interesting day for the tech press if both Google and Microsoft hold duelling keynotes on that day.
Registration for Build will open on February 15. Developers who want to be in the running for a ticket for Google I/O will have to put their hat in the ring between February 22 and 27.
Yesterday, we noted and reported on how Apple was sending developers emails with install and ad spend details for other developers’ apps. Today, Apple has sent out a note to developers to say that a processing error caused the problem, and that in the future it will only send alerts by email, but that developers will have to log in to their accounts to see any actual numbers or other details.
Louis D’hauwe, one of the developers we spoke to yesterday, shared a note from Apple apologising for the error, which appeared confined only to some developers using the company’s new Search Ads Basic tier of advertising:
The company said that to avoid any problems like this in the future, developers will need to sign into their accounts to view their dashboards with details. Yesterday’s snafu revealed details about apps using the Search Ads Basic service, including how many installs each app was getting, the cost-per-install, and the total spend.
As we pointed out yesterday, not only may developers not want to share sensitive information like this with other developers, but they may be particularly unhappy if the data was shared with direct competitors.
Search Ads Basic was launched only a couple of months ago, in December 2017, and is aimed at both smaller developers and those who are just starting out with search ads. Apple has a second tier of service that is aimed at bigger ad buys, with more data points and priced at a cost-per-impression rate rather than cost-per-install.
Both products underscore a couple of interesting trends around Apple and the App Store.
With millions of apps now on the market across both web and native formats, discoverability remains a challenge for many developers, and hoping for virality or publicity in other forms alone to promote an app will not be enough for most. Coupled with that, in a landscape rife with dodgy “growth hack” networks, Apple has seized the moment to provide something that it can control more itself, and make a little revenue from it to boot. Its carrot: installs gained from these ads are weighted more heavily than other installs from other promotional means, as Apple considers them “high quality.”
Getting the security for developers right around those ads and ad services should and will be considered part and parcel of whether Apple truly is delivering on that quality assurance.