As a developer, my favorite sessions were about a new Facebook tool called Infer. This tool allows the Facebook team to use automation to deliver reliable code at scale.

Infer is a static analysis tool for finding bugs in the code without executing it.

Inter goes line by line and analyzes the code, if there’s a function call, it goes to that function and reads through it and provides summary. There’s no need to reanalyze a method again if it has not changed. Infer gives back a code review, so developer can quickly fix the bug.

Infer can find over 80 types of bugs, here’s a list of the most popular ones:

· Null dereference

· Resource leak

· Race conditions

· Taint analysis

· Starvation

The tool is constantly being developed, and new features are added. Recently they released the Cost Analysis feature that tells you the complexity of your code.

Infer works with any project written in Java/C/C++/Objective C.

If you’re curious about Infer, this is a good place to start https://fbinfer.com/. Also take a look at this awesome hands-on lab that was demoed by Jules Villard during F8.

Facebook for Developers - Getting Started with Infer: Find and Fix Bugs in Your Code

Going forward I’ll be sure to include Infer in the set of tools I use for my Java projects.

It was also great to hear all about how Facebook changes their products by making them smaller and lighter, so they could be used across the world, especially in developing countries where the Internet connection is not great and people don’t have the newest Pixel phone or an iPhone. If you’d like to learn more about those topics, I encourage you to look at this video.

Facebook for Developers - Building Android Lightweight Apps at Facebook

You can find all the sessions here.

Cool things from F8 |2019 was originally published in Systers on Medium, where people are continuing the conversation by highlighting and responding to this story.

As the event grew closer, I got additional emails to put those worries out of mind though. A Facebook admin created a private Facebook group for all of the Hackathon attendees. This made finding a group much easier. Everyone started introducing themselves and their skill sets. Days before the hackathon, I already had a group who was interested in “doing something with VR”. I was pretty excited!

The first day of the hackathon, we were told about the Sustainable Development Goals. At first I thought this was something that was specific to Facebook. They aren’t. You can read more about them here: https://www.un.org/sustainabledevelopment/. According to the site, these goals are “a call for action by all countries — poor, rich and middle-income — to promote prosperity while protecting the planet.” Facebook chose 4 of these goals and used them as the theme for the hackathon.

After announcing the theme, we got to get into groups and brainstorm. In the end, my group was probably one of the most diverse groups. We had 6 people. Three men, three women. We decided to focus on goal four: Ensure inclusive and equitable quality education and promote lifelong learning opportunities for all. At first we didn’t know much other than we wanted to make some sort of learning environment. Soon though, we had our idea and got to work.

We wanted to ensure balance with how much work we were putting in. On the first night, we all decided to go back to the hotel and sleep and come back in 7 hours. (A solid 6 hours of sleep is amazing during a hackathon!) The second and final night, we decided that we were going to go to the F8 All Voices networking event. In the past, this “F8 All Voices” event was a separate event that was by invite only for underrepresented groups. This time, however, it was just the name of the networking event for people picking up their badge early. We all decided we’d stay for a couple of hours, relax, network, and then get back to work. Honestly, it was a great idea to get away from the code and sitting in a chair for a bit. Nothing beats taking a walk after sitting for hours!

After a full 48 hours, we had a rather buggy prototype. The result was a multifaceted application that would help people learn to build and code robots.

One half of the web interface was a code editor where users could enter lines of code to run on their robot. On the other half was a WebVR application. Users could enter VR and build their robot using struts and wheels. Then they could come out of VR, write their code, go back in VR and see their code in action. They could also export their robot to AR to share it with others on Instagram, Facebook, etc. Complete with code they wrote animating the robot.

What I absolutely loved about the whole experience is that the team was able to pull together a variety of applications and make them work cohesively in on application. A WebVR project built in Unity. A React app for the code editor and displaying the WebVR view. Finally, the ability to take something from the WebVR app, package it up in the React app, and then pass it to a SparkAR project to see it in AR. It was ambitious and, like I said, very buggy. We were proud of our work.

We managed to have something to submit at 8 am the first day of F8. We all thought that we’d get to go to the sessions and that would be the end of it. We were wrong. Instead, we were told that the first day of F8, we’d be escorted to the keynote and then back for judging. We all thought the “escorting” was a little much. I mean, we were all grown adults. Turns out we had reserved seating during the keynote. We were within the first 10 rows right in the center.

After the keynote we were escorted back to the hackathon area for judging. We were told to “expect a surprise” while we were rushed through our lunch. After much speculation, we learned that Mark Zuckerberg was going to come through to see a few of the hackathon projects. Not only that, but our project was one of the ones that he wanted to see, so we got to demo our project for him!

At this point, not only were we sleep deprived, but we were also a little awe-struck. (How many people can say that Mark Zuckerberg saw something they built?!) We still had to go through the official round of judging! This involved giving our 30 second pitch to groups of two judges at a time — several times. By the end, we saw about 10–12 judges. We were told to go to a specific session at 4:30 pm that day to learn which groups made the finals.

Our group was one of 8 that made it to the finals. We found out at the beginning of that session, and then had to give our pitch with very little preparation or warning. My anxiety was not happy about the way this happened. I would have liked a little more time than 30 minutes to mentally prepare. We were told initially that only one person from each team would be allowed up on stage to present. My group picked a speaker and then made the case that at least one other person should be up there to actually demo. During the pitches though, the rules shifted and entire teams started going up. I somehow missed the memo, so I stood in the back taking pictures of my amazing team doing our pitch.

The following day, the winners were announced during another session. While my team didn’t win any additional prizes, we still accomplished amazing things. We built something ambitious and amazing, and we demoed for Mark Zuckerberg. Honestly, I don’t think we could have been any more proud of our achievements here!

F8 Hackathon was originally published in Systers on Medium, where people are continuing the conversation by highlighting and responding to this story.

My romance with MWC started on the wrong foot.

When I signed myself, I had to fill up a registration form with different types of information. One of them was my face image - for image verification at the conference itself. This year MWC decided to use a face recognition technology which was suppose to help them with a faster security checkup. Well…

when I entered my face image it refused to accept it, claiming it can’t recognize a face. I tried different face photos of myself, from different angles and lights but unfortunately it didn’t help

And the interesting part only starts here

As you can tell from my profile photo, I look a bit different (hi five to all the people there who look different!) so I thought maybe it could be the problem (although I never had this problem with any other app before). So I decided to try and troll the system. I uploaded images of monkey faces and guess what?

It didn’t recognize them either!

There wasn’t any contact information (mail address or phone number) and the only option for solving it was like in the good old days- going physically, praying for the great lord and casting a spell on the system.

(btw, in retrospect I understood that the image wasn’t an essential part in the form but it wasn’t written anywhere…)

So remember folks, if you are going to work with a face recognition technology which is embedded in your product, make sure it works… and if something isn’t crucial, please specify it (or put it in a different section)

Face recognition technology and UX was originally published in Systers on Medium, where people are continuing the conversation by highlighting and responding to this story.