Category: Review

Posted on

Language Learning Tools

Language

As we start a new school year, many students – both in high school and college – will be learning a new language. For most of my life, I have been fascinated with foreign languages and spend time every day honing my skills. While I spend much time learning from books, there are some great tools for the more tech savvy that can make language learning more fun.

Pimsleur Language Programs

One of my favorite tools, particularly for working toward conversational skills, is Pimsleur. Their product typically provides 30 half-hour recordings that guide the listener from no language skill through the path to conversational knowledge. I used Pimsleur Chinese 1-4 to help improve my Chinese proficiency – and was pleased with the results. While Pimsleur can be expensive, it’s probably one of the best tools if your ultimate goal is fluency. Note, while they have traditionally been a CD-based program, they have have an app too.

Duolingo

Another favorite for language learning is Duolingo. This is probably the best-known mobile application for language development. Duolingo offers a variety of languages from typical European languages such as German, French, and Spanish to less common languages such as Yiddish and even fictional languages like Klingon. In my experience, Duolingo works best with languages sharing our alphabet. But if you’re willing to work a little harder, their Asian language tools are pretty awesome too. Duolingo has some really nice features such as weekly competitions on the leaderboard and some social aspects too.

Japanese Dungeon

As I’ve started trying to learn Japanese, I came across this app. It’s particularly fun as it’s a good mix of game and learning. This particular tool is great for learning the Japanese scripts. In addition to Japanese Dungeon, versions exist for Korean, Russian, and a variety of European languages. What makes this app so fun is that it plays like an 8-bit game where you fight orcs. Get a character right and you hit an orc. Otherwise, you will be hit. As you progress, you can unlock more characters with different stats and level your character up to deal more damage.

Other Tools

The above are my current favorites. However, there are countless other tools out such as Rosetta Stone, Hello Chinese, Drops, and too many others to enumerate. So, if you are looking to expand your language skills, head to your mobile app store and get learning!

Posted on

The NVIDIA Jetson Nano

During the last few weeks, I’ve had the opportunity to play around with the NVIDIA Jetson Nano. As an artificial intelligence enthusiast, I was very eager to see what the Jetson Nano could do, and I was not disappointed.

The Jetson Nano is a small AI computer, about the size of a Raspberry Pi. Just like the Pi, the Nano runs Linux and contains a 40-pin header for interfacing with other components. With 4 USB ports, ethernet, and HDMI, it’s a self-contained computer. What separates the Nano from the Pi, however, is the NVIDIA GPU. With 128 cores, the GPU is capable of performing floating point operations substantially faster than a typical machine. It’s those very floating point calculations that are the bulk of AI processing applications.

Taking the Nano for a spin, I downloaded their OS image and installed on an SD card. Visiting NVIDIA’s website, I found that they offered a free online course to learn the basics. Through their course, I was able to create simple AI to perform real-time image classification. Unlike a typical Haar filter generation, the Nano was able to create a classifier in minutes. And, even more exciting, the PyTorch framework they demonstrate contains pre-trained networks that allow for far fewer images than a typical Haar filter requires. Additional NVIDIA courses include video processing using C++, medical image processing, and countless other AI tasks.

What makes the Nano so exciting is just how fast it works. Having developed AI models in the past, I’m accustomed to lots of waiting. However, the Jetson Nano functions like a mini AI supercomputer. For beginners, this means you can easily generate your neural network model quickly and perform validation tests with substantially less effort than traditional computer architectures.

I highly recommend the Jetson Nano for anyone wanting to learn more about artificial intelligence or wishing to experiment with potential consumer or industrial applications with computer vision.

Posted on

WiFi Calling

Vintage Telephon

During the last few years, WiFi calling has become an option for cellular users. This is particularly exciting for those who use smaller cellular providers, those in rural areas, or those in building with poor cell coverage. But how good is it?

I have two cell phones – one that I use strictly for development and one for general use. A few months ago, I purchased a cell plan for the development phone to use Google Fi. My personal phone uses Verizon. My office is in a concrete building that does a pretty good job blocking cell signals. I do, however, have pretty solid WiFi. So, it only made sense to setup both phones for WiFi calling in my office. Now, no matter what, I can make calls from anywhere in the building.

My Experience

Unfortunately, the only option for WiFi calling on my Google Pixel is on or off. I can’t specify which should receive priority – WiFi or 4G – and I can’t specify what access points to enable WiFi calling on. So, when I’m in my car (with cellular WiFi), my phone uses the car’s WiFi for calling. Hardly optimal given that I pay per megabyte for data. When I’m at home, I have an excellent cell signal – so there’s no reason for WiFi calling. But that won’t stop my phone from defaulting to WiFi anyway.

This wouldn’t be a big deal if WiFi calling was as good as the cell network. But, unfortunately, it’s not. Even on an access point without any other users, my WiFi calling tends to break up or be delayed. Sometimes, it’s so bad that the other party complains and I’m forced to call back later. Of particular annoyance is that there’s typically a several second delay after the other party answers.

While I think WiFi calling is a great idea, it’s not reliable enough – at this time – for me to rely on. If I’m in an area without cell service, I may switch it back on, but until then I’m keeping it shut off.

Posted on

MATLAB

Code

Running my own software company, I run into people trying to solve all kinds of different problems with software. As such, I end up using a variety of different technologies, languages, platforms, and frameworks. Recently, a client project required me to download MATLAB in order to experiment with several existing toolkits needed for the R&D phase of the project. The first thing I noticed was the price tag. Today, the overwhelming majority of development environments are free. However, MATLAB carries a hefty price tag of $800/year for commercial customers. After installing the application, I found out that the plugins (called toolboxes) averaged an additional $500/year. All told, this development environment was going to cost me around $3000/year after all necessary plugins were purchased.

Is It Worth It?

Seeing that MATLAB is so expensive, and so highly regarded by many in the research and academic community, I assumed that the MATLAB user experience was going to be epic! How wrong I was. The code I was examining contained dozens of files across numerous folders. Of course, this isn’t abnormal for any large project and I expect any decent IDE to be able to handle it. Any IDE except MATLAB. Need to find where a function is defined? Don’t expect any tools in MATLAB to accomplish that! Of course, you can do a find in files for all references of a string, but that’s not particularly helpful when a function is called dozens of times or has a short name. How about displaying a call hierarchy so that you can see where functions are being called? Nope – that’s not possible either. Ever aspect of the IDE proved to be well below what was commonly available in other IDEs.

If you’re not intersected in the user experience, I guess MATLAB is great. Apparently, its ability to perform all kinds of mathematical wizardry is beyond compare. It’s great for solving linear algebra problems and other advanced math tasks. Unfortunately, I wasn’t interested in that.

Open Source Code

I managed to find several open source projects and toolboxes for MATLAB. I was interested in understanding how these particular toolboxes work, so I started examining the code. MATLAB, like so many other languages, is a weakly-typed language. While weakly-typed languages do offer many benefits, readability is rarely one of them. MATLAB also does not require variables to be defined before use. This can be very difficult to follow particularly when the bulk of the code I saw contained neither useful variable names nor adequate comments to understand what’s going on. Very typical of people without a development background.

Conclusion

I was very disappointed with everything I saw in the MATLAB community. I assume most of the users are mathematicians and have no idea the kinds of features that are available in modern environments, so they have no idea what they’re missing. Likewise, I assume they’re used to digging through poorly documented code with variable names like g, f, c, and y. For me as a developer, this is neither acceptable nor maintainable long-term. For number crunching, I much prefer R. It may not have the most advanced IDE, but it doesn’t set me back anything either!

Posted on

Samsung Galaxy Watch

Galaxy Watch

Over the last several years, I have owned several smart watches. Unfortunately, I was never impressed with any of them. Early smart watches had horrible battery life. At the end of the day, it was not uncommon for my watch to have less than 20% battery  remaining. Use your watch more heavily, and you could expect to recharge mid way through the day. While early smart watches were nice toys, they lacked any real use case beyond being able to read your text messages.

How things have changed. I bought a Samsung Galaxy Watch hoping to give smart watches another try. Not only is the battery life amazing, but so are the features. At the end of the day, I typically have 70% or more battery remaining. I can now go multiple days without charging! In addition to being able to send and receive text messages, I can use my watch to make calls! The Samsung Galaxy Watch includes WiFi, NFC, optional 4G, and is even waterproof!

The Samsung Galaxy Watch is also an amazing tool for runners, hikers, bikers, and anyone who wants to keep track of their exercise or heart rate. With GPS, altimeter, barometer, step counter, and heart rate monitor, the Samsung Galaxy Watch is the perfect tool for any fitness enthusiast.

The Samsung Galaxy Watch has rewritten the book on smart watch technology. Their watch is absolutely amazing and is the first smart watch I have ever really loved.

Posted on

Surface Pro 6

As a developer, I tend to need a lot of hard drive space for projects and tools. As such, my MacBook Pro was starting to run out of drive space. This was only exacerbated by the fact that I was also running Windows 10 on the same machine for .NET projects. I decided it was time to replace my MacBook Pro, and looked at the newest product line. Unfortunately, I was very disappointed by the lack of ports on the new MacBook Pro. With a sticker price of nearly $4,000 for a machine with all the bells and whistles, the lack of ports is simply unacceptable.

What to Do?

After much deliberation, I decided to get a separate Windows machine and reclaim the space on my MacBook Pro for MacOS. I had been looking at the Surface Pro for a while as it looks like a great machine. So I went to the store and purchased the computer, a keyboard, and the pen for around $1000. Given a price tag one fourth that of a new MacBook Pro, how does it compare? So far, I am very impressed. The Surface Pro includes numerous features that the MacBook Pro can’t touch including: facial recognition for login, touch screen, functionality as both a tablet and a laptop, and he use of an optional pen for drawing.

Conclusion

While the Surface Pro does not compete with a top-of-the-line MacBook Pro in terms of raw horsepower, it is lightweight, multipurpose, and more than adequate for the occasional .NET project I need to support. Even more exciting, it will work far better than my MacBook Pro for taking onsite to gather customer requirements or take quick notes.

Note to Apple

Apple, your decision to remove standard USB from your laptop cost you $4,000 I would have gladly paid. Instead, I paid $1,000 to buy a laptop from your competitor. While you may think you’re a trendsetter, a large number of developers I’ve spoken with are less excited by your choice. For us, computers are a tool to solve a problem – not a trendy accessory. You managed to capture a growing portion of developers, but focusing on trends instead of solutions will negatively impact that growth in the future.

Posted on

Microsoft AppCenter

Several years ago, I needed to find a way to distribute mobile applications to test users. Ideally, the solution should allow for anyone to access the test application. This would include internal users as well as key stakeholders within the client organization. The distribution channel should support Android apps as well as iOS and, optimally, any other application type I would like to deploy. After searching, I found HockeyApp. What an amazing tool HockeyApp was. Not only did it meet those requirements, but it also allowed for integration of their API to include advanced features. For years, I used HockeyApp without issue. Apps are built by my automated build platform, pushed to HockeyApp, and users are notified of new test versions.

Fast forward to today, and HockeyApp is now owned by Microsoft and being moved to AppCenter. To be honest, I’ve never been much of a Microsoft fan. I could point to countless reasons such as lack of good native development tools, poor support for scripting or automation, their hatred of Linux (even calling a cancer), Internet Explorer which worked differently from every other browser for developers, and now I can add their horrible transition to AppCenter to the list.

Today, I received a notification from Google that one of my apps was removed from the store. It appears to be a mistake on their end, but I need to verify and respond to Google. So, I uninstall the copy of the app from my phone and go to AppCenter to download the app onto my phone. The first thing I notice is that I have multiple copies of the app but none of them are tagged with the appropriate build type. I can’t tell free version from paid version or beta version or anything else – just a bunch of apps with the same name. After clicking on the individual versions, I find out that none of them are available for me to download. After digging around, it appears I need to release the app to a distribution group. It had been setup properly for years, but now it’s broken. So, I go to the website to check the distribution groups. Guess what? There is no way to change them in AppCenter. I can see that I have a few groups, but I can’t even tell who is in them. So, to fix the problem, I will need to go to every single app I have (a total of 28), add them to a new distribution group (with unknown members), and rerelease them (generating 28 emails to each member of an unknown distribution group). What a horrible experience for both me and my team members. As a small business owner, I don’t have the time to waste to make all the necessary changes, updates, reconfiguration of continuous development pipelines, etc. What an absolute nightmare.

Maybe my experience will be better once I get everything transitioned to AppCenter. Perhaps these are just short-term growing pains. Regardless, this is just another example of why developers have always been skeptical of Microsoft – and another reason I’m glad I use Bitbucket instead of GitHub.

Posted on

AI Resources

Today, there are countless AI products and resources available for the developer. I’d like to review a few of the resources I’ve used.

AIFH

The first resource on my list is AI For Humans (AIFH) by Jeff Heaton. Of all the books on programming artificial intelligence, his books are by far the best. Written without any heavy math, Jeff’s books explain just about everything you could want to know about artificial intelligence. Additionally, Jeff has his own framework called ENCOG which can be used in Java, or can be run as a standalone GUI for development. Of all the resources I’ve used to date, this is without hesitation the best.

AWS DeepLens

I was recently sent an AWS DeepLens by a client for an artificial intelligence project. It’s always exciting to receive hardware from a client, so I was certainly looking forward to this! Unfortunately, my experience with it has been a bit less exciting. For starters, I’m stuck trying to figure out what exactly the purpose of this box is. Am I to use this to learn artificial intelligence? Is this box intended to be embedded within a product? I have absolutely no idea. While the DeepLens is a neat toy, the setup is far from simple. An internet connection to AWS is needed, certificates need installed on your machine, and everything is configured through your Amazon account. Even worse, the service does not appear to be free. (Note, you do get one free year on AWS, but after that you pay for the service. After that, I have no idea what it costs to run the DeepLens, but I’ve noticed warnings about costs associated with some actions). I applaud Amazon for trying to bring deep learning to the masses, but I think this product is a dud.

DL4J

DeepLearning4Java (DL4J) is a Java library for deep learning AI. Written by SkyMind, DL4J is one of the most well-known AI libraries for Java. With the earlier versions of DL4J, the user would have to install various native libraries such as ND4J (N-Dimensional Arrays 4 Java). This proved to be more difficult than it sounds as various libraries were dependent on other libraries, documentation was scarce, and error messages were cryptic at best. Fortunately, with the 1.X version of DL4J, the install process is streamlined to sampling cloning a repository and running a maven build. Native libraries are managed within the maven build, saving users the trouble from earlier versions. With these changes, DL4J is an excellent framework I would recommend for any AI project.

Posted on

Getting Started with Artificial Intelligence

It seems that artificial intelligence is in the news more and more. Most larger companies use AI for something within their business, and more and more businesses are finding ways to improve their organization with AI. Purchase recommendation systems, self driving cars, video games, language translation apps, and route mapping software are just a few examples of artificial intelligence we see every day.

But where does someone interested in AI get started?  AI can be very complicated, but that doesn’t mean you can’t get involved too! One of my favorite resources for learning AI is the series Artificial Intelligence for Humans by Jeff Heaton. In his books, he covers a variety of topics including genetic algorithms, machine learning, clustering, linear regression, swarm algorithms, and so much more. While these topics can be complicated, Jeff presents them without all the math in a way that is far more readable than most texts on the topic.

If you are interested in AI, one of the first questions to answer is what programming language you want to use. While any language can be used for AI, the bulk of the tools and frameworks available exist for Java, C++, Python, and R. If you’re big into number crunching and Big Data, R may be the obvious choice. If you’re not a programmer, Python may be easier. For existing developers, Java or C++ may be best.

What type of AI do you start with? Try genetic algorithms or swarming algorithms. Genetic AI assumes the answer to be like a genetic genome and, through a series of mutations and genetic splicing, attempts to find an answer.  Swarming algorithms look at groups of objects and attempts to have them behave like a cohesive team. Swarming algorithms are great for games with AI controlled enemy armies and are commonly used by game developers. Other common, and simple, AI algorithms include K-Means clustering (used to group objects by similarity),  linear regression (used to predict unknown values using relatively simple algebra), or path finding algorithms such as Dijkstra’s Algorithm.

Once you understand the basics of AI, you can move onto frameworks like DeepLearning4J or TensorFlow to help create Neural Networks (a far more advanced type of AI) or look into libraries like OpenCV for tinkering with Computer Vision.

Whether it’s creating an emulation of a fish tank using swarming algorithms, solving the Traveling Salesman Problem using genetic algorithms, or calculating a path through a maze using Dijkstra’s Algorithm, artificial intelligence is loads of fun.

Posted on

.Net Core 2.0 Woes

One of my favorite technologies is Microsoft’s .Net Core 2.0 framework. Some of the things I love are that it’s cross-platform, it has an excellent framework for interacting with databases, and their MVC model is straight forward to implement. However, with all that Microsoft has managed to accomplish, there are still some major shortcomings. (Before going forward, let me point out that I develop on a Mac using Visual Studio – so a native Windows user may have a different experience.) For example, while the Entity Framework is amazing – the support for stored procedures is less stellar. After days of searching for solutions, I finally came across a working answer. This is the biggest problem with .Net Core – lack of answers on the web. Or more properly, the lack of the right answer. Look up any question on how to do something using .Net Core and you’ll find a dozen different answers – none of which actually work. This problem was repeated over and over again. When I tried to store a simple session variable, I saw tutorial after tutorial – none of them worked. Next on the list is the NuGet package manager. As a Java user, I have grown accustomed to pulling libraries from a remote repository using Maven or Gradle – so I was happy to see similar functionality for .Net. The problem? It’s impossible to find what package contains the thing you’re looking for. Online code never includes the appropriate imports, so you’re left to guess as to what you need. This problem is exacerbated by the horrible idea of creating class extensions. Every package you import can add extensions to classes in other packages. This sounds great, but in reality it only complicates figuring out what to import. Last, the way you configure the web server is the worst I have ever seen – you configure it in code! Want to use a different port? Hard code it. Need to run the app in a subdirectory? Hard code it. Need to setup database support? Hard code it. So, as the app goes from machine to machine it actually has to be rewritten to support the configuration required on the new server. I have seen various examples of how to use configuration files to accomplish this, but like everything else the documentation is poor and inconsistent at best. In the end, while I do love what Microsoft has accomplished with .Net Core 2.0 and their MVC framework, it still has a long way to go before I would consider it to be able to seriously complete with more mature frameworks.

Proudly powered by WordPress