Last fall, I ran into a dilemma. A sophomore in high school, I found that my honors history class entailed a good number of lengthy readings for homework every week. The readings were quite interesting; however, my crammed schedule rowing crew left me with limited time in which to complete my homework. Having written computer programs before, I wondered if I could write a program that would show me just the sentences that matter most in a given text.
I’d read about Summly in the news when it was announced that Yahoo! had acquired it for upwards of $30 million. I thought it sounded incredibly cool when I heard of it and I made sure to download Yahoo News Digest when it was announced as a sort of successor to Summly in January 2013. However, as awesome as existing summary news aggregators were, I found that they offered no way for me to summarize content that I needed to read, such as my history readings. I’d written a basic Java program that extracted the first four sentences from a given text and attached an ellipsis (that’s how many news aggregators tend to summarize content); however, I wanted to take that further and create a mobile experience for my program. Then I began to research the field of Natural Language Processing and, in the months that followed, developed an algorithm that would extract the four most important sentences in a given text and display them as bullet points.
I created a basic iPhone app called Points and set it up with my algorithm so it could take the text that a user had copied onto their clipboard and summarize it in bullets. After spending the winter fine-tuning the algorithm and working on some other side projects, I finally decided to submit Points 1.0 to the App Store for review in early March 2015.
When Apple finally approved Points 1.0 for the App Store, I shared a download link on Facebook and waited to receive feedback from my friends. Just about everyone I talked to loved the idea of being able to summarize their homework, news articles and other texts. However, many of those people also explained how cumbersome it is to copy text onto the clipboard using a touch screen and suggested that I make it easier for users to summarize their content. I understood where they were coming from and thought about how I could potentially simplify the workflow of the app so users would have an easier time using it.
Through the feedback I received, I realized that most of the content users were trying to summarize came from sources on the Internet. To that end, I opened up my text editor and wrote a program in PHP that would extract relevant text from a given web page. I then decided to rewrite my entire algorithm in PHP and push it to my Web server so users could summarize web pages and news articles using just a URL. Many iOS apps offered built-in functionality to copy a web page URL, so users would have a much easier time using Points once it supported URL summarization. I also figured it made sense to get eliminate the ability to summarize text from the clipboard and created a redesigned iOS client (aptly called Points 2.0) to reflect the new Points experience. When I felt that Points 2.0 was ready for prime time, I submitted it to Apple review and waited.
While Points for iOS 2.0 was waiting for review by Apple, I occupied myself with the creation of a web-based Points client that would make use of the same PHP algorithm I had written for Points for iOS. The Web client took a few days to develop and was online in late April. After I pushed the initial build of Points for Web to my Web server, I submitted it to the popular product discovery site Product Hunt. The next day, it received 98 up-votes on Product Hunt and got more than a thousand page views.
Following the launch of Points for Web, Points for iOS 2.o was accepted by Apple review. A couple weeks later, I received an email from Charlie Egan, who was working on a news reader for various popular technology-focused sites called Serializer. He told me how he’d discovered Points while looking to build a similar service and had integrated the service into Serializer in such a manner that users could click a button next to each article to read a quick summary. It was really exciting to see how Points could be integrated into third-party experiences such as Serializer to provide contextual summaries for others’ specialized purposes.
I also had the privilege of giving a TEDxYouth talk about my experience developing Points. Additionally, Points went on to be downloaded more than a thousand times and was featured on AppAdvice and MacStories Weekly.
It was a really great experience developing Points and teaching myself the ins and outs of launching and marketing a product. However, as of now, I am finished with my work on Points. I set out to create an app that solved a problem I was dealing with (and, as it turns out, a lot of other people were dealing with it as well) and in the process I learned so much about the product development process. In that sense alone, I consider Points to have been a success. That said, there is so much more to be learned and as I enter my summer break from school I’ll be pursuing exciting new projects with a particular focus on education and connectivity.