Articles from businessmen from Silicon Valley. Alexey Fedoseev: “Make the structure in chaos! And how to approach this ... "(Part 2)

With this article, we continue our series of publications from the authorship of “honored entrepreneurs” from Silicon Valley, in which they talk about their start-ups, the areas in which they work, and other things that fall within their area of ​​expertise. Earlier we already acquainted the inhabitants of Habr with interesting people and places of Silicon Valley with video interviews and reports from Silicon Valley .

About the author. Alexey Fedoseev , founder of 1World Online, previously worked as a product development director for Motorola, which sold his previous startup 4HomeMedia. In this part of the article, Alexey talks about the technical side of his company.
')
In the first part of our series, which tells about the product of 1World “ Brave and new, but not too clear world ”, I gave a brief overview of how we began to approach the problem “So what do people really think?”. To do this, we put together surveys and expert opinions (so that people can better understand the problem and both points of view on it), and also separated the facts from opinions, adding to the discussion Data Points (“data elements”), very brief and accurate, with links that you can check and see the details.

It would seem that a simple introduction of structure into the subject of discussion immediately changed this discussion both quantitatively and qualitatively. After all, it is very difficult to defeat chaos and come to a meaningful conclusion, tackling any problem, if you do not apply a structural approach. So, let's summarize what exactly the others did to 1World :
1. In each important survey, we add two expert opinions "for" and "against"
2. We separated opinions from facts by adding Data Points as separate objects.
3. We gave people the opportunity to vote on the question plus evaluate and experts, and Data Points

Thus, the discussion expanded and acquired a three-dimensional character. In other words, it is possible to win an argument in it, but lose on points in your expert opinion.

When the company, or rather another group of founders and first key programmers began the project at the end of 2011, several problems arose:
A) Which interface to start with - Web, iOS or Android? For a number of reasons, we settled on Android (since there was the most solid experience and the possibility of rapid development), and we made it, starting to demonstrate in the middle of 2012 (we worked, as usual, at night and on weekends, because during the day main job).
B) On the back end, there was a discussion of options, but there wasn’t any serious alternative to Amazon AWS. Its use was so obvious (straightforward), plus we dealt with them in a past life, so the infrastructure was made just for their reference guide.

By the way, when we officially launched the product in March 2013, at the SxSW exhibition, our booths were nearby, and we talked closely with our colleagues from Amazon, even received special discounts as manufacturers of a promising app.
As for the Application Server, our architect chose GlassFish (open source, launched at the time Sun), because he used it in many previous projects and has great confidence in him.

After some time, there is an urgent need to develop a website. After heated discussions, the majority of votes decided to write a so-called one-page application, where all content is requested from the backend via AJAX requests. The main advantages: interactivity for users and simplification of the server infrastructure (by eliminating the server (s) responsible for rendering web pages - even then we were aiming to create a mass service). There are also shortcomings in this approach - an increase in development time, problems with SEO and sharing in social networks. On the basis of the same web application over time, we developed a facebook application and a widget for installing our questionnaires on third-party sites. Looking back, it is impossible to confidently say that a one-page application was the right choice for our project, but this solution cannot be called a failure either. The answer to this burning question will only give time.

In parallel with the website, we began developing an iOS application. For a start, I also had to pretty much break my head, choosing between native and cross-platform applications. The latter were very popular at that time, and of course there was a huge temptation to kill all the hares with a single shot. But here in time came Mr. M. Zuckerberg, who described the ordeals of Facebook in this area. Perhaps this was the decisive argument in favor of the native iOS application, which we do not regret at all.

To be continued

Thank you for your attention and for the positive feedback. Subscribe to our blog on Habré , Twitter , Facebook , YouTube-channel to follow the updates. Previous posts can be found here .

If you have a question for Alexey Fedoseyev, you can ask him here .