Let’s start with the first one. Information silos are pretty common in big corporations where in information gets stuck between individuals, groups or departments. This in turn creates difficulties when proper turn over is necessary for your replacement when ever you leave a job or go up the hierarchy. This is a pretty common scenario, and most corporations, the solution is creating documentations. Tons and tons of documentations that are not only redundant, but pretty tedious to maintain, and because of this, most of these documentations will be severely outdated, unless you actually dedicate a person in keeping these documentations up to date. We cannot really forgo the need for documentation, but we can forgo the redundancies in them. Most of the time what a employee is looking for is contextual to his/her needs and these are not captured in the documentation. This is where a Q&A system comes into play, capturing these contextual requirements and making sure that it can be easily tracked, solved and archived for future reference.

Now on to our second point, any new enterprise system will need to be integrated to the company’s other systems, making sure that we can not only provide single sign on support, but also make the data that is archived in the system available to other systems via API. A Q&A system that is properly curated can be a gold mine of contextual information for a company. Building on top of this information can be very beneficial to a company’s knowledge management system.

Although I have several concerns, like:

1. Do I need a sales guy?
2. Where do I get a business savy co-founder?
3. Most importantly, how do I get that first sale?

I guess, i’ll worry about those when I get my first beta version out.

So with that in mind, and after spending the entire day finding and testing different ORM frameworks (Doctrine, Propel, php.activerecord), that I decided to write my own instead. Don’t know what to call it yet, nor have I decided whether to opensource it via github. Its basically, at its core, very similar to how rails does it (well that’s the holy grail of awesome implementation). Still have a lot more work to be done though, for the time being it will remain as a postgres only ActiveRecord implementation, but I will, hopefully when time allows, have it refactored to use different types of RDBMS.

I’ve always been a messenger bag kind of guy for the past few years, but shoulder problems can be really painful specially if you’re walking around Singapore/Philippines for hours on end. So I did my research and found one of the best selling bags for 15/13 inch macbook pro’s. I love that it has a lot of space inside without looking like a turtle outside, and the fact that it doesn’t feel like you’re carrying a 15 inch macbook pro + an ipad helps specially on those long walks.

Incase Compact Backpack

I’ve never been a fan of puzzle questions during interviews, it defeats the purpose of the interview itself since the questions doesn’t relate to the what he will be actually doing. So i’ve been using FizzBuzz to filter out our candidates and choose between those that have successfully answered the question. The question goes like this:

Create a program that will loop 1 to 100. Print “Fizz” if it’s a multiple of 3, print “Buzz” if it’s a multiple of 5, print “FizzBuzz” if its a multiple of 3 and 5, otherwise just print the number.

At first I was a little skeptical about using this question because of its sheer simplicity, but the sad part is it actually works. I usually just ask the candidates to use pseudo code or whatever programming language that they’re most comfortable using. I sometimes wonder how they did their task in their previous job or how they got a computer science degree without being able to pass such a simple programming question.