Meetings

Nov 25, 02:01 PM by Eric Allen

Some people are detail-oriented, and some are big-picture types. A healthy mix of the two is crucial for any successful team. However, I’m not sure putting the two types in the same meeting is ever a good idea. I just spent two hours listening to discussions of minute details for a project that we don’t even want to do right now. Before even answering the most basic questions, like “can we afford to do this?” my detail-oriented teammate went crazy on all of the details of the project. It was amazingly boring and unproductive. At least our manager had the sense to realize we’d wasted at least $250 of developer time screwing around!

I’m going to take this as a lesson to avoid putting detail-oriented and big-picture people in the same meetings. Keep the communication between the two types asynchronous and mediated if possible. Putting them together just wastes time.

Comment

First Place

Nov 8, 10:19 PM by Eric Allen

A couple of weeks ago my research advisor for my open source project encouraged us to enter a programming competition sponsored by several groups at RPI. I figured it would be a interesting way to spend a few hours on a Saturday afternoon, and hey, I might just win an iPod touch!

As it turned out, I won a $100 Amazon gift certificate, an 8GB iPod Touch, and an interview with Bloomberg! Not bad for three hours of effort. Oh, and I got a free T-shirt, too. There were eight problems, and I was one of two people who solved four of them in the three-hour period. I believe I was the only one using Java who placed in the top five, and I would argue that my choice of language is what gave me an edge. We got to choose between C, C++, and Java (I know, terrible choices), and I was by far most comfortable in Java. The big advantage came in the libraries, though. I had set operations baked in, didn’t have to waste much effort parsing input, and had full OO at my disposal. I guess my real-world experience helped, too.

So where do I go from here? I’m sort of thinking about trying out TopCoder now that I’ve shown that I can solve these kinds of problems pretty well. On the other hand, I still like building stuff better. Back to coding on real problems!

TipJoy Textpattern plugin

Feb 13, 01:08 PM by Eric Allen

Yesterday I quickly hacked TipJoy buttons into the articles on my site, but my solution was not very elegant or flexible. I have now gone back and put that code into a Textpattern plugin that can be downloaded here. All you have to do is install it like any other Textpattern plugin, and then drop a <txp:tj_button /> or <txp:tj_banner /> tag where you want them to show up.

My Computer Science Rant

Dec 28, 04:44 AM by Eric Allen

If you go to RPI, you’ve probably gotten my Comp Sci rant a few too many times. I was pretty convinced that computer science does not prepare programmers for software engineering before I got to college, but now I’m sure of it. Why don’t we have a software engineering major here?! Anyway, somebody else beat me to blogging about it.

Programming for fun

Dec 2, 04:36 PM by Eric Allen

I must admit I’m a bit of a Risk addict. When somebody had the idea of making it a Facebook app, I jumped on it. So, I’ve been playing a lot of Risk lately, but I’ve been plagued by a question: if you’ve two equally sized hordes against each other, is it better to attack or defend? The defender wins ties, but the attacker gets to roll more dice. I probably could have figured this out mathematically, but I’m a programmer. I chose Ruby because it’s fun and easy to use and whipped up a little program to calculate the odds by facing off two very large armies (100,000 in my version) and seeing who loses more. I was pleasantly surprised (yet again) at the clarity and simplicity of Ruby. I’ve made the source code available online. The project took me maybe half an hour, and it has answered the question conclusively: on average, the attacker will lose only 85 armies for every 100 the defender loses. Therefore it’s better to attack while you have the chance instead of waiting to be attacked if you ignore all other factors.

A Mac virus!

Jun 12, 09:35 PM by Eric Allen

A virus on a mac? No way! you say. Wrong. Microsoft has provided our friends the virus writers with a wonderful cross-platform virus environment: Word.

Read more...

Previous