Ok, up till now I’ve been talking about me and my experience, it’s time I began talking about the starting point in my current role and the developing environment that existed. Basically, that can be described in one word: Inexistant.

The problem is that there were already more than 10 applications being developed, maintained and ran. All done by  different contractors, some of them quite good, most of them with just a basic knowledge of what they were doing.

The company wasn’t much help with this. It’s focused in giving infrastructure services to a (much) bigger corporation, just recently created (like 4-5 years ago) whose main job is related to running servers and mainframes. For this company the software was a second thought and most of the apps running on those servers had been inherited from the other companies (“entities” in the local lingo) that formed the corporation.

What this means is a developer hell.

Most of the people working in that company are system or network administrators, these people don’t really use their computers, just use them to run a few local applications and to connect to the different servers, routers, switchs… The people that work with the mainframes do the same, so the basic spec for a “workstation” here is a P4 with 512MB of RAM. Try to run a local database, a local webspehere AS and your choice of Java IDE (among other tools and apps like the mail client, document editor…) and things can get really exasperating.

Then there are all the different policies. Basically, I have been the first “normal” employee (outside the reduced IT dept, I mean) to get local administration rights to his machine. Unfortunately it’s something basic if you have to be installing-uninstalling several development tools, attaching debuggers to running processes and things like that. Perhaps you don’t need them every time, but you need to have the option. In the case of the company chosen tools, it’s a must. The environment is heavily based on IBM tools (Websphere AS, Websphere Studio, DB2, MQSeries…) and to run the integrated Websphere AS in the Websphere Studio you absolutely need admin rights. A bug in the program? Something that can be fixed with specific permissions? Probably, but I haven’t had the time to play around with it enough and getting admin rights solves the problem.

This way of thinking is extended to the servers. Don’t get me wrong, I’m totally in favour of having strict security policies in the production servers. Even being an experienced developer (or perhaps because of that) you can easily break something so it’s better to only allow access when and where you need it. But how about the test machines? Does having to ask for permission to update an app in the test machine looks right to you? It is, after all, a test environment, it is meant to change quickly and to always have the last versions (developing versions) of everything. At least now it is automated, but still, it means that there’s no easy way to set up a continuous integration environment.

I must say at this point that, even if the environment is not the best and some policies seem tailored to hinder development, the managers usually keep an open mind and have actually been my best allies to make things easier for everyone. Just five minutes looking at my screen and I got 2 extra GB of RAM. One mail explaining what are admin rights and why I needed them and I got them the next day (and loose them the next week and had to start over with the mail until I got them permanently). Both changes have now been included in the standard needs for the java department and I hope that people in other departments (which are usually running through hoops to be able to do their job) will start asking for the things that they need.

Well, I think that this is enough for now. I’ll continue next week with this, but focused in the software side (and I’ll probably have a bit of a rant about contractors).

I feel that, to really explain the things I want to write of in this place, there’s some kind of introduction needed, someplace where a visitor can read a bit about me and perhaps understand better the reasons for me saying whatever stupidity I’m currently saying.

First of all, I’m a computer engineer. In my country, in the way univerities work there, that means that I had a lot of theory of all things computer related (from software development and engineering, compiler’s theory, data structures, algorithms, artificial intelligence but also hardware design, electronics, robotics, signal treatment up to organization of a data center, enterprise administration and laws and down to physics, algebra, calculus, numeric and mathematical analysis…) but very little practice and almost no possibility of specialization (not even being able to focus more on software than hardware or vice versa). All the practice we had was a couple of laboratories focused on circuits and several programming exercises through the five years of classes.

Most of these classes where taught by people who stuck to some kind of book and didn’t wanted (or weren’t able) to go further than that or even counsel the alumni that wanted to learn more than what was in that book.

I began working on my second year, in a dot com company (this was the year 1999), and to me that was a complete new world with a lot of freedom to learn, to investigate and to propose ideas (I suppose that most people will find strange that I found more freedom to do this in a job than in class, but that’s how it felt like). It wasn’t long before I was on charge of small projects (and the only one working on them, for the most part), forcing and motivating me to study and learn much more than all the classes and teachers combined.

Then the dot com crisis came and this company folded (well, not really, it was bought by another, bigger company, but changed the way of working so much that the things that made me like it disappeared). I’ll probably talk again about this place in the future, we had some interesting ways of working that I’ve only heard of in other companies since then (Google, with their 20% of the employee’s time for other projects, for example, is one). At the same time, an opportunity came by to work on a big project for a car manufacturer, with applications and systems used worldwide, so I took it and had to deal with a completely different environment, full of legacy code and very strict rules.

Since them I’ve moved a lot: I’ve been in big companies and start-ups (one of them my own, only lasted two years, though), in green field projects and maintaining old legacy code, involved in simple internal web apps and huge banking systems, all in about ten years of experience, about the same number of companies and three different countries. I’ve always tried to see both the good and bad decisions taken in those companies and learn from them, sometimes I’ve even succeeded, I think… Well, I hope, at least, and will continue trying.

So here I am  now, with a good opportunity to try and set things up my way (look at my last post) and see if I’m better than all the managers and bosses I’ve been bitching about all these years. At least I hope I won’t be worse.

Either I’m busy, or tired, or there are more important things to do… Enough of that.

Hopefully starting now I will begin to turn this into a real blog.

One of the main themes of this web will be software development, after all, I have been working as a developer for the last 10 years (mostly with java, but I’ve also done things in several other languages, from cobol (urgh!) to .Net).

I am currently creating a java department from scratch in my job which has had nothing like it until now. So I’m writing procedures, setting up tools, helping other developers, dealing with other departments and, at the same time, trying to tame several java applications that are running wild (and that have been a really good source of what-the-fuck? moments lately). I’ll try to comment on my experiences on this both for my own use and in case it is useful to someone.

Cheers!

The thing is that I’ve decided more or less what I’m going to use this blog for (yes, it took long, I’m slow, so what?) and the first step has been to import all the posts I had in a weblog in barrapunto (all of them in spanish, I’m afraid, will be translating them as I go), my first blog and the only thing I have that deserves that name.

For the rest, I will try to write more frecuently, but I can’t promise anything…

Here is where I should be explaining a bit of what is this and what I’m going to talk about. The problem is that I’ve created this blog just to play around and do some tests, so, at this moment, I haven’t got the faintest idea of what I’m going to do or say here (or even if I will actually get to do or say something). If you have arrived here searching for something, thank you for your visit and I apologize for the time you’ve lost.