coder . cl

As a software developer, who has a side project — open source or not — is a software developer that can make products. If the developer can identify a community need, and produce a product, there is no problem for that developer to understand a specification and meet the desired specification. Also, the developer will be able to satisfy the correct solution and will be able to meet any observations about the requirements and the source code.

All the time I am reading tips on how to become a good developer, and there are many programming courses. But I am pretty sure that there is no strict method to become a good developer. The main reason that there are very good developers — almost on every place in the world — is not about their education, also is not about their methodologies and even is not about their passion for programming. There is a renewal trick, because they are always seeking new stuff to learn, and that keeps their mind fresh and clear about how they can solve each problems that they are facing. Almost the best developers that I know, are always learning, reading books and they keep searching information about related topics.

The basic definition of a state machine is a set of states S, with a subset of initial states I, another subset of final states F, and a set of transitions between states T, and every state s’ ∈ S ∧ s” ∈ S where I ⊆ S ∧ F ⊆ S, and each element on T { s’, s”, h }, with s’ as the current state, s” as the next state and h as the state action. On my current automation project where I am using ProView, I have each device controlled by state machines, and each user or automatic control request is queued on a priority queue, where the highest priority is executed replacing any previous execution state, because we have only one communication channel. So the basic state machine definition is not enough to hold every environmental variable that we are using to control each device.

I have started a new project on Haskell. It is called CSDS, or Core Stateful Data Structures. It will hold common functional data structures, mostly based on the implementation described on the book “Purely Functional Data Structures”, but on Haskell I am using type classes and implementing some common interfaces for that kind of data structures. Is very nice to work on project like this, I think that I will use this kind of project on integrations with projects like Cloud Haskell, for distributed processing. For example some of them like the StackLT and QueueLT are providing a class interface to work with the Foldable and Monoid type classes.

How many times did you get an “amazing job offering from an amazing company with an amazing environment, but they want the cheapest employee that they can get”. I have stopped trying to handle that kind of job offerings. Also, I am filtering all of them. Specially that job announce where you read in emphasis “market reference salary”. If they are expecting to pay an average salary, why should I work for them if am over the average? I am a real polyglot programmer that can handle more than three programming language paradigms and over 10 programming languages, and my understanding of programming goes much over the average.

As freelancer I am constantly looking for new projects. Sometimes is hard to find good projects, mainly because not all customers are technical customers, making hard to communicate some aspects of the project, among other stuff like programmer evaluation, without too much technical evaluation, mostly based on his accreditations. One of my latest projects was earned by handling a programming problem on the middle of the interview, I was using Skype to share my screen and allow the customer to review the source code and watch me how I was solving the problem. That was great.

We are doing pretty good advances on the ICS/DCS that we are doing. The current code base is using the STL std::priority_queue as operation request queue, but we are changing the task request priority each time that the communication channel is being used by any transition that requests the Profibus channel. We cannot saturate the communication channel with requests, because it runs on a very low bandwidth connection, so each request should be correctly measured, where all operations on the machinery is being directed by states machines with dynamic priorities of execution, and the STL priority queue is doing a good job. But was a little bit complex to think in terms of source code with the model presented as state diagrams and flow control diagrams on a whiteboard, and took me about one week to finish the final skeleton of the primary source code.