I have been developing software for over 5 years now, and one of my largest pitfalls that gets me is my tendency to overengineer.

Overengineering, for those who have yet to experience, or recognise it in their own work, is creating additional complexity to a solution with no, or negative results stemming from it. Usually added in for the sake of abstraction, preferred software patterns or sometimes irrational levels of encapsulation.

I started overengineering when I finally “got” Object Oriented Programming (OOP) back in 2007. Just like many of us, I saw the power in representing solutions in terms of objects. My reaction was to convert all of my current projects into Object Oriented format within the PHP language.

Anyone who knows PHP will know why PHP is bad. At this point it is important to note that PHP has never been designed for Object Oriented Programming.

I was developing an image host that I have mentioned in the previous post called “The Rotating Signature Host”. When I completed the first edition of the site in OOP, I looked back at the code and saw it had patterns, that parts could be abstracted and reused. And so I set out to abstract all of these behaviours so that I could reuse them. Everything from page loading to session management, to language, to every little aspect was consolidated into it’s own module.

This might seem ideal at first, as this is essentially what a framework does. It might seem that all of my future products would have benefited from this awesome framework.

That was not the case (more…)

So I just had my holiday in Kuwait for 7 weeks with my fiancée. It wasn’t a productive time code wise, but it was a time for personal growth.

In more relevant matters, I have made a decision regarding one of my projects;

I have dropped the mod called SurvivalCraft.

I’m going to admit that it was partially down to diffiulty, and as a result, partially down to time. It was my first time writing code that I have yet to fully realise a decent and capable solution for. I felt I did quite well, as when it worked; it worked well. When it didn’t, it was clearly a bug. The algorithm seemed to be fine, but my integration of it into Minecraft was about as poor as you could get.

It would be obvious that I designed and coded the algorithm completely separately from the game, then forced it in. It may also be obvious that I didn’t do any form of unit testing or integration testing. Call that noobishness, I’m still not one to do automated testing on a hobby project. I might one day though.

It’s good to know when to quit a project you can’t complete on a realistic schedule. I have the will to finish it, and maybe, one day I’ll skip enough classes to find the time between hobbies to actually do so, but that’s demanding and pulls from time more productively spent improving my skills on projects closer to my ability. If I were to reattempt an A* algorithm, I’d reattempt it in a closed, non-minecraft, environment so I can ensure I understand it before using it in a live environment.

In wake of this, I have started a new project (more…)

Well, here it is, the most useful update MOAPI has had in a few months.

I’ve finally decided to add the all-important key binding options and text input fields, something it was lacking sorely before. I postponed these features to flesh out NatureOverhaul and Survival Craft.

Now they are out I can imagine more developers choosing to develop with ModOptionsAPI rather than other solutions. Which would be nice, as I want to see something I have created put to the test in the wild. Something other than a website or web app anyway.

As for the future, version 0.8 of MOAPI will not be backwards compatible. I will be restructuring aspects simply because it was my first API and I didn’t do a great job of it. By my standards at least.  I mixed up supporting libraries with the API, I used too-long names, and used Strings to pass information too much when a perfectly good object which *contained* that very string was already present, and more useful, amongst other mistakes. There is also the matter of the secret aspect of 0.8 that I’ve vaguely referenced in the past.

I will also be adding IDs to options in order to abstract the name from the storage mechanisms a little more. It only makes sense, and I originally shot it down because I felt it could be added later, and it can.

Expect good things from 0.8, which may be here in a week or a month, I can’t say.

SurvivalCraft has hit 0.6b2 today and I’ve decided to stop developing for the night. This Minecraft mod is one that creates a much more deadly environment within the Minecraft world.

Today I managed to get a fair amount done in improving the A* Pathfinding search that the mod is implementing, but it seems like there are problems with the goal determining and the heuristic, or something along those lines, as some severe bugs exist causing mobs to take erratic paths, spin on the spot.

I have a feeling it’ll be a day of solid developing before I can reign in performance too. I first of all need to fully develop some way of visualising the search tree that results from the A* search. Of course, this is my first A* algorithm, so it’s pretty much going to be partially broken somewhere in implementation or understanding.

Once I finish the path finding section, I have a lot of fun toys to add to this mod! Stay tuned.

I’m going to publish stuff related to any mod or game I am currently producing. It only makes sense

If you’re a developer, you’ll probably have, at one point, programmed way, way, beyond your bed time.

I often get called out on it with people asking me “Why do you sleep so late Clint?” and “Why do you spend all night programming?”. I’ve always struggled to fully justify it to them until today.

My daily routine when freelancing or under studies for my degree is usually this: Wake up – Chores – Spend time with fiancée – Socialise – Work – Sleep.

You might think “Isn’t that cycle backwards?”. Usually work comes between wake up and chores. So why do I pick so sleep after work and work the night shift?

Here’s why: Most modern professions require primarily simple repetitive tasks. Some physical, some mental, some so simply they are essentially effortless. Then there are the science professions, one that require a lot of mental work, a lot of consideration, but also a lot of general writing too, and I hear it can get very repetitive. Software Engineering, implemented in the form of programming requires primarily mental activity. Design, structure, creation, testing, planning, preempting. Especially if you freelance and have to handle most of the work yourself.

This lends itself to a nocturnal pattern. I wake up in the morning, do the physical chores early when I still have the spirit and energy. I shower, wash, eat. Then I socialise, I spend time with people, become forfilled in my existence and life. This is all very easy on the mind. I will then work from around 9pm till midnight. Then I will socialise with those nearing sleep. By this time I am still very mentally active, but physically the day is drawing to a close. I will then work until around 5am. I wake at 1pm. It is flexible, from midnight till 8am I can sleep, ensuring I can drift over timezones to meet client timezones.

The main benefit here is that you spend your physical energy during the afternoon. Spend your social energy in the evening then spend your mental energy at night. Leaving yourself thoroughly exhausted by morning. You then sleep until afternoon.

I am writing this post after a long day of work and play. I’m thoroughly exhausted and am only writing this blog post as a testament to the fact.

I know this is going to see pedantic, but I must address this small grudge I have.

Software Developers and programmers often have strong feelings about naming conventions, from all_understores, to CamelCase, so capsCase and ALL_CAPS and whoThe_hellcareany-way.

I am going to address an issue I’ve found in the two varieties of camel case.

When writing a method like getHTML, or getHtml, it usually implies you have a consistent attitude towards acronyms. Either treat them like a word, or treat them like an acronym or .

Yet, for some reason, a fair portion of developers ignore their own convention, or the company wide/ group wide convention for one specific variable name.

Yes, just one. It’s very infuriating.

ID. Identification. In life, we often write ID or Id. Whichever is correct is not the purpose here. But people either treat it as one or the other.

Yet in code, I am not unfamiliar to seeing;

getHTML in the same package as getId, with an internal variable of int ID.

I just want to say; keep it consistent guys. ID should be kept the same as acronyms on your system. The fact it is an abbreviation is beside the point, in common English it’s treated as “ID” and should be treated in the same way as acronyms.

So it’s either getHTML and getID or getHtml and getId, not a combination of either.

Cutting deadweight is one of the best things you can do for your life as a programmer.

In 2007 I created clintonio.com as a free service to anyone. Over the past 3 years it consumed all spare “code time” I have to just maintain it. It killed all motivation, drive and passion for me as a programmer, and it was not the least bit successful financially.

This is not the only time this has happened; every time a new project was added on top, my drive, motivation and interest dried up slowly. Yet, each time I binned a project I would be euphoric. Yes! I would say knowing that time for a new language, project, design, or few hours with a book were ahead. And not long after would I be called up by another friend for a project. We’re just gonna build a CMS for …, and I’m back in the cycle.

I’d advise anyone who is currently struggling to maintain interest in their hobby, especially enthusiast programmers, to first cut the deadweight before anything else.

Today, after culling Clintonio.com and the Rotatr project, I left them both with a new, fledging framework in my hand, which is pretty functional (Rotatr Framework) and within just two days completed two new projects, even though they were small. A wordpress gallery extension I promise to make for my fiancée’s website back in August, and a new small startup group site called Digital Uppercut. I also made two blog posts in one day!

A new ethos has come out of killing my gargantuan project; fast development. I don’t just mean that in a buzzterm sense either. I’m talking “stop over-engineering solutions”. While my Rotatr Framework is my favourite piece of software yet, it still didn’t help me release my products, even it isn’t released yet. Today, rather than spending time creating tonnes of widgets for both of the projects I completed, I said “Fuck it, I’ll use the least code possible to get to the solution, and as much pre-existing code as possible”.

The gallery was approximately 100 lines, maybe less and the project site was about 50 lines of PHP inside the Rotatr Framework and 2 rewrite rules in lighty, and 5 lines of CSS on top of the Rotatr Framework.

I recommend both of these approaches! Have fun developing. Please comment with opinions and suggestions! All are welcome.

A little self promotion here; digitaluppercut.com is a new project group currently containing two experienced developers. We are using it to showcase our work, projects and skills. Consider it a group portfolio.

If you wish to join a team of application and web developers for projects that will be for profit or free and open source, then you are welcomed to email us at admin@digitaluppercut.com.

Clintonio’s Personal Blog now posts to his Twitter.

Read my other posts  here