Development Delay / Moving to new codebase

[The Lady Shael]

Hello little warlords,

We are putting our current development on hold. We can tell you guys have really been enjoying the new features we've been putting out on Turbo, and we've been excited to reveal more next round. But we will have to delay those for a little while.


For the more technically inclined, we are updating RWL's codebase. (History lesson time!) RWL was built off of QM Promisance's code (version 3) back in 2002.  Since then we've been building all of our new features off of this code for 8 1/2 years, and technology has certainly improved since then. So we are doing what we should have done years ago.

We will be updating to version 4.4 of QM Promisance. This is going to be a long process, but lately Shadow and I have been cranking out new developments right and left, and windy and Peace are on board to help out as well. So we hope the delay will be less than a couple months.

In the meantime, all the current features of Turbo and Reg will stay as they are (and don't worry, both servers will still be up and running), and we will be sure to keep putting up awesome themes on Turbo to keep you guys occupied. =)

Ask if you have any questions.

~The Lady Shael
RWL Administrator

[Shadow]

In the meantime, we will keep tweaking the balance on the existing code and refining the new features so that we can drop them into the new code without any delay. So don't hesitate to keep reporting any bugs you find or suggestions you have regarding the code, as fixes will still be put in as fast as ever - we will just be delaying the introduction of several new features we have in mind until the code migrations is done.

[Yellow Eyes]

Why the attraction to that?

[The Lady Shael]

Why are we updating our code or why did we choose QMT?

[Yellow Eyes]

Both. What do you think is wrong with the code now and why does it need updating? And why choose the QMT 4.4 version?

[The Lady Shael]

I think a more appropriate question is, "What isn't wrong with the code?"

Here's just a short list:

-it isn't object-oriented
-it is not well-documented (or even just documented)
-it had SQL injection vulnerabilities (that I covered but in duct-tape fashion)
-it's just very messy in general. And some of that is my fault, since I was still learning PHP when I became in charge of the code

We are choosing the 4.4 version because it is QMT's latest stable version. And the reasons why we decided to go with QMT:

-it's our parent site, therefore their features make up the basic skeleton of our site. They haven't seem to have added any new developments or features in the last 8 years.
-it's well-documented
-it's object-oriented
-it will most likely provide the quickest transition time compared to our other options

[Shadow]

We are currently hanging out in the wrong loop ^_^

[Sharptooh]

I've heard a lot about object oriented code (I have yet to learn about it in my course) but what kind of benefits does it give? Faster? Safer? Better organised?

I also found that picture pretty funny Shadow XD it's disturbingly true

[Shadow]

credit to XKCD, click the image for a link

OO code is generally more intuitive, easier to maintain, better organised, etc. I am only just leaning about it now myself, so Shael can give you a better answer I'm sure. I don't know if it offers performance enhancement or not, but for an application like this, speed is not really that important since most of the time consumed is spent waiting for user input ^_^

[Sharptooh]

Thanks for the answer Shadow

OO code is generally more intuitive, easier to maintain, better organised, etc.

I can see why that would be helpful considering how much code actually makes up RWL

Unorganised code looks so scary 0_0 especially if it doesn't have indents (well I think so anyway)

[Gen. Volkov]

Is this an April Fool's joke?

[The Lady Shael]

Actually, no, this one we're definitely serious about.

[bjornredtail]

Wait, there's an OO Promisance these days? Wow, I have been out of the loop for a while.

The other question is: did Boze/Retto/Shael/Peace/Windy/Shadow/myself really add that many features? I mean, besides Peace's awesome front-end stuff, I can't think of all that much that RWL does that differently. Of course, I haven't looked at or played RWL in nearly 3 years, so chances are things probably have changed somewhat.

[Shadow]

Well, the balance has changed a lot, as well as many of the equations that make up the basic game (in income and workers, for example). There are also a lot of RWL-specific features like race spells that need to move over.

All in all it shouldn't be too bad to switch over, but it will require a careful look at all of the equations to make sure everything lines up, which will probably be the most time-consuming part.

I have taken a peek at the QMT code and it is rather nice. It has done away with most of the global variables, and while it still has a few hard coded constants and magic numbers lying around, it has a lot less than it could have ^_^

I think that development will actually be faster after the switch, just because the code will be much more modular and therefore predictable.

[Gen. Volkov]

Actually, no, this one we're definitely serious about.

Oh, cool. Bout time.