Arda -- Version 1.13.2 -- Delayed | Veterancraft Dedicated Server | Forum

My apologies, I am venting a bit. Life, these past two years (and more), just seems to have been one mishap followed by another; well beyond coincidence. I am beginning to suspect that somewhere in the world an adversary or disgruntled individual is sticking pins willy-nilly into a voodoo doll with my likeness.

The good news is that the surgery was textbook perfect. At my follow-up examination this morning, the doctor was astonished how quickly my eye was recovering. I cannot see squat out of it; it is akin to looking through a bottle of milk. This was to be expected and will clear up; how quickly remains uncertain. I was told it could be as short as a week or as long as 6-8 weeks. I have another follow-up visit later this week to gauge the progress.

The bad news still is that to promote proper healing, I need to be lying on my left side regularly to ensure that the gas bubble is pressing against the retina as it heals and attaches itself again. The silver lining is that rather than being up for 1 hour followed by lying down 2 hours, I now can be up for 2 hours and lying down for 1 hour; rinse and repeat. The pain is still there, but slow getting less. Anyone that has had the misfortune not dodging several left hooks against their face should be able to relate.

More vertical time gives me more time to try making some progress. Unfortunately, given this latest malady, family is coming to visit and staying to help out. My inner voice is shouting “no, no, please stay at home, I am okay”. Alas, bless them for caring, but it will impact what up-time I have and what I can do.

The good news is that I can do stuff, albeit slowly. With one reasonably good eye and a patch over the other, this pirate hacker can still read and write code. Lots of typos and mistakes, but it is getting better. No better time to improve my typing skills. Sadly, even with having set GNOME (my main system is running CentOS 7.6) and all my apps to use dark themes, I do tire easily.

Normally, Mudwog would be able to carry the ball farther, but despite my panache for documentation, a lot of the details are still locked in my head. Plus, it is the holidays, and he has his things to do too. We will be touching base after XMAS to see what we can do to get back on track as best as we can.

Could we have Arda up and running today? Yes, mostly. However, it would not be a pleasant experience for our members and us. Chances are that it may crash, items get lost, maps may get corrupted. All of us have waited this long, it is better to wait a little longer and do it as right as possible. Granted, this may not sit well with some. As much as I too can succumb to the wants of instant gratification, there are other driving forces that must win out. Furthermore, I make enough mistakes on my own. There is no need to tempt fate to add more.

Given that I have an ample supply of forced downtime, I have been catching up on a lot of YouTube videos about Minecraft, Java, unit testing, and cool ideas for mini-games, though I have to be careful not to let that distract me from the mission at hand.

If anyone has any specific questions burning in their mind, please feel free to ask and I will try to answer them as best as I can. I cannot promise a satisfying answer, but at least you will know more.

Okay, that is it more now. Back to lying down and watching a few more videos.

Best wishes for the holiday season,

Frelling

Greetings all:

Thought I would post a vent… err, update on my condition and the Arda progress. My optimistic view from earlier this week was a bit too optimistic. I think the pain medication made me feel a bit euphoric and overconfident. Not that anything is bad. My checkup yesterday showed that the eye was healing very well; better than expected.

While that was great news, the problem still remains. Being someone that spends a lot of his waking time in front of monitors both at work and home, having an eye problem had a greater impact on productivity than imagined.

While at first, I could not see anything out of my eye; akin to looking through a glass of milk; it has been getting clearer. After the vitrectomy, the eye was filled 90% with an inert gas bubble to help keep the retina in place. As mentioned before, that required me to lay down in a certain position regularly to ensure that the bubble was in the right position. Over time, the bubble is absorbed and replaced with fluid by the eye.

After the surgery, the doctor mentioned that as the eye fills with fluid again, it will be like looking through a fish bowl until the gas bubble is gone, which could take anywhere from 2-6 weeks. No kidding! After a couple of days I started noticing a boundary layer between the fluid and gas; it was, as suggested, like looking through a fish bowl. When I moved my eyes or tilted my head, that layer jiggled and moved too.

At first it was interesting. For example, I “see” the fluid at the top of my vision and slowly grow downward since images on the retina are inverted just like in a camera. However, as the fluid level increased, it became more noticeable and disorienting; more so distracting. My comprehension of anything visual was affected more than I thought it would. Then again, I could never read a book in a moving vehicle without getting queezy.

It got to the point that I stopped trying to read or watch anything and switched to audio books. Obviously, it also affected anything I could do in front of a computer. Answering work email took long enough for me to even consider doing other things. I am timing how long it is taking me to write this post.

As of today, the fluid level is about 50%; right smack in the middle of my vision. The good news is that I can start to see clearly at times, even better than before the surgery, if I shift my head just right. Unfortunately, that makes it even more distracting. Wearing an eye patch, sealed with opaque tape makes things easier. Despite my eye being completely “blacked out”, I can still see the boundary layer move due to the physical action on the retina between gas and liquid. It is kind of neat, though I would never would want anyone have to experience it. Unfortunately, it is about as annoying without a patch.

Mudwog has been doing as much as he can under my direction, but progress is limited since he has very little time during the Holidays. However, we did have ample time to discuss tasks and goals in far more detail than normal. The silvering lining is that we have a pretty damn good process in place to make the Arda upgrade process run smoothly. A learning point of this is that we should probably spend more time talking than doing; something we both know, but rarely experience until forced to do so.

The key task for Arda is getting the Shoppe plugin completed. The problem with the old LocalShops and our JcShop was that they were closely coupled with persistent storage, making it hard to maintain or even switch storage systems (e.g., file, database, object storage, etc.). Initially, we made the same mistake with the new Shoppe plugin.

A driving point for separating storage from shop functions other than being a better and more flexible design, is testability. All of our plugins have been redesigned and updated to support maximum testability to ensure that when changes are made or features are added that everything else still works. This is something that third-party developers rarely do well or even at all, leaving side effects or problems to be discovered by server operators, many times with great frustration.

We have always strived for as much testability as possible. Unfortunately, Minecraft/CraftBukkit/Spigot were never conducive to testing, short of just running plugins on a test server to see if they work. It required us to create “patches” and dozens of mocks to make it as usable as possible in a test environment. While the “test jigs” that we had created did good enough initially, they were not perfect and often got in the way of testing or required complex setups. This is bothersome, since tests should be simple.

Thus, we had decided to revamp our test jigs and take advantage of new features provided by Java 8, JUnit 5, and Mockito 2, back in November. That really is the last step to updating Arda. Shoppe is the most complex plugin in that it touches on all plugin boundaries – command line processing, text displays, parsing user queries, networked storage, integration with economy plugins and others, search algorithms, etc. Its diversity also stresses testing and our tools as much as possible.

The process of writing and updating Shoppe tests truly showed how “messed up” our existing Spigot test jigs were, but more important, what we needed to change. My current downtime has given me much opportunity to read, watch videos, and ponder on best practices for making untestable libraries/applications (i.e. Spigot), work in a test environment consistently, rather than just winging as before. I have a lot of action items in place to spur this process along to get to our goal. It is now just a matter of me being able sit in front of a computer (a.k.a. screen time) to implement.

Prior to my surgery, I had also become intimately familiar with the Spigot contribution process while submitting a couple of bug fixes that were discovered as we were examining Spigot code in detail for the purposes of updating our test utilities. Matter of fact, we have several design changes for Spigot itself that would make it better for testing; but, that is a whole beast of its own and will have to wait until Arda is done.

This current situation, and a few in the past, clearly demonstrate that while having our own plugins provided us much greater possibilities and flexibility, it also hampers us by having them depend on a single (or two) individual, when changes need to be made. For the most part, our plugins rarely had to be changed after a Minecraft update other than to take into account new game features. Most of our headaches still come from third-party plugins. However, when our time is limited or sidetracked by unforeseen mishaps, it limits forward progress.

We do not envision a major rewrite of everything as with this 1.12 to 1.13 transition on every Minecraft update. Even so, a lot of unforeseen issues in my life over these past 2 years have greatly delayed and limited my ability to deploy the update. LWK and I have discussed bringing on a Junior Tech a few times, especially when things took longer than expected. It would be someone that would be immerse dnot just in server operations, but plugin maintenance and development.

This should not be construed as a position posting at this time, but rather as a heads up that we will be looking for a pragmatic self-starter, that wants to enjoy the darker side of code development in a production environment and help toward the general well-being of VeteranCraft. It will definitely be different from armchair/weekend coding most hacks are used to. It will definitely require an open mind and understand that administration and bureaucracy are a necessary evil. It will also help us put into place development tools and processes that currently are “unwritten” and contained in the minds of two individuals that have worked closely together for decades.

So to sum it up. I am doing well. I keep testing myself every day to see how much I can do. Mudwog and I have plans and tasks in place to move forward as soon as I am able to code more effectively. Nothing is at a standstill at this point; rather it is just moving very, very, very slowly; then again, relatively speaking, the overall update is not much different; we should have been done 6 months ago.

P.S. It took me 3 hours and 42 minutes to write this.