Getting Turbine's attention
I write software for a living, as I suspect a lot of the other plugin authors do as well. In my job I often deal with bug reports from various sources, and let me tell you, not all bug reports are created equal. There are good bug reports and bad bug reports, and a developer is generally much happier to spend his time addressing the former than the latter.
A poor bug report is one in which the problem isn't completely described and I can't tell from the report how to reproduce the problem. Among my colleagues, we have a name for this kind of bug report: "Broke, fix!" (It often comes from foreigners who don't speak English very well, and we're racists so we mock them.) We generally have to follow up and waste a bunch of time going back and forth with the bug reporter to pry the necessary information out of them. We still fix the bug, but it takes longer and we're less motivated to work with someone who hasn't made a conscientious effort to make a good bug report.
A good bug report is when the person reporting the bug takes the time to thoroughly describe the symptoms and circumstances of the problem, and ideally, develop a "minimal test case" for me and make it as easy as possible for me to reproduce the bug. [A minimal test case is the absolute smallest number of steps that can be followed to demonstrate the problem.] This way I can quickly diagnose the problem, and when I've fixed it, quickly verify that it's fixed.
Well I was thinking, if the Lua API really is an unpaid side project of one of the Turbine Devs, it's probably even harder for him to get motivated to fix stuff. After all he's not getting paid for it; the only reward is pride in a job well done, and it's already pretty neat the way it is, warts and all. Whatever marginal extra pride that would be gleaned by fixing a bug here and there is apparently not enough to motivate him to put in the hours of his free time necessary to research the problems, fix them, and test them. I know many of you have probably submitted excellent bug reports through the official channels (Garan probably 20 of them) but it still takes a lot of unpaid work to address them.
So how can we make it really easy and, to the extent possible, take on some of the burdens of this tedious process ourselves? I propose
1. We make a "LoTROInterface Bug Tracker" plugin. It's basically just a listbox that contains a list of known bugs. Click on a bug, and a window opens that describes the bug. The window contains a button to execute a "minimal test case" that manifests the bug. Each test case is in its own separate .lua file (e.g. "bug_123.lua") so you can easily find the source code using the bug ID number. Our Turbine Dev could use this plugin to easily reproduce bugs and verify that they are fixed.
2. When there's an update, our Dev tells us the ID numbers of the bug(s) he has fixed. We then go on Bullroarer and verify that they are fixed, and that nothing else has been inadvertently broken as a result (regression testing). Basically, we all try out all of our own existing plugins thoroughly and make sure they still work the way we think they should. We have to be diligent about this so that we actually catch any regressions. If we let new bugs slip through, then our Dev's boss will get pissed off and be less likely to allow him to mess with the Lua API in the future. Management might take the position that it's not worth the risk of introducing bugs that could affect all players, even those that don't use plugins.
Does this sound like something people would be willing to participate in? If so, then we could run it by Turbine and see if our Dev is willing to go along with it. That's assuming he hasn't been laid off. :(
1- Bug reports that explicitly manage to tell you how to reproduce the bug are a definite plus.
2- A list of bug "tests" are an absolute necessity for regression testing. Especially if the tests can be scripted!
3- According to Sapience, someone is "in charge" of Luas, not laid off. (I think that was a question asked in one of the Isengard runs.) I don't believe he stated "who."
4- Can we get the Dev interested -- probably -- but it seems that Devs who communicate with players "disappear."
5- Our "new/old" Executive Producer has indicated that he will likely do a monthly letter to the players. That might mean a change in policy about communications, I don't know.
6- It's a good idea, and worth a try.
|All times are GMT -5. The time now is 07:15 AM.|
vBulletin® - Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.