|
Shannon V. OKeets -> RE: Easy to reproduce? (3/3/2014 9:37:25 PM)
|
I spend a lot of time keeping track of 'bug' reports. There are a lot of things reported as bugs that aren't.
Misunderstandings of the rules, misunderstanding of the player interface, and disagreements about both come up frequently. Throughout my work on this game, there have been continual discussions about rule interpretations. At various times over the years in the development forum there have been numerous threads about individual rules. And as I recall, Paul put together a single thread listing all the (at that time) issues about the rules. Rule discussions also appeared in the open forum (i.e., this forum).
So rules questions appear in some of my task lists, but they aren't technically bug reports.
Then there were (and are) discussions about how the player interface should work. Similar to rules discussions, these spanned many years and easily more than 100 separate threads. In the beginning I was very open to suggestions for how the player interface should work. In fact I aggressively solicited advice, alternatives, and engaged in dialogues about the pros and cons of different choices. Eventually a consensus appeared, or I simply cast my vote on how things would work. As in all decisions in life, when there are difficult choices to make, the alternatives not chosen never completely disappear, but come back off and on at times as "you should have ...". Now that the game has been released, these items are mostly gone from my task lists. But there are still a dozen or so of "it would be a good idea to some day ...".
Suggested improvements are more items on my task lists, but not bugs.
Duplicate reports on a bug are made all the time. This is not a bad thing, except in the sense it is terrible that many players are encountering the bug. A list of current bugs with 30 items might really only have 4. That fact is not obvious at a glance. Only after delving into all 30 can I determine the true number.
The vast majority (over 90%) of items on my task list fall under the heading "Investigate XYZ".
Once I can confirm that a bug exists, my natural tendency is to see if I can fix it immediately. Roughly half the time I can. Those that require more time and effort remain on my task list, grouped into loosely defined groups (e.g., NetPlay, supply, production planning, naval combat, etc.). I prefer to work on these more difficult problems in batches, because they require me to refamiliarize myself with the relevant code (typically tens of thousands of lines of object-oriented Pascal code).
So when people ask for a list of things to 'fix', it isn't as easy as just posting a bunch of text. Each item has its own little history: from whom, when, which version, about what, preconditions, manifestations, file references, references to posts in the forum, references to emails, possible causes, things investigated, what to do next, etc.
And all of this is in constant flux. Getting my various task lists organized simply so I can understand them takes me a lot of time. Putting them into shape so other people could make sense of them would double the amount of work needed. I use to do that for the beta testers a few times a month, but I had a lot more time available then. Since the game was released, the volume of my 'communications' has skyrocketed, albeit I am mostly on the receiving end.
Anything I posted as a list of "current bugs" would be more a historical curiosity than helpful. It would also be seriously out-of-date within 24 hours.
|
|
|
|
