UB bug still present in v16, [split by Baronius] Options

[Baronius]

The bug was known for v15 already. Wouldn't it be easier to fix it in UB itself?

[This topic was split from this]

[Baronius]

It's just surprising, that there are ANY fights when it comes to modding. It's hard to earn money or real "social power" in modding. It's just something you do, because you are enjoying it. So why care about "lies and accusations"?

1) Some people consider this as a competition (or something like that) and/or want to make their site more popular, (so) they want to spread their mods (even by convincing other authors to change their own mods). A few modders even spread lies, just to cause harm to the other author. Or simply due to ignorance, envy or some other thing. They believe they don't need any more knowledge, and that they are right under any circumstances.
2) There are modders who invest a lot of time to their modding work (and this is true to many modders), and some of them becomes the victim of lies and attacks. Reasons detailed in (1) above.
3) There are forum users who have never made any mod, but love reading stupid topics and enjoy "fights". They enjoy the drama, the strain. Some of them prefer a certain modder, and defend that modder's interests/opinion even if they know jack about the topic. These users are the best troll army to support the people who I characterized in (1).
4) There are forum users who believe all the stuff they hear. Everything, even if they don't know the topic. Some of them belong to the group who I characterized in (3). Others aren't modders, and they get misled due to the things they hear.

But what is the answer to your question, "why to care about the lies and accusations"? The answer is above. Modders of group (2) invest a lot of time, and many of them are sensitive to lies (as most people). Jab, a mod is an intellectual work, intellectual property. They invest a lot of time, and is it a wonder they are furious after seeing that the people in group (4) believe all the lies told by (1) or (3)?!

In other words, a person has full right to reply accusations and reject lies about his or her intellectual work. Especially if there are naive users who may believe those lies unconditionally. I "love" the scenario when a modder tells total nonsense or perhaps a subtle lie, several trolls start to support him, but someone dares to defend the mod/feature in question, and what is the result? "You are going to get a heart attack", "You are over-defending it!" etc. Are they surprised that someone rejects lies? But they are hopeless. They enjoy the strain, enjoy ridiculing other's work, and so on. And the envy. Human envy is a big lord.

Naturally, and I do wish it were fixed, as well as other issues. I like having UB and QP in my install, and it is rather annoying when well-made, well-liked and popular mods have issues. Hopefully the authors will have some free time in the future - I, for one, am looking forward to it.

Version 16 was released, yet the bug wasn't touched. It is about changing three lines of its TP2:

CODE

EXTEND_TOP    ~AR0021.bcs~ ~ub/crane/AR0021.BAF~

COPY_EXISTING ~ar0021.are~ ~override/ar0021.are~  /* Associates correct area script */
WRITE_ASCII   ~0x94~       ~AR0021~               /* Corrected Crooked Crane, Level One */

For example, it can be changed to this:

CODE

EXTEND_TOP    ~AR0004.bcs~ ~ub/crane/AR0021.BAF~

So it has nothing to do with free time. It's about principles (or ignorance, sometimes). Modders often don't realize (or just don't care) that they may break "backward compatibility" when they modify the potential dependencies of mods by altering the original game.

In this specific case, mods that use ar0004.baf will be broken (for example, these mods use EXTEND_TOP in order to summon a creature in ar0021.are). A little illustration, specifically about the bug of Unfinished Business:

An "unmodded" game:




Five mods are installed, Mod1 and Mod3 depend on ar0004.bcs:



If the reference from ar0021.are to ar0004.bcs is removed, the content/plot of Mod1 and Mod3 may be broken.

Mod6 (e.g., Unfinished Business v16) is installed. It replaces the reference in ar0021.are (this is where we "fix" something and actually introduce a bug by violating backward compatibility):



The file ar0004.bcs is now a lonely file among the game resources of BG2. It has no function.

Note that it doesn't matter if Mod6 is installed before or after other mods; if installed, it may break Mod1 and Mod3. It's also important to note that ar0021.bcs has no function in the original game (it doesn't exist), so UB doesn't restore any original but unreachable game content here.

There are several other types of dependencies as well. This was just one example.

The best example for such (and similar) changes is the G3 BG2 Fixpack, of course. These changes cause severe incompatibility risk for other mods, and can't guarantee backward compatibility even with the original game itself. It can be acceptable for real bugfixes where the incompatibility risk should be ignored due to the severity of the problems (in other words, don't make mods that rely on bugs). On the other hand, it's unacceptable in case of subjective changes and quasi-fixes (quasi-tweaks) such as "Keys should be consumed after use" (may easily break certain mods) or "That person cannot be neutral because he is a smuggler and attacks you when you reveal him" (alignment changes may break certain mods' scripts). I brought up G3 Fixpack because it's the best example. But I know the usual reaction: "You don't like the mod so you try to discourage players from playing it" (no comment).

So now I hope it's clear why I keep emphasizing all the time that mods (especially those which are meant to be installed with many other mods, such as fixpacks and small mods) should only modify as much original content as *needed*. And no more. If the mod has lots of dependencies (such as Improved Anvil) due to its nature, its compatibility with other mods should be examined with even more attention, and proper documentation should be provided to help players and the author too (Improved Anvil offers all this).

If an IE mod is a plug, the IE game is the socket. You can imagine what would happen if the standard socket was modified...



Yes, all plugs (i.e., all devices with such a plug) would have to be modified. All existing copies. Funny.

"That is why bug reports and tests are needed". Funny statement. Wouldn't it be easier to build a reliable mod instead of mistaking players for testers and expecting others to change their earlier and current mods as well? (Needless to repeat that G3 Fixpack is the best example: it's not an ordinary mod, it's a fixpack which requires mods to constantly fix themselves to be compatible with the fixpack )


Ensuring backward compatibility can be tough. One of the main problems of big systems (as well). It's not hard to introduce new, shiny features, but what about the software or hardware components, tools, source code etc. that were produced for/with the earlier version of the system/language?

Finally, a concrete example again. Quote from the WeiDU readme:

NO_IF_EVAL_BUG this action solves a long-standing bug with the IF_EVAL action. Since solving it is not directly possible without damaging backwards compatibility, you have to use this action (preferably in an ALWAYS statement) to solve the bug for yourself.

Yes, that's the only way to solve it. (For those who aren't familiar with this topic: the quote isn't a negative example to criticize WeiDU. It's an example how WeiDU solves a problem -- correctly. A problem that arises quite often in this field.)