New policy for the Community Prefix List, We need your feedback Options

[Baronius]

We had a discussion about the Prefix List with SConrad. He assured BWL that their intention isn't to boycott the Prefix Reservations list, and he offered a suggestion.

There are several registered prefixes which have never used to release any Infinity Engine game mod. Many prefix owners haven't informed the community that they have a mod under development. On the other hand, sometimes mod developers return to the community to release a mod for the public, which development they started earlier with a registered prefix.

It also seems to be true that many mod developers like to choose prefixes that they favour for some reason (monogram, initial mod name abbreviation etc.). I'm not really convinced that this should be an important aspect (as the problem is purely technical: identifiers, filenames shouldn't collide), but I have no right to decide this (neither the BWL Council or any department of BWL). If you believe it's an important aspect (or the opposite), please do post your opinion in this topic.

The Prefix List originally supported prefixes of any length, but since this easily leads to collisions, BWL decided to fix the length of 2 characters. I don't think this needs to be changed (especially because there are already very many two-character long registrations). On the other hand, due to the high number of prefixes (which include many apparently abandoned registrations), it's not easy for those to choose a prefix who want some typical, favoured prefix (monogram etc.) -- many variations are already registered by others, and often abandoned. To improve the situation, the best way would be to estabilish a new policy for registrations.

The new policy would assign a time limit for each registered prefix, and if no mod is released with that prefix for the specified time limit, the prefix would be moved to second list, and its place would be freed on the primary list. This means that it becomes available again for others' to register, and if someone registers it, the original owner would lose it. If no one registers it before the original owner returns, the original owner can re-register it (i.e. extend the time limit, but the added time limit would be much shorter than the base limit for prefixes). The number of such time limit extensions for prefixes which aren't used in any released mod could also be limited (for example, after all extensions are used, the prefix would remain in the second list and could be registered by anyone, and if it's not registered by anyone else and the original owner would release a mod with it, then it would return to its original owner; but these details require more examination than what I've done here).

If a mod is released with a registered prefix, it would be reasonable to allow the prefix to be valid for an unlimited time. On the other hand, it's important to deal with the situation when a single project with a prefix is released, the author leaves the community and there are no releases for years -- while the prefix is taken. In such an extreme case, I think it's enough to free the prefix for the public again, and simply designate the original (and old) project as incompatible with those mods who use the re-registered prefix (but "designation" probably wouldn't be the duty of the Prefix List maintainers).

I also have an idea about a restriction for registrations: a mod developer who registers a prefix will have to specify an Infinity Engine game where he or she intends to use the prefix. However, I am not convinced that this is necessary -- it might make things more complex, while wouldn't offer much additional advantage (most mods are for BG2 and BGTutu/EasyTutu anyway).

Assuming that the majority will support the initiative detailed above, it's important to emphasize that this requires technical changes on the management system of the Community Prefix List. As far as the implemenation changes are concerned, neither the Department of Graphics nor the Department of Mod Development of BWL has any free capacity (i.e. member who would volunteer) to my knowledge. I also don't have the time, however, someone will have to do it, if the initiative gets approved and accepted in this topic. If it's accepted, I will try to contact drake127, the developer of the current system, and ask him to extend the system. I suspect that the database scheme will have to be modified too.

Please note that the above descriptions (about new policies) are just based on quick ideas I had while writing this post, so feel free to suggest different or modified versions for them.

If you believe you have anything constructive to say (or you have any questions), please don't hesitate to send a reply to this topic!

I said to SConrad that this matter requires public discussion, and (more importantly) public approval. BWL has always taken the fact into consideration that not all mod developers (who may release public mods) are active in the message boards of the community, so the approving feedback of active forum members should not be considered as authoritative in all cases, but in this case, I think it is. The Community Prefix List has been estabilished and maintained by active IE mod developers (active in the respect of community forums as well), so making a common decision here is justified in my opinion.

[kulyok]

Prefixes ensure compatibility. I am pro-protected prefixes - if people start releasing mods with my prefix, and we have multiple O#Dream variables and O#(areanames), it'll be a nightmare. Even if I had only one mod released with O#, and not eight, I'd still be against that.

Not to mention that there are project prefixes - RE prefix, that is being used by all RE authors; X# prefix for BG1 NPC material. These are even more important, especially given how many people play, for example, BG1 NPC or Banter Pack. Screw these mods, and thousands of players are in trouble.

If a modder wishes to create conflicts, and uses someone else's prefix intentionally, I'd say "they are in the wrong". If they were intentionally killing my mods(using my prefixes, causing, eventually, more bug reports for me, less time for writing), I'd be - what is the word - Stalingrad?

Yes, if a prefix's been registered two years ago, and no mod's been released, it is a Bad Thing. But combinatorics tells us that there are many, many prefixes available, and it's possible to find many new ones still.

No, it is a Bad Idea to say "hey, Anne only released one mod, her nice prefix A! should be free", because Anne may be working on her own mod far away from BWL. She will read this topic two years later, after she finds out there is a conflict, and, again, somebody will catch hell.

"Please, use your own prefix. It will ensure that your variables, area names, tags and filenames are unique, and there is no conflict with other mods by other mod makers. It is important both to maintain technical compatibility and goodwill in the community" would be my reply.

[SConrad]

I think a lot of people are taking this the wrong way. Let me restate my views on the issue:

I think mod prefixes is a great idea, don't get me wrong. But the argument that I should avoid using MPI for my mod because of the reason that it's registered to Wolfmoon (Neil) seems a bit... weak. I don't even recognize a fraction of the people in the list as modders with released mods, and a vast majority of them won't be coming back to finish the mod idea they had several years ago and registered a prefix for. As a result of this, most "attractive" prefixes are already "taken," and I think it's a shame that modders are discouraged from using the prefix they want because somebody who never will release a mod already has registered it. Even if it's not explicitly said that they can't register it as a conflict, they will nevertheless be reluctant to do so anyway.

In the past it has been suggested that modders who are inactive for two years in the community is removed from the list, but that's a bad idea too: I haven't seen japeth, Ajoc or BobTokyo in the past 24 months, but it would be a bad idea to remove their prefixes as they have mods released. And if I really, really want to use EV for my mod, I can't--because one "DragonJewel13" registered it in August last year, but hasn't been seen or heard from ever since. My money's on that he'll never release a mod.

Another idea was to add a modder's prefix after they have released at least one mod, since it's only after a mod has been released that incompatibility errors appear. However, this does not take into account the possibility of two modders simultaneously work on separate mod with the same prefix, not knowing that somebody else is using it.

Therefore, my suggestion is that some sort of timeframe is established for how long a modder have the right to reserve a prefix. If they haven't released a mod within six months after registering the prefix, they lose the registration. At this point, however, they may re-register the prefix if they are working on a large project--this would also mean that the modder is still active in the community. As soon as a mod is released by the modder, the prefix stays without any time limit.

This would also give incentives to modders to release smaller mods, while they're learning how to. Even if they only release a minor kit, spell or item mod, a side-effect would be that their prefix stays. Hopefully, this would mean that more modders would release learn-to-mod mods, and those releases (and how they are received in the community) may result in the modder being encouraged to continue modding. I know that for me, it was very helpful to release Timestop Tweak, because I felt like I had contributed--and it made me more eager to continue working on other projects. I'm honestly not sure I would have stayed in the IE community if it wasn't for that. Besides, more mods has never hurt the community, either.

Needless to say, I do think there are good reasons for freeing up prefixes that never will be used. I am in support of a revised version of the prefix list, however that may be. Considering that people seem to be... disagreeing, I would at least say the suggestion Kulyok listed in post 13 looks reasonable (and not too far off my original suggestion). I'd like to include the the parenthesized revision to 0), though--no, I'm not going to release a mod with EX, and Kingrames is not going to release a mod with K!.