User:Ser Amantio di Nicolao/Sandbox/Bot proposals


 * Caveat: I'm not very good at propounding suggestions for things like this. However, I've identified a couple of issues that I think it would be helpful to address, and I'd like to go ahead and lay them out below.

Wikipedia is one of the most mind-bogglingly massive projects on the planet, in my mind; as of this writing, there are more than 2.58 million articles on legitimate subjects in the English Wikipedia alone, and more are constantly being added. It's a brilliant idea - I don't know of any other encyclopedia in which articles on American lighthouse keepers, Australian internet personalities, African archbishops, and Italian opera singers can exist side by side with comprehensive listings of communities in Great Britain, France, Germany, Poland, and so forth. The scope alone is incredible. Yet despite this, there are two or three areas in which Wikipedia is, in my view, lacking somewhat, and I would like to see them addressed somehow. Here are the three I can think of for right now:

Places
Articles on various kinds of places are a large part of this encyclopedia, but there are still millions of them missing. Many are inhabited places; a bot account has been set up to deal with them, so I'm not considering them right now. But there are more geographical features that need coverage, and while I see a comprehensive attempt to cover them manually, I can't help feeling that another bot would be most useful to pick up the slack. What about a bot dedicated to the rivers of the world? One dedicated to mountains? Gazetted databases of such information exist, I'm sure - why not create a bot, patterned after the geobot, that would gather this information, create stub articles from it, and enrich the encyclopedia? I don't see why it wouldn't work, especially if it received manual input from a group of dedicated overseers, much as the geobot is supposed to.

Also, as of this writing, a geobot has been allowed to fall by the wayside. There's no reason for that; a lot of the groundwork has been done, and I think it would be very good to see somebody pick up the torch and continue with the project. As is, articles on geography divert too many valuable resources from the project to be of much use.

Species
As with places, so with species. I've seen a lot of little stub articles on the subject being created of late - a bot would be awfully helpful in getting more coverage more quickly. Isn't there some kind of substantial species database we could work from?

Translation
This, perhaps, is the biggest bugbear of the three for me. I honestly don't think the current translation system works very well; I know of two places in which translations can be requested, but there's a lot of material I know I haven't seen there, and it takes a while for anything to be taken care of. From my perusals of other Wikipedias, I know of many things that are well-covered in other languages that are missing in English - indeed, without looking too closely I'm sure I could compile a list of well over a thousand. I've already got a huge list of Russian-language articles requiring translation, and that's without having looked too closely.

Here's my proposal: I know bots exist on other wikis that search out inter-wiki links and create more based on them. For instance, I translated an article from the French, and included a link back to the original; the bot picked up on this, and made a corresponding link from the French to the English. Why not, then, create a bot capable of searching out the opposite, English-language links in other wikis? It could then compile a list of all articles in which such a link has not been found, leading to a sort of master database of articles which require translation. Such a project would certainly require human supervision, weeding out the wheat from the chaff - I know some articles exist in multiple languages but are missing interwiki links. And I don't think an automated translation system would be any good, for obvious reasons. Still, with a dedicated Wikiproject working on the subject, I think good inroads could be made into the numbers of articles requiring at least basic translation from other Wikis.

The uses of bots
I'm not always in favor of using bots for article creation. However, I think that in some cases they are a necessary evil. So it is here. As it stands, users are taking up valuable time creating stub articles on all manner of things that could be handled automatically. There's a lot of other work to do, and I think it would be wonderful to see some time freed up by the creation of such bots to allow people to use their talents more constructively. When it comes to translation, I think that the creation of a comprehensive database of untranslated articles would also save a lot of time; as it stands now, I know the discovery of such articles can be rather haphazard.