Should the Latin Wikipedia employ article creation bots?

[Iohannes Aurum]

Should the Latin Wikipedia employ article creation bots? An example of an article creation bot is Lsjbot, which created millions of articles for the Cebuano, Swedish, Dutch, and Waray Wikipedias; Cebuano and Waray are Filipino languages.

In fact, the Cebuano Wikipedia is well on its way to surpass the English Wikipedia, thanks to Lsjbot.

The Latin Wikipedia is a rather small Wikipedia with very lax article creation rules, which would significantly reduce opposition to article creation bots such as Lsjbot.

You can discuss with the bot's creator here: https://en.wikipedia.org/wiki/User_talk:Lsj

However, his home Wikipedia is the Swedish Wikipedia: http://sv.wikipedia.org/wiki/Användare:Lsj

We could ask him if the Latin Wikipedia is open to using Lsjbot.

 

[Iáson]

I think a major worry with this sort of approach is that machines simply aren't that good at generating correct and readable Latin, as Google Translate has repeatedly proved. Is this different with this sort of bot, or would the result just be a tremendous number of articles formed of nothing but incomprehensible gibberish?

 

[Iohannes Aurum]

I think a major worry with this sort of approach is that machines simply aren't that good at generating correct and readable Latin, as Google Translate has repeatedly proved. Is this different with this sort of bot, or would the result just be a tremendous number of articles formed of nothing but incomprehensible gibberish?

Bots such as Lsjbot create articles that almost always have this formula (this one is for geographic locations):

[Infobox containing basic information about the topic]
 
[Topic] is a(n) [geographic feature] located in [administrative division], [country].

[References]
 
[Navigation template]
 
[Geographic coordinates]

The bot just reads HTML tables and extracts certain fields. Often, the article has only one reference: in the case of geographic locations, it would be from GeoNames.

 

[Araneus]

Better to have a handful of Latin articles written by competent people, than 10 000 000 000 000 Latin articles mindlessly generated by a machine. How on earth will you then be able to single out the ones worth reading. I haven't looked at many of the existing Latin Wikipedia articles, but I guess they are written/translated by actual people, and not by computers - so far?

 

[Iáson]

If you want a large number of geographical etc. articles, maybe just divide up & copy and paste the entirety of Pliny's Natural History into suitable articles.

 

[Callaina]

I can't see any possible benefit to employing bots to generate Latin Wikipedia articles, I'm afraid.

 

[Pacifica]

Neither can I. It would probably be disastrous,

a tremendous number of articles formed of nothing but incomprehensible gibberish

, as Iason said.

 

[Iohannes Aurum]

There's an extremely good reason why many editions of Wikipedia based on major languages (such as English, French, Spanish, German, Russian, and Mandarin) have numerous strict requirements to create articles, such as every article having at least one reliably sourced section explaining the significance of the topic.

Wikipedias with smaller communities of various article counts such as Cebuano, Waray-Waray, and Latin have much less strict requirements and bot-generated articles are generally encouraged.

I seriously doubt that a Latinist would be interested in reading about a tiny uninhabited insignificant islet in the middle of Nunavut.

 

[Godmy]

Without reading the dicussion: absolute no. Usually the conlang wikies that have bot created articles are utter failure and Latin wiki already suffers by sometimes not so quality content by amateur latinists (while there are some articles done by epxert latinists which are flawless, but one needs to search harder).

This is more or less why one shouldn't use the Google Translate for Latin.

 

[Iohannes Aurum]

Without reading the dicussion: absolute no. Usually the conlang wikies that have bot created articles are utter failure and Latin wiki already suffers by sometimes not so quality content by amateur latinists (while there are some articles done by epxert latinists which are flawless, but one needs to search harder).

This is more or less why one shouldn't use the Google Translate for Latin.

At least amateur Latinists are better than robots, based on current technology.

 

[Callaina]

An amateur anything tends to be better than robots for any cognitive task that isn't pure data mining.

 

[Godmy]

Sure! That's what I wanted to say. What I wanted to say was that we need to decrease, not increase the amount of rubbish on the Latin Wikipedia.

 

[Callaina]

Seconding that!!