John wrote: The Cunctator wrote:
The RamBot was a bad idea, and the IsraelBot is too. Wikipedia articles should be generated individually, not en masse.
A semi-reasonable compromise would be to tag the auto-generated
entries
so that they didn't show up in the site statistics, etc. That is,
the'd
act as preformatted resources for editors who wanted to create an
entry
on a town, but wouldn't be considered real entries.
--tc
Your argument makes no sense. There are certain things simply taken as facts that should be included in the Wikipedia. To wait for someone to enter all these things by hand would take forever. A bot can enter the data much more quickly and with a constantly consistent format for
easy
reading. Sure, someone should go back and look at all these bot entrys to add more information they might have from personal experience but
to
not let bots seed the 'pedia with factual stubs first would lessen Wikipedia's content.
As long as the information in a bot's stubs is factual, why not count
it
in the site statistics?
Your argument, sir, makes as much sense as mine. The need for Wikipedia seeding is highly debatable; and the advisability of doing it even more debatable, as the more that bots are used, the less likely it is that someone will go back and look at all those bot entries.
There should be a balance between encyclopedia size and contributor participation--the best way to ensure that is to not auto-generate entries.
My suggestion was to let bot-seeding happen, but for those seeds to remain "hidden" until someone goes and looks for the information. Take, for example, [[Leominster, Massachusetts]].
My proposal would be as follows: 1. If you went to [[Johnny Appleseed]], for example, the link to Leominster would be in new-page or stub-link format (? or !), not a standard link, until someone actually made a change other than the Rambot entry. 2. If you click to that link, you see the Rambot content. 3. If you search for Leominster, the Rambot entry would show up in the listing (perhaps tagged). 4. The page wouldn't count in the statistics until someone's made at least one edit to the entry.
That way you get the benefit of the content-seeding without the distortion of the size or intent of Wikipedia, even if hundreds of thousands of such seedings are done.
There's also the side argument that all Rambot did was take information that should be in tables and graphs and put it into paragraph form to make it seem like Wikipedia style. It probably sould have done better just to supply the tables and graphs for eventual inclusion in real entries about the towns.