Template talk:Searchguidelines

From The Urban Dead Wiki
Revision as of 13:50, 18 August 2007 by Morgan Blair (talk | contribs) (→‎Ruined buildings worse then ransacked?)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigationJump to search

Ruined buildings worse then ransacked?

I haven't got enough data points yet to say for sure, but doesn't anybody else get the feeling that ruined buildings get worse search rates then merely ransacked? --Morgan Blair 00:52, 18 August 2007 (BST)

I updated a couple of Search Odds pages for specific locations where toolboxes are now found, and I don't want new data going in under the same variable, when it may properly be two, so I'm going to go ahead an added R(r) for Ransacked & Ruined. If you think this is incorrect, or think it should be done another way, let's discuss... --Morgan Blair 14:50, 18 August 2007 (BST)

Is LR Possible?

I realized looking at the wiki today that I actually haven’t seen an LR building. Does anyone know if this is even possible? Gamer.g33k 12:34, 19 May 2006 (BST)


I'm not sure that altering the timestamp to reflect the time when the searches were performed is really that big of a deal. Only time I could forsee it being a problem is when there's a major change to the search system, and people are recording data from before the change with a timestamp indicating that it was gathered after. Though, that would screw things up anyway. --Raelin 05:49, 14 Oct 2005 (BST)

Someone set it up in the original guidelines because apparently they wanted to be able to do statistical analyses to figure out whether Search results change over time (or between contributors). I don't know who posted that originally, as a note. If we're not going to bother with change issues (and there doesn't seem to be an issue looking at the analyses), then you're right, we can probably remove the comments on changing timestamps
However, I've been making the assumption that the largest majority of contributors are recording data that they've just collected, thus they wouldn't need to worry about it. It's more for people who had been collecting data previously, and decided to contribute to the project with a large sum of data, or who wanted to collect large numbers of data throughout the week then flood the tables with it at the weekend (which seems dangerous to me anyway, but that's neither here nor there). -- Odd Starter 05:55, 14 Oct 2005 (BST)
Ah, I see. Perhaps, instead, we can request that large sums of data, collected over a long period be marked as such with a range of dates. The way it's worded now though, people may assume they need to be double editing to shave off a few hours from the time stamp, when it's really not needed. --Raelin 06:13, 14 Oct 2005 (BST)
Maybe it would be better to note that we're only looking for date, rather than time? That seems to be the real issue here. -- Odd Starter 06:16, 14 Oct 2005 (BST)
Yep, that seems fine. --Raelin 06:22, 14 Oct 2005 (BST)


Powered Generator vs. Ransacked data collection

I may have goofed, or broken wiki etiquette or something, in which case I apoligize. I changed the text regarding building condition, deleteing the bit about it being ok, even preferable, to mix search data from powered and ransacked buildings. Perhaps I should have posted here first. Mixing these two data sets makes absolutely no sense to me. The whole point of collecting post 29 April data is to see how each of these two changes affect search odds. Unless someone has an easy (something automated based on the L or R entry?) way to tally the data seperately, mixing L and R data in the same table will make it extremely difficult to ascertain the difference between the two. We should be collecting post 29 April data in two tables (powered generator vs. ransacked) and talling each table seperately.

I see your point (whoever you are J) but that would mean that we would need to create two extra tables on each search odds page (one for L, one for R, and one for neither). And then we would have to move the already corrected data to the new tables. If the consensus is that the data should be sorted like this maybe it would be possible to put a couple of rows in the existing table that say something like “ransack below this line lights above�? to act as separators. (personally if I would just copy the table into excel and sort it that way.) Gamer.g33k 12:32, 19 May 2006 (BST)