Jump to content

User talk:Paultyng/Archives/2007/July

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia


NRHP Bulk Stubbing

Hey, sorry I took so long. Distractions in the real world, doncha know. :)

What you suggest would be wonderful! Automating the process could be tricksy, though. I can tell you that the List of Registered Historic Places in Florida is done, 'cause I did it. :)

Let me throw out potential problems, just to make you aware. One thing is the multiple listing of the same name. Like Webster County Courthouse or Washington County Courthouse for example, but there's loads of others. They'd need to be filtered out so it'd only be listings that show up one time in the whole NRHP database.

There's also the thing of alternate names. I know there are articles for NRHP sites on Wikipedia under names that aren't the same as the gov't database has. That's not too bad an issue, though. Those stubs could always be changed to redirects, and the info added to the existing article.

It'd probably be best to do one state at a time, and contact the state's WikiProject to give them a headsup. I'd suggest starting with Illinois, as they have a good active group, and User:IvoShandor is a member of our group and that one.

That's just off the top of my head. I'd say the main thing to do is break it into manageable chunks, like by state. Also the automation method. Hmm. I'm not too familiar with bots, though I'd think that would be the ideal way. There's always AutoWikiBrowser, but that requires pushing a button. Though that might be better, as letting a bot run hogwild could cause problems.

Anyway, hope that helps, and let's see if we can get 'er dun. :) -Ebyabe 20:27, 30 June 2007 (UTC)

Yeah, I have been just trying to run ohio. Here is my current plan, you are right about the posibility of dupes and bad naming and so forth, so:
  1. Clean up the database (I'm working on this now, getting clean significant names off of properties and so forth, making sure the data all matches up correctly)
  2. Run a number of different searches for potential existing matches to determine if the article already exists. (this is the next thing I'll be doing).
    1. There are a few things I can do for this, like search for the listing number, search for the name or the signifcant names of people associated with the building, the architect, etc.
  3. Hand sift through the possible existing articles and determine if one does exist, etc.
  4. Write up a pretty decent stub article generator (this is kind of independent, but I need to determine a good stub template to plug the data in to to get a good article).
  5. Etc.
Obviously a lot of work still to do, but I think this may be a good path as there are many missing articles and a ton of data in the NRHP database that could be put to good use. pw 20:45, 30 June 2007 (UTC)
Check out User:Paultyng/NRHPStubTemplate to help me work on a good template for the stubs. I don't have a handly list of the data available but you can see most of it on their search pages. pw 01:49, 1 July 2007 (UTC)
I proposed a new stub specifically for NRHP articles, so hopefully there are no disagreements there and I can create in a few days. pw 02:08, 1 July 2007 (UTC)
I found these structs: NRHP-Struct search, we can probably just do one per state. pw 04:13, 1 July 2007 (UTC)
Yeah, those were my fault, since last year I stubbed the thousand or so Florida ones that didn't exist previously. You can see from the stub-types that were created what would probably work best. A district one is good, especially since there's a task force for them at WP:NRHP, and each state usually has a few dozen (Florida has over 200). For Ohio, there could probably be county ones for Erie, Franklin, Hamilton, and Summit, since they have their own separate lists.
Btw, I'm not sure if the more recent stuff is in the database you have access to. I don't think new listings get entered immediately. Maybe not for months or longer. But there's enough of the older ones to work on, so we can always worry about that later. Keep on keepin' on, dude! :) --Ebyabe 15:52, 1 July 2007 (UTC)
A thought: accuracy. I've found that town names aren't always spelled correctly. Or places are listed as being in one city (or county), when in fact they're in another. I can tell you that'll be fun to weed through. You probably thought of that, but I wanted to put it out there, whilst I'm thinking about it. :) --Ebyabe 15:56, 1 July 2007 (UTC)

The NRIS is a few weeks (months) out of date with some listings never entered - Spirit Mountain, Nevada comes to mind and it was added in 1999. Some uniformity should be adopted. I went through the state lists and the articles range from god-awful to quite good. Einbierbitte 21:39, 1 July 2007 (UTC)

Spirit mountain actually is in the listing I have, it lists some significant person for it as Mastamho which is apparently some Mohave guy so it actually has more data that in that article in the database. I think their site just has problems with some of the bad links in the data, I hand cleaned it all up, so that I could get at all the info. pw 22:22, 1 July 2007 (UTC)

Hi - a stub category which you created has been nominated for deletion or renaming at Wikipedia:Stub types for deletion. The stub type does not meet the standard requirements for a stub type, either through being incorrectly named, ambiguously scoped, or through failure to meet standards relating to the current stub hierarchy or likely size, as explained at Wikipedia:Stub. Please feel free to make any comments at WP:SFD regarding this stub type, and in future, please wait until discussion is completed! Grutness...wha? 02:12, 3 July 2007 (UTC)

Historic District Boundary Increases etc.

You bring up a good question. I only do the historic district listing because 1) the guideline for doing so exists on the Project page and 2) the district has a unique NR reference number. Logically, a boundary increase would include more contributors, which may merit a separate listing. But you raise an interesting question nonetheless. I'll defer to majority rule on this - if these merit a separate listing or not.

Thanks for the update on Spirit Mountain. Einbierbitte 21:41, 5 July 2007 (UTC)

I dealt with it in two of the following Florida articles, Lloyd Historic District and Palmer-Perkins House. Just to give you an idea. And yeah, it's kind of weird when a building listing gets a boundary increase. Took the pics for the Monticello historic spots, and near as I can figure, the boundary increase was the grounds surrounding the place. It's a B&B, btw, and one of the classier ones, just from the exterior. If you haven't already, I highly recommend NRHP picture taking roadtrips. You've got some great places up there in Ohio to photograph, I know. Maybe I'll get back up there someday. :) -Ebyabe 00:02, 6 July 2007 (UTC)
As far as the lists, I recommend consolidation. I went through ages ago and tried to merge the links on as many lists as possible. I know there's some with two boundary increases, a few I think have three. It just doesn't seem necessary to create separate articles for each increase/decrease. If enough info is available, sure, make separate ones. But it's easier to do one per "listing", and break them out separately if and when more's found out about them. My two pence, for what it's worth. :) -Ebyabe 00:07, 6 July 2007 (UTC)
For the stubbing (assuming the bot gets approval, I'm still waiting on that) I combined all the boundary increases in to a single article, and it will just have a list of numbers. I would agree with how you handled it Ebyabe, i think thats the best way as well. I will build in some sentence generation like you did for the increases, just a little paragraph with the date, etc. pw 00:10, 6 July 2007 (UTC)

Hello. Thanks for doing some newpage patrol. When you run into a newpage that is clearly a copyright violation from some external website, it's more appropriate to tag the article with the speedy deletion template {{db-copyvio}} than to report the article to the copyright problems noticeboard. Also, please make sure to notify the creator of the page that his new article will be deleted for those reasons. (Oh and all of this is referring to Portsea Surf Life Saving Club which I've just deleted!) Cheers! Pascal.Tesson 03:51, 6 July 2007 (UTC)

Cincinnati Population Numbers

First of all, to address the urban population numbers which you are contesting--those can be found on another Wikipedia page which has the references. I'm not just pulling these numbers out of thin air. Most of the American cities on Wikipedia have their urban population listed which correlates with the Wikipedia article which lists them. None of these pages have any specific reference to the urban population sources however--what's good for the goose is good for the gander. You can check the numbers and references for yourself, but I'm not going to argue on this point. If you insist on leaving it out, then so be it.

Secondly, the population densities that I have repeatedly changed reflect the new population numbers of the city. The numbers that you keep reverting back to are not correct and, as such, I will be reverting your revision back to the correct numbers that I have calculated. Population densities were never listed by the Census Bureau to begin with so I don't know what information you are referencing. If so, you should include your reference in the footnotes as well.

Thirdly, each time that this has happened you have simply reverted it back to the former state without considering that there may have been other changes made. I had cleaned up the code so that it was in the proper order and with proper spacing & nomenclature. I would appreciate it if, in the future, you would just change that which you have a problem with. To avoid this in this situation, I will be sure to do my format revision before changing back the population densities. tpetross 08:34, 6 July 2007 (UTC)

I don't believe that the density numbers that have been traditionally listed under the city population have been for the urban clusters. If so, these numbers are terribly off on all of the city pages. Nonetheless, those that I have added are correct if you divide the 2006 total estimate by the total area. I am, however, confused as to what the technically-correct way to calculate this is--should the total area be used or should just the land area be used (considering the chances that people live on water is slim to none)? I don't know if it's even possible, but we should look into finding a way to automatically calculate the density through coding. tpetross 08:33, 7 July 2007 (UTC)

Question

Hey there. I see you run a bot that uses C#. What do you type into cmd or whatever you use to launch the bot? Is there a specific compile and run that you use that you could point me too? Thanks! ~ Wikihermit 02:36, 10 July 2007 (UTC)

Its actually a little windows application I set up. You can see all of my code at this link. pw 16:26, 10 July 2007 (UTC)

NRHP stubs/articles

No problem--feel free to revert the stub change part. When I'm 'head's down' new pages patrol I often don't notice editor's 'grand schemes'. I had no idea people were doing county-granularity stubs nowadays--seems excessive to me, but I'll leave that to people more involved with stubs than I am. One comment on the project, articles that don't need disambiguation in their titles shouldn't have disambiguation in the titles. I don't know how difficult it would be to automate, but it (theoretically/for sake of argument--I just noticed another issue with the title) should have been created at Cedric G. Boulter and Patricia Neils House, instead of having a redirect from there. The second problem I just noticed is that your bot seems to have mis-interpreted the comma order in the title from NRIS--based on this it appears the actual name is probably Cedric G. and Patricia Neils Boulter House--that's probably even more difficult to remedy in an automated way, I imagine. Anyway, I think its great you're trying to tackle this void in WIkipedia, and hope you can fine-tune any wrinkles away. Seattlenow 02:09, 11 July 2007 (UTC)

Your recent bot approvals request has been approved. Please see the request page for details. When the bot flag is set it will show up in this log. --ST47Talk 12:12, 12 July 2007 (UTC)

Question

Could you write a quick bot script in C# for me? I'll explain what I need as best I can. I use a pywikipedia bot to search special:newpages for copy vios. However, it just saves it to a text file. I need the bot to upload the new text in the text file to a userpage in its subpages for other users to see. Do you think you will be able to do this? Do you think you could do it to replace old sections with new ones so the page in the user space doesn't get too big? Thanks! ~ Wikihermit 19:37, 13 July 2007 (UTC)

NRHP bot

I noticed you just started up a bot to create stubs on National Register of Historic Places articles. I created this infobox generator a few months ago, so I'd be willing to help with any questions you have about the database.

One thing you might be interested in is some code that reads the geographic coordinate database and converts those locations to degree/minute/second coordinates, as used in the infobox. It reads the geographic database and converts the Universal Transverse Mercator coordinate system coordinates to degrees. There's some tricky math involved, but fortunately I got the tricky math from another site. The infobox code is in PHP, but it should be easy enough to read it and convert it to C#.

Let me know if you're interested in it. Also, you might want to mention at Wikipedia talk:WikiProject National Register of Historic Places that your bot is now up and running. --Elkman (Elkspeak) 18:05, 17 July 2007 (UTC)

Thanks, I've been ironing out some bugs before I start it up some more, I'll take a look through that code and see what I can incorporate in, I'll let you know if I have any questions. pw 19:27, 17 July 2007 (UTC)

I've posted the code for the infobox generator at http://www2.elkman.net/nrhp/infobox.txt (the main infobox generator) and http://www2.elkman.net/nrhp/utmconvert.txt (the UTM to degree/minute/second converter). You can look at the calculations in utmconvert.txt to see how I'm converting UTM to degree/minute/second coordinates.

As far as the geocoding in your example, it's possible that the people who did the geocoding for the buildings didn't use exactly the same datum or the same addresses. In fact, I've never seen the database of geocoded buildings before. (If I had, I could have saved myself some effort in learning how to convert UTM to degree/minute/second coordinates.) I don't know if those discrepancies from the example are a big deal, since I generally haven't gone back to check on the coordinates in infoboxes that I've created, but I might check on them in a little more detail now. --Elkman (Elkspeak) 16:29, 19 July 2007 (UTC)

Hamilton County NRHP

Those articles look really good. Einbierbitte 19:57, 22 July 2007 (UTC)

Disambigs

You are absolutely welcome to restore the disambigs, when there are reasons for there to be disambigs. Disambigs are for navigating between multiple pages on the project. At the moment, the ones I am reverting only have one destination on the project. Redirects are the proper navigation tool for single items. When you have multiple destinations, disambigs will be proper. Until then, the redirects are the correct form. I'm finding these because they are showing up on the short pages reports, and I regularly do short pages patrolling. Anyway, given that they are not proper disambigs, why do they need to be broken disambigs right now, instead of proper redirects that you can convert back to disambigs as you actually create the extra articles that make them proper disambigs? - TexasAndroid 14:59, 23 July 2007 (UTC)

If a destination is not notable enough for a page on the project, then it is unimportant as far as incoming links. If there are bad incoming links about non-notable subjects, then the incoming links should be fixed, not a broken disambig page created. I'll be back with more comments after I look up the actual instructions on disambig pages, which are for specific purposes, one of which I'm pretty sure is not faking a situation where a redirect is the proper tool. - TexasAndroid 15:10, 23 July 2007 (UTC)
I'm looking up the requirements for disambig pages. If you are indeed violating them, and I am not totally off-base, then I'm sorry, but I really do not care if a badly-programmed bot has problems. If the bot is having problems with properly formed redirects, and no problems with broken disambig, then the bot is also broken. - TexasAndroid 15:12, 23 July 2007 (UTC)
I'm not saying that the pages need clean-up. I'm saying that, when there is only one destination notable enough to have an article on the project, that they should not be disambigs at all. I do lots of disambig clean-up myself, as I did on the one that I formatted. These are not a disambig clean-up issue, but an issue of whether they should be disambigs in the first place. I keep getting distracted with reading and typing responses. One of these moments I'll be able to sort through the disambig instructions to read if I'm right or wrong. For now, I'm not going to revert any more, as we are in the definition of a reverting edit-war, which neither of us really should be doing. So I will back off for now. But this is far from over. - TexasAndroid 15:27, 23 July 2007 (UTC)
I'm not assuming bad faith. At what point in this discussion have I indicated in the slightest that you are acting in bad faith? I may disagree vehmently with your methods, but I have no reason to question your motives here. So please, do not imply I am assuming any sort of bad faith about you, when I have done no such thing.
I'm still a bit confused over what the point is over you repeatedly mentioning the bot speed. The bot is creating redirects, but you then come through and convert the redirects to the broken disambigs, undoing the bot's work. You also have still never answered why the broken disambigs are so important that they must remain broken disambigs until the extra articles are actually created. Leaving them redirects until there are actually multiple items to point to, even if that takes a few weeks, would solve the situation, but that is obviously not an acceptable solution to you for some reason.
I'm not finding the text I thought I would find in the Disambig pages. My "Not over" comment meant I would persue this in other ways. And the next step will be to open up a more public discussion on this on one of the general disambig talk pages. I need to get other feedback on this, because your "broken" disambigs are far from the first I have converted to redirects. You are just the first to get me into a revert-war over it. :) If I am operating on a false assumption of proper disambigs, then I need to know so I can adjust my general MO on the issue. If however you are wrong, then hopefully someone else from the disambig regulars can better argue the situation to get you to understand why these are bad. Or maybe the situation is somewhere in the middle. (it always seems to be in the middle in these things.) Anyway, I will drop you a link to the discussion once I start it. I'm currently looking at Wikipedia talk:Manual of Style (disambiguation pages) for the discussion, but I will let you know where it ends up. - TexasAndroid 15:48, 23 July 2007 (UTC)
Discussion started at Wikipedia talk:Manual of Style (disambiguation pages)#disambigs vs redirects. - TexasAndroid 16:16, 23 July 2007 (UTC)

Balch House (disambiguation)

Why did you revert me on this one? By Wikipedia:Manual of Style (disambiguation pages)#Introductory line your disambig is not properly formatted. I standardized the format. This has nothing to do with our other debate, as this one has multiple destinations just fine. But it was badly formatted, I fixed the formatting, and you reverted my fix. - TexasAndroid 15:53, 23 July 2007 (UTC)