Talk:National Register of Historic Places listings in Connecticut/Archive 2

NRHP town name issues
User:Polaron has made many changes of locations in CT list-articles, including:
 * Various town names in Fairfield County NRHP list-article, changed in this diff
 * Various town names in Hartford County NRHP list-article, changed in this diff
 * Various town names in Litchfield County NRHP list-article, changed in this diff
 * Various town names in Middlesex County NRHP list-article, changed in this diff
 * One town names in New Haven County NRHP list-article, changed in this edit
 * Various town names in New London County NRHP list-article, changed in this diff
 * Various town names in Tolland County NRHP list-article, changed in this diff
 * Various town names in Windham County NRHP list-article, changed in this diff
 * Possibly other changes within these county articles not captured in these single edit diffs
 * Various other changes in which an NRHP row has been moved from one NRHP list-article to another
 * Possibly hamlet/location name changes within New Haven city, Hartford city, and Bridgeport city articles.
 * Others? doncram (talk) 19:11, 12 June 2009 (UTC)


 * The lists were merely adjusted to use town names uniformly. They were all inconsistent, likely from the nominators themselves not being consistent in the use of the town name or the neighborhood/village name. I don't see why it would be preferred to have inconsistent place identification style. --Polaron | Talk 20:12, 12 June 2009 (UTC)


 * My first concern, stated first at your talk page, is: "Thinking about it, I am concerned about these, that these changes in some cases will be introducing incorrect information. In some cases, your knowledge of towns and perhaps of specific NRHP places will help uncover specific errors in NRIS information, which is fine and good if that is made public and addressed (as by covering at wp:NRIS info issues).  But I wonder if you are aware that the NRIS-based tables in these list-articles show in the location field, in many cases, towns or cities that are "near to" the NRHP sites.  That is different from what is shown in a full NRHP infobox from Elkman's NRHP infobox generator for a given site, which shows whether NRIS identifies the place as actually being in a given town or whether the hamlet/town/city it gives is identified as "nearest".  In the NRIS-based tables from Elkman's table-generator tool, the two kinds of locations are mixed together.  It will sometimes be the case that the near-to hamlet is across a boundary line, in a different town or even across a county line.  So if you are basing your decisions to replace hamlet locations on just that list of "near-to"s, you will sometimes be putting in a different town name which applies for the hamlet but not for the NRHP, and is then just not correct.  And these would then be unsourced errors that we as wikipedia editors have introduced, not NRIS errors which are at least sourced and not our fault.  Also, your identification of various places as being in a different county and moving them from one list-article to another can likewise be in error, if the actual location of the place is not reviewed.  So, at the moment I think the mass change edits should be reverted, and individual changes be discussed one by one."


 * You replied at my talk page that you "didn't change the town names. In some cases, the town names were not used and I only adjusted them so that everything is a town. As I said, the town level is how place information is primarily thought of locally. There is no "near to" in Connecticut. A point location is in one town or another. I can assure you I have checked all these changes and the list entries are in their appropriate towns." Well, i can't immediately tell whether your changes introduced any outright errors due to misunderstanding the "near-to" nature of the NRIS table locations.  I don't necessarily agree that official towns are the way to describe all places in Connecticut, and that no one uses more specific place names.  I also don't understand your assertion that there is no "near to" in Connecticut.  Clearly the National Register and many past Connecticut nominators and state staff have used "near to" type locations.  I don't want to cause offense, but I also don't immediately believe that you are familiar with the exact locations of each of these NRHP places and their relations to town lines.  For many of these NRHPs there is no article yet, and therefore documentation about them including nuances of their locations has not been developed, uncovered.  In some previous discussions we've had I thought you were sometimes overconfident about the exact boundaries of historic districts when neither you nor I had the NRHP documents which would show maps of the actual historic districts, without which it would be impossible to know whether the districts spanned certain boundary lines.


 * I'm now looking at some changes where you seem to have replaced a hamlet name by a town name, such as for Bridge No. 1603 and Bridge No. 1604 in the Middlesex County NRHP list article. For those, you replaced Millington, Connecticut, which is, I guess, a nearby hamlet, with the town name East Haddam, Connecticut.  First of all I don't know for sure that the bridges are in East Haddam.  They may or may not be in Millington, which may or may not be entirely in East Haddam.  Unfortuntately the article Millington, Connecticut is a redirect to East Haddam, but there is no mention of Millington in the East Haddam article.  (I'll start an article on Millington now, replacing the redirect.) It may be that the hamlet and the bridges are all located in the town of East Haddam, which looks pretty large, but even then I am not sure that is an improvement to say East Haddam rather than Millington.  It seems less specific and therefore less informative to lose the more specific hamlet location.  A reader should be able to click on Millington and find an article about it which explains it is wholly or partially within East Haddam, whatever is the case.  I currently think it would be better to revert this specific change for these two NRHP places.


 * I am going to ask now at wt:NRHP and at Wikipedia talk:WikiProject Connecticut for other editors to take a look at these questions and join this discussion. doncram (talk) 21:35, 12 June 2009 (UTC)

The column does say "City or Town", so Polaron was correct to change it. (By the way, it shouldn't be hard to verify the town using the coordinates.) If that column is supposed to have unincorporated locations, it should read "Location" - and then it might make sense to list both the town and the smaller community if applicable. --NE2 22:00, 12 June 2009 (UTC)
 * I must agree with Polaron. I've done lots of work on the list for my hometown, Newton, Massachusetts, which has over 180 NRHP listings in 13 villages, one of which is the same as a post office name which spreads into Brookline and Boston. I had to use the Newton assessor's online street address search to determine whether a particular listing was or was not in the city and if it was, which one or more of the villages was it in. Polaron has put the neighborhood names in the remarks section which is good. clariosophic (talk) 22:36, 12 June 2009 (UTC)
 * I don't see where the hamlet names are being captured in remarks sections. Perhaps these edits could be revisited to do that systematically?  It wasn't done in the Middlesex county town changes, for example.  That would at least partially address my concerns.
 * The column says "City or Town" because that is the way it came in from the Elkman table-generator tool. Perhaps that label should be changed in all of the CT list-articles, according to what makes sense for CT.  I don't believe NRIS names of places are being changed systematically like this in any other state.  I've made a lot of corrections, e.g. for municipalities in Puerto Rico, but that is different.  There's no question that any outright errors in NRIS should be corrected in wikipedia (and hopefully reported at wp:NRIS info issues, which Polaron has been doing for several CT NRIS errors).  What makes sense for that column.  I guess Polaron has gone on a campaign to change all entries in that column to be Connecticut's legal towns, which span all of the state I believe (I believe there are no unincorporated areas).  Polaron, is that essentially correct? doncram (talk) 22:53, 12 June 2009 (UTC)
 * If it's including unincorporated communities, "City or Town" isn't correct in any state. It's also miscapitalized. --NE2 01:02, 13 June 2009 (UTC)

I've rechecked all the lists and made a few corrections to the table locations that were there from the table's original inception. If the neighborhood names are preferred over town names, I can make those changes. However, the disadvantage is that neighborhood/village boundaries are not always clear-cut like town boundaries. My point is we should be consistent - either use village/neighborhood/town section names as a general rule or use town names as a general rule.

Regarding the Millington issue, the village of Millington is wholly within East Haddam. The bridges you mention are in Devil's Hopyard State Park, which is east of what is generally thought of as Millington. The park itself is wholly within East Haddam. So using Millington as the location itself isn't perfectly correct either. --Polaron | Talk 23:10, 12 June 2009 (UTC)


 * Based on my knowledge of Connecticut geography/government and a quick skim of the edits that were highlighted, I think that almost all of Polaron's changes are "right on" (i.e., appropriate).
 * Every square inch of Connecticut is divided into legally constituted towns, which are far less ambiguous as locations than the informal hamlets and villages that may exist within those towns. (Polaron is correct in saying that there is no "near to" in Connecticut.) However, the state has some incorporated cities, too, most -- but not all -- of which are coterminous with their associated towns. Accordingly, the column heading "City or town" (or "Town or city") would indicate that all entries in the column are identified by the name of the local governmental body for the property location.
 * The one place where I take exception to Polaron's changes is in changing Winsted to Winchester (in Litchfield County). Winsted is an incorporated city within the Town of Winchester (which is otherwise fairly rural), and the properties in Winsted are properly identified as being in Winsted, not Winchester.
 * Similarly, Willimantic used to be an incorporated city in the town of Windham, and it seems odd to identify it as "Windham," but it is no longer an official place...
 * Additionally, I think that a few unofficial places such as Mystic and West Cornwall are sufficiently distinct from their associated towns that they probably could be identified by those names rather than their associated towns, but for consistency it makes sense to list everything by an official name and avoid using obscure village/hamlet/section names like Millington, Centerbrook, and Abington. Thus, I can concur with Polaron's changes for these places. --Orlady (talk) 01:03, 13 June 2009 (UTC)
 * Good point about Winsted and Willimantic. News reporters do use these places in bylines so I think those are indeed locally distinct enough to be listed as the location. I've already somewhat highlighted Mystic by listing the town as "Stonington (Mystic)" but a better presentation is probably warranted. Would indicating distinctive subtown areas in the Summary column be an acceptable alternative? I think places in South Norwalk already do these (South Norwalk was also a former city). I would prefer leaving the town name in the "City/town" column as there is merit in grouping listings by town. For those worried about the accuracy of my changes, I basically used this state document as a reference. --Polaron | Talk 02:06, 13 June 2009 (UTC)

Followup: I just looked at the state of the list prior to my changes and just realized that the NRIS location information itself is inconsistent. Some places in Willimantic are listed as in Windham and some places in Winsted are listed as in Winchester. That tells me that the particular place level (town or subtown) used in the NRIS database is completely random (probably depends on who filled the nomination form). --Polaron | Talk 03:40, 13 June 2009 (UTC)
 * In my opinion, listing all sites by municipality is a great idea. For states with minor civil divisions (whether the MCDs are municipalities, as in CT, or not, as in OH), it's much more precise and easy to verify that a site is in one location or another by its municipality/unincorporated MCD.  To take an out-of-state example, why should we be so certain that the McConnell's Mill Covered Bridge in PA should be listed in Princeton, several miles away, when Rose Point is just across the highway?  Since everywhere in PA is incorporated, like in CT, it's much easier to say that this bridge is in Slippery Rock Township, which is obvious from a map.  Situations such as Winstead are ambiguous, but if it's a municipality, I think it's more precise to say that it's in the city, just as I'd think it better to list a Vermont site that's in an incorporated village in the village rather than in the town.  Nyttend (talk) 04:10, 13 June 2009 (UTC)

arbitrary break for convenience
As a further note, see Talk:National Register of Historic Places listings in Pittsburgh, Pennsylvania — because some sites near Pittsburgh are listed in Pittsburgh, they were erroneously split out with the city's sites when the Allegheny County list became too long. If there's a need to split a city from the list because it's too large, such as has been done with Hartford and New Haven here in CT, unless we're careful to identify sites by the correct municipalities this type of error is likely to occur. Nyttend (talk) 04:56, 13 June 2009 (UTC)
 * Sorry for the confusing wording — I think I might have been falling asleep when I wrote this. My point is that, if we always list sites by the NRIS if they're in another municipality, they may end up in the wrong list in cases such as Pittsburgh.  Nyttend (talk) 14:26, 13 June 2009 (UTC)
 * That Connecticut state government list that identifies all NRHP listings by town looks to be more reliable than the NRIS. That list should make it possible to reliably slot everything by town. In view of the existence that list (and the confusion regarding Winsted and Winchester that Polaron has observed), it seems to me that it makes a lot of sense to list everything by town -- no exceptions. --Orlady (talk) 15:14, 13 June 2009 (UTC)


 * The CT state document does seem like a very good source. I assume that it includes some errors, too, but I think it will be better than NRIS, perhaps perfect, about correctly identifying the town locations of CT NRHP-listed properties.  Checking the CT list for the Georgetown HD, I note that the CT list does list it properly, in agreement with NRIS, in the two towns it spans.  By the way, with several Oregon editors, I went through a detailed reconciliation between a similar Oregon state NRHP list-document vs. the NRIS-based list-tables of Oregon NRHPs.  That reconciliation, which I think is still visible at Talk:National Register of Historic Places listings in Oregon, turned up numerous NRIS errors of location, as well as some NRIS errors of omission of NRHP-listed places.  The Oregon list also had a number of errors.  Where there were location discrepancies the Oregon state list turned out to be more accurate in most but not all cases. doncram (talk) 18:42, 17 June 2009 (UTC)

(outdent) This is not quite resolved. Assuming that all the town/village location changes are accepted, which in general seems to be the consensus view (I am not entirely convinced, but won't argue unless i review the list and see problems), there remain some cleanup-type technical matters. Briefly these are:
 * 1) incorporating the CT state document as a source into the list articles, perhaps to be footnoted from the column heading that now reads "Town or City" or whatever
 * 2) clarifying, probably, about NRIS as an initial source
 * 3) reviewing the list of all these changes to update, also, every corresponding disambiguation page that gives the previously suggested article name. These can be found by checking each specific change (not too many).  Otherwise they could be found by scanning the entire list of NRHP places for redlink or bluelink entries that are common-looking names (Smith House, etc).
 * 4) creating redirects from every previous NRIS-based name, to make the connection for readers who may be looking for the site at its previous location. Note, there are many sources on the internet which state that Smith House is in Podunk Hamlet, Connecticut rather than being in Town, Connecticut, because the NRIS database is public domain and copied entirely into NRHP.COM and Archiplanet.Org, and copied partially into various tourist-related webpage systems. (Note also we lack many Podunk Hamlet, Connecticut articles so a wikipedia reader can't even look up what Town it is in or sort out which County list-article it will appear in.)
 * 5) Since redirects to redlinks will be deleted by a semi-automated process, create a stub article for each one of the renamed places. In other state lists, renaming of many articles has occurred but usually after an article is created, so the redirect is created naturally and there are not redlink redirect issues.  (This step can be combined with previous step by creating stub articles at old name and then moving it to new name.)
 * 6) perhaps capturing the list of deleted locations of type Podunk Hamlet, Connecticut and noting that at WikiProject Connecticut or somewhere that articles can be created for redlinks among them, including with mention of the location of NRHP-listed places within them. In other state lists, the redlinks in the NRHP list-tables have promoted / facilitated the process of creating hamlet articles.

This may seem boring or technical or whatever, but these are steps which should go onto a worklist and addressed, in my view. They are all consequences of going with different-than-NRIS-based naming conventions, which have elsewhere been used to generate the disambiguation pages and other structure, and to conform wikipedia to the potential reader expectations due to NRHP.COM etc.'s listings. These steps are all done in all other states gradually as new articles are created and moved, just here this is all to be done upfront due to the mass changes. The one other comparable state situation I am aware of is Pennsylvania, where editors have also chosen to move to (Township, Pennsylvania)- rather than (Hamlet, Pennsylvania) type names, and where some similar cleanup like this is also needed (but i think less, because I think most of the renamed places were created as articles first at the old names). doncram (talk) 23:00, 18 June 2009 (UTC)


 * The main part of this is creating stub articles for each moved place, hopefully at the old name and then moving it to the new name, to get the redirect set up easily. Note another reason to create all these stub articles is to address, within them, the discrepancy between what comes in as the NRHP town/city location in the infobox, vs. the name of the article.  Since in most cases the NRHP-given town is a small hamlet, it will probably be best to state in the article that it is located in the hamlet (which may be a redlink) within Town, Connecticut.  I don't care whether or not the town/city location in the infobox is changed. doncram (talk) 23:16, 18 June 2009 (UTC)
 * Hmm, I was thinking every one of the town name changes was an article name change, which is not the case, only a few may be. In most cases it was just a change of the town name in the City or Town column.  Polaron, can you please help identify any changes that did change an article name (redlink or bluelink I suppose), from format Smith House (Hamlet, Connecticut) to Smith House (Town, Connecticut), so that these can be addressed?  It is hard for me to detect these changes in the edit histories which involve many moves of rows to new city list-articles.
 * I'll check through all my changes tonight. The ones that might need to be relinked/renamed are the ones where the borough name was listed and possibly Winsted, Mystic, Willimantic. In any case, I'll post a complete list tonight. --Polaron | Talk 21:26, 19 June 2009 (UTC)

Here's the list of names that will be affected:
 * Main Street Historic District (Willimantic, Connecticut) --> Windham
 * Prospect Hill Historic District (Willimantic, Connecticut) --> Windham
 * Rockville Historic District (Rockville, Connecticut) --> Vernon
 * Carpenter House (Norwichtown, Connecticut) --> Norwich
 * Brookfield Center Historic District (Brookfield Center, Connecticut) --> Brookfield

I'm more inclined to leave the Willimantic ones alone as Willimantic is a well-known place name that is distinct from Windham. --Polaron | Talk 03:36, 20 June 2009 (UTC)


 * Trying to follow up...


 * Main Street Historic District (Willimantic, Connecticut) set up, with redirect from Main Street Historic District (Windham, Connecticut)
 * Prospect Hill Historic District (Willimantic, Connecticut) article set up, with redirect from Prospect Hill Historic District (Windham, Connecticut)
 * Rockville Historic District (Rockville, Connecticut) redirects to Rockville Historic District (Vernon, Connecticut)
 * Carpenter House (Norwichtown, Connecticut) redirects to Carpenter House (Norwich, Connecticut)
 * Brookfield Center Historic District, no disambiguating needed in title.
 * I guess that does it. This whole big Talk section can be archived, i think.  doncram (talk) 19:46, 11 August 2009 (UTC)

Using neighborhood names for city lists
Would there be objection to using neighborhood names in the city-specific NRHP lists (Bridgeport, Hartford, and New Haven). These cities have well-defined naighborhood planning areas that can be used to assign specific locations. As of now, the city lists use the city name as the location, which is not so useful. I know a few city NRHP lists also list neighborhood names so there is precedent for this sort of thing. --Polaron | Talk 18:02, 19 June 2009 (UTC)
 * Are you proposing to replace the list column that currently is called "City or town" and just gives the city name (kinda useless for an article that is about just one city) with a column for neighborhood/section names? That makes a lot of sense to me. Similar things have been done for large cities in other states (including Pittsburgh, as cited above) -- NRHP regulars (like Doncram) could give you a list and give you good advice on carrying this out. Also, some city NRHP list articles have actually been divided into separate articles for individual neighborhoods.
 * What are your sources for assigning NRHP listings to specific neighborhoods? --Orlady (talk) 19:52, 19 June 2009 (UTC)
 * Sounds good to me: no point in listing every New Haven site with a New Haven column, but a clearly defined neighborhood is much better than no community location column at all. Sources for assigning listings per neighborhoods are the official maps:  simply look at the street, plot on neighborhood map, and type; or if that's unclear, go by the coords.  You might want to ask Dtbohrer's input; s/he was the one who sorted the Pittsburgh listings.  By the way, are these boundaries stable, as I think you mean, or do they change every few years?  I at least have fallen into the hole of suggesting using districts that change virtually every year, primarily legislative districts or other districts with boundaries dependent on the legislative ones.  Nyttend (talk) 04:49, 20 June 2009 (UTC)
 * I have New Haven and Hartford official neighborhood boundaries (will post link to maps tomorrow). I'm pretty sure these are stable as these neighborhoods have existed for at least the past century. The online neighborhood maps for Bridgeport are currently unavailable on the city website so I might need to check out the hard copy that the Planning Department of the city has. --Polaron | Talk 05:33, 20 June 2009 (UTC)

I'm starting off with New Haven today. The official neighborhood maps are here. I also have a photocopied map of historic districts (national, state, and local) within the city limits as an additional reference. --Polaron | Talk 17:07, 22 June 2009 (UTC)

UPDATE: New Haven is done. I found an entry (added to the Register in 2009) that was mistakenly split out because it had "New Haven" in the town name when it is in a different town in New Haven County. I'll begin working on Hartford soon. Hartford neighborhood boundaries can be found here. --Polaron | Talk 19:51, 22 June 2009 (UTC)

I hope to finish Hartford by tomorrow. I don't think I'll be able to work on Bridgeport anymore. I also created a new category for historic districts as a subcategory of NRHP entries. Several states do that so I hope that doesn't ruffle anyone's feathers. With that I think I'll minimize my involvement in NRHP and only do occasional cleanup work. Sorry for all the trouble. --Polaron | Talk 01:52, 24 June 2009 (UTC)

UPDATE: Hartford is done. Someone else will have to do the work for Bridgeport. --Polaron | Talk 15:23, 24 June 2009 (UTC)


 * The Hartford and New Haven lists look great! Two questions:
 * Is it really necessary to include "(city)" in their names (e.g., National Register of Historic Places listings in Hartford (city), Connecticut)? Doesn't "Hartford, Connecticut" or "New Haven, Connecticut" imply the city? (I doubt that users would expect these names to refer to the counties.)
 * In the New Haven article, I found it odd that "The Hill" neighborhood is alphabetized under T. Should these links be piped to force alphabetization under H?
 * --Orlady (talk) 15:38, 24 June 2009 (UTC)


 * I didn't comment earlier because this was a secondary-at-best issue for me, but I agreed with Orlady at least on the first question. Now, I want to archive this discussion section, so have to revisit.  On the first question, I believe there are no NRHP list-article names, anywhere, with (city) included, and it is exactly as Orlady states:  few readers would expect to reach a county article when clicking on the New Haven or Hartford list-article names.  And it is very clear at the top of the city list-articles, that other NRHPs in the counties are covered in linked county articles.  And I note no objection or response to Orlady's question/suggestion.  So, towards resolving this, I moved the New Haven and Hartford city NRHP list-articles just now, per Orlady's suggestion.  On the second question, within the New Haven city article, I see that Polaron responded along the lines of Orlady's suggestion, a while back, by revising/moving the The Hill neighborhood entries to Hill neighborhood.  Unless there is further discussion soon, this discussion section could/should be archived as a resolved topic. doncram (talk) 02:40, 22 July 2009 (UTC)