Wikidata:Property proposal/rebuilt

rebuilt edit

Originally proposed at Wikidata:Property proposal/Place

Not done

Description	year or date when the structure was reconstructed, rebuilt, repurposed or replaced by similar one (if the new and old structures have separate items, use P167/P1398 links instead)
Represents	reconstruction (Q1370468), reconstruction (Q2478058), repurposing (Q21021151), replacement (Q23009439)
Data type	Point in time
Domain	item
Example 1	bridge of road III/27011 over the Panensk potok in Brniště (Q115973713) → around 1989 (a stone bridge replaced with a concrete one)
Example 2	former Protestant church in Vraclávek (Q81914020) → 1980s (a church repurposed to a garage)
Example 3	Bethlehem Chapel (Q831362) → 1952 (a replica of a formerly destroyed building)
See also	structure replaces (P1398), structure replaced by (P167), point in time (P585), inception (P571), dissolved, abolished or demolished date (P576)

Motivation edit

I often come across this problem especially with bridges, which are often replaced by a new bridge of the same or different construction, but at the same place and with the same function. When it comes to encyclopedically and architecturally significant structures, it is advisable to establish a separate item for each version of the bridge, and link them with structure replaces (P1398) / structure replaced by (P167). However, if it is a less notable structure, or the new structure is essentially a replica of the original one, or as a stopgap solution before the item is cloned into two items, it may be expedient to include both structures in one common item.

Currently, such a date can be recorded using significant event (P793) with a value reconstruction (Q2478058), repurposing (Q21021151), replacement (Q23009439) or similar one with a qualifier point in time (P585). However, this method is to complicated and hidden in relation to the fundamental importance of such a date. Similar significant events have separate properties: inception (P571), dissolved, abolished or demolished date (P576), date of official opening (P1619), date of official closure (P3999), service entry (P729), service retirement (P730), time of discovery or invention (P575) etc. A fundamental reconstruction or replacement of a structure is certainly an event of comparable importance.

I can imagine the application of this property especially for structures: buildings, bridges, roads and railways, dams, parks, towers etc. It can also be admitted for some unique movables, as sculptures, jewelry, vehicles etc., maybe also for some abstract entities (projects, activities, organizations).

The question is whether the property should cover a wide scale of events (reconstruction (Q1370468), reconstruction (Q2478058), repurposing (Q21021151), replacement (Q23009439)) or be more specific for every (or only some) of these cases. --ŠJů (talk) 01:55, 20 February 2023 (UTC)[reply]

Discussion edit

Oppose I disagree with the statement that using P793 is complicated, I think it's in fact easier and simpler, when you are adding important dates for an item, you don't have to worry about if there's a property for the event type you want to model. In a ideal database, P571, P576, etc. would not exist, and all important dates would be modeled using P793 with P585 qualifier, but sometimes we make exceptions in order to make easier for newcomers, but it complicates things when consuming the data. --Tinker Bell ★ ♥ 21:48, 20 February 2023 (UTC)[reply]
Oppose. Rebuilding is a big issue for fortifications. Timber castles are replaced in stone, forts and gun batteries are often changed, some changes involve adding, some repurposing, some complete obliteration and rebuilding. significant event (P793) should be the master source for change information. While a inception (P571) of 1500 is odd for a 19th century stone fort that replaced a earth gun battery, or a historic house that replaced a castle, that's the nature of an evolving site, I don't think a new top-level property 'rebuilt' will help. More useful would be a 'current structure largely dates from' property, and I would be more inclined to support that proposal than this one. Vicarage (talk) 19:47, 27 February 2023 (UTC)[reply]
Oppose This is nicely covered by P793. I'd actually suggest revising current properties which might be deleted and instead remodelled via P793 (such as possibly date of official opening (P1619)). Vojtěch Dostál (talk) 10:24, 2 March 2023 (UTC)[reply]
Oppose per the above. But @Vojtěch Dostál, Vicarage, Tinker Bell: I'd welcome some advice: do we have a good value for significant event (P793) to say that something was rebuilt rather differently, as opposed to reconstructed to restore it close to what it was? (eg best P793 value for a railway bridge with a timber superstructure that was replaced with an iron girder structure?) Thanks, Jheald (talk) 12:03, 3 March 2023 (UTC)[reply]

Just to add to the above, a few weeks ago Multichill suggested using renovation (Q2144402) for this, perhaps with applies to part (P518) = deck (Q2042353) or superstructure (Q15715941) as appropriate; but to me that suggests something more like limited repairs rather than wholescale reconstruction. So if anybody has thoughts, that would be useful. Jheald (talk) 13:05, 3 March 2023 (UTC)[reply]

Notified participants of WikiProject Built heritage -- Jheald (talk) 13:06, 3 March 2023 (UTC)[reply]

I think if the process is not a simple restoration, but a major change, renovation (Q2144402) is fine, and if is a completely new construction, maybe it should have its own item. --Tinker Bell ★ ♥ 13:36, 3 March 2023 (UTC)[reply]

I think they are better as two element if they are two different building. At least it was I done for Grasslawn II (Q105187074) and Grass Lawn (Q5597443). Fralambert (talk) 15:53, 3 March 2023 (UTC)[reply]

Plaque marking the reconstruction of Brotherton railway bridge

Sometimes yes, sometimes no (in my opinion). e.g. I made two separate items for Brotherton Tubular Bridge (Q116914119) and Brotherton Railway Bridge (Q116914074) because the thing that made the first one so distinctive was its tubular cast-iron construction (reflected in how the bridge was referred to), which was not continued in the 1903 structure, even though it shared the same alignment, sits on the same masonry abutments, is approached by the same original masonry-edged embankments either side, etc.

On the other hand, if we consider Duke's Cut Railway Bridge (Q116940837) I believe that has had a number of different decks in its life -- the present concrete deck cannot be particularly old, and old maps show that the canal was previously crossed by four tracks rather than the present two (the junction used to be further south). I thought I'd found somewhere some dates about particular stages of the bridge's evolution, though I haven't seemed to be able to find them again in a quick seach just now. But either way, I'm not sure there would be so much value in distinguishing the different stages of the bridge's life with different items, given that most railway bridges don't have any item at all. (And similarly for most of the other railway bridges over the Oxford Canal as documented at eg [1], almost all of which have been rebuilt or had a new deck in their lives).

Ness Viaduct (Q112243616) / original Ness Viaduct (Q112243687) and Angarrack Viaduct (Q4761980) / Angarrack Timber Viaduct (Q116357286) are a couple more cases where separate items did seem right -- the first because the new bridge is so different from the original, the second because the piers of the original still exist and are visible parallel to the piers of the new viaduct, and are specifically mentioned in the listing as a listed building.

On the other hand for Stonehouse Pool Viaduct (Q116304364), which is another of the en:Cornwall Railway viaducts, I kept all in the one item, even though it went from five stone piers to 10 brick piers (albeit on the same line).

Perhaps I'm just inconsistent; or lazy. But I do think for most bridges (if we have an item for them at all), we maybe won't even know the full history to be able to record it. And even if we do, it is very tempting both work-wise and accessibility-wise to keep everything together in the one item.

But keen to know what more people think. (General Ship of Theseus (Q1050837) problem). Thx, Jheald (talk) 00:15, 5 March 2023 (UTC)[reply]

Currently we don't even handle London Bridge (Q130206) well, the most famous rebuilt bridge. All language wikipedias seem to go for a single page for the London crossing, a dozen have Arizona ones. But if you read the WD entries, the detail is a right muddle. One item for all versions in a place would be provide a much cleaner narrative, even if nearly every statement would need a date range qualifier. Vicarage (talk) 05:12, 5 March 2023 (UTC)[reply]

Not done, no consensus of proposed property at this time based on the above discussion. Regards, ZI Jony ^(Talk) 12:31, 24 January 2024 (UTC)[reply]