Introducing the AARoads Wiki

[D-Dey65]

Just so many of you know, Wikipedia still has articles on the former New York State Route 25C and 25D, which the AARoads Wiki doesn't have yet. Since the articles on the Wiki so far swipe most of the material from the existing wikipedia articles, including many of the photos we've posted in them, I'd also like to point out that I've recently been posting pics of Adirondack Drive between Farmingville and Selden, New York to the Wikimedia commons. Having said that I don't think an article on Adirondack Drive would be worthy on either Wikipedia or the AARoads Wiki.

[Bruce]

Just about any file uploaded to Wikimedia Commons can be used on AARoads Wiki as well, so feel free to dump photos there. Missing articles from Wikipedia can be requested for import on the Cleanup page.

We've had some discussion on the inclusion criteria for non-numbered roads in the past, and will be refining them as we go along. I'm going to be writing some more articles on historic Washington highways as well (including never-built roads) given that I never had the opportunity to do so on Wikipedia.

[Scott5114]

What is the goal of the Wiki?  Is it to eventually become basically a better Wikipedia?  One in which anyone and everyone is free to contribute and discuss and edit?  Or will there be a group of gatekeepers?

In addition to what was contained in Jonathan's reply upthread, the goal is also to branch out somewhat from what Wikipedia allowed. Our wiki is not necessarily tied to the traditional "encyclopedia" model that Wikipedia seeks to emulate. (The term that we've decided best fits what we're shooting for is "gazetteer".) Accurate coverage of more or less anything pertaining to (real) highways is allowed, within reason (coverage of things like local streets or section line county roads isn't really within our remit, for instance). If it doesn't fit in the typical "encyclopedia article" mold, we have an Annex section that can hold the content. So, for example, I stashed the list of Oklahoma county codes you put together for me in there, since it might be useful to a roadgeek: https://wiki.aaroads.com/wiki/Annex:List_of_Oklahoma_county_codes

Anyone who is interested in editing the wiki can join. (In fact, we'd love more help, especially in states we don't have active editors in!) In practice, probably only roadgeeks would get a whole lot out of editing it, so our editor base is likely to be drawn from this forum and/or any other roadgeeks that happen onto the site from Google or whatever. The core team of editors are the same editors that have been maintaining the articles for years on Wikipedia, so we do have opinions about things like formatting and article structure that we'd like to keep consistent from article to article. But the mantra of "nobody owns the articles" has been drilled into us from years of Wikipedia, and since that has worked well for us for years, I would expect that would continue rather than shifting to a formal "gatekeeper" model.

I used to participate in an internet forum that eventually died a pitiful death because basically one guy was running it, he couldn't afford to run it anymore, and nobody else in the 'community' was able, ready, or willing to do anything about it.  Is there any danger of this Wiki eventually dying because whoever is running the show just throws in the towel (similar to what could imaginably happen to this forum someday)?

Currently myself and rschen7754 have direct backend access to the server, and there are currently nine people (including myself, rschen, and Alex) with administration rights in the wiki software (and we have a few others who have expressed interest in requesting rights eventually). Much of the core team of wiki editors have been working on this set of articles for decades at this point (rschen7754 and I both started in 2005, for instance; the earliest version of many of the Oklahoma articles were illustrated with circle shields because the meat cleaver didn't exist yet, and I was bugging KTA for information to use in the Kansas Turnpike article when I was in high school) so I would expect that if any of us were to get sick of it we could just pass the reins off to someone else.

1.  What is the goal of the Wiki?  Is it to eventually become basically a better Wikipedia?  One in which anyone and everyone is free to contribute and discuss and edit?  Or will there be a group of gatekeepers?

I can't answer the latter questions, but a "better Wikipedia" for roads, yes, that's the goal. Bulbapedia was cited by myself and some others, and is a great model. There are articles about every single Pokemon, as well as every single trading card, every single anime episode. And then highly technical articles dealing with the really complex formulas for calculating damage, predicting RNG, possible stat ranges, and so on. Something like 50,000 articles on just one video game franchise. It seems that is the end goal here. Every numbered highway has an article, gives at least some cursory information. Perhaps articles about historical alignments, historical highways, etc.

Is it limited to numbered highways?  I didn't see any indication that the AAroads Wiki is beholden to such a standard.  If an unnumbered highway is notable, then why not add an article?

We do not have exact inclusion guidelines written down just yet. Historically, the US Roads project on Wikipedia went for a 'state maintained numbered highways only' criteria because that was a clear, unambiguous standard that was set by a body separate from the project, so we could justify inclusion by saying "well, it's not us saying the road is important, it's the DOT, if you don't like it, take it up with them".

Now that we are on our own wiki, we could potentially expand the inclusion criteria. However, every article that is included in a wiki does impose a bit of a maintenance liability on its editors; links in references go dead over time and need to be replaced or removed, when templates used in articles are upgraded, articles need to be checked for breakage, and so on. So any expansion needs to be carefully considered to ensure that the benefit of additional coverage outweighs the maintenance costs. Also, it is somewhat tricky to write an inclusion guideline that would allow things like, say, Nipton Road that would be of general interest, while excluding obscure roads nobody cares about like Las Grande Ranchoristo Terrace in the McMansion Heights subdivision (and yes, we have encountered people that will write articles on such things even under Wikipedia's rules–recently there was an instance of someone writing an article for a 2,300-ft-long street in Melbourne, and they were so adamant about its inclusion that they managed to get themselves sitebanned for abusing multiple accounts to try to obstruct the deletion discussion).

[Max Rockatansky]

I can think of dozens of unnumbered roads just in California and Arizona alone that have way more significance from a historic standpoint than a high percentage of state highways.  This isn't directed at this Wiki project but we as a community (emphasis this is my opinion) don't do such a great job documenting these types of highways.  There certainly is an opportunity for this project to break the mold from mentality of only numbered routes are important.  Granted I know in the near term this isn't the priority but I would urge those involved in building this Wiki to consider things like this. 

To clarify, I'm talking about unnumbered roads along the lines of say "San Francisquito Canyon Road"  and not the subdivision loop road.

[Scott5114]

It's certainly a possibility. One part of the problem, as alluded to above, is the difficulty in defining this class of roads, such that someone without direct familiarity with them (like, say, an editor from the East Coast) could say with some accuracy whether a road is one of these historic unnumbered roads, or something not worth covering. Do you have any suggestions for how the line might be drawn?

[rschen7754]

Currently myself and rschen7754 have direct backend access to the server, and there are currently nine people (including myself, rschen, and Alex) with administration rights in the wiki software (and we have a few others who have expressed interest in requesting rights eventually). Much of the core team of wiki editors have been working on this set of articles for decades at this point (rschen7754 and I both started in 2005, for instance; the earliest version of many of the Oklahoma articles were illustrated with circle shields because the meat cleaver didn't exist yet, and I was bugging KTA for information to use in the Kansas Turnpike article when I was in high school) so I would expect that if any of us were to get sick of it we could just pass the reins off to someone else.

I was a Wikimedia steward with full access to the software at one point, so I'm pretty well versed in how the application interface operates. Where there's a bit of a struggle is keeping the application running with acceptable performance. I'm a software engineer professionally but I don't have a lot of PHP or Apache experience. Scott has a bit more Linux experience. We're open to suggestions on how to get it running better (https://github.com/aaroads-wiki/aaroads-config).

[Max Rockatansky]

It's certainly a possibility. One part of the problem, as alluded to above, is the difficulty in defining this class of roads, such that someone without direct familiarity with them (like, say, an editor from the East Coast) could say with some accuracy whether a road is one of these historic unnumbered roads, or something not worth covering. Do you have any suggestions for how the line might be drawn?

Part of the fun in this hobby (at least for me) is discovering things about roads and places I didn't know existed or fully understood.  I don't really know if you really can have a fully hardened rule and get a good mix of unnumbered highways, my thought would be that is a subjective measure.  Certainly, the lower end of where that line is going to be pretty obvious as we've already discussed.   I am certainly not suggesting subjectively looking at everything submitted for unnumbered highways is a perfect system.  Part of the problem I see with Wikipedia is that they are now too structured and rigid (hence why this project exists to begin with) with their rules in general.  Why replicate the worst aspects of the existing Wikipedia model when something that may require more subjective effort may yield a fuller product?  Wikipedia definitely was a far more interesting place when it wasn't burdened by a bunch of stodgy rules created by the editorial elites.

[kphoger]

Also, it is somewhat tricky to write an inclusion guideline that would allow things like, say, Nipton Road that would be of general interest, while excluding obscure roads nobody cares about like Las Grande Ranchoristo Terrace in the McMansion Heights subdivision (and yes, we have encountered people that will write articles on such things even under Wikipedia's rules–recently there was an instance of someone writing an article for a 2,300-ft-long street in Melbourne, and they were so adamant about its inclusion that they managed to get themselves sitebanned for abusing multiple accounts to try to obstruct the deletion discussion).

What, exactly, is the problem with allowing such an article?

[Scott5114]

Also, it is somewhat tricky to write an inclusion guideline that would allow things like, say, Nipton Road that would be of general interest, while excluding obscure roads nobody cares about like Las Grande Ranchoristo Terrace in the McMansion Heights subdivision (and yes, we have encountered people that will write articles on such things even under Wikipedia's rules–recently there was an instance of someone writing an article for a 2,300-ft-long street in Melbourne, and they were so adamant about its inclusion that they managed to get themselves sitebanned for abusing multiple accounts to try to obstruct the deletion discussion).

What, exactly, is the problem with allowing such an article?

The problem is that a wiki article takes a non-zero amount of effort to maintain over the years, even if it is not expanded at all. Most articles will include one or more templates, which are basically small bits of software that take input and turn it into a particular type of output (the familiar "infobox" that appears on most articles on Wikipedia is an example of this). Over time new features often need to be added to these, which can cause the required input to get a certain result to change, meaning that someone has to go through all of the articles and update them to match the new way the template works. There is also the problem of links being used in references breaking over time; someone has to go through and check these (and either fix or remove them). There is also, of course, the fact that just because someone types some text into a wiki page, that doesn't mean it's true; it is a good idea to have someone periodically spot-checking that the references actually supports what the text claims.

This all takes a non-zero amount of time to accomplish, of course, and with a smaller editor base than Wikipedia has, the opportunity cost from time that we could be spending researching to expand existing articles is a bigger concern. This time sink is justified if it is an article on a major highway that a reader is likely to actually want information about, since that's what wiki editors are for. But if I were to, say, make an article on the cul-de-sac I live on, I'm probably the only person who is ever going to care about that article. This means that the updates and verifications to that article are going to spread the other editors thinner than if the page didn't exist at all, and it will help fewer readers than if that editing time was spent researching, say, the history of K-15 or something like that. (There is also the problem that the more obscure the subject, the harder it is to find references on it; this means that we have no real guarantee that anything written about a minor residential street is accurate or not beyond what can be verified by looking at a map.)

[kphoger]

But if I were to, say, make an article on the cul-de-sac I live on, I'm probably the only person who is ever going to care about that article.

Hey, I'd probably read it!  I'm sure you could come up with something interesting, such as when and why each of your neighbors found it necessary to advertise their favorite sports teams by painting their logos on the curb next to the house number.  Is that practice allowed or prohibited by local ordinance?  Did it start with just one guy?  Is there now a friendly rivalry between residents of your cul-de-sac?  An unfriendly rivalry?  Are non- OSU fans allowed to move in?  Inquiring minds want to know...

[bulldog1979]

Since the core group behind the fork comes from Wikipedia, we start with that framework as a baseline for the new AARW. We aren't rigidly beholden to that framework of policies and guidelines going forward though. We have a prime opportunity to review all of those rules just to see what does or doesn't make sense for us.

Much has been said so far about loosening inclusion standards from the rigid application of the General Notability Guideline (which is supposed to supplement, not supplant subject-specific guidelines). For example, the article on M-144 from the 1930s in the Lansing area was demoted and deleted off Wikipedia, but it was restored onto AARW and polished up. So where some of the naysayers have been picking at the edges to declare certain topics as ineligible for articles, we're able to say that every state highway, current or former, warrants an article to have a full set in the future, just as we had tried on WP. From there, we can expand to other categories where we can articulate some sort of guideline.

One other angle is that we can reexamine the concept of a reliable source. We have ideas in mind to allow some additional sources that WP doesn't currently accept, like the long-standing well-regarded roadgeek websites. Because we aren't going to worry about "notability", we won't have to focus on the independence of a source in the same fashion as WP has.

[freebrickproductions]

If a county has numbered county routes/roads, a list of those might not be a bad idea, IMHO, with articles for the more "notable" ones.

I also know SABRE has an article on nearly every classified road in the UK and Ireland, might not be a bad idea to at least link over there for a more in-depth look at British and Irish roads, an maybe just cover the more basic/important bits as we see fit on this wiki.

[rschen7754]

https://www.tiktok.com/@bmacs001/video/7277383641397857579

I also know SABRE has an article on nearly every classified road in the UK and Ireland, might not be a bad idea to at least link over there for a more in-depth look at British and Irish roads, an maybe just cover the more basic/important bits as we see fit on this wiki.

One day we do want to expand to covering roads in every country. One problem with SABRE is that due to their licensing, they cannot import the current Wikipedia articles.

[Scott5114]

https://www.tiktok.com/@bmacs001/video/7277383641397857579

Has AARoads ever gone viral before?

[rschen7754]

Also https://en.wikipedia.org/w/index.php?title=Wikipedia:Wikipedia_Signpost/Next_issue/News_and_notes&curid=74677104&diff=1175001390&oldid=1174586075 (draft of story for internal newsletter The Signpost)

https://www.facebook.com/groups/wikipediaweekly/posts/6561226760591766

[freebrickproductions]

I also know SABRE has an article on nearly every classified road in the UK and Ireland, might not be a bad idea to at least link over there for a more in-depth look at British and Irish roads, an maybe just cover the more basic/important bits as we see fit on this wiki.

One day we do want to expand to covering roads in every country. One problem with SABRE is that due to their licensing, they cannot import the current Wikipedia articles.

I was more of linking/referencing to SABRE for information on British/Irish roads. Especially since it's probably gonna be more in depth on that subject that the current Wikipedia articles are.

[bulldog1979]

If a county has numbered county routes/roads, a list of those might not be a bad idea, IMHO, with articles for the more "notable" ones.

I also know SABRE has an article on nearly every classified road in the UK and Ireland, might not be a bad idea to at least link over there for a more in-depth look at British and Irish roads, an maybe just cover the more basic/important bits as we see fit on this wiki.

Long term, we may get county road lists. It's a matter of people finding the time to write them and finding the sources for the information.

As for SABRE, we currently don't have any British articles. By design, we focused our time and energies on importing US and Canadian articles first. To put things in perspective, we got the server in the middle of June, and we only now felt like we had done enough work to launch after almost 3 months. There's still work to be done as we need to create/convert maps over to the new mapping format and edit articles to fix the links and other artifacts from the import process. We currently have over 15,000 articles on the server, and many still haven't been touched from the initial import process. Importing was automated, while the rest is semi-automated at best.

If there is a demand and a user base to expand our coverage, we can import more articles. We've  already discussed expanding the concept of a reliable source to encompass well-regarded roadgeek sites, but I don't know that we'd stretch that to include other wikis. We could encourage editors to consult the sources listed for a topic on The Roader's Digest, but I wouldn't be in favor of citing that wiki directly.

[Chris]

I'm reading this with interest.

In the Dutch road enthusiast community, we decided to do something similar, already back in 2007, when the Dutch-language 'roads wiki' (Wegenwiki) was launched. It now has nearly 23,000 articles, but none of those were copied from Wikipedia due to the poor quality of the road articles on the Dutch Wikipedia.

The problem we had with the Dutch Wikipedia at the time was the poor quality of road articles (and the NL Wikipedia in general) and the excessive bickering and edit wars over names, notability and style.

Setting up your own Wiki has some advantages, you can retain some creative control, can use relevant sources, don't have to worry about intricate Wikipedia guidelines and 'common' knowledge about a million conventions and notability issues. A smaller, dedicated team also allows for better and more consistent article structure. Many outsiders update road articles only if they hit the news headlines and this tends to mess up the article quality, or overemphasize whatever is the latest (anti) road hype.

A problem with copying Wikipedia articles en masse may be that Google sees the AAroads Wiki as a mirror site instead of its own thing, which could rank it lower in the search results.

[J N Winkler]

A problem with copying Wikipedia articles en masse may be that Google sees the AAroads Wiki as a mirror site instead of its own thing, which could rank it lower in the search results.

This has been cited as a reason to allow the roads articles to disappear from the English Wikipedia over time.  However, the AARoads Wiki was announced just eight days ago and we are now seeing imported articles floating up to within the top five results on their own.

(I did find the Wegenforum Wiki useful in understanding why the NBd was created, by the way.)

[D-Dey65]

I recently tried to find a source that Wikipedia might find suitable regarding evidence of the colored US shields in Florida. The source I thought would have this evidence didn't have it.
https://www.fdot.gov/gis/floridatransportationmaparchive.shtm

However, on the 1956 map, which I thought would have it, I found some interesting info.
https://fdotwww.blob.core.windows.net/sitefinity/docs/default-source/co-gis/past_statemap/flstatemap1956.pdf?sfvrsn=c0a0c1a1_0

1)Florida SR 100 east of Bunnell used to be part of SR 11, which I learned earlier this year.


2)Florida SR 44 east of DeLand actually used to be part of SR 40. Evidently, SR 40 ran south along an overlap with US 17 then turned east along where SR 44 is today.

[NE2]

Are there plans to copy over the wikiproject pages? (Or has this already been done?) There are some good resources in there.

[rschen7754]

Are there plans to copy over the wikiproject pages? (Or has this already been done?) There are some good resources in there.

The pages have been copied over but are still being cleaned up.

Signpost story went live: https://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2023-09-16/News_and_notes (also comments are possible)

[Scott5114]

Are there plans to copy over the wikiproject pages? (Or has this already been done?) There are some good resources in there.

The pages have been copied over but are still being cleaned up.

Additionally, we have an "Annex" namespace that is intended for resource pages that would be of general interest to roadgeeks rather than only being useful for project maintenance (so it would be ideal for, e.g. history resources or notes on mass renumberings). Many of the WikiProject subpages will probably eventually end up there, rather than buried under the WikiProject structure.

[Quillz]

You know, just to bring up Bulbapedia again: the latest mainline games got some DLC released 48 hours ago, and already there are about 1,000 new articles detailing it. The amount of dedication in just a single day is astounding. And I'm talking the articles dealing with data mining, stuff like that. Really gives me hope this wiki will be similarly passionate. Getting new articles up and running in just a few days. Pretty neat. (For example, an article about the new Crosstown Connector in Bakersfield).

[vdeane]

I can't help but wonder if the "reading a map is original research" thing is being at least partly fueled by the fact that such is a skill that is dying out in the era of GPS navigation.  I find that people don't even try to figure out where they need to go; they just plug the address into their phone and trust it to tell them where to go (which can be a huge problem if the data has issues or is outdated, if there's a malfunction in the device, or the road network is a little tricky in the area and the algorithms for generating the directions can't produce something useful).