The Wikipedia roads thread

[NE2]

Bleh. Old exit numbers are definitely useful information, e.g. for cross-referencing something that was written before the renumbering.

[Michael in Philly]

Hi -

I just visited the Wikipedia page for I-95 in Maine and was disappointed to see the old exit numbers have been removed from the exit list.  (I mean, of course, they used to appear alongside the current ones.)  On the chance that that was done by someone here, I think that we should really hesitate before removing historic information that someone may be interested in.  Just my two cents.

That said, I'm very non-participative on Wikipedia - I think I've done one edit ever - but know how to get at the history page.  Would clicking on "undo" in the right places make deleted information visible to me without changing it for the public?

The best way to do this is often to display the history and then click the "cur" link to the left of the past revision you're interested in seeing. It will highlight the differences between the current version and the old one. If you use "prev" it will show the differences between that version and the immediately-preceding one.

It's not a foolproof method, but it's often the easiest way to display changes. You can always view the full text of any old revision by simply clicking on the time and date in the history—all old article versions are saved. But doing that isn't always helpful because it won't always be apparent where something was changed.

The "undo" idea may not work because often it will say "this edit cannot be undone," usually if there are multiple conflicting edits since then. But yes, you can always try clicking "Undo" and then simply do not save the page after you click it—instead use your "back" button or click somewhere else.

I figured it out - thanks.

[Michael in Philly]

Hi -

I just visited the Wikipedia page for I-95 in Maine and was disappointed to see the old exit numbers have been removed from the exit list.  (I mean, of course, they used to appear alongside the current ones.)  On the chance that that was done by someone here, I think that we should really hesitate before removing historic information that someone may be interested in.  Just my two cents.

That said, I'm very non-participative on Wikipedia - I think I've done one edit ever - but know how to get at the history page.  Would clicking on "undo" in the right places make deleted information visible to me without changing it for the public?

The general guideline looks to be that old exit numbers will be removed from an article when the state DOT has removed transitional signage, no longer publishes the old numbers on maps distributed to the public or about 5 years. One of Wikipedia's core policies is that "Wikipedia is not a collection of indiscriminate information". Once the transition is complete, the specific old exit numbers aren't really useful for the general public that reads our articles. Roadgeeks might different levels of interest, but Wikipedia is written for a general audience. (Trust me, we get enough crap from some people on the site for even writing highway articles in the first place...)

Hmm.  Not trying to make an issue of this (personally, as long as the old version of the page is available to me, as it turns out it is, that's cool.  IF - IF - I can really trust them not to at some point purge old versions.).  But, well, as to the idea that Wikipedia's not "indiscriminate, " I'd say there's a fine line between "exhaustive" and "indiscriminate," and Wikipedia pushes right up to it - and that's not a criticism:  that's why I love it!  And what non-roadgeek looks at exit lists?  (And the fact that "we get crap for writing highway articles at all" carries no weight at all with me.  Not coming from you but coming from the crap-givers:  no one's holding a gun to their heads forcing them to read them.  Unless there are server-capacity issues....)

THAT SAID, has anyone considered an off-Wikipedia Roads Wiki, we're we could be as geeky and indiscriminate as we damn well please?

[Scott5114]

Hi -

I just visited the Wikipedia page for I-95 in Maine and was disappointed to see the old exit numbers have been removed from the exit list.  (I mean, of course, they used to appear alongside the current ones.)  On the chance that that was done by someone here, I think that we should really hesitate before removing historic information that someone may be interested in.  Just my two cents.

That said, I'm very non-participative on Wikipedia - I think I've done one edit ever - but know how to get at the history page.  Would clicking on "undo" in the right places make deleted information visible to me without changing it for the public?

The general guideline looks to be that old exit numbers will be removed from an article when the state DOT has removed transitional signage, no longer publishes the old numbers on maps distributed to the public or about 5 years. One of Wikipedia's core policies is that "Wikipedia is not a collection of indiscriminate information". Once the transition is complete, the specific old exit numbers aren't really useful for the general public that reads our articles. Roadgeeks might different levels of interest, but Wikipedia is written for a general audience. (Trust me, we get enough crap from some people on the site for even writing highway articles in the first place...)

Hmm.  Not trying to make an issue of this (personally, as long as the old version of the page is available to me, as it turns out it is, that's cool.  IF - IF - I can really trust them not to at some point purge old versions.).  But, well, as to the idea that Wikipedia's not "indiscriminate, " I'd say there's a fine line between "exhaustive" and "indiscriminate," and Wikipedia pushes right up to it - and that's not a criticism:  that's why I love it!  And what non-roadgeek looks at exit lists?  (And the fact that "we get crap for writing highway articles at all" carries no weight at all with me.  Not coming from you but coming from the crap-givers:  no one's holding a gun to their heads forcing them to read them.  Unless there are server-capacity issues....)

THAT SAID, has anyone considered an off-Wikipedia Roads Wiki, we're we could be as geeky and indiscriminate as we damn well please?

For legal reasons, old revisions can't be purged, as then Wikipedia would be in violation of the CC-BY-SA license which requires documentation of everyone who's ever edited a page. Besides, they have no reason to do so, as a wiki with no old revisions is sort of pointless, because then you can't revert. (Look up Metababy for an idea of what a wiki with no revisioning capability is like.)

The Wikipedia roads editors have considered forking from Wikipedia but there are enough problems with getting editors on Wikipedia, with the name brand and all the infrastructure that draws people to work on the project. All of the time that would need to be spent importing the pages from Wikipedia (as long as all of the templates, photos, and shield images that the pages depend on) wouldn't really be worth it. There's over 5000 articles alone...

I tried putting a request in for a map of "Texas State Loop 306" but nothing has happened yet. I guess it's a low priority? But could someone help me with this? (I'm "DCBS18" on Wikipedia"
BigMatt

Maps take a long time to make (especially if nobody has a setup of TX data...setting up a new state is a labor-intensive process in QGIS). That means that they're prioritized, by means of the article's assessment (the higher articles will get maps first). If you want your article to get a map quicker, improve it to B-class (or even better, get it a GA rating).

You can also learn to make maps yourself. If you go to USRD's IRC channel they will generally be glad to help.

Edit: Where did you put the request at?? I'm not seeing anything on the map request page...
Edit 2: Oh, okay, it was over a year ago. Yeah, it looks like the map request was deferred until the article is improved beyond the state it's in. When that didn't happen the request was archived. Improve the article and then put in the request again.

[english si]

www.sabre-roads.org.uk/wiki is basically the British road geeks way around all the annoying wikipedia guys talking about 'notability', 'original research' and 'unreliable sources' on one hand, while others plunder the members' sites and plagiarize them, so that added together works out to perhaps 60% of content that isn't original research on WikiProject:UK roads' better articles has been plagiarized off of a SABRE member source.

http://routes.wikia.com and www.wegenwiki.com are also non-wikipedia roads wikis (French and Dutch languages respectively)

Notability and original research are the big problems with roads on Wikipedia - is some small road notable? Unlikely and it only takes one jobsworth to notice it's existence to go bye-bye. Can you make the article decent easily without adding material you yourself have researched? It's rather hard without ripping off someone else's research in the process. You can't play by Wikipedia's rules and make a comprehensive road directory, IMV.

[NE2]

Not to mention that there are some kinds of original research/non-notable information that we wouldn't want, e.g. everything in fictional highways. Wikipedia's dispute resolution sucks, but it's better than nothing.

Note: I used to be a Wikipedia editor, but quit a few years ago after the MMORPG element got to be too much, with people pumping out shitty articles and removing important but uncited stuff from existing articles to get 'good article' and 'featured article' badges. Now I edit OpenStreetMap, which is based on original research (mainly of factual information) and has an obviously almost nonexistent notability bar.

[Scott5114]

Notability and original research are the big problems with roads on Wikipedia - is some small road notable? Unlikely and it only takes one jobsworth to notice it's existence to go bye-bye. Can you make the article decent easily without adding material you yourself have researched? It's rather hard without ripping off someone else's research in the process. You can't play by Wikipedia's rules and make a comprehensive road directory, IMV.

I agree with some of your points. A lot of the decisions that the USRD folks have to make are influenced by the non-roads part of Wikipedia. It is a difficult balancing act between keeping roadgeeks happy and keeping non-roads Wikipedians happy.

I am not really familiar with the state of the British road articles, but in the US we perpetually face the problem of too many roads and not enough editors. It is problematic when someone creates thousands of county road articles (C-roads would be the closest UK equivalent) and fails to maintain them or improve them to an acceptable standard. Additionally there is the hazard that someone will happen on a small road article such as that and attempt to get all roads articles removed, which just wastes the editors' time having to defend their work. This is why "notability" is often enforced on the lower class local roads. It doesn't happen so much with Interstates/motorways, state highways/A routes, etc.

In my experience direct plagiarism of roadgeek websites is rare, though if you have performed research into some historical facet of the road or calculated mileposts or something like that, that information may be written into the article and a citation given to your work, but this is not objectionable because it is not plagiarism. If blatant plagiarism has occurred there are processes to deal with it and get the copied work removed.

[NE2]

I am not really familiar with the state of the British road articles, but in the US we perpetually face the problem of too many roads and not enough editors. It is problematic when someone creates thousands of county road articles (C-roads would be the closest UK equivalent)

http://en.wikipedia.org/wiki/Wikipedia:Centralized_discussion/B_roads_in_the_United_Kingdom

[bulldog1979]

We've had members of the roadgeek community that maintain websites attempt to remove information from Wikipedia articles under copyright violation or plagiarism claims. Remember, that under US copyright law, you can't copyright a fact or a date. If your website says that X road opened on Y date, and a Wikipedia article repeats that information, no copyright violation has occurred, especially if your site is listed as the source. Now, if the sentence is lifted word for word, that's different. Let Scott5114, myself or someone else connected to the project know so that we can deal with it.

Then again, there's also the issue that most roadgeek websites aren't supposed to be used as sources in articles on Wikipedia anyway. For all of the respect and esteem I have for those that run and maintain their own websites, there's a policy on Wikipedia against using "self-published" sources. I'm not saying that roadgeek websites are publishing any incorrect information, but since there is not a formal editorial process at work to vet the information published on each website, they're not considered "reliable sources" under the policy. The same policy also excludes any book published with a vanity/on-demand publisher, most blogs, etc. Once again, I didn't say that any roadgeek website is unreliable, just that the policy says we have to stick to sources by acknowledged experts or sources published by groups with editorial oversight and a reputation for accuracy. Certain roadgeeks' sites may be accepted as sources if we can demonstrate that reputation.

[bulldog1979]

Bleh. Old exit numbers are definitely useful information, e.g. for cross-referencing something that was written before the renumbering.

The issue only recently came up for discussion, but given our need to write for a generalized audience, old exit numbers may still fall outside of the realm of specific details left out of articles going forward except during a transition period. The point you expressed wasn't brought up in that discussion, so I'll pass it along and see if that concept changes anything. The templates that can be used to general junction/exit list templates were recently changed to create old exit numbers. We might update them to minimize or differentiate the old numbers differently using the template, and they might be added back in the future.

[rschen7754]

The UK roads project has their own issues entirely.

Typically, the only old revisions that get deleted are ones containing copyvios, or where something has to be removed for legal reasons (someone's contact information got posted, death threats, etc). Administrators (Scott5114 and myself) can at least give you a clue as to what the deleted content was in the first case. I don't think we can just post deleted revisions, but we can at least comment about it.

As far as forking from Wikipedia... we've had this discussion quite a few times. And what it boiled down to was that the benefits of being on Wikipedia are better than the disadvantages, right now. If that ever changed (Wikipedia passed a guideline that caused serious harm to the project, for example) I suppose we could revisit the discussion, but then we have to discuss server hosting, migration, etc.

[J N Winkler]

In my experience direct plagiarism of roadgeek websites is rare, though if you have performed research into some historical facet of the road or calculated mileposts or something like that, that information may be written into the article and a citation given to your work, but this is not objectionable because it is not plagiarism. If blatant plagiarism has occurred there are processes to deal with it and get the copied work removed.

What if both text and citations are lifted from a roadgeek website and lightly edited so that the verbiage is no longer a word-for-word match, and there is no attribution to the website.  What recourse does the site owner have?  Even if his words have not been stolen, he has been deprived of recognition and credit for his hard work excavating in archives to bring the information to light.

[hbelkins]

The Wikipedia roads editors have considered forking from Wikipedia but there are enough problems with getting editors on Wikipedia, with the name brand and all the infrastructure that draws people to work on the project. All of the time that would need to be spent importing the pages from Wikipedia (as long as all of the templates, photos, and shield images that the pages depend on) wouldn't really be worth it. There's over 5000 articles alone...

I'd be more inclined to participate if it was NOT a part of Wikipedia.

[Alex]

I had a contributor ask to start up a wiki through AARoads, and I started setting up Mediawiki for him to try out. He then disappeared after just copying the main site index onto the Wiki main page and it has sat idle since then. If anyone is interested in messing with it, let me know.

[bulldog1979]

In my experience direct plagiarism of roadgeek websites is rare, though if you have performed research into some historical facet of the road or calculated mileposts or something like that, that information may be written into the article and a citation given to your work, but this is not objectionable because it is not plagiarism. If blatant plagiarism has occurred there are processes to deal with it and get the copied work removed.

What if both text and citations are lifted from a roadgeek website and lightly edited so that the verbiage is no longer a word-for-word match, and there is no attribution to the website.  What recourse does the site owner have?  Even if his words have not been stolen, he has been deprived of recognition and credit for his hard work excavating in archives to bring the information to light.

US copyright law does not recognize the "sweat of brow doctrine" anymore, only the creative expression. In a court case called Feist v. Rural, the Supreme Court ruled on this very issue. In that case, Feist Publications copied the contents of a phone directory published by Rural Telephone Services Co. In that case, the copying entries from Rural's phone books into Feist's directories was ruled not to be a copyright violation. From the Wikipedia article, "The fact that Rural spent considerable time and money collecting the data was irrelevant to copyright law, and Rural's copyright claim was dismissed."

http://en.wikipedia.org/wiki/Feist_v._Rural

Now, plagiarism is separate from copyright violations. If text is copied and only minimally reworded, it is can still be plagiarism even if the changes are sufficient to avoid copyright issues. That's a different issue, and Wikipedia's policies also prohibit plagiarism in addition to copyright violations. Taking an exact wording, swapping in a few synonymous words but maintaining the original sentence structures, is still plagiarism. There is an exception though for phrases that are so simplistic that the information can't be expressed any other way without changing the meaning.

The key difference here is that you can't copyright or plagiarize pure facts (like a name and a phone number in a phone book or the date a road opened) but you can copyright or plagiarize the expression of that fact. Some expressions will be so basic though to fail to gain protection, but the bar on that is low. In either case, anyone can edit a Wikipedia article to change or remove wording, or tell rschen7754, Scott5114 or myself to look into a specific situation.

[rschen7754]

The Wikipedia roads editors have considered forking from Wikipedia but there are enough problems with getting editors on Wikipedia, with the name brand and all the infrastructure that draws people to work on the project. All of the time that would need to be spent importing the pages from Wikipedia (as long as all of the templates, photos, and shield images that the pages depend on) wouldn't really be worth it. There's over 5000 articles alone...

You mean 10,000 in the US alone.

[NE2]

More of the same bullshit: http://en.wikipedia.org/wiki/Wikipedia:Articles_for_deletion/I-73/74_North%E2%80%93South_Corridor

[bulldog1979]

Just in case anyone is curious, but Wikipedia got some press today related to highways:

Reimink, Troy (November 20, 2011). "Paul B. Henry Freeway (M-6) lands on Wikipedia front page on 10th anniversary of first phase opening". The Grand Rapids Press.

[D-Dey65]

Who cares about the articles as long as they have pretty pictures?

Oh, you're funny.

In any case I've been struggling to prove a lot of what I wrote on Wikipedia. I've got into some pretty heated arguments with administrators over certain facts about roads that don't fit in with their standards, and citing sources for articles about roads on Long Island is quite difficult when you're stuck in Florida, and are surrounded by relatives who trash all your research!

[bulldog1979]

In any case I've been struggling to prove a lot of what I wrote on Wikipedia. I've got into some pretty heated arguments with administrators over certain facts about roads that don't fit in with their standards, and citing sources for articles about roads on Long Island is quite difficult when you're stuck in Florida, and are surrounded by relatives who trash all your research!

Well, it's not just personal standards... make sure you're complying with the website's policies. If you aren't, anything you add is subject to deletion, per policy.

[BlueNacho]

In the old days, I was a USRD editor, but I didn't have the time to edit pages so I had to stop. Good luck on the destubbing goal!

[empirestate]

My biggest critique of the Wikipedia roads articles is that they are at once too specific, yet too general. For example, I've lately been browsing through the US Highway system, trying to better comprehend it, since it escapes me in a way the Interstates never have. While WP does have comprehensive articles on all the US routes, I can't say they've helped me grasp how the routes fit into the system, or into the nation's infrastructure generally.

Picking, at random, the US 12 article, in the first paragraph we have two statements of the highway's termini along with an approximation of its length and the Interstates that supplanted it. It seems to me that this paragraph should instead mention what general corridor US 12 serves nationally, which major cities it connects, and which broad regions it serves in the states or groups of states it traverses. Perhaps there are unifying aspects to its route that make it a cohesive whole, something more than just the string of its waypoints (this is very satisfyingly true of US 11, in my opinion, as it follows the spine of Appalachia and aligns much of the nation's coal and steel belt, and similarly reflects its socio-cultural surroundings)...

We then follow with a series of excessively detailed routing descriptions by state, but these read largely as turn-by-turn instructions and don't give a sense of the overall scheme. At the least, these sections ought to be preceded by multi-state or regional assessments of the route's course.

The history section is, I think, most important to understanding what US routes are. Some of the articles have wonderful history treatises, but others, like the one in question, are kind of just lists of facts and dates.

I don't mean to criticize excessively, as I do appreciate the effort and contributions that go into these articles, but when I refer to WP on other topics I do it first for a general grasp of the topic. And, despite being a long-time roads scholar, sometimes I want to read about roads in the same way.

[Scott5114]

I think the issues you're having are attributable to a number of things. One of them is Wikipedia's sourcing requirements–it is difficult to look at a length of road and say what its "purpose" is; to do so we would have to have a citation from an independent article in a newspaper or something saying "US 12 was built to serve X purpose", we can't just try to come to our own conclusion and state that as fact. In some cases it's blatantly obvious, like a state highway spur to a small town that would otherwise be off the beaten path, but in that case it is generally preferred to state the facts and leave the rest for the reader to workout himself. Social geography is a territory that is hazardous to wade into because things like "the Steel Belt", "the Bible Belt", "Appalachia", "the Ozarks", "the Upper Midwest" et al tend to have ambiguous borders and it is very easy to run into a situation where people want to bicker back and forth about whether the road really serves the Steel Belt or whatever.

Another issue is sort of the way that work is organized on Wikipedia. The roads project is broken down into subprojects by state. Makes sense in a lot of cases, but for the comprehensive multi-state route articles (we call these the "main article" for the route) it sort of breaks down because there might be half a dozen state subprojects involved, some of which only exist on paper. Say for I-35: the TX, OK, KS, MO, IA, and MN projects are all partly responsible for keeping the I-35 article awesome. However, the KS and MO projects (and maybe the MN project) basically exist as dormant shells, with nobody actually actively working on those states' articles unless some charitable active editor from a bordering state takes a couple hours off from editing their preferred state to improve a couple articles. That means the main articles for the routes that traverse multiple states can become uneven in quality from state to state.

Editing outside of their familiar state is a major barrier to most people–for each state you have to learn what the DOT makes available in terms of historical maps, other historical data like changelogs, mileposts, etc., and what non-DOT resources are handy, like what newspapers are likely to have covered highway construction, whether their archives are free or not, how easily searchable they are, etc. Each editor carries a lot of this information in their mind for their particular state, and the prospect of having to "relearn" that information to become proficient in multiple states is unpleasant. There are some people that put in the effort, though. Also the prospect of editing in a state far from home that you have no connection to is deathly dull to many people and it shows in the quality of the articles produced, which also suffers from the fact that the editor has little first-hand knowledge about the conditions on the ground.

In addition to the main articles on a route, there are also "state-detail" articles (like "U.S. Route 52 in Illinois") that examine a segment of route in a particular state. These are often improved by the relevant state project quicker than the main articles because there is less route that needs to be covered and you only have to deal with one state's worth of resources for the article. However, the issue of certain states having no active editors to cover them remains apparent. Someone is supposed to go through at some point and merge the relevant information from the state-detail articles to the main article, forming a cohesive summary, but often this is forgotten.

Also, what usually makes or breaks a particular main article is simply whether it's been touched by a proficient editor or not. Well-done history sections have to be researched and written. The history sections that are just smatterings of facts tend to be the result of random passersby adding their notes to the article or an editor finding out a fact while researching something else and adding it to the article so they have the fact handy for when they fix up the article later. What we are aiming for eventually is something more resembling the articles on US 113, US 131, and US 491–these are the current gold standard of US route articles–do those fall more in line with what you'd like to see?

[mukade]

My biggest critique of the Wikipedia roads articles is that they are at once too specific, yet too general...

We then follow with a series of excessively detailed routing descriptions by state, but these read largely as turn-by-turn instructions and don't give a sense of the overall scheme. At the least, these sections ought to be preceded by multi-state or regional assessments of the route's course.

I echo the comments of empirestate. The excessive detail, in particular gets me. Very few are going to read the turn by turn instructions. A map is much more efficient and understandable way to convey that type of information, but also routes are tweaked frequently enough that these instructions easily get out of date.

Another issue I have is that there are errors that seem to be caused by the author not personally being familiar with the subject road. I have seen articles that seem to be based on little more than what Google Maps shows - and that is not necessarily correct. I guess this is a result of an effort to have an article for every route - even the most minor ones - in whatever system.

Another funny thing is that because articles must have citations, you sometimes see more recent changes called out even when they are minor. If a certain route was significantly truncated or altered in the 1960s, but was slightly moved in 2010, you tend to see the recent change called out because an online press release or article can easily be cited.

It all depends a lot on how knowledgeable and thorough the author really is, and that is the crux of the issue for me.

It is a tough call. I applaud the effort, but nothing will ever be perfect.

[empirestate]

What we are aiming for eventually is something more resembling the articles on US 113, US 131, and US 491–these are the current gold standard of US route articles–do those fall more in line with what you'd like to see?

Yes, those are definitely much better.

I see how the process and requirements of Wikipedia are going to limit what the articles can do, and ultimately, my own ideal of what they should be is probably beyond what would be allowed. I'd envision an article about a highway to be an overall assessment of the road's function in the abstract first, digested into a nice summary by an observant editor, and that smacks an awful lot of Original Research to me.