To which someone do you refer?

It seems to be an accident. I welcome any corrections you or anybody else more familiar than I am with the Jefferson City Subdivision can offer.

OK, I believe I have changed all bicycle=shoulder tags to cycleway=shoulder.

This was an old-fashioned way of tagging way back when (circa 2009 when the CASIL and Santa Cruz County GIS imports happened). I agree with you (now, 2015) that it is incorrect to put admin_level=4 on a State Park and so I have removed that tag.

Of course, Minh: thank you for your corrections and especially for documenting here the correct way to tag this so I know how to do this if/as I find such construction in the future.

OpenRailwayMap's Infrastructure and Maxspeed styles now render these changes accurately. I consider this resolved.

Not a problem, took me just a couple of minutes. Thanks for your cooperation and good attitude. Go OSM!

In changeset 30077144, I have deprecated the highspeed=yes tags from all NEC segments. However, previous changesets have set maxspeed= tags. As a net result, on segments where maxspeed>=160, highspeed=yes is "back" to being set. These segments include the great majority of the NEC, absent primarily around Newark, New York City and New London and between New Rochelle/New Haven.

It isn't too far a stretch to say NEC is "somewhere between orange and red." Subtle, huh, yeah, I know. As we best know how to tag.

Looks like you cracked open (disconnected) a (rather admittedly complex, I will concur) landuse=conservation multipolygon while editing around Camp Ben Lomond and Empire Grade a couple of weeks ago. As one of the original authors of this, I have fixed it.

Yes, as Nakaner (and ORM tagging) document, ORM truly IS unable to "make the link between" (render) the relation, UNLESS the track is so tagged.

AGAIN, (I repeat) AS IT IS DOCUMENTED, the "highspeed=yes" tag literally means: "Is this line a high-speed line?" Emphasis on the LINE to which that track segment belongs.

An analogous situation we seem to be touching upon (with our motorway example having a maxspeed=65 mph tag) would be if an Interstate highway had segments within it which were not quite up to Interstate highway standards. For example, perhaps a few km were highway=trunk instead of highway=motorway. In that case, AASHTO would not grant full Interstate status to that highway (or specifically exclude that segment, as is I-210 and CA-210 in southern California). But the difference is that Amtrak regarding conferring "high speed status" is not as strict as AASHTO is at conferring "Interstate status." In the latter case, the higher-class tag should not be applied, as the official designation has not been granted. But in the former case (Amtrak saying Acela on NEC is high-speed), the official designation HAS been granted. Ipso facto, the tracks can be (and should be) tagged with highspeed=yes, as each track segment meets the documented definition: "Is this line a high-speed line?"

The route (Acela Express, route=train) already is marked high speed, with the service=high_speed tag. This is precisely how ORM tagging instructions say it should be done. The tag is not applied (again, exactly as instructed) to the route=railway (NEC) relation. This is because ORM's tagging sections of Railway Route and Train Route do not document to put a high_speed tag on the former, but to one should put a high_speed tag on the latter. So, that has been true for some time.

I hear you when you say you object to each element of track being labelled highspeed=yes. Again, the world over already does this on several (dozens?) of high speed train rail infrastructure segments to indicate that these segments are part of a high-speed line. I am merely following their lead. Do you object to those as well?

As 1) this is already done the world over, 2) the author of the primary renderer for this tag says (effectively) "no clear consensus yet" and 3) Tagging_for_the_renderer wiki's Clarification section supports me, my determination to leave these tags in place seems OK.

Finally, a road (highway) with a maxspeed tag (let's say it is 65 mph, because of posted signs that say so) doesn't mean that cars will always travel over it at 65 mph. They might travel over it at 75 mph (and violate the Vehicle Code) or they might travel over it at 5 mph in heavy traffic (as was the point of my earlier effort to point this out). However, the maxspeed tag of 65 mph remains correct on that road, doesn't it? Yes, it does.

And there you have it from one of the authors (and a true rail expert from an OSM tagging perspective): "there has not been reached any consensus." What this says to me is that we have a bit of a tempest in a teapot here. Especially as other lines (in Europe, Asia) are tagged as I have tagged NEC (highspeed=yes along the length of the infrastructure), signifying that track segment is part of a high speed line.

To show my continuing good faith and hopefully to assuage the situation somewhat, I have applied a few maxspeed= tags on those sections of NEC track where its Wikipedia article (footnote 64) allows me to determine these. These are displayed using ORM's Maxspeed style (radio button). They begin to yield a more comprehensive OSM visualization of the changing speeds that Serge experienced while riding the Acela. These are early data, but they are all I am easily able to determine now, and I encourage more maxspeed= tagging where it is known.

To Paul's comment: Amtrak's "upgrades" are intended to make an already-exists high speed line into an EVEN HIGHER speed line.

To Frederick's comments: a "more correct" way to capture that certain segments of rail have a limiting speed is with a maxspeed tag. I welcome these, as then ORM's Maxspeed style would show what Serge both wants to see and experienced on the train in person.

I am in email contact (today, actually) with Michael Reichert (one of the authors of ORM) awaiting answers to several questions, among them whether tagging a route=train with highspeed=yes will render. If it will, I will happily change tagging to do that, removing them from the infrastructure. However, it may be that because of CSS limitations on relation rendering, it will not work. I may also experiment with this without waiting for an answer from Michael.

If this (tagging the route=train relation) does not work, I stand by what "Tagging_for_the_renderer" states: "If a specialist map renders a particular specialist tag...then using the tags the renderer understands is a perfectly reasonable thing to do."

Frederick, as I look at high speed rail in UK, it is as I suspect: the highspeed=yes tag is on the rail element, not the relation. However, there is a "service=highspeed" tag which may render in ORM (though I doubt it). I can experiment with this service=highspeed tag on the Acela route=train as well. But if it doesn't work, nor if putting highspeed=yes on the route=train relation works, then I really must do what the whole rest of the world does in OSM to tag high speed rail that renders in ORM: tag each element of the railway to which it belongs (NEC in this case for Acela) as highspeed=yes. Just as ORM documents, just as other highspeed lines in the world are tagged.

Serge, individual tracks are tagged this way because that is how the tag is documented: it means the LINE of which this rail segment is a member is capable of supporting high speed route=train service. So, it is correct. (I repeat myself here, not a good sign).

Are there individual track segments which do not always see a particular train (even an Acela), traveling at high speed? You say yes. However, that does NOT contradict that the tag is being used properly as it is documented, which it is: as a MEMBER of a LINE which SUPPORTS high speed route=train service. There are numerous examples of these in OSM. But only one in the USA: this one. And it is tagged as are other high-speed rail lines, because IT IS ONE.

If you wish to see "more accurate" tagging on this rail line, please enter maxspeed= values where you know them. I greatly encourage that task.

Addressing Serge's specific request to "correct (my) tagging on...which track segments are high speed and which aren't," I continue to assert that ALL of the track segments of NEC are high speed. Again, this particular tag (highspeed=yes) is a correct answer to the semantic of the tag: "Is this a highspeed line?" Amtrak says Acela is, Acela runs on NEC, ipso facto, NEC is a highspeed line, therefore tagging highspeed=yes on the segments of track infrastructure which make up the entire line is correct.

Once again, this is NOT the same thing as saying that a particular segment of track (e.g. through a town or on a tight curve) means that trains will travel over it at high speed. It DOES mean that trains which travel on the line it belongs to CAN travel at high speed on that line. This is similar to tagging an entire Interstate highway=motorway even though a driver could reasonably enter at one access point, travel at 5 MPH in stop-and-go traffic, get frustrated and exit at the next access point. That segment of (and the entire infrastructure of Interstate) is still highway=motorway, capable of high speed travel, high speed travel just doesn't always happen.

And again, it is the maxspeed tag (of which I have no knowledge on NEC, so I haven't tagged with maxspeed there) which indicates which rail segments allow particular speeds. That is likely a much better tag to consult when you find your Acela experience to involve both high-speed and low-speed travel. Not the highspeed=yes tag, which simply says that the LINE of which this rail segment is a member is capable of supporting high speed route=train service. In the case of NEC, that is true, hence the tags.

Thanks to your request that I re-read Tagging_for_the_renderer, I did: I find nothing there to contradict my correct tags. In fact, the last section (Clarification) actually supports me. It says "if a specialist map renders a particular specialist tag...then using the tags the renderer understands is a perfectly reasonable thing to do." I agree.

Serge, the discussion at hand is: Amtrak says Acela uses the Northeast Corridor. Amtrak says Acela is highspeed. I have put 2 and 2 together and come up with 4: the tracks of the NEC are highspeed, because they support highspeed service, exactly as the tag is documented.

There are different tags which specify the maximum speed for each individual track segment, for example, where a speed limit sign actively changes the maxspeed. As I do not have knowledge of these, I have not added those tags.

I do have knowledge of the NEC being a highspeed line: Amtrak tells me so. Our minds allow deduction, the inference of a particular instance by reference. There are examples of this deduction all over OSM.

Steve

OK, Caltrain publishes web pages and printed schedules that say Gilroy is a station on the Baby Bullet line.

If you would like one, here it is: http://www.caltrain.com/schedules/weekdaytimetable.html

If you would like me to endeavor to be quite exact with web links and exact publication names in source tags (well, changeset source comments, not source=* tags -- although I endeavor to put web links in source=* tags, and have for years), I can do so. However, a messy URL doesn't easily paste into my JOSM upload dialog, so instead I type in what is a reasonable expectation.

If "Caltrain" doesn't make sense to you as the name of a public train route sufficient for you to say "Oh, Steve simply completed the last station in the route by consulting the operators of the route itself -- he must have consulted one of their schedules or a web site containing one..." then I guess we have different senses of what it means to come to a logical conclusion.

I will endeavor to be more exact in my sources in future, though I believe I've been pretty good about them (Personal knowledge, Bing imagery, a specific book, a survey I made myself, a GPS/GPX track...) for most of the life of OSM.

If replying to you via email doesn't make sense for this sort of "changeset comments" (i.e. you think I should be typing this into the changeset comment "process"), please let me know that and I'll cut-and-paste.

Regards,
Steve

Hello Serge:

The "infrastructure on rail" tag of "highspeed=yes" is documented as widely used on the OpenRailwayMap (ORM) wiki here: osm.wiki/OpenRailwayMap/Tagging#Tracks It says "Is this line a high-speed line (with permissible speeds greater than 200 kph)?" The boolean allowed values are yes or no. 200 kph translates to about 122 mph.

Amtrak says that the Acela Express line offers "Superior Comfort, Upscale Amenities, Polished Professional Service, at Speeds up to 150 mph" here: http://www.amtrak.com/acela-express-train . Clearly, 150 > 122. Also, at http://www.amtrak.com/train-schedules-timetables Amtrak lists its first route as "Acela Express - see Northeast Corridor timetables." So, taken together, Amtrak provides us with these data.

Please note that it is the Acela Express LINE (route=train) which offers these speeds, so effectively the entire infrastructure (route=railway) meets the "yes" definition for ORM's highspeed tag. ("Is this LINE...high-speed...with PERMISSIBLE speeds..."). To me, this means that this tag can and should be applied to the underlying rail infrastructure.

Again, this infrastructure is the Northeast Corridor, whose tracks ALLOW trains at speeds > 122 mph. That is why the tag is correct. Please note (by ORM wiki section) that the highspeed=yes tag does not go on route=train relations (or elements) but rather track/infrastructure elements. I believe I have done this correctly by asserting the tag on all track elements which make up the Northeast Corridor route=railway relation: the entire "subdivision" (route=railway) supports highspeed trains, so highspeed=yes on its elements is correct.

ORM's default display is the "Infrastructure" radio button, but there are also its "Maxspeed" and "Signalling" map styles. If exhaustive tagging were correctly applied to each track element of the Northeast Corridor (including maxspeed=245, correctly in kph, about 150 mph), the train would rocket through curves and small towns in ways it actually doesn't. (As you personally experienced). And as it is early days in USA rail tagging (and VERY early tagging in USA speedlimit rail tagging), we haven't applied this exhaustive tagging. But it would be correct to ALSO apply to each and every track segment where it is known (and on-the-ground verifiable, often with a speed limit sign that can be seen by train passengers) a speedlimit= tag.

Correctly completed, this would accurately result in ORM displaying (with Infrastructure style) a solid red line on Northeast Corridor, properly reflecting highspeed=yes tags. It would also accurately result in ORM's Maxspeeds style displaying a "rainbow" of colors representing different speeds: some straight and no station segments would have speedlimit=245, others (near stations, around curves) will be lower. Try this, but not in the USA (there are just a few spots of color representing Maxspeeds displayed on ORM in USA -- early knowledge and a tiny start of data entry by novice rail folks still finding our legs on how to do this). I think a Maxspeeds view of a high-speed rail line in Germany or China will give you a better sense (through color representing speedlimit) what you experienced while riding the Acela Express. But the whole line being red in Infrastructure view because the entire infrastructure supports highspeed=yes? Yup, that's how it's done.

Relax, Serge. This is the only line in North America where the highspeed=yes tag is appropriate. I nor anybody else are going to assert it anywhere else, unless more highspeed trains are built. Please do use ORM's Infrastructure view (and documentation) to see (and understand) that this is the correct way to do it for highspeed trains in Europe and Asia. We have exactly one highspeed train in the USA -- Acela Express -- and I believe now is as good a time as any that we allow ORM to properly display this by tagging it as such.

Rail in the USA has a long, long way to go. But if we correctly tag each and every brick in the wall one by one by one, eventually, we'll really have something. THAT is an important essence of OSM. I hope you feel the same way about it as I do, as Rome wasn't built in a day.

Thanks for your comments, and I welcome continued discussion on the topic if you wish,
Steve

I have "knowledge" of these segments of freeway, also. They are not highway=trunk, they are highway=motorway (fully controlled-access "freeway" in California). They are even all named "Freeway." Please consider changing back to highway=motorway.