CT 2 Western End

[vdeane]

Shouldn't the western end of CT 2 be at Columbus Boulevard rather than I-91?  There don't appear to be any direct ramps between I-91 and CT 2.

[yakra]

Agreed.

Although, ramps existed in decades past, with their movements replaced by the Charter Oak Bridge.

On the one hand, Keeping a closed *I-91 point there retains a graph connection.
OTOH, that would mean 15 people losing their 100% clinches & having to re-edit their .lists.

[Jim]

In this case, I think we should lose the graph connection.  Unless traffic can go directly from I-91 to CT 2 or the other way around, and it doesn't appear that that's the case.

[rickmastfan67]

I have a similar case with SH-592 in Ontario.  There used to be a direct connection there before ON-11 was turned into a full freeway there.  However, SH-592 still officially ends at that overpass, thus, I have the graph connection between the two. https://goo.gl/maps/YSKvhuakQp9jRhzA7

There don't appear to be any direct ramps between I-91 and CT 2.

However, I clearly see 2 ramps still.  Exit #31 SB (not signed for CT-2 WB, but does connect before Columbus Blvd) and an onramp to SB I-91 from CT-2 EB.

I think those 2 ramps alone, justify keeping a graph connection with I-91, even if you want to have it end @ Columbus Blvd.

[rickmastfan67]

OT (sorta): Here's some other fixes I noticed that should be addressed with CT-2.

4A: Shouldn't there be a graph connection here with US-5/CT-15?  There's a clear direct loop ramp from NB US-5/CT-15 there.  I think that's more than enough justification for the graph connection and having it centered on the overpass for all 3 routes.  Maybe relabel it as well, since no off-ramps here?
5A: Hmmm, with a WB offramp there to Main Street (the other part of the OWR setup there), I think it's should be centered between High Street & Main Street there instead of just being centered on High Street.
15: This is clearly off-centered, and should be centered on the overpass there for South Main Street.

** The following 3 also effect CT-32
26: Should really be 27 (EB/WB), as there is no Exit #26.
27 (current): Turn this into the new 27A, as it's not in-use, and neither is the current 27A (listed below).
27A: Should be WewSt, since no offramps there.  Just a solo EB on-ramp.

[yakra]

I have a similar case with SH-592 in Ontario.  There used to be a direct connection there before ON-11 was turned into a full freeway there.  However, SH-592 still officially ends at that overpass, thus, I have the graph connection between the two. https://goo.gl/maps/YSKvhuakQp9jRhzA7

Then shouldn't it be SunPassRd (Are we in Florida?), or is Sunset Pass considered to be concurrent S of there?

There don't appear to be any direct ramps between I-91 and CT 2.

However, I clearly see 2 ramps still.  Exit #31 SB (not signed for CT-2 WB, but does connect before Columbus Blvd) and an onramp to SB I-91 from CT-2 EB.

I think those 2 ramps alone, justify keeping a graph connection with I-91, even if you want to have it end @ Columbus Blvd.

So it becomes a question of, does CT2 exist W of I-91 for its ramps to connect to?
In the shapefiles I've been using, the geometry is too rubbish to say for sure, but ConnDOT's Hartford map makes it look like Columbus.
With the geometry of the way those ramps connect, it really walks a tightrope on 1PPI. Which it seems, 1PPI was probably why it was done the way it was in 2009 or whenever.
My current thinking is to extend all the way to Columbus, which would mean having a not-closed connection @ I-91. So, we'd all lose those 100% clinches.


Blue = local edits, still uncommitted.

OT (sorta): Here's some other fixes I noticed that should be addressed with CT-2.

4A: Shouldn't there be a graph connection here with US-5/CT-15?  There's a clear direct loop ramp from NB US-5/CT-15 there.  I think that's more than enough justification for the graph connection and having it centered on the overpass for all 3 routes.  Maybe relabel it as well, since no off-ramps here?

Coords: going no-build.
Label: Changed to US5/15. Doesn't quite meet the Two separate interchanges numbered 4 criterion. Other guidelines seem to require it actually be signed 4A, so yeah, my reading off the manual suggests named not numbered for this one, as its own entrance-only interchange.

5A: Hmmm, with a WB offramp there to Main Street (the other part of the OWR setup there), I think it's should be centered between High Street & Main Street there instead of just being centered on High Street.

Nah. 3 out of 4 ramps intersect here. Not worth changing.

*5B was reported as permanently closing last June-ish, but GMSV from January shows it still open. I've asked what's going on here on AARoads before editing this line and taking the * back out.

15: This is clearly off-centered, and should be centered on the overpass there for South Main Street.

Repositioned.

** The following 3 also effect CT-32
26: Should really be 27 (EB/WB), as there is no Exit #26.
27 (current): Turn this into the new 27A, as it's not in-use, and neither is the current 27A (listed below).
27A: Should be WewSt, since no offramps there.  Just a solo EB on-ramp.

WawSt, but done.
However, (new) 27A (Yantic/Otrobando) could arguably be removed per double half interchanges < 0.5 mi apart. Who knows how it'll all shake out after mileage-based renumbering though. I guess keeping it in & changing the label is closer to no-build for now?

[mapcat]

So it becomes a question of, does CT2 exist W of I-91 for its ramps to connect to?
In the shapefiles I've been using, the geometry is too rubbish to say for sure, but ConnDOT's Hartford map makes it look like Columbus.
With the geometry of the way those ramps connect, it really walks a tightrope on 1PPI. Which it seems, 1PPI was probably why it was done the way it was in 2009 or whenever.
My current thinking is to extend all the way to Columbus, which would mean having a not-closed connection @ I-91. So, we'd all lose those 100% clinches.

Would anyone really lose that clinch, though? The way the ramps are configured now, it seems that anyone who drove CT 2 to/from the I-91 point must have actually entered or exited at Columbus Blvd, not I-91. I'm reading your earlier comment to mean that any direct connection from I-91 to CT 2 east of the interchange must have been lost long ago, so the I-91 point as it currently exists is essentially a fudge of Columbus anyway.

[yakra]

I'm reading your earlier comment to mean that any direct connection from I-91 to CT 2 east of the interchange must have been lost long ago,

Between 1992 and 2004 -- 90s sometime?

so the I-91 point as it currently exists is essentially a fudge of Columbus anyway.

Yes.

The way the ramps are configured now, it seems that anyone who drove CT 2 to/from the I-91 point must have actually entered or exited at Columbus Blvd, not I-91.

Anyone driving it relatively recently, yes. At any time CHM/TM has existed.

Would anyone really lose that clinch, though?

They'd have driven the whole road, but it'd disappear from their stats until their .lists are updated if the label at the end (I-91) is no longer at the end.

[Duke87]

Would anyone really lose that clinch, though?

They'd have driven the whole road, but it'd disappear from their stats until their .lists are updated if the label at the end (I-91) is no longer at the end.

This, combined with the fact that a direct connection has not existed in over 20 years, makes me inclined to think that having separate points for I-91 and Columbus Blvd isn't worth the mess it'd make.

Whether the end is moved to Columbus or left at 91 I have no real preference on. Don't think the graph connection matters since the two roads, well, don't connect.

[yakra]

the two roads, well, don't connect.

I accept rickmastfan67's argument.
Lack of [2] signage from I-91 notwithstanding, it's the physical configuration.

[froggie]

Regardless of what's done with 2's western end, the current 91/2 point should be recentered based on both OSM and ESRI imagery.

[rickmastfan67]

Blue = local edits, still uncommitted.

So yakra, are you ever going to submit those changes?