dmaze

Consecutive routes

2021-08-21T23:12:24Z

Pittsfield has US-7, MA-8, and MA-9 (and there are signs from US-20 that list all three). Routes 7 and 8 don't connect. But, if you're on MA-8, then you can go east to MA-9, and when you get to Northampton, you can turn on to MA-10. If you go north a very long way, staying on Route 10, then NH-10 intersects NH-11 in Newport, NH; go west and you'll reach NH-12 before you cross the Connecticut River; go back south into Massachusetts and MA-12 will reach MA-13 in Leominster.

So: what is the longest chain of consecutively-numbered routes you can connect? You can cross a state line, but only if the road keeps the same number (NY-7 turning into VT-9 breaks the chain). No specific need to start with Route 1. No subsidiary routes (MA-2A, "business" routes, &c.; 3-digit routes like I-290 or US-202 are legitimate but have to stay in sequence). Distance doesn't matter (starting in Key West, taking US-1 to US-2 in Maine, then to US-3 and US-4 in New Hampshire and then US-5 in Vermont is a legitimate start) (but if you start off this way, US-7 runs parallel to CT/MA/VT-8 and does not intersect anywhere).

comments

Farewell, Porter commuter rail signals

2021-03-07T15:28:06Z

If you stood on the commuter rail platform at Porter Square and looked towards Boston, there has always been a pair of signals, one for each track. They were replaced around 2015 as part of the general Fitchburg Line upgrade project, at the same time the line was double-tracked through Acton and Littleton: they moved east behind the car wash, and with a block signal near Union Square removed, became a distant signal for Swift Interlocking at the McGrath Highway overpass.

Those signals got removed this week as part of the T's Positive Train Control project.

There's very established technology to tell a train what the next signal will be by sending signals through the rails, and displaying the corresponding signal in the operator's cab. Of note, Amtrak's Northeast Corridor uses a variation on this system that displays both a signal and a speed limit, and so MBTA trains that operate on the Providence Line are already equipped with the ACSES hardware. In the wake of the 2008 Metrolink crash, Congress mandated that all passenger trains be equipped with a positive train control system that can stop a train before it passes a red signal, and in turn this has caused the T to install cab signaling everywhere on the commuter rail system.

If the train already has a display in the cab telling it what the next signal should be and how fast it should go, you don't also need signals next to the track displaying the same thing. So the wayside automatic signals are also disappearing across the MBTA commuter rail network. The Old Colony lines never had them; if you drive on the Pike next to the Worcester Line, you'll also notice the signals there have been removed.

Still, I have a fondness for seeing these signals as I go past. This is one of a couple of locations where I said "...so what does yellow-over-green mean?" and actually looked it up. As my...well, "home signal" is a technical term, but you get the idea...I'll miss seeing them.

comments

Derailment-pocalypse

2019-06-14T00:34:59Z

Are there any good sources that have current actual peak headways on the Red Line? I haven't verified it myself but from maps like this one it looks like the T just isn't publishing real-time vehicle data for the Red Line at all, which means nobody who's not willing to hang around a T station for an hour or so can find this out.

The good news is that the conductors on the Fitchburg line seem to have gotten the memo that any CharlieCard will get you to Porter, and the 5:55 Wachusett train had lots of open seats leaving North Station. But, the walk from South Station to North Station is just a little bit longer than I think it is, which is an issue if the 5:35 train leaves on time.

comments

Bike log: urban February edition

2019-02-17T23:05:11Z

Yesterday rode the Minuteman as far as Lexington; today explored the various Boscamberville roadway improvements.

In Boston proper, the cycle track around the North End is actually pretty awesome. But navigating through downtown Boston, especially the Scollay Square/Post Office Square area, is horrific. (I at least think I know how I'd get to work.) The bridge into Charlestown would be worth avoiding if there were any other way to go that direction.

The cycle track along Beacon Street in Somerville is somehow less impressive; maybe because the driveway crossings aren't as smooth and it's not totally clear what to do when it stops and starts and stops again.

"Sharrows" feel better than I give them credit for. They tend to communicate "sorry there's not bike lane for this block" to me, a little more than "you can take up this space in the lane which you legally could have done anyways".

comments

The T: the past month or so

2018-01-14T02:55:14Z

Incidents of either dead trains or track problems extracted from http://www.universalhub.com/mbta ...where it is impressive that the editor can still come up with dead train rhymes.

	Su	M	T	W	Th	F	Sa
Dec 2017	17	18 O R	19 G O R	20	21 O	22 O	23
	24	25	26	27 R	28 R R R R	29 G O O	30 G G O R
Jan 2018	31	1 G	2	3 G R	4	5 B G O R	6 O R
	7	8 G O R R	9 G	10 R R	11 G	12 O R	13

(Manually extracted. Didn't count "station closed due to flooding" incidents, of which there were multiple. And yes, on December 28, according to UHub, Red Line trains separately died at Alewife, Kendall, Porter, and Davis.)

comments

Library review: Cowboy, Sumo Rest

2017-07-13T14:40:43Z

The other end of the Erlang experiment is the libraries I was working with. My application was built on the Sumo Rest stack, built on top of the Erlang Cowboy HTTP server.

I found this finicky and prone to vague runtime errors. That's not important, though. The overall style of the system was pretty reasonable, and is probably a good way to write REST services in general.

( Read more... )

The layout implications of this are useful and generic: one source file per type, where each type file defines its storage, serialization, and documentation; and one source file per route, where each route file has its own URL path, complete machine-processable documentation and metadata, and understands the standard failure cases. You just need the right high-level REST library in your language of choice to support this.

comments

Programming language review: Erlang

2017-07-13T13:02:06Z

What do you get if you combine Haskell's basic functional style, Lisp's type system and object representation, a unique concurrency system, and a very good runtime?

( Read more... )

In short, Erlang looks like a good language, a little dated, but I'd rather have good static checking and a robust library stack than an excellent runtime with okay libraries, especially if I can avoid peeking "under the hood".

comments