Live data from Hacker News

Never Missing the Train Again

lilymara.xyz

131–140 of 147 posts

Re: Never Missing the Train Again

#131

Earlier quoted context omitted.

I think more of the issue might be the eink screens. As far as I can tell, there just aren't 5+ inch eink screens for cheap.

yes they are: https://www.waveshare.com/product/displays/e-paper.htm?___SI... A 4.37inch E-Paper in 3 colors is $24, problem is need you to program yourself (they have code sample in python, for raspberry pi), and you need a raspberry pi, case, cables, etc. Also, these cheap epaper displays are, of course, of lower quality (slower, lower resolution) than an kindle display.

They look kind of cool, and now I'm trying to come up with a project such that I can justify buying one.

Re: Never Missing the Train Again

#132

Earlier quoted context omitted.

If you live in a city that already has a OneBusAway server How do you find a list of places it's deployed. I tried Googling "onebusaway cities" which got me this page https://onebusaway.org/onebusaway-deployments/ but that doesn't list cities like Adelaide that you mention.

See the footnote in GP's comment.

This one?

[1] New York, Seattle, Washington, D.C., San Diego, Alexandria, Victoria, Adelaide, Buenos Aires, etc.

I was trying to figure out what "etc." meant.

Re: Never Missing the Train Again

#133
post #71

Maybe I'm not understanding the use case. I don't want to "take the train". I want to get from A to B. If the train is broken, the workers are on strike, the route is blocked between 1pm and 4pm, or it's outside of operating hours then I want an alternative. So for me, I actually want what Google Maps gives me (or tries to give me). I do not just want to know about trains. Maybe I want to get from the Ferry Building…

> So for me, I actually want what Google Maps gives me (or tries to give me). One thing I find is that Google Maps really, really assumes that walking is very very bad and no-one wants to do it. So for instance, if I'm going to a particular place I will get a 39 bus and then walk 15 minutes. If I ask Google how to get there, I'll get a route with two transfers; if the stars align it'll take about as long as one-bus p…

> Google Maps really, really assumes

Indeed, they assume all kinds of things - with no option to customize. For example, they assume that a 1 or 2 minute transfer will happen. Which you might want to override, but can't. You have your reasons to want to route through Caltrain instead of BART... and you can't. This is a characteristic of the brand really: Google people know what's good for you. I mean, them.

Re: Never Missing the Train Again

#134
Wow what a cool writeup. Two things stuck out for me (I'm only halfway so forgive me if you address these issues). I have two comments on processing the 511.org data. First, you generally want to use a streaming parser rather than one that allocates memory to the entire response. Second, you should filter out the most data first (in this case, dropping the stops you don't care about) and then filter the least data second (dropping the fields you don't care about). This idea is similar to how, in SQL, you want to order your WHERE clauses such that the most impactful comes first.

This may not be strictly required by your use-case at this scale - 27MB of data is not a lot, after all. And the filter ordering performance is probably trivial given its all in memory (I'd be curious to see a benchmark!) However, a) efficiency is always good, especially on a Pi, and b) if you make the code more efficient it makes it easier to scale later if you want to.

Also, regarding the BOM problem you had, wouldn't it be nice if all APIs had a "developer feedback" mode built into it? That is, you can send feedback to the ppl who own the API endpoint by...posting to the endpoint. In this case you could send "Please remove the BOM. k thanks."

Re: Never Missing the Train Again

#136

I'm kind of surprised that no one in here seems to have mentioned https://oeffi.schildbach.de/index.html so far. It does exactly what seems to be wanted here.

These are also worth a mention:

https://play.google.com/store/apps/details?id=nl.skywave.ovi... / https://apps.apple.com/nl/app/ovinfo/id1144468923

https://drgl.nl/

https://www.rijdendetreinen.nl/

https://spoorkaart.mwnn.nl/

Re: Never Missing the Train Again

#137
post #72

Earlier quoted context omitted.

Ehh... Cough, tough, bough, lough, slough, and rough called, they wanted to know when English phonology became logical and consistent, and why no one informed them about it?

Rough and cough have different sounds so... buses is the winner? Not seeing the connection.

Sorry, I thought it'd be obvious that all those -ough words have different pronunciations (for most people, anyway, I think) and that I'd be making my point clearly and lightheartedly. I was just saying that English spelling isn't always "guessable", or how you think it logically should be. [Even though it may well be guessable the majority of the time, as another responder points out, for some reason].

The fact you think "busses" is a preferable spelling to "buses" because it might help you pronounce "buses" differently to "fuses" is only relevant to you yourself. I would have thought this was tautological, myself.

In summary, we could avoid all these fusses with a bit of effort to adhere to accepted usage.

Re: Never Missing the Train Again

#138
post #59

Earlier quoted context omitted.

In this case, we really, really don’t want to continue using “busses” as a plural noun. Merriam-Webster is the authority. We’d rather fight our holy war over labor and center.

I sympathizzze he-ugely with what you're saying, as "busses" looks very strange to me, I admit.

"Buss" is an archaic word for a kiss, so "busses" looks like a verb straight outta Shakespeare to me!

Re: Never Missing the Train Again

#139
post #54

BART and MUNI both support the General Transit Feed Specification.[1] There's a standard way to obtain this data. [1] https://gtfs.org/documentation/overview/#gtfs-realtime

Do you know how reliable is the realtime info? I find that Google maps very often says “bus coming in 4 min” as im watching the bus drive away. Or “delayed by 5 min” and it’s early by 2, etc.

It depends on the specific transit system.

In my experience in NYC, the subway data is extremely accurate in terms of the minimum time until the next train. The subway virtually never arrives "early".

Buses seem to have a problem where their location transmission sometimes fails for a few minutes. The system always assumes the bus is still stuck at its last reported location rather than moving. That's why you get a bus arriving when the feed says it's 4 minutes away.

A good rule of thumb is that if you see the minutes away change, e.g. from 7 to 6 minutes, it's accurate. If it's not changing (e.g. just sits at 7), it might be because it's genuinely stuck in traffic, or because it's stopped transmitting. (Which explains the "delayed by" situation you describe.)

Re: Never Missing the Train Again

#140
post #71

Maybe I'm not understanding the use case. I don't want to "take the train". I want to get from A to B. If the train is broken, the workers are on strike, the route is blocked between 1pm and 4pm, or it's outside of operating hours then I want an alternative. So for me, I actually want what Google Maps gives me (or tries to give me). I do not just want to know about trains. Maybe I want to get from the Ferry Building…

For your daily commute, you really do just want to "take the train".

If the train is broken or the workers are on strike, that will be reflected in the unexpected absence of live trains listed.

I pull up Google Maps and plug in my destination when I'm running around the city. I don't pull it up when I'm leaving my home to go somewhere I've been 500 times before.

Post reply on HN