Share good ideas and conversation.   Login, Join Us, or Take a Tour!
veen's comments
veen  ·  2 days ago  ·  link  ·    ·  parent  ·  post: Pubski: February 14, 2018

You mention width, which makes me realize something: there's no reason for a bus-lane AV service to stop everywhere. This means that cars should be able to pass at stops, which is possible when the road is as wide as bus lanes are. Bus lanes here are around 6.5-7.5m, and car lanes can be as slim as 2.2m, so bidirectional bus lanes now could be three-lane AV lanes in the future. Overhead clearance is a good point, although I think clearances are pretty well-standardized at 4m.

kleinbl00  ·  2 days ago  ·  link  ·  

This is not my area of expertise, but if I were trying to serve access to a new neighborhood, I'd much rather cut in a 2.2m road than a 7.5m one.

veen  ·  2 days ago  ·  link  ·    ·  parent  ·  post: Pubski: February 14, 2018

    The real map of the world is just too complex and changing for anything other than a true AI to be able to operate within.

    What i see is that all AVs will be interconnected, and constantly communicating with each other.

Funny, I think of it exactly the other way. Mapping the real world is within the realm of possibilities - I mean this article is 5 years old by now. Besides, sensory input will always trump map knowledge. I talked to someone from TomTom a while ago. IIRC, when a dozen of their users drive over a new road they'll push that update to other users.

Permanently connected and highly reliable wireless connections that risk taking depends on? I'll believe it when I see it. When you have such a train of V2V connected cars, it only takes one malfunctioning / package-not-arriving car to screw it up for everyone behind it. 4G LTE (or even 5G) might be fast, but whenever I am at a busy train station it chokes the fuck out. My Bluetooth gets choppy when I move my head too quickly, and that's at a distance of less than a foot from my phone. Wireless tech is worse the denser your urban area is, while for AV's the opposite needs to be true. It exists on production lines because those are static, isolated environments, while (urban) roads are much more dynamic and prone to errors and interference.

    Segregating AV traffic from human-driven vehicles is key, though. So reconsidering road widths, composition, merging, etc, you wind up with something much more like bicycle paths, than the streets and highways we use today.

I think so too, at least for the near future, unless Google seriously gets their shit together. There's also a handful of public road tests, which could prove hopeful, but they are quite far removed from the ideal high-frequency, high-capacity bus replacement that I'm looking for.

veen  ·  2 days ago  ·  link  ·    ·  parent  ·  post: Pubski: February 14, 2018

I don't know yet! I have been told that we (the Dutch) are world-leading with the technology, so a venture that's a subsidiary of my work wants to export that knowledge and want me to ride along. Hopefully I can get a neat trip out of it. East coast 'ski meetup anyone?

kantos  ·  2 days ago  ·  link  ·  

Interesting :o the man I shadowed under was from Germany, and Waterfurnace's parent company was Scandinavian iirc. Y'all European's got it down pat, I'm sure. Would be sweet to get a ride across the pond, too!

Not actually sure who from the 'ski is up there.

veen  ·  3 days ago  ·  link  ·    ·  parent  ·  post: In honor of Valentine's Day, tell everybody your tale of heartache and woe

Got the word that I got the apartment today. Which means I’ll have my place to settle, to build a new social circle. And I feel like I’m the opposite of collapsed right now.

    Just 'cuz it didn't hit you this time don't mean it'll never hit ya.

Can’t fuckin wait. :)

veen  ·  4 days ago  ·  link  ·    ·  parent  ·  post: Pubski: February 14, 2018

Well, for one, the idea that the leftmost lane is reserved for speciality traffic is something that you guys have in the US but that hasn't permeated to a lot of other countries. Highway bus lanes are generally on the right over here. In almost every country the left lane is a passing lane, so it doesn't hamper your idea, just something to note.

What people do often forget is that AVs are heavily sight-based, and will not risk something they can't see. So while I can see the first part of the ramp being shorter, any yaw or negative pitch change will have to be taken significantly slower than we would take them. The benefit of not having to cross all lanes is obviously a great one, but the fact that many urban highways don't have a significant median might make it really difficult to construct.

My idea was not at all focused on highways, really. Here's my elevator pitch now - I invite anyone to gimme some feedback because it's a relatively early idea. (kleinbl00?) The context is European city centers.

    Autonomous vehicles are coming, but their biggest challenge will be to conquer the densest urban areas. Precisely the places that most people will want to go, are also the most difficult to traverse on public roads. Some estimations put dense urban autonomous vehicles at at least 40, 50 years into the future. Infrastructure changes, most notably dedicated AV lanes, will likely be a necessary step in the transition to autonomous vehicles.

    Autonomous vehicles present a threat to public transport, so it is wise to design autonomous infrastructure in such a way that it also strengthens the public transport network instead of only competing with it. At the same time, due to the high uncertainty of the precise development path of AVs, it is important that any infrastructural investments are already useful now, instead of waiting for the revolution to happen.

    Thus: the idea to build bus lanes or expand the bus lane network in cities so that they connect the most important locations with highways. In the current PT paradigm, this would allow regional buses to easily drive on into cities and compete with cars more easily. In that sense, it's a 'no regret' investment. In a highway-AV-only scenario, the highway location could be an important transfer hub for the last miles into the city. In an AV-on-separated-lanes scenario, the bus lanes provide the AVs with the separated environment they need to maintain high speeds in dense urban areas. And because they enhance already existant high-quality PT with intermodal transfer hubs, they can significantly strengthen the PT network.

You know why I have an iPad? This is why:

kleinbl00  ·  4 days ago  ·  link  ·  

If the goal is prepping the layout for autonomous vehicles, designing for buses forces you into some compromises you wouldn't ordinarily make. Your roads have to be wider, for example. Your overhead clearance is higher. And you have to select routes that maximize foot traffic.

Vanpool? You could build for that now and assume it's future-proof. After all, any AV fleet isn't going to seat four passengers. It's going to seat ten. There are lanes in Vancouver BC that are limited to 8 passengers or more.

I would say expand the bus network such that the definition of "bus" includes lower-capacity vehicles but doesn't include carpool. It'd be an easy pilot project to roll out.

veen  ·  2 days ago  ·  link  ·  

You mention width, which makes me realize something: there's no reason for a bus-lane AV service to stop everywhere. This means that cars should be able to pass at stops, which is possible when the road is as wide as bus lanes are. Bus lanes here are around 6.5-7.5m, and car lanes can be as slim as 2.2m, so bidirectional bus lanes now could be three-lane AV lanes in the future. Overhead clearance is a good point, although I think clearances are pretty well-standardized at 4m.

kleinbl00  ·  2 days ago  ·  link  ·  

This is not my area of expertise, but if I were trying to serve access to a new neighborhood, I'd much rather cut in a 2.2m road than a 7.5m one.

keifermiller  ·  2 days ago  ·  link  ·  

    Highway bus lanes are generally on the right over here.

Here in KC, the bus lanes are actually in the marginal "shoulder" area to the right of the actual road.

I know High Occupancy Vehicle lanes are to the left in the Dallas Fortworth area, though.

We're improvising all of this I think.

goobster  ·  3 days ago  ·  link  ·  

Ah. I see we have different base assumptions about what an autonomous vehicle is going to be.

For me, the self-driving, smart-as-a-chauffeur, version of the AV is really a pipe dream. The real map of the world is just too complex and changing for anything other than a true AI to be able to operate within.

What i see is that all AVs will be interconnected, and constantly communicating with each other. So they will be able to "see" several blocks, even miles, ahead, and adapt their operation for the conditions. This means following distances between vehicles of inches, rather than car lengths. Road widths with tolerances measured in less than a foot.

Visualize independent train cars that are virtually linked together, rather than physically.

This seems far more likely of a future for AVs. This already exists on production lines today. Scaling it up, and providing safe operating zones for these transport pods, are really the only limitations.

Segregating AV traffic from human-driven vehicles is key, though. So reconsidering road widths, composition, merging, etc, you wind up with something much more like bicycle paths, than the streets and highways we use today.

That's my pipe dream. (And the AVs can operate someone autonomously - but slowly - in the suburbs, until they pool together and move as a unit to the AV-only road.)

veen  ·  2 days ago  ·  link  ·  

    The real map of the world is just too complex and changing for anything other than a true AI to be able to operate within.

    What i see is that all AVs will be interconnected, and constantly communicating with each other.

Funny, I think of it exactly the other way. Mapping the real world is within the realm of possibilities - I mean this article is 5 years old by now. Besides, sensory input will always trump map knowledge. I talked to someone from TomTom a while ago. IIRC, when a dozen of their users drive over a new road they'll push that update to other users.

Permanently connected and highly reliable wireless connections that risk taking depends on? I'll believe it when I see it. When you have such a train of V2V connected cars, it only takes one malfunctioning / package-not-arriving car to screw it up for everyone behind it. 4G LTE (or even 5G) might be fast, but whenever I am at a busy train station it chokes the fuck out. My Bluetooth gets choppy when I move my head too quickly, and that's at a distance of less than a foot from my phone. Wireless tech is worse the denser your urban area is, while for AV's the opposite needs to be true. It exists on production lines because those are static, isolated environments, while (urban) roads are much more dynamic and prone to errors and interference.

    Segregating AV traffic from human-driven vehicles is key, though. So reconsidering road widths, composition, merging, etc, you wind up with something much more like bicycle paths, than the streets and highways we use today.

I think so too, at least for the near future, unless Google seriously gets their shit together. There's also a handful of public road tests, which could prove hopeful, but they are quite far removed from the ideal high-frequency, high-capacity bus replacement that I'm looking for.

veen  ·  4 days ago  ·  link  ·    ·  parent  ·  post: Pubski: February 14, 2018

eventually, yeah... only to go back home a few hours later.

It made national news though! Unlucky me. ;)

veen  ·  5 days ago  ·  link  ·    ·  parent  ·  post: Pubski: February 14, 2018

My train ripped a broken piece of overhead line. We got stuck without power or safe evacuation method, so my commute, which I'm still in, enters it's fifth hour soon. Eep.

Work is going super great. It's as close as I can get to working in a start-up: the freedom to make great things happen but without all of the failings and insecurity that that usually brings. One of my ideas is to help municipalities prepare for autonomous vehicles by building dedicated bus lanes that connect to highways. Pitched it to a few coworkers and we're now going to brainstorm about an MVP. There is also a large chance I'll get to work on a project for mapping the potential for geothermal heating NY, Baltimore and Washington DC, so there's that. Cool stuph, just getting started. :)

nowaypablo  ·  4 days ago  ·  link  ·  

So uh, you ever make it out of the train?

veen  ·  4 days ago  ·  link  ·  

eventually, yeah... only to go back home a few hours later.

It made national news though! Unlucky me. ;)

kantos  ·  4 days ago  ·  link  ·  

West New York by any chance? Too early to disclose with whom?

My internship over the summer there working with Ground Source Heat Pumps was insane. The guys distributing Waterfurnace's gshp's on the west side are hungry and smart fellas.

veen  ·  2 days ago  ·  link  ·  

I don't know yet! I have been told that we (the Dutch) are world-leading with the technology, so a venture that's a subsidiary of my work wants to export that knowledge and want me to ride along. Hopefully I can get a neat trip out of it. East coast 'ski meetup anyone?

kantos  ·  2 days ago  ·  link  ·  

Interesting :o the man I shadowed under was from Germany, and Waterfurnace's parent company was Scandinavian iirc. Y'all European's got it down pat, I'm sure. Would be sweet to get a ride across the pond, too!

Not actually sure who from the 'ski is up there.

kantos  ·  4 days ago  ·  link  ·  
This comment has been deleted.
goobster  ·  4 days ago  ·  link  ·  

    ...help municipalities prepare for autonomous vehicles by building dedicated bus lanes that connect to highways...

LEFT LANE ON/OFFRAMPS for autonomous vehicles!

I've been thinking about this a lot, and the left lane (in countries that drive on the right) is largely reserved for "specialty" traffic, like HOV, dedicated bus lanes, or even just the passing lane.

Left lane exits for vehicles that are "smarter" than the average driver, would allow shorter on/off ramps (acceleration/deceleration lanes), easier merging into the normal flow of traffic (because human-driven cars are generally turning right), and eliminate the need to negotiate lane changes with humans (moving right across multiple lanes to a human-driver exit).

Am I wrong about this?

veen  ·  4 days ago  ·  link  ·  

Well, for one, the idea that the leftmost lane is reserved for speciality traffic is something that you guys have in the US but that hasn't permeated to a lot of other countries. Highway bus lanes are generally on the right over here. In almost every country the left lane is a passing lane, so it doesn't hamper your idea, just something to note.

What people do often forget is that AVs are heavily sight-based, and will not risk something they can't see. So while I can see the first part of the ramp being shorter, any yaw or negative pitch change will have to be taken significantly slower than we would take them. The benefit of not having to cross all lanes is obviously a great one, but the fact that many urban highways don't have a significant median might make it really difficult to construct.

My idea was not at all focused on highways, really. Here's my elevator pitch now - I invite anyone to gimme some feedback because it's a relatively early idea. (kleinbl00?) The context is European city centers.

    Autonomous vehicles are coming, but their biggest challenge will be to conquer the densest urban areas. Precisely the places that most people will want to go, are also the most difficult to traverse on public roads. Some estimations put dense urban autonomous vehicles at at least 40, 50 years into the future. Infrastructure changes, most notably dedicated AV lanes, will likely be a necessary step in the transition to autonomous vehicles.

    Autonomous vehicles present a threat to public transport, so it is wise to design autonomous infrastructure in such a way that it also strengthens the public transport network instead of only competing with it. At the same time, due to the high uncertainty of the precise development path of AVs, it is important that any infrastructural investments are already useful now, instead of waiting for the revolution to happen.

    Thus: the idea to build bus lanes or expand the bus lane network in cities so that they connect the most important locations with highways. In the current PT paradigm, this would allow regional buses to easily drive on into cities and compete with cars more easily. In that sense, it's a 'no regret' investment. In a highway-AV-only scenario, the highway location could be an important transfer hub for the last miles into the city. In an AV-on-separated-lanes scenario, the bus lanes provide the AVs with the separated environment they need to maintain high speeds in dense urban areas. And because they enhance already existant high-quality PT with intermodal transfer hubs, they can significantly strengthen the PT network.

You know why I have an iPad? This is why:

kleinbl00  ·  4 days ago  ·  link  ·  

If the goal is prepping the layout for autonomous vehicles, designing for buses forces you into some compromises you wouldn't ordinarily make. Your roads have to be wider, for example. Your overhead clearance is higher. And you have to select routes that maximize foot traffic.

Vanpool? You could build for that now and assume it's future-proof. After all, any AV fleet isn't going to seat four passengers. It's going to seat ten. There are lanes in Vancouver BC that are limited to 8 passengers or more.

I would say expand the bus network such that the definition of "bus" includes lower-capacity vehicles but doesn't include carpool. It'd be an easy pilot project to roll out.

veen  ·  2 days ago  ·  link  ·  

You mention width, which makes me realize something: there's no reason for a bus-lane AV service to stop everywhere. This means that cars should be able to pass at stops, which is possible when the road is as wide as bus lanes are. Bus lanes here are around 6.5-7.5m, and car lanes can be as slim as 2.2m, so bidirectional bus lanes now could be three-lane AV lanes in the future. Overhead clearance is a good point, although I think clearances are pretty well-standardized at 4m.

kleinbl00  ·  2 days ago  ·  link  ·  

This is not my area of expertise, but if I were trying to serve access to a new neighborhood, I'd much rather cut in a 2.2m road than a 7.5m one.

keifermiller  ·  2 days ago  ·  link  ·  

    Highway bus lanes are generally on the right over here.

Here in KC, the bus lanes are actually in the marginal "shoulder" area to the right of the actual road.

I know High Occupancy Vehicle lanes are to the left in the Dallas Fortworth area, though.

We're improvising all of this I think.

goobster  ·  3 days ago  ·  link  ·  

Ah. I see we have different base assumptions about what an autonomous vehicle is going to be.

For me, the self-driving, smart-as-a-chauffeur, version of the AV is really a pipe dream. The real map of the world is just too complex and changing for anything other than a true AI to be able to operate within.

What i see is that all AVs will be interconnected, and constantly communicating with each other. So they will be able to "see" several blocks, even miles, ahead, and adapt their operation for the conditions. This means following distances between vehicles of inches, rather than car lengths. Road widths with tolerances measured in less than a foot.

Visualize independent train cars that are virtually linked together, rather than physically.

This seems far more likely of a future for AVs. This already exists on production lines today. Scaling it up, and providing safe operating zones for these transport pods, are really the only limitations.

Segregating AV traffic from human-driven vehicles is key, though. So reconsidering road widths, composition, merging, etc, you wind up with something much more like bicycle paths, than the streets and highways we use today.

That's my pipe dream. (And the AVs can operate someone autonomously - but slowly - in the suburbs, until they pool together and move as a unit to the AV-only road.)

veen  ·  2 days ago  ·  link  ·  

    The real map of the world is just too complex and changing for anything other than a true AI to be able to operate within.

    What i see is that all AVs will be interconnected, and constantly communicating with each other.

Funny, I think of it exactly the other way. Mapping the real world is within the realm of possibilities - I mean this article is 5 years old by now. Besides, sensory input will always trump map knowledge. I talked to someone from TomTom a while ago. IIRC, when a dozen of their users drive over a new road they'll push that update to other users.

Permanently connected and highly reliable wireless connections that risk taking depends on? I'll believe it when I see it. When you have such a train of V2V connected cars, it only takes one malfunctioning / package-not-arriving car to screw it up for everyone behind it. 4G LTE (or even 5G) might be fast, but whenever I am at a busy train station it chokes the fuck out. My Bluetooth gets choppy when I move my head too quickly, and that's at a distance of less than a foot from my phone. Wireless tech is worse the denser your urban area is, while for AV's the opposite needs to be true. It exists on production lines because those are static, isolated environments, while (urban) roads are much more dynamic and prone to errors and interference.

    Segregating AV traffic from human-driven vehicles is key, though. So reconsidering road widths, composition, merging, etc, you wind up with something much more like bicycle paths, than the streets and highways we use today.

I think so too, at least for the near future, unless Google seriously gets their shit together. There's also a handful of public road tests, which could prove hopeful, but they are quite far removed from the ideal high-frequency, high-capacity bus replacement that I'm looking for.

veen  ·  6 days ago  ·  link  ·    ·  parent  ·  post: What are your favourite podcasts?

Related:

elizabeth  ·  6 days ago  ·  link  ·  
This comment has been deleted.
elizabeth  ·  6 days ago  ·  link  ·  

Ah damn, I missed that thread. Didn't meat to create a duplicate so soon.

veen  ·  6 days ago  ·  link  ·    ·  parent  ·  post: Why Millennials Are Making Memes About Wanting to Die

I do think it can be argued that younger generations tend to like absurdist humor more than before. Absurdism is subversive, so it requires you to understand what's normal, which is easier in our age of easily accessible knowledge online.

kleinbl00  ·  6 days ago  ·  link  ·  

I would argue that we'll never be able to make a comparison because distribution is effortless now while previous generations had to find a printing press.

https://en.wikipedia.org/wiki/Andre_the_Giant_Has_a_Posse

That's just 30 years ago.

veen  ·  6 days ago  ·  link  ·    ·  parent  ·  post: 265th Weekly "Share Some Music You've Been Into Lately" Thread

Been enjoying a few Drake songs, Logic's The Incredible True Story, and Kanye finally somewhat clicked for me with MBDTF. Also this juicy beat ft Tyler:

veen  ·  7 days ago  ·  link  ·    ·  parent  ·  post: Everything Easy Is Hard Again

Bookmarked your Python link!

(Welcome back by the way.)

I have had multiple goes at teaching myself some web frameworks but have frustratedly quit almost every time. Last web project I just resorted to HTML/CSS with only the bare minimum JS/jQuery, outsourcing the hosting to Github Pages. But then again I wouldn't call myself good at any of those things.

rene  ·  7 days ago  ·  link  ·  

I was where you are a few months ago. I recommend playing around with Digital Ocean droplets to understand web servers, they make it very easy (and their documentation is excellent). With nginx you can have a static site up in <15 min.

I have a peripatetic posting style....Let me know if you have any questions about web frameworks, I had several conversations with friends that led to eureka moments.

veen  ·  7 days ago  ·  link  ·    ·  parent  ·  post: The Autonomy Ecosystem: Where and How It Begins

It took me a bit to realize it's from Andreessen Horowitz. There wasn't a lot of new info in it for me, but it summarizes (my position in) a bunch of discussions I've had here and in private. Thanks, I've mailed it to some of my coworkers. :)

veen  ·  10 days ago  ·  link  ·    ·  parent  ·  post: FALCON HEAVY IS A GO! Tuesday 1:30PM EST

This is great to watch too

veen  ·  11 days ago  ·  link  ·    ·  parent  ·  post: Art lovers of Hubski, post landscape images, just because

You can get an official print or official high-res image here. Bottom right, only requires you to make a free Rijksstudio account. :)

veen  ·  11 days ago  ·  link  ·    ·  parent  ·  post: Pubski: February 7, 2018

Skiing

My ski trip to Kirchberg/Kitzbühel was awesome. It was my first ski holiday ever. (You know, telling that fact to people I know instantly lets me know how well-off someone is. Europe is so tiny that anyone wealthy enough to afford a second holiday has fled to a ski resort in the Alps at one point or another.)

I had great lessons and went from angstily making 'pizza curves' to whoosing down a hill in the span of an hour or two. At the end of the second day and the beginning of the third day, my unexperienced ski buddy and colleague made the mistake of picking awfully difficult 'blue' routes. It was disheartening and scary, because you start to doubt every ounce of technique you think you have when you stumble down for the twenty-sixth time. But after that we went to the easy blue tracks, I finally 'got' how to make corners (lean forward and outward!) and made real, tangible progress. We ended the day by going to one of the tops and skiing down a route that I would have never thought I'd be able to conquer just days earlier. That third day was beautiful, not just because of the satisfaction that learning a new skill brings, but also because of the drop-dead gorgeous views. And lighting (see above).

There was also plenty of time for aprés-ski, I had some great food and got to know a surprisingly large amount of my new colleagues. Which brings me to the second heading...

Work

I technically haven't worked more days than I've gone skiing. I also don't really know yet what I will do, but I have a bunch of good ideas already and I'm already enjoying this work tremendously. Today I had this realization that I don't feel a student anymore - that I'm starting to feel like a professional.

The only issue now is my god-damned commute. It's looonngg. I walk to the bus 07:30, take the bus to a train station, take a stop train to an intercity station, transfer to an intercity train and walk fifteen minutes to my destination. Total time: almost 2 hours to traverse 60 miles. It's an hour by car but if I'd do that, I would be caught in the traffic-jam-vortex of 4 different cities so it would not be a whole lot faster. In a weird way I am proud of the proper timetable planning that's behind my commute: each transfer has around 8 to 10 minutes of transfer time. My bus is meant to arrive just before my stop train, which is synchronised with the IC train. It means I don't have to rush and small delays aren't a big deal. Plus, with my tablet I can read and shoot off some emails on the go.

And it's only temporary, or so I hope. We sent the final documents today - well, if they don't pester us again and just write the damn rental contract for us to sign, that is.

someguyfromcanada  ·  11 days ago  ·  link  ·  

Ski holidays are the best holidays IMO. The physical exhilaration, the views, the feeling of zipping down a mountain, the apres ski... it is perfect for me. It also gives one the opportunity to sit in a snow bank and take in the experience of being in a place like your picture.

Lessons are the way to go no matter how long you have been skiing or how good you are. I take 2 half day lessons on the first 2 days of every trip and I am a double black diamond skier who has been doing it for 40 years.