2,418 karma · joined March 8, 2017
As far as I can tell, the general consensus among urban planners is that building pedestrian overpasses/Skyways/skywalks is a _really bad idea_.
They suck pedestrian life away from the ground plane and surrender it to cars, which has all sorts of second order effects.
The ground level becomes darker, and more dangerous with less "eyes on the street". Car focused transport planners are more bold to increase traffic volume (with increases in pollution/crashes/noise). Businesses on the street struggle.
> Isolating large vehicles from fragile human beings is the best engineering solution.
It may be the best _engineering_ solution (don't get me wrong, I love some Brutalism and textured concrete), but it's usually a horrible outcome for street life. The best solution for a healthy, livable street is to build large spaces for pedestrians and cyclists, reduce car speeds to 30kph/20mph, or remove cars altogether.
One argument I saw somewhere (I can't remember where) is that if there are enough people to make a scramble worthwhile (and you don't care about car throughput) - you probably shouldn't have any cars there at all!
This comment covers some of the downsides: https://www.strongtowns.org/journal/2018/11/27/a-pedestrian-...
You may recognise Australia's parking button sound from a sample in Billie Eilish's hit Bad Guy, I've linked a video at the end of the blog post!
https://developer.nvidia.com/blog/getting-started-with-nvidi...
Large path:
- OpenDroneMap: 4.54 metres
- Tape measure: 4.56 metres
Small path:
- OpenDroneMap: 1.89 metres (though this varies slightly due to distortions)
- Tape measure: 1.89 metres
> I was just wondering if it’s possible to get the +/- 10cm accuracy
I'm quite confident this is possible - maybe do a few sanity checks though!
I believe OpenDroneMap uses OpenSfM which Mapillary open sourced:
https://blog.mapillary.com/update/2016/10/31/denser-3d-point...
> OpenSfM is also used internally by the OpenDroneMap project. OpenDroneMap creates 3D models and orthophotos from drone imagery. It includes all the steps of the pipeline so one can go from images to 3D models in one command. It uses OpenSfM to get the camera positions, and can now use it to compute the dense point clouds as well.
Good question! I'll go out there later today and get an on the ground measurement and get back to you on this thread (and upload the blog post).
> great tool for advocacy groups to identify where a road is wide enough to accommodate cycle lanes
Agreed :D
Wow that is some awesome software! I know OpenDroneMap does have a library of lenses but it doesn't seem to be able to detect the GoPro Fusion. I can't find the code with the presets right now, but https://docs.opendronemap.org/arguments/camera-lens/ is the documentation for selecting the lens type.
My understanding is that if it can't find a lens preset it's not an issue; it will instead figure out the lens geometry which it also can output. I haven't tinkered with the code though and I'm still getting to know the software!
How far apart were the photos taken? I think OpenDroneMap needs lots of overlap with drone photos, and even more with spherical photos to extract features well.
I haven't experimented with a non-360 degree camera - the benefit of a 360 degree camera is you effectively double the photos you get (front/back). Cycling or walking also ensures huge overlap.
I'm hoping this makes mapping curb/street/parking data more accessible.
For example, OpenStreetMap has a new street parking spec which could make use of lots of street imagery! https://wiki.openstreetmap.org/wiki/Street_parking
Sounds reminiscent of asbestos.
Might be a bug if the latter!
> have you considered graphing trains too?
I'm very keen to! It's actually an identical data format, I just need to change the endpoint from /buses to /trains essentially. Though I would also need to store 2 times the number of protocol buffers (which come in every 7 seconds) and do some UI work.
So many things to do, so little time :D
> However, I suspect there are some bugs:
Yep I wouldn't be surprised! The routes are statically generated so if there are updates (I'm not sure how often that happens) the progress lines won't be correct.
Thanks for the report though, I'll try looking into it when I have time.
> The buses were constantly late
Yep still happens, though the frequency is quite high so it's not too frustrating
> Unbelievable noisy. Walking through the CBD on a workday morning was deafening.
Yep still very loud on the hills, though electric buses are being rolled out. George St is now pedestrianised with the tram so it's less noisy.
> Gave up trying to ride a bike. Just too dangerous and aggressive bus and car drivers :(
Cycling infrastructure is getting a lot better! BikeEast (https://www.bikeast.org.au/) do a lot of advocacy in the eastern suburbs. Clover Moore has done amazing things for the CBD.
If you had asked your friends to get a Matrix client instead (eg. Element), at least you'd be able to move to a new client [2] if you didn't like the direction it moved in - and you'd still have your group chats and keep your contacts because it's federated.
Unfortunately you're now stuck with stories (or maybe crypto payments one day [3]) because Signal doesn't federate, and likely never will, and the amount of effort to ask your friends to move _again_ is just too big.
This is _exactly_ the argument I make when friends ask why Matrix over Signal. I'm exhausted of picking up and moving again.
[1]: https://signal.org/blog/the-ecosystem-is-moving/ [2]: https://matrix.org/clients/ [3]: https://www.xda-developers.com/signal-crypto-payments-featur...
This is the beauty of Matrix - you can change your client, or even your server and keep all your contacts!
From the FAQs at https://www.tomtom.com/tomtom-maps-platform/:
> I don’t want to use the new maps with OSM, will my application continue to work if I use an older library version?
> Yes. Although we encourage you to switch to the new enhanced map to take full advantage of the new map features integrated by the OSM community.
This update is light on details around licensing of the new map.
The map is open source on Github: https://github.com/jakecoppinger/sydney-advertising-map
There's a fair bit of backlash in the community, here's some reporting by the Sydney Morning Herald: https://www.smh.com.au/national/nsw/balance-isn-t-right-clov...
FluffyChat also has quite nice UX and a bubble layout by default, but threads are still a while off [2]. On iOS it worked flawlessly through the iOS 16 betas while Element had some show stopping bugs, a couple of my friends moved over if they were on the beta.
I haven't had any friends ask me about the verify session buttons. I don't see any prompts on latest iOS Element but it's still too prominent on Element desktop for my liking.
SchildiChat [3] is my daily driver and feels more friendly than Element on desktop (unified DMs & group chats, no verify UX, chat bubbles), but it doesn't have any update mechanism built in, so I'm wary to recommend it to non-technical friends. It was also my goto recommendation on Android before the Element redesign.
I'm confident the ecosystem is moving in the right direction though, and so thankful for the amount of choice.
[1]: https://github.com/vector-im/element-x-ios [2]: https://gitlab.com/famedly/fluffychat/-/issues/881 [3]: https://schildi.chat/
Or another paid provider: https://matrix.org/hosting/
What costs? Element (a popular Matrix client) has recently improved their onboarding greatly!
https://element.io/blog/all-aboard-better-ftue-for-less-wtf/
It really is just as easy to onboard to Element as Signal these days, the UX has come a long way. And you'll never have the move them again, because you can choose any client you like.
> our friends at Bluesky announced their application protocol for building decentralised social media called AT. While not based on Matrix, there are some parallels, and some stuff we may be able to get inspiration from around portable identity :) https://blueskyweb.xyz/blog/10-18-2022-the-at-protocol
Sounds like some there are some parallels, and the protocol might inspire portable account identities in Matrix!
The only account migration tool I know of at the moment is proprietary [1], I'd love to see portable identity in Matrix come to life.
This seems like a good idea for a bot built with https://github.com/turt2live/matrix-bot-sdk (which is very easy, I've used it for a few small projects).
The SDK has a callback for each new message - you could just put it into a DB and then render to static text.