Hacker Newsnew | past | comments | ask | show | jobs | submit | more ConceitedCode's commentslogin


Apparently they tested just that -

Säll said: “There is no electricity on the surface. There are two tracks, just like an outlet in the wall. Five or six centimetres down is where the electricity is. But if you flood the road with salt water then we have found that the electricity level at the surface is just one volt. You could walk on it barefoot.”

They also mentioned they turn off each segment if there is not a car on it. Each segment is 50 meters.


That's actually reassuring, I'm glad they looked into this.

My follow-up concern is, what if, under these salty and wet conditions, something causes a short circuit between the powered rails?

e.g., either a defective vehicle-side contact where the anode and cathode allow for the water and salt to close a much smaller gap, or if someone deliberately sticks a piece of metal on there.


You'll get a small steam explosion but there are lots of ways for the pressure to escape, and if the circuitry protecting the 50 meter stretch is fast enough that might not even happen.


But will slugs be able to safely cross the road?


No. But in a few generations they'll learn not to. In a few more they'll be expert jumpers.


They said nothing about the wear characteristics of the rail when covered in salty road crap.

Stray current wreaks all sorts of havoc on electric light rail systems. I don't see why this would be different.


It is Sweden. If anyone knows salty road crap it is Scandinavians.


This looks very simple and easy to use.

Are there any differences with using this instead of CSVKit?

https://csvkit.readthedocs.io/en/1.0.3/

It includes a tool called csvsql.

Example usage -

  csvsql --query "select name from data where age > 30" data.csv > new.csv


Was wondering the same. I've been using csvkit a lot lately. This looks like a subset of functionality using more or less the same approach. Csvsql uses sqlite under neath and you can do some nice things with this like joins, using sql functions, etc. There is also stuff like amazon athena that allows you to do similar things in s3 at scale.

Csvkit is great with pipes and you can also easily convert between csv tsv and even stuff like json.


Python's pandas library can also do sql-like queries against csv data.

https://pandas.pydata.org/pandas-docs/stable/comparison_with...


Although csvsql is great, the real reason I love csvkit is that most of the tools feel like standard *nix tools that just operate on CSV files (csvcut, csvgrep, etc.)


It would depend on your definition of "succeeded", but Docker (formally dotCloud) made a great transition after releasing docker and just raised a very nice round of funding.

"In 2013, recognizing the need for flexible, PaaS-like environments inside enterprises and across clouds, the company released much of its PaaS container technology as the open source Docker project. Docker is an open source engine for deploying any application as a lightweight, portable, self-sufficient container that will run virtually anywhere. By delivering on the promise “Build Once…Run Anywhere,” Docker has seen explosive growth, and its impact is being seen across devops, PaaS, and hybrid cloud environments.

The success of the Docker project led the company to change its name from "dotCloud, Inc." to "Docker, Inc." in October 2013 in order to reflect its focus on the new product. Docker, Inc. continues to run the dotCloud Platform, supporting thousands of containers running applications for a wide variety of businesses."[0]

[0] https://www.dotcloud.com/about.html


Yeah, Docker seems to be more popular than dotCloud ever was, but do we actually know how the pivot affected their finances?


Derek Sivers talks about how ideas don't necessarily have value, but are multipliers of value. This happens to me at least once a week.

https://www.youtube.com/watch?v=QgaBvEO2LYY


No particular reason. I just now came across it.


Correct me if I'm wrong, but I think it is more than just messaging and might include SMS. From the Play store description, "Not just for Facebook friends: Now you can message people in your phone book. And it's easy to add new contacts—just enter a phone number.". The android app also requires permission to edit/create SMS messages and make phone calls.

If so, does this mean we can be able to send an SMS from facebook now or just from our phones?

It's worth noting that the screenshot from the android phone also shows you missed calls.


I assume (but don't know!) that the "find people by entering their phone number" feature simply tries to find Facebook users with lax enough privacy settings such that (a) they expose their phone numbers to maybe friends or friends and (b) they allow non-friends to message them. Otherwise I feel like they would phrase it as "message people who aren't on Facebook!"

What Android screenshot are you looking at that makes you think it shows missed calls? I see two icons: a sort of bolt and a facebook "f". I think that the bolt means the user is using "Messenger" (i.e. fb on mobile) whereas the fb "f" means facebook.com.


This seems to answer the question: https://www.facebook.com/help/458736564237313


Can anyone confirm or deny? I no longer have a facebook account and don't want one so I can't check.

The first android screenshot on the linked page, shows that there is 1 missed call from Mac Tyler. Just worth noting, not sure if it means anything but it is interesting.


At least on Android, you can send SMS through the Facebook app, which is what I assumed they were referencing here. It doesn't send the message through Facebook in the sense of web-to-SMS gateway, though; it's just that Android apps are allowed to send SMS, and Facebook has built that functionality into their app, so that 'all of your messages are in one place'.


Now you can message people in your phone book

Doesn't this mean FB will be hacking your phone book? Nice backdoor...


When you launch the updated Messenger app on iOS, it asks for permission to use your contacts. The alert message was sufficiently creepy that I didn't give it permission to "sync contacts" with Facebook.


Yep. Just another bite at the apple. It only takes one mistake.


What constitutes "internet property"? Is this exclusively search or does this include other products such as Tumblr?


Has anyone figure out how to use this when switching between views?


If you take a look at this video: http://www.youtube.com/watch?v=P6KITGRQujQ and associated tutorial, it will give you an example of binding to the RouteChange events for exactly this use case.


Can you show how you are using the views? Are they directly in the body?



Has anyone figured out what caused this yet? I'm interested in knowing what took out all of Google.


It was a bad DoS config push that caused a general GFE crash.

The outage lasted less then 5 minutes, so not that interesting.

On a more interesting note, the Helpouts leaker has been identified and fired: http://www.linkedin.com/in/georgearison

And in related news you might want to ask Jean Baptiste Quéru how feels about having been fired after having spent years working on AOSP.

That's Google for ya.


It wasn't all of Google, but just their search.


Well according to a friend of mine who works at the Goog it affected them internally as well, things like C.I. infrastructure and the like. So I doubt something that large goes down "all at once" but it seems it was more than search.


Oh, my apologies then. It seems odd that everything was affected, and simultaneously too. Perhaps Skynet is online now and it's testing the reaction time and impact of its reach.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: