Rolandt

Originally published on the GDS User Research Blog

Once every 5 years, when UK general elections come round, we’re given the opportunity to research the experience of voting. Although voting is not something we’re really working on, the recent general elections offered an opportunity impossible for the GDS user research team to resist.

We conducted a small study

With some help from colleagues in the Home Office, we conducted a small study in the days before, during and after the general election. Our goal was to see if there were opportunities to make the voting experience better.

Our data comes from:

• 30 phone interviews with young people who were mostly voting for the first time. We spoke to them the day after the election.
• 44 participants from a varied demographic participated in a diary study using dScout.
• We triangulated the data from these 2 studies with data that our social media team found using Brandwatch.

Recruitment constraints

We were recruiting during Purdah (the pre-election period) and without a budget, both of which were significant constraints. As a result, the participants in our study skew towards being more engaged in the democratic process than we’d expect to be typical. Still, we think many of the findings are widely applicable, although you’d probably unearth a lot more opportunities and insights if you did the same study with less democratically engaged participants.

Here’s what we found out.

Voting is emotional

Young people told us that casting their first vote felt like a rite of passage to adulthood. They felt compelled to vote because of the history of sacrifices made to be able to vote, and how hard-won democracy can be.

It is exciting that I could vote – afterwards I felt like an adult and I can get a mortgage next. – Ben

Many people talked about the act of voting as feeling historic, and an important moment for them and their participation in society.

Some people interpreted the paper and stubby pencils as signifying the ‘tradition’ of voting, but many others felt that the experience of voting was quite antiquated.

When leaving the polling station it is hard to believe that we are in 2015… I think next time I will wear a Victorian costume to fit the experience. – EV

The experience of voting can be a bit of a let down

I expected it to be exciting but there wasn’t much of an atmosphere – maybe because I went early. – Hatti

The general elections are a moment in time where people are more engaged in their community and in the democratic process than they might normally be. This may offer an opportunity.

People are looking for voting to be more of an experience, but the experience often turns out to be a bit of a non-event.

Many people are surprised at how little time it takes to vote. On the one hand, they feel this should be communicated more widely so that more people know that voting is not time consuming and difficult. On the other hand, they sometimes feel ‘rushed through’ and the importance of the act of voting is lost.

Several people felt confused and intimidated by people wearing rosettes and asking for their polling number. People taking exit polls don’t identify themselves, or what they’re doing, leading some people to mistake them as polling officials. For some people, this made the voting experience more stressful.

[I’m] At the school to vote. People outside asking for my polling card number. Don’t know why. They are wearing political party rosettes. Does this mean my details will be used by the party? I wouldn’t want that. How are you feeling now? Anxious. – DS/BK

Making sure you’re registered to vote is not always simple

For people who do want to vote, making sure you’re registered to vote is not always simple. Most people we spoke to were registered to vote, but they told us about friends and family who were not aware that they needed to register until it was too late.

The need to provide your National Insurance number when registering to vote is more difficult for young people – they don’t receive a physical card with their number on it, so it’s more difficult to provide that information.

Some people indicated that they weren’t sure whether they were successfully registered to vote, or where they were registered. This meant telephone calls to local authorities, which often went unanswered as the election drew closer, and people re-registering just to be sure.

I registered online but I didn’t feel confident that it had been done as I didn’t get a confirmation email, so I called the local council a few days before to confirm. – Elizabeth

Moving house or having two addresses (eg students) was particularly problematic. This meant that some people delayed registering to vote until the last moment because they weren’t sure where they were going to be at election time. Or, sometimes they’d discover that they were enrolled a long distance from their current address.

I registered on the last day. I wasn’t sure where I was going to be [on election day]. I moved placement during my degree and forgot where I had registered. – Scott

Polling cards are important but often fail

People use polling cards to confirm they are registered to vote, to know when and where to vote, and to find the polling station. Polling cards were a point of failure for many people, especially as they seemed to arrive at different times or not at all.

The polling card never arrived, so I’m not sure if I’m registered to vote. – Shad

They were often misplaced because they’d arrived very early, or caused stress because they arrived very late or not at all. Also, the polling station map on the cards seemed to be unreliable.

The map on the polling cards is to the wrong place – people in the queue are furious, they’ve been wandering about lost. One man asks them to put a poster up to direct people. They say they can’t – they only take notes and tell the council for next time. Bit awkward. – DS/RA

Working out who to vote for can be difficult

Most people we spoke to took the decision as to who they would vote for quite seriously. Most of them struggled to find information that was useful to help them make a decision. Many people told us that they used the online tools that anonymised the policies in the manifestos and then told them who to vote for. They found these tools useful.

People weren’t happy to discover new candidates in the polling booth at the moment of voting.

When I got to the sanctum of the booth I was amused to see that there were ten, TEN, candidates. So where were their leaflets? How am I supposed to consider voting for them if the first time I hear about them is in the bleeding voting booth? I mean, FFS! – DS/SW

Young people told us that they talked with each other openly about who they were going to vote for and who they did vote for – they’re aware of this being more of a taboo for their parent’s generation, but feel that talking openly was an important way to help them decide who to vote for.

People thought government could do a better job of helping them know when, where and how to vote

Young people in particular thought there was a lot that government could be doing to help more people feel more confident and knowledgeable about the process of voting. A lot of people felt they learned about voting in the polling station when it should have been taught at school.

They don’t teach you that at school. I know how to draw plant cells but not how to vote. – Yiannis

It would be nice to explain the details about voting – to say, this is your first election, when to register, you don’t need to bring ID etc. – Michael

People also thought that the government should be more proactive about messaging people to remind them that the election was coming up and when and where they should plan to vote. Less consumption of TV and newspaper content seemed to mean that it’s more possible for young people to not notice there is an election coming up.

What would have been useful is an email a week before the election saying ‘you are registered, remember to vote, this is your polling station’. – Sophie

Voting for two elections at once can be confusing

A number of people were registered to vote in locations where both the general and local elections were held at the same time. For two main reasons, this often caused confusion.

First: the focus on the general election often meant that people were not aware that the local election was being held and had not considered who to vote for.

Went with my husband to vote. Surprised there were council elections as we haven’t heard anything about them. No idea who the candidates are. – DS/MX

Second: there are different methods of voting for each type of election. This caused some confusion and people felt they had to be careful about voting to ensure their vote was valid – some weren’t sure that they had in fact voted successfully.

All done. Always find the mixing of local and national elections tricky. Two crosses on one, definitely one cross on the other. – DS/JS

The’localness’ of voting can be frustrating

Voting is very local – you can only vote in one location where you live and you vote for local members who represent that area. Many people found this difficult to understand and frustrating.

The requirement to attend a single location to vote is a point of failure for people who intend to vote but aren’t able to be in the correct place on the day – they often don’t know this sufficiently in advance to arrange for alternative ways of voting or don’t know that other methods are available. To some people, this seemed particularly unnecessary and archaic and is often an unexpected discovery for people who are voting in the UK for the first time.

I thought I could vote anywhere, my friend explained to me I had to vote where I was registered. I missed the deadline to change address. It is irritating that can’t go to any polling station, that it is linked geographically. – Shad

People were often very frustrated that the local representation limited their ability to vote for the party they wished to represent. Most people seemed to think about voting for a party rather than for a particular Member of Parliament – a mental model that’s perpetuated in the way the media talks about the election – then they arrive at the polling booth to discover they can’t vote for their party of choice. This is a frustrating and disenfranchising experience.

I live in Buckinghamshire so I was only offered Conservative, Green and UKIP. It was annoying not to have more choice. Instead of voting on my ballot I just wrote ‘none of the above’ because I was very annoyed at the lack of choice I had. – Joanna

Finally, people who lived in safe seats felt that their votes were much less valuable than those who lived in closely-contested seats.

Part of the reason [I didn’t vote] was that the Tory seat where I live isn’t going to change, so my vote felt a bit pointless anyway. – Liam

Plenty of opportunities for the future

So, it turns out there are lots of opportunities to make the voting experience better, which will in turn result in more people voting in a more informed way.

You can download the deck for more details. We hope you find it useful.

#SpeakForMe is a campaign to petition Indian MPs, Banks, Mobile operators and other service providers to speak for you against the Aadhaar linking coercion. You can go to #SpeakForMe to send your petition. As part of campaign I built a quick and dirty dashboard for the emails sent. This is a quick note on how I did that.

Part of #SpeakForMe dashboard showing emails sent to MPs on a PC map.

#SpeakForMe has a twitter account @bulletinbabu which used to tweet updates in a standard format, at regular intervals. At first I started parsing these tweets and started plotting them on a graph. The parsing script would run every hour find all the tweets and then parse them and insert them into a CouchDB. Parsed CouchDB document is very simple and can be used to directly for charting

{  
   "_id":"2017-12-13T18:20:02+05:30",
   "_rev":"1-67c8a405a19f3a787f42640fa1ac9aef",
   "govt":32,
   "stat":"email_sent",
   "mps":780,
   "campaign":"#SpeakForMe",
   "others":13,
   "mobile":69,
   "tw":940926800292540417,
   "total":1000,
   "banks":106
}

Scraper code is pretty standard too

#!/usr/bin/env python
# encoding: utf-8
import couchdb
import tweepy #https://github.com/tweepy/tweepy
import csv
import re
import arrow
import time

# The consumer keys can be found on your application's Details
# page located at https://dev.twitter.com/apps (under "OAuth settings")
consumer_key=""
consumer_secret=""

# The access tokens can be found on your applications's Details
# page located at https://dev.twitter.com/apps (located
# under "Your access token")
access_key=""
access_secret=""

#you will have to change this
couch_url = "https://username:password@mycouchdb.url.com"

remote_server = couchdb.Server(couch_url)
bulletinbabu_db = remote_server['bulletinbabu']

def get_all_tweets(screen_name):
	#Twitter only allows access to a users most recent 3240 tweets with this method
	
	#authorize twitter, initialize tweepy
	auth = tweepy.OAuthHandler(consumer_key, consumer_secret)
	auth.set_access_token(access_key, access_secret)
	api = tweepy.API(auth)
	
	#initialize a list to hold all the tweepy Tweets
	alltweets = []	
	
	#make initial request for most recent tweets (200 is the maximum allowed count)
	new_tweets = api.user_timeline(screen_name = screen_name,count=200,tweet_mode="extended")
	
	#save most recent tweets
	alltweets.extend(new_tweets)
	
	#save the id of the oldest tweet less one
	oldest = alltweets[-1].id - 1
	
	#keep grabbing tweets until there are no tweets left to grab
	while len(new_tweets) > 0:
		break		
		
		#all subsiquent requests use the max_id param to prevent duplicates
		new_tweets = api.user_timeline(screen_name = screen_name,count=200,max_id=oldest,tweet_mode="extended")
		
		#save most recent tweets
		alltweets.extend(new_tweets)
		
		#update the id of the oldest tweet less one
		oldest = alltweets[-1].id - 1
		
		print "...%s tweets downloaded so far" % (len(alltweets))
	


	for tweet in alltweets:
		print "--------------------------------------------------------------------------------------------"
		bulletinbabu = {}
		bulletinbabu['tw']=tweet.id
		bulletinbabu['campaign']="#SpeakForMe"		
		bulletinbabu['_id'] = arrow.get(tweet.created_at).to('local').format('YYYY-MM-DDTHH:mm:ssZZ')	
		text =  tweet.full_text.encode("utf-8")
		print str(text)
		if text.startswith("Emails from #SpeakForMe to:"):
			bulletinbabu['stat']="email_sent"
			regex_search = re.search('MPs:(.*) ', text, re.IGNORECASE)
			if regex_search:
				mps = regex_search.group(1)
				mps = mps.replace(",","")
				print str(mps)
				bulletinbabu['mps']=int(mps.strip())

			regex_search = re.search('Banks:(.*) ', text, re.IGNORECASE)
			if regex_search:
				banks = regex_search.group(1)
				banks = banks.replace(",","")
				bulletinbabu['banks']=int(banks.strip())

			regex_search = re.search('Mobile service providers:(.*)\ ', text, re.IGNORECASE)
			if regex_search:
				mobile = regex_search.group(1)
				mobile = mobile.replace(",","")
				bulletinbabu['mobile']=int(mobile.strip())

			regex_search = re.search('Government services:(.*)\ ', text, re.IGNORECASE)
			if regex_search:
				govt = regex_search.group(1)
				govt = govt.replace(",","")
				bulletinbabu['govt']=int(govt.strip())

			regex_search = re.search('Others:(.*)\ ', text, re.IGNORECASE)
			if regex_search:
				others = regex_search.group(1)
				others = others.replace(",","")
				bulletinbabu['others']=int(others.strip())

			regex_search = re.search('Total:(.*)\ ', text, re.IGNORECASE)
			if regex_search:
				total = regex_search.group(1)
				total = total.replace(",","")
				bulletinbabu['total']=int(total.strip())
			print str(bulletinbabu)
			try:
				bulletinbabu_db.save(bulletinbabu)
			except couchdb.http.ResourceConflict:
				print "Already exists"
				break
			time.sleep(0.1)
		elif text.startswith("Top recipients of #SpeakForMe emails:"):
			#bulletinbabu['stat']="top_rcpt"
			pass
	
	

if __name__ == '__main__':
	#pass in the username of the account you want to download
	get_all_tweets("bulletinbabu")

Since CouchDB provides http restful access to data, there was no issue in pulling the data from the database using standard AJAX requests for plotting. Couple of days later #SpeakForMe team wanted to see how many emails were sent to MPs. So I asked them to post the aggregate analytics ¹ they were collecting to my CouchDB. They started posting two types of documents. One aggregate at services level, second aggregates at individual receiver level. Posting would be a simple web POST using python requests for them. Just like posting to any webhook

import requests
couch_url = "https://username:password@mycouchdb.url.com"
data = {'stat': 'email_sent', 'total':2339 , 'campaign': '#SpeakForMe', 'mobile':198 , 'tw':941038005199998978 , 'govt':78 , 'mps': 1824, 'others':6 , 'banks': 233, '_id': u'2017-12-14T01:41:00+05:30'}
r = requests.post(couch_url, json = data)

First document is similar to what I used to scrape. Second one is a bigger document. It has number of emails at the level of service provider or MP. Attribute “stat” differentiates the two types of document. _id which is primary key is just a standard time-stamp. As you can see in the partial “mailbox_email_sent” document below. The key has two parts “type of provider” and “provider name”, separated by “/”. For airtel it is “mobile/airtel” etc. For mps it starts with mp and then has state code and parliamentary constituency number, Eg: “mp/mh-47”. Here is the copy of full document if you like to see.

{  
   "_id":"2017-12-20T13:10:03.243684+05:30",
   "_rev":"1-b9ed7d5104bf0c8e1fbd341743829084",
   "gov/pan":344,
   "mobile/mts":1,
   "mp/ar-2":4,
   "mp/mh-47":28,
   "bank/bkid":24,
   "mp/ka-14":19,
   "mp/ka-15":65,  
   "mp/wb-41":1,
   "campaign":"#SpeakForMe",
   "bank/orbc":10,
   "mp/ke-20":167,
   "total":32354,
   "bank/lavb":2,
   "bank/synb":6,
   "bank/ibkl":19,
   "mobile/airtel":595,
   ...
   ....
   ...
   "mp/or-12":4,
   "mp/pb-8":17,
   "mp/wb-30":5,
   "bank/indb":14,
   "mobile/idea":183,
   "stat":"mailbox_email_sent",
   "bank/vijb":7,
   "mp/bi-38":18,
   "mp/bi-40":3
}

On the client side its just static html and javascript. I used Parliamentary Constituencies Maps provided by Data{Meet} Community Maps Project. They are displayed using leaflet and d3. In fact I borrowed parts of code from DataMeet maps project. I use Lodash, to query, filter and manipulate the documents returned by CouchDB. For example

let all_rows = _.reverse(returned_data.rows);
//Filter emails sent
let rows =  _.filter(all_rows, function(o) { return o.doc.stat == "email_sent" && o.doc.campaign == "#SpeakForMe"});
let latest_row = _.last(rows);

let rows_mailbox_email_sent = _.filter(all_rows, function(o) { return o.doc.stat == "mailbox_email_sent" && o.doc.campaign == "#SpeakForMe"});
let latest_mailbox_email_sent = _.last(rows_mailbox_email_sent);

You can see the code that does everything here. I used Frappé Charts for charting. I love them. They are simple and look great.

Basically analytics data gets stored in a CouchDB and served as standard restful that CouchDB provides to browser. Running couchdb to receive external authenticated webhook post and then serve the data as restful service worked like a charm. Of course I didn’t have much traffic to test under load. But since CouchDB is behind a CloudFront (Amazon CDN), I guess is it can take quite a bit of load. Usually the team pushes data every 5 minutes if there are updates. So it shows live status (see the last updated time stamp).

At some point I will create graph of email traffic (daily emails sent etc). Any other graphs you would like to see? I will be happy to answer any questions if you have.

Source

The way that machine learning works is basically this: you input some models, let’s say of what tables look like, and then the code generates some things it thinks are tables. You click yes on the things that look like tables and the code reinforces the processes that made those and makes some more attempts. You rate again, and with each rating the elements of the process that produce table-like things are strengthened and the ones that produce non-table-like things are weakened.

It doesn’t have to be making things — it can be recognition as well. In fact, as long as you have some human feedback in the mix you can train an machine learning process to recognize and rate tables that another machine learning process makes, in something called a generative adversarial network.

People often use machine learning and AI interchangeably (and sometimes I do too). In reality machine learning is one approach to AI, and it works very well for some things and not so well for others. So far, for example, it’s been a bit of a bust in education. It’s had some good results in terms of self-driving cars. It hasn’t done great in medicine.

It will get better in these areas but there’s a bit of a gating factor here — the feedback loops in these areas are both delayed and complex. In medicine we’re interested in survival rates that span from months to decades — not exactly a fast paced loop — and the information that is currently out there for machines to learn from is messy and inconclusive. In learning, the ability to produce custom content is likely to have some effect, but bigger issues such as motivation, deep understanding, and long-term learning gains are not as simple as recognizing tables. In cars machine learning has turned out to be more useful, but even there you can use machine learning to recognize stop signs, but it’s a bit harder to test the rarer and more complex instances of “you-go-no-you-go” yielding protocols.

You know what machine learning is really good at learning, though? Like, scary, Skynet-level good?

What you click on.

Think about our tables example, but replace it with headlines. Imagine feeding into a machine learning algorithm the 1,000 most shared headlines and stories, and then having the ML generate over the next hour 10,000 headlines that it publishes by 1,000 bots. The ones that are successful get shared and those parts of the ML net are boosted (produce more like this!). The ones that don’t get shared let the ML know to produce less along those lines.

That’s hour one of our disinfo Skynet. If the bots have any sizable audience, you’re running maybe 20,000 tests per piece of content — showing it to 20,000 people and seeing how they react. Hour two repeats that with better content. By the next morning you’ve run millions of tests on your various pieces of content, all slowly improving the virality of the material.

At that scale you can start checking valence, targeting, impact. It’s easy enough for a network analysis to show whether certain material is starting fights for example, and stuff that starts fights can be rated up. You can find what shares well and produces cynicism in rural counties if you want. Facebook’s staff will even help you with some of that.

In short, the social media audience becomes one big training pool for your clickbait or disinfo machine. And since there is enough information from the human training to model what humans click on, that process can be amplified via generative adversarial networks, just like with our tables.

It doesn’t stop there. The actual articles can be written by ML, with their opening grafs adjusted for maximum impact. Videos can be automatically generated off of popular articles and flood YouTube.

Even the bots can get less distinguishable. An article in the New York Times today details the work being done in ML face generation, where believable fake faces are generated. Right now the process is slow, partially because it relies solely on GAN, and because it’s processor intensive. But imagine generating out a 1,000 fake faces for your bot avatars and tracking which ones get the most shares, then regenerating a thousand more based on that and updating. Or even easier, autogenerating and re-generating user bios.

You don’t even need to hand-grow the faces, as with the NYT article. You could generate 1.000 morphs, or combos of existing faces.

Just as with the last wave of disinformation the first adopters of this stuff will be the clickbait farms, finding new and more effective means to get us to sites selling dietary supplements, or watch weird autogenerated YouTube videos. There will be a flood of low-information ML-based content. But from there it will be weaponized, and used to suppress speech and manipulate public opinion.

These different elements of ML-based gaming of the system have different ETAs, and I’m not saying all of this is imminent. Some of it is quite far off. But I am saying it is unavoidable. You have machine learning — which loves short and simple feedback loops — and you have social media, which has a business model and interface built around those loops. The two things fit together like a lock and a key. And once these two things come together it is likely to have a profoundly detrimental effect on online culture, and make our current mess seem quite primitive by comparison.

Several recently-published research articles have demonstrated a new class of timing attacks (Meltdown and Spectre) that work on modern CPUs.  Our internal experiments confirm that it is possible to use similar techniques from Web content to read private information between different origins.  The full extent of this class of attack is still under investigation and we are working with security researchers and other browser vendors to fully understand the threat and fixes.  Since this new class of attacks involves measuring precise time intervals, as a partial, short-term, mitigation we are disabling or reducing the precision of several time sources in Firefox.  This includes both explicit sources, like performance.now(), and implicit sources that allow building high-resolution timers, viz., SharedArrayBuffer.

Specifically, in all release channels, starting with 57:

• The resolution of performance.now() will be reduced to 20µs. (UPDATE: see the MDN documentation for performance.now for up-to-date precision information.)
• The SharedArrayBuffer feature is being disabled by default.

Furthermore, other timing sources and time-fuzzing techniques are being worked on.

In the longer term, we have started experimenting with techniques to remove the information leak closer to the source, instead of just hiding the leak by disabling timers.  This project requires time to understand, implement and test, but might allow us to consider reenabling SharedArrayBuffer and the other high-resolution timers as these features provide important capabilities to the Web platform.

Update [January 4, 2018]: We have released the two timing-related mitigations described above with Firefox 57.0.4, Beta and Developers Edition 58.0b14, and Nightly 59.0a1 dated “2018-01-04” and later. Firefox 52 ESR does not support SharedArrayBuffer and is less at risk; the performance.now() mitigations will be included in the regularly scheduled Firefox 52.6 ESR release on January 23, 2018.

The post Mitigations landing for new class of timing attack appeared first on Mozilla Security Blog.

Sports are growing more international with respect to the athletes. Gregor Aisch, Kevin Quealy, and Rory Smith for The Upshot show by how much, with a focus on leagues in Europe and North America.

I like how: The dominant home country in each chart doubles as background and a layer; the tooltip shades the country you moused over while still showing the other countries; and the missing data and gaps are shown clearly but don’t obstruct the overall view.

Tags: sports, Upshot

How we sleep is unbelievably important. Getting too little sleep not only makes you feel lousy and cranky, but it’s also linked to obesity, diabetes, hypertension, and even early death. (No pressure.)

But it’s amazingly hard to measure our sleep, as a population. Sure, one person at a time can stay overnight at a sleep lab, hooked up to scalp electrodes — but try sleeping normally that way, away from home and wired to strange equipment. Other sleep studies use self-reporting, where you write down each morning how you slept, but that data is famously unreliable.

Now, though, there’s a new way to study our sleep: Fitness bands, worn by millions of people. Most of Fitibit’s bands, for example, have built-in heart-rate monitors, which produce much more accurate sleep-measurement results than earlier bands. These bands track your sleep automatically, in your own bed, on your normal schedule, under normal conditions.

Since Fitbit began tracking sleep stages in March 2017, it has collected data from 6 billion nights of its customers’ sleep. This is a gold mine — by far the largest set of sleep data ever assembled. (This data is anonymous and averaged; it’s not associated with individual customers’ names.)

“It’s a really, really exciting and really rare data set,” Fitbit data scientist Karla Gleichauf says. “It’s probably the largest biometric data set in the world.”

The measurements include not just how long you sleep, but what stages of sleep you experience. Each morning, the Fitbit app shows which parts of the night you spent in REM sleep (the vivid-dreams stage, good for mood regulation and memory processing), in deep sleep (good for memory, learning, the immune system, and feeling rested), in light sleep, and awake. (It’s always disheartening to see how much of the night you waste in little one- or two-minute wake-ups that you don’t even remember.)

But wait, there’s more. The Fitbit app also knows your gender, age, weight, height, location, and activity level. Therefore, the company’s data scientists can slice and dice its massive sleep database in fantastic ways. They should be able to tell us who sleeps more: men or women. Northerners or Southerners. East Coasters or West Coasters. They should be able to calculate our national average bedtime. They should be able to draw all kinds of conclusions about the way we sleep — and what’s good for us.

Now, for the first time, they have. Gleichauf and her boss, Conor Heneghan, Fitbit’s lead sleep research scientist, agreed to mine that vast sleep database to unearth some of its secrets. Some of their findings reinforce what sleep scientists have already studied; some have never been measured before.

Here’s what Fitbit discovered — a Yahoo Finance exclusive.

Men vs. women

Women sleep 25 minutes longer a night than men. They average six hours and 50 minutes of sleep a night, whereas men get only six hours and 26 minutes. Neither group gets anywhere close to the recommended eight hours a night.

Women also get about 10 minutes more REM sleep than men every night, too — a gap that widens after age 50.

Why these differences? “It’s really not known if it’s a physiology thing, is it a cultural thing, who knows,” Heneghan says. “I think that would be super exciting over the next 10, 20 years for people to really get into why.”

The news for women isn’t all good, though: They’re 40% more likely to suffer from insomnia — trouble falling asleep — than men.

(Those two findings could be related, too: Since women’s sleep is less efficient, they have to spend more time in bed.)

Old vs. young

Getting older also affects your sleep. In this graph from Fitbit’s sleep study, you can see that we get less deep sleep as we age. When you’re 20, you’re getting half an hour more deep sleep a night than when you’re 70.

North vs. South

Yes, it’s true: Northerners go to bed five minutes earlier than Southerners. They wake up earlier, too.

That may seem like a very small difference, but on the scale of billions of data points, it’s significant.

On the other hand, Heneghan points out that statistics can be tricky. “I think North/South may be an artificial divide; urban/rural is probably a more meaningful divide,” he notes. In other words, there may just be more big cities in the North.

East vs. West

East Coasters, according to the data, stay up seven minutes later than West Coasters (and wake up five minutes later, too).

“I personally find this consistent with my experience of American culture,” Heneghan says. “I lived in New York. I lived in California. You get to 9 p.m. here in California, and the restaurant staff are kind of looking at you funny. There’s a great quote from Yogi Berra: ‘It gets late real early around here.’”

The national bedtime

Here’s a data point that no amount of sleep-lab studies could have unearthed: The average American goes to bed at 11:21 p.m.

Bedtime consistency

The biggest finding in Fitbit’s data may be the link between sleep quality and bedtime consistency.

That, Gleichauf explains, “is this idea that your bedtime varies.”

And in America, it really does vary — by an average of 64 minutes. You might go to bed at 11 p.m. on weeknights, but stay up after midnight on the weekends.

The Fitbit data shows that your sleep suffers as a result. If your bedtime varies by two hours over the week, you’ll average half hour of sleep a night lessthan someone whose bedtime varies by only 30 minutes.

And you’ll pay the price.

You know how jet lag works, right? “When you have jet lag, it’s the mismatch between the actual time, in the zone you’re in, and your circadian rhythm,” Gleichauf told me. “You’re not on the right part of that curve to make you fall asleep.” So, at night in your new city, you lie there for hours, unable to fall asleep — and then in the middle of the next day, you’re overcome by exhaustion.

When your bedtime varies over the week, then, you’re creating self-induced jet lag. Gleichauf calls it social jet lag: On Monday, when you have to go back to work (and drag your bedtime backward), you feel crummy and you’re more likely to get sick.

(Dr. Till Roenneberg, professor at the Institute of Medical Psychology at the University of Munich, calculates that every hour of social jetlag increases your risk of being overweight or obese by about 33%.)

“I’m super excited about this data,” Heneghan says. “For the first time ever, we were actually able to show the link between consistency and how long you sleep.”

Social jet lag, by city

Gleichauf dove into American geography to see if there were differences in bedtime consistency — and there is.

Can you guess which city has the most widely varying bedtimes over the week?

It’s Boston — probably because it’s a huge college town, with a huge population of young people.

Can you guess which one has the least variation in bedtimes?

It’s Las Vegas. “People who live and work in Las Vegas — if they’re in the industry of nightclubs and casinos, their schedule is going to be much less weekend-dominated,” Heneghan says.

Wake-up times also vary. This time, Seattle is the winner, with the least variation across the week. The losers here are New Yorkers, whose wake-up times swing an average of 73 minutes over the week. (Well, it is the city that never sleeps.)

The takeaways

The reporting that Fitbit’s sleep scientists offer in this exploration is only the very, very beginning. The company has amassed big data — big sleep data — that could provide some incredible answers. We just have to ask the right questions.

Why do men and women sleep differently? Why do we get less deep sleep as we age? Beyond the “party on the weekend” effect, why do our bedtimes vary so much? We know that exercise is good for our sleep, but when should we exercise for the best sleep? When should we eat if we want to get the most deepest sleep? Is the kind of sleep (REM sleep, deep sleep) more important than the total time asleep? Should the nation’s school hours and work hours be adjusted to fit the way we actually sleep?

Fortunately, Fitbit plans to share its data, both with other scientific institutions and in science journals; it’s thrilling to think of the new knowledge that may result from it.

Until then, consider trying to get to bed at a more consistent hour throughout the week. You’ll sleep better, you’ll sleep longer, and you’ll feel better once you’re up.

Or just move to Las Vegas.

David Pogue, tech columnist for Yahoo Finance, welcomes non-toxic comments in the Comments below. On the Web, he’s davidpogue.com. On Twitter, he’s @pogue. On email, he’s poguester@yahoo.com. You can sign up to get his stuff by email.  

Read more:

Tech that can help you keep your New Year’s resolutions

Pogue’s holiday picks: 8 cool, surprising tech gifts

Google’s Pixel Buds: Wireless earbuds for the extremely tolerant

Study finds you tend to break your old iPhone when a new one comes out

Rejoice: Sonos Speakers are finally voice-controllable

As Simon mentioned in his recent blog post about Raspbian Stretch, we have developed a new piece of software called PiServer. Use this tool to easily set up a network of client Raspberry Pis connected to a single x86-based server via Ethernet. With PiServer, you don’t need SD cards, you can control all clients via the server, and you can add and configure user accounts — it’s ideal for the classroom, your home, or an industrial setting.

Client? Server?

Before I go into more detail, let me quickly explain some terms.

• Server — the server is the computer that provides the file system, boot files, and password authentication to the client(s)
  
• Client — a client is a computer that retrieves boot files from the server over the network, and then uses a file system the server has shared. More than one client can connect to a server, but all clients use the same file system.
• User – a user is a username/password combination that allows someone to log into a client to access the file system on the server. Any user can log into any client with their credentials, and will always see the same server and share the same file system. Users do not have sudo capability on a client, meaning they cannot make significant changes to the file system and software.

I see no SD cards

Last year we described how the Raspberry Pi 3 Model B can be booted without an SD card over an Ethernet network from another computer (the server). This is called network booting or PXE (pronounced ‘pixie’) booting.

Why would you want to do this?

• A client computer (the Raspberry Pi) doesn’t need any permanent storage (an SD card) to boot.
• You can network a large number of clients to one server, and all clients are exactly the same. If you log into one of the clients, you will see the same file system as if you logged into any other client.
• The server can be run on an x86 system, which means you get to take advantage of the performance, network, and disk speed on the server.

Sounds great, right? Of course, for the less technical, creating such a network is very difficult. For example, there’s setting up all the required DHCP and TFTP servers, and making sure they behave nicely with the rest of the network. If you get this wrong, you can break your entire network.

PiServer to the rescue

To make network booting easy, I thought it would be nice to develop an application which did everything for you. Let me introduce: PiServer!

PiServer has the following functionalities:

• It automatically detects Raspberry Pis trying to network boot, so you don’t have to work out their Ethernet addresses.
• It sets up a DHCP server — the thing inside the router that gives all network devices an IP address — either in proxy mode or in full IP mode. No matter the mode, the DHCP server will only reply to the Raspberry Pis you have specified, which is important for network safety.
• It creates usernames and passwords for the server. This is great for a classroom full of Pis: just set up all the users beforehand, and everyone gets to log in with their passwords and keep all their work in a central place. Moreover, users cannot change the software, so educators have control over which programs their learners can use.
• It uses a slightly altered Raspbian build which allows separation of temporary spaces, doesn’t have the default ‘pi’ user, and has LDAP enabled for log-in.

What can I do with PiServer?

Serve a whole classroom of Pis

In a classroom, PiServer allows all files for lessons or projects to be stored on a central x86-based computer. Each user can have their own account, and any files they create are also stored on the server. Moreover, the networked Pis doesn’t need to be connected to the internet. The teacher has centralised control over all Pis, and all Pis are user-agnostic, meaning there’s no need to match a person with a computer or an SD card.

Build a home server

PiServer could be used in the home to serve file systems for all Raspberry Pis around the house — either a single common Raspbian file system for all Pis or a different operating system for each. Hopefully, our extensive OS suppliers will provide suitable build files in future.

Use it as a controller for networked Pis

In an industrial scenario, it is possible to use PiServer to develop a network of Raspberry Pis (maybe even using Power over Ethernet (PoE)) such that the control software for each Pi is stored remotely on a server. This enables easy remote control and provisioning of the Pis from a central repository.

How to use PiServer

The client machines

So that you can use a Pi as a client, you need to enable network booting on it. Power it up using an SD card with a Raspbian Lite image, and open a terminal window. Type in

echo program_usb_boot_mode=1 | sudo tee -a /boot/config.txt

and press Return. This adds the line program_usb_boot_mode=1 to the end of the config.txt file in /boot. Now power the Pi down and remove the SD card. The next time you connect the Pi to a power source, you will be able to network boot it.

The server machine

As a server, you will need an x86 computer on which you can install x86 Debian Stretch. Refer to Simon’s blog post for additional information on this. It is possible to use a Raspberry Pi to serve to the client Pis, but the file system will be slower, especially at boot time.

Make sure your server has a good amount of disk space available for the file system — in general, we recommend at least 16Gb SD cards for Raspberry Pis. The whole client file system is stored locally on the server, so the disk space requirement is fairly significant.

Next, start PiServer by clicking on the start icon and then clicking Preferences > PiServer. This will open a graphical user interface — the wizard — that will walk you through setting up your network. Skip the introduction screen, and you should see a screen looking like this:

If you’ve enabled network booting on the client Pis and they are connected to a power source, their MAC addresses will automatically appear in the table shown above. When you have added all your Pis, click Next.

On the Add users screen, you can set up users on your server. These are pairs of usernames and passwords that will be valid for logging into the client Raspberry Pis. Don’t worry, you can add more users at any point. Click Next again when you’re done.

The Add software screen allows you to select the operating system you want to run on the attached Pis. (You’ll have the option to assign an operating system to each client individually in the setting after the wizard has finished its job.) There are some automatically populated operating systems, such as Raspbian and Raspbian Lite. Hopefully, we’ll add more in due course. You can also provide your own operating system from a local file, or install it from a URL. For further information about how these operating system images are created, have a look at the scripts in /var/lib/piserver/scripts.

Once you’re done, click Next again. The wizard will then install the necessary components and the operating systems you’ve chosen. This will take a little time, so grab a coffee (or decaffeinated drink of your choice).

When the installation process is finished, PiServer is up and running — all you need to do is reboot the Pis to get them to run from the server.

Shooting troubles

If you have trouble getting clients connected to your network, there are a fewthings you can do to debug:

1. If some clients are connecting but others are not, check whether you’ve enabled the network booting mode on the Pis that give you issues. To do that, plug an Ethernet cable into the Pi (with the SD card removed) — the LEDs on the Pi and connector should turn on. If that doesn’t happen, you’ll need to follow the instructions above to boot the Pi and edit its /boot/config.txt file.
2. If you can’t connect to any clients, check whether your network is suitable: format an SD card, and copy bootcode.bin from /boot on a standard Raspbian image onto it. Plug the card into a client Pi, and check whether it appears as a new MAC address in the PiServer GUI. If it does, then the problem is a known issue, and you can head to our forums to ask for advice about it (the network booting code has a couple of problems which we’re already aware of). For a temporary fix, you can clone the SD card on which bootcode.bin is stored for all your clients.

If neither of these things fix your problem, our forums are the place to find help — there’s a host of people there who’ve got PiServer working. If you’re sure you have identified a problem that hasn’t been addressed on the forums, or if you have a request for a functionality, then please add it to the GitHub issues.

The post The Raspberry Pi PiServer tool appeared first on Raspberry Pi.

The Gordie Award for 2017 for happiest transportation story goes to the Arbutus Greenway. After a bitter battle which spanned a decade and a half an agreement was finally made for the City of Vancouver to buy the nine kilometer long railway bed for 55 million dollars from CP Rail. At the time in 2016 the Mayor of Vancouver got pretty evocative, calling this “Vancouver’s chance to have a New York-style High Line”. Of course Vancouver does not have the supporting density around the Arbutus greenway~yet.

The City of Vancouver has embarked on a public process to look at interim measures for the greenway and to examine proposed long-term measures, which include a tram line, a pedestrian linkage and a bike way. But go to any part of the Arbutus greenway in any weather and you will see Vancouverites running, walking, and pushing baby carriages along this new space. You can also connect with the City’s plans for the Arbutus greenway here. 

As well Price Tags Vancouver editor Ken Ohrn has been following the process and and you can review his remarks here.

This is getting curiouser and curiouser—these chips would be quite handy on new mid-level Macs.

Previously.

Mike Monteiro writes of Twitter CEO Jack Dorsey:

Jack let it happen. He watched as a once-entertaining, once-illuminating, once-vital network to global communication became a garbage fire of hate. He did nothing to stop it. Or curb it. He didn’t see a problem.

Our current crises of democracy and good faith did not just blow in with the wind and transform the air without our knowledge or consent.

These crises were made by people, and we knew what they were doing, and we agreed to this.

Jack Dorsey is one of those many people. Just one. But one with a kind of power that nobody in the world should have: the power to directly control a vast amount of the world’s communication.

It’s not that Dorsey failed to consider the good of the world. Or, really, it’s not just that. It’s that this kind of power should not exist at all.

But we agreed to it. We’re still agreeing to it.

Twitter — and Facebook, and the power of tech companies — is not our only problem.

But I have no doubt that had Twitter not become a loving home for hate, Trump would not be President now. In that universe we’d still have big problems, yes, but not like this.

How we can stop agreeing to this

The great social network is, or ought to be, the web itself.

The unruly web — unregulated and uncontrolled — is, perhaps paradoxically, the easiest place to limit hate. Not because we can stop people from publishing, but because we don’t have to live by Dorsey’s and Zuckerberg’s rules and designs.

I don’t know all the details of how we get there, or what it will be like once we do. That’s fine: that’s part of what makes the journey fun.

Software

Consider a few apps.

Overcast and Castro and others help ensure that podcasting is not just a vital and exciting medium of independent publishing but is also open and built on standards. Anybody can write any kind of podcasting software they want to — but nobody can control podcasting.

And nobody can force you to listen to hate. You pick the shows you want to hear.

MarsEdit lets you write whatever you want to write and publish it on the web. You’re limited only by the law and whatever terms of service your hosting provider may have.

All the words you read in MarsEdit are your own. And nobody can make you read what other MarsEdit users write.

Evergreen (which I’m working on), NetNewsWire, Reeder, Unread and other RSS readers work like Overcast and Castro but for written words. You choose what to read, and if a blogger you like suddenly turns hateful, you hit the Delete key.

Then there’s Manton’s new service, which needs its own section…

Micro.blog

It’s a publishing platform and a social network, based on standards.

You don’t even have to use Manton’s Mac or iOS apps: you can write posts in MarsEdit or other blog editor, or read your timeline using an RSS reader.

People could, though, sign up for it and flood your mentions with hate. In theory. So I asked Manton about that, and he wrote:

Micro.blog is similar to MarsEdit in a way in that it can be used to write hateful posts, etc. What we have to do as a social network is limit the damage. So, by default, if someone writes something terrible… No one sees it. It doesn’t automatically show up in trends (because we don’t have them, for this reason) and it doesn’t show up in Discover (because that’s curated by a human). Replies are where it’s an issue, and that’s where good tools and automatic flagging and reporting are needed.

In other words: the rules are different, and the easy exploits on Twitter are not so easy on Micro.blog. And there’s an actual committment to fighting this.

Manton also writes on his blog:

Imagine instead a service based on blogs, where the internal posts on the platform were the same format as the external posts. The curators of the platform would have more freedom to block harassing posts and ban nazis because those problematic users could always retreat to their own web site and leave everyone else in the community alone.

That’s how the web is supposed to work. It’s a core principle of Micro.blog.

(I’m @brentsimmons on Micro.blog, by the way. Here’s my microblog. I plan to post there more often than on Twitter in 2018.)

I should also mention…

Slack

I’m not sure where to put Slack in all this — except to say that admins control who’s in their groups, and I’ve never seen hate and harrassment there. I run a few groups, and I would have zero tolerance for this, and so would everybody I know who runs groups.

Well. One more thing about Slack: it meets some of the needs that Twitter used to meet. The talking-with-friends-and-family needs are very well covered there.

And the more we find ways outside of Twitter and Facebook to meet those needs, the less we’ll use Twitter and Facebook.

Maybe, back in 2012, Twitter did five important things for you that only Twitter did. I bet, in 2018, that that’s down to one or two.

Fun

The period from 1995-2008 (roughly speaking) was fun. It seemed like everybody was coming up with new things, and people were experimenting, and we were finding new joys in new connections, both human and technological.

Then, as Facebook and Twitter (and Google Reader; can’t forget that thing) grew, it’s as if we froze.

And those things were fun for a while, but they’re not now, and it’s obvious we made a mistake in allowing that much power to concentrate.

It’s time for the thaw: it’s time to get back to having fun. You’re free to make whatever you want.

What I’m Not Saying

Rebuilding the social open web is not the one cure that we need for all our ills. I’m fully skeptical of technological solutions to problems of culture and politics.

But it is an important thing we can and should do.

My small hope for 2018 is the knowledge that I’m not the only person thinking that way.

Related Reading

Me in 2011: What we talk about when we talk about RSS

Me in 2013: Why I love RSS and You Do Too

Anil Dash in 2012: The Web We Lost and Rebuilding the Web We Lost

IndieWeb is a thing I need to learn more about.

Tantek Çelik (video) - The once and future IndieWeb

Buch des Tages. New York Magazine hat einen Ausschnitt.

More >

On New Year’s Day, Gene Munster, former Piper Jaffray analyst and currently a venture capitalist at Loup Ventures, predicted that Amazon will buy Target. This prediction got picked coverage in hundreds of publications yesterday. Let’s look at why analyst types make predictions like this, how accurate those predictions are, and what they do for the … Continued

The post Analyst Gene Munster says Amazon will buy Target. Is he right? appeared first on without bullshit.

Mike Caulfield, Hapgood, Jan 06, 2018

"In short," writes Mike Caulfield, "the social media audience becomes one big training pool for your clickbait or disinfo machine." True. But even without social media, there will be no shortage of data for the machines. Consider, for example, the content from our learning management systems. Or our loyalty card programs. Or the telephone listings. There are issues with the input end - biased training data, for example - but the real problems are happening at the application end. That's when this data is used by machine learning algorithms to perpetuate stereotypes, create fake news, teach falsehoods, etc., and this can't be solved by the technology. Nor even by teaching people how to spot fake news. It's a social problem. It's a governance problem.

Jan 06, 2018

The Columbia Journalism Review is not above writing click-bait headlines, it seems (rememeber when clickbait headlines were the biggest problems in social media?). The idea has merit on first glance: make sure people have read the story before they can comment by asking them simple questions about what the story said. This did reduce the number of comments, but did not keep commenters on track. And critics point out that making it more difficult to engage with the story does not encourage people to engage with the story. When my high school English teacher used the same tactic on me I boycotted a year's worth of content quizzes.

Charlie Stross, Charlie's Diary, Jan 06, 2018

This is the text of science fiction writer Charlie Stross's address to the 34th Chaos Communication Congress in Leipzig in December. The speech rambles a bit but there are interesting reflections on how to predict the future (the key is combining the 85 percept of trends that will continue as expected and the small percentage that leave you wondering what happened), the role of corproations in society, the question of what AI wants, and what went wrong with it all (his explanation: the "mistake was to fund the build-out of the public world wide web—as opposed to the earlier, government-funded corporate and academic internet—by monetizing eyeballs via advertising revenue."). 

Matthew Rascoff, Learning Innovation, Jan 06, 2018

This article combines some much-needed optimism about educational technology (which has been in short supply lately) with some useful links. There's the HAIL Storm Network, Tsugi and NGDLE, Authorea (built on top of GitHub), and closer to the author's home at Duke, the OSPRI Lab's open source education technology project. 

Security researchers have discovered flaws in Intel, AMD and ARM Holdings computing chips that put nearly every modern computing device at risk to hackers.

According to researchers, there are two major bugs that are affecting devices — Meltdown and Spectre. The former flaw — which specifically targets Intel chips — allows hackers to bypass the hardware barrier between applications and a computer’s memory, which can leave any stored passwords exposed.

Spectre, meanwhile, affects chips from all three manufacturers and tricks applications into revealing private information.

Because the issues are at the chip level, devices that run Linux, macOS and Windows are all susceptible. However, researchers say Microsoft and Apple had security patches ready for affected desktop computers.

Intel has also acknowledged the flaw and said it is working on a solution with AMD and ARM Holdings. Incoming patches will need to be downloaded to update the operating systems and fix the issues, although doing so may slow down devices.

Fixes might not come so easily, though, said Daniel Gruss, one of the researchers from Graz University of Technology that discovered the Meltdown bug. Gruss told Reuters that Meltdown is “probably one of the worst CPU bugs ever found.”

Gruss said that Meltdown is the more serious current problem, although the bug’s Intel-specific nature makes it easier to patch with software. Spectre, on the other hand, will be harder to fix, he says, because it applies to nearly all computing devices. However, Gruss said Spectre is harder for hackers to take advantage of.

Via: Reuters

The post Major Intel, AMD, ARM security bugs may affect most phones and PCs appeared first on MobileSyrup.

(cross posted from NewCo Shift)

Ahead of CES 2018, General Electronic (GE) has revealed two new entries in its ‘C by GE’ line of smart lightning products. The new ceiling fixture and light switch are slated to arrive later this year and can be controlled with Alexa and Google Assistant.

The ceiling fixture is shaped like a disk with a circular speaker in the middle. As previously mentioned, the light supports voice control feedback, which allows users to give voice commands in order to control other smart gadgets in the home.

The ceiling light is also capable of answering questions with the assistant of the user’s choice and play music and trivia games. The warmth of the light can also be changed through voice commands as well.

GE’s new light switch supports the same features as the ceiling light, though one might find it odd to listen to music from an in-wall switch. The switch is also capable of connecting to GE’s C-Life Bulbs or C-Sleep Bulbs. The switch can also control regular lightbulbs it’s connected to and supports built-in motion, temperature and humidity sensors. The feature also suggests that GE might be considering working on its own products that control heat and humidity in the home.

Both of GE’s new devices feature built-in smart home hubs that allow them to control other C by GE lights as well. As of right now GE’s smart lights are not supported by HomeKit, though GE plans to add support for Apple’s smart home ecosystem in a few months.

GE will show off both products at CES 2018 starting this Sunday. GE has not shared Canadian pricing details yet, however, it’s likely it’ll retail for more than $180 CAD, considering the price of the C by GE Sol lamp. The company has confirmed that the products will arrive in Canada in Q2 2018.

The C by GE Sol lamp also works with Amazon Alexa and can utilize apps such as Spotify and SiriusXM.

The post GE’s upcoming smart ceiling light and switch support Alexa and Google Assistant appeared first on MobileSyrup.

This is from Radiohead’s recent A Moon Shaped Pool, which I’ve been listening to a whole lot, and oh my goodness what a beautiful song.

I came to Radiohead late, missed them completely on the way up. For years all I knew was they were the band behind the OK Computer hits on the radio, but I guess I’m a fan now. I like pretty well all of Moon Shaped Pool, and it’s an absolute production triumph, the kind of thing I put on when people ask why I still have a big hulking old-school audiophile system down at that end of the room. The songs are strong but the whole album is a carefully-conceived end-to-end piece of work, a huge wide swatch of beautiful sonic fabric. Kid A was trying to be the same thing, but the tunes and sounds are just more beautiful here.

Identikit mixes multiple melodies, all strong, and deploys the most beautiful instrument currently existing in popular music, namely Thom Yorke’s voice. I can’t make sense of the lyrics, but I only put about fifteen seconds into trying to make them out which is about ten seconds more than I usually devote to pop words.

Another thing to like here is the song’s clean emphatic ending. Maybe it’s because I listen to and sometimes play classical music, but I do think a great song should come to a great end; few pop musicians pay enough attention to this.

This is part of the Song of the Day series (background).

Links

iTunes, Spotify, Amazon, live video - more powerful, more official. It’s interesting to watch the song develop in the videos over the years.

Rumours that Spotify has plans to go public have been circulating for some time now, with the move long-expected to occur sometime between 2017 and 2018.

While the venture didn’t happen in 2017, the streaming service company is now one step closer to launching its IPO this year. According to Axios, Spotify confidentially filed IPO documents with the US Securities and Exchange Commission in late December.

Bloomberg, the filing suggests that Spotify is looking to list its shares in the first quarter of the year.

However, the news comes right as Spotify faces a new $1.6 billion USD lawsuit from Wixen Music Publishing over allegedly using thousands of songs without license or compensation. Therefore, it’s currently unclear if Spotify’s plans will be hindered at all by the pending legal dispute.

As it stands, though, Spotify is valued at around $16 billion and has over 60 million paying subscribers, putting it in a strong position to go public.

Source: Axios Via: Engadget

The post Spotify has quietly filed to go public appeared first on MobileSyrup.

On our way back from a lovely family afternoon out at the local streetcar museum, we hit up a McDonald’s for a quick bit of dinner on our way home. There we found a new promotion:

So far so blah.

But then I looked closer:

I’m upgrading my phone shortly, so hopefully you won’t have to suffer such poor-quality images in the future. But for now, allow me to transcribe:

“Not valid with any other offer. At participating McDonald’s restaurants in Ontario, Quebec, Atlantic Canada, Alberta, Northwest Territories and Lloydminster, SK.”

First: What is Atlantic Canada? I know it’s a colloquial designation for the four Eastern provinces (New Brunswick, Prince Edward Island, Nova Scotia, and Newfoundland and Labrador), but I didn’t expect to see it referenced in legalese at the bottom of promotional copy. Apparently the term is semi-legitimate, as there is an official arm of the federal government called the Atlantic Canada Opportunities Agency/Agence de promotion économique du Canada atlantique. Neat.

Second: Oh good, Quebec gets the promotion, too. Quebec gets left out of many things advertised to the whole of Canada because of stricter laws governing gambling (including sweepstakes. Tim Hortons has to bend over backwards to make Roll Up the Rim To Win work there) and advertising (especially to children).

Third: What did Manitoba, Saskatchewan, British Columbia, and the other two territories (Yukon and Nunavut) do to be left out? Nunavut seems obvious: there are no McDonald’s restaurants there. But there’s at least two golden arches in Whitehorse, and scads in Manitoba, Saskatchewan, and BC. Must be some sort of legal something.

And that brings us to the titular mystery: Why, of all of Saskatchewan, is Lloydminster spared? Is it the one city where there’s competition? Is it to do with that urban legend about an older burger restaurant in Western Canada someplace that was called McDonald’s first? Is it because it’s licensed under a special food services employer contract that….

No.

It’s because Lloydminster Saskatchewan is here:

It straddles the border between the provinces Alberta and Saskatchewan. Alberta has the McDonald’s promotion. Saskatachewan doesn’t. Lloydminster, AB has two McDonald’s restaurants. Lloydminster, SK has one. It would be unfair to deny the Albertan restaurants the promotion, and unfair to exclude the Saskatechewanian restaurant. So what is McDonald’s to do?

They have to add a rider on their promotion to all corners of the Great White North that proclaims that there is one city (barely) in Saskatchewan in which you can partake of a five dollar meal deal.

Yeesh.