Datasets:
Bingsu
/
openwebtext_20p

Dataset card Data Studio Files
xet
Community
text
stringlengths
1
100k
The result of the voting on the election bill is no longer relevant to the future of Hong Kong’s democracy.
If it is passed, we will all have our shameful share of the big swindle. If it’s voted down, it merely reaffirms what we’ve been saying all along — we prefer nothing to something fake.
Rejection of the bill won’t lead us back to the right track either, and the Hong Kong government will simply continue with its naked lies about freedom, democracy and universal suffrage.
Life will go on but the years to come may not be the kind of future we want.
The odds of seeing a truly democratic Hong Kong are tremendous. And we may be seeing the last of Hong Kong as we know it — stable and prosperous — as we stagger into an even more uncertain future.
This article appeared in the Hong Kong Economic Journal on June 17. It was written before the Legco vote on the election bill.
Translation by Frank Chen
[Chinese version 中文版]
– Contact us at [email protected]
RA
Introduction
On Feb. 1, 2017, the United States led Coalition was accused by local activists and journalists that it had bombed the headquarters of the Syrian Arab Red Crescent (SARC) located in the Carlton Hotel in the city of Idlib, Syria. Photos and videos allegedly showing the bombed headquarters quickly emerged online. This open source investigation aims to verify the various claims made surrounding this incident.
Content
Summary
Claims
Methodology
Open Sources
Findings
Summary
Based on open source research, it can be confirmed that the Syrian Arab Red Crescent (SARC) utilised the Carlton Hotel in Idlib city as a headquarters. It can also be confirmed that the Carlton Hotel was recently severely damaged. The US-led Coalition was accused of conducting an airstrike on the building in the early morning of Feb. 1, killing one SARC member, but the Coalition denies any involvement. Neither the exact date of the attack nor the perpetrator can be established based on open sources. What can be observed, however, is that a bomb appears to have pierced at least four floors before detonating.
Claims
The following claims are circulating on social media:
One or more airstrikes conducted by the US-led Coalition destroyed a part of the Carlton Hotel in Idlib city. The Carlton Hotel was used by the SARC as headquarters.
This open source investigation aims to verify those claims by answering the following questions:
Is there a Carlton Hotel in Idlib city? Was the Carlton Hotel in Idlib was used by the SARC as headquarters? Was the Carlton Hotel in Idlib was bombed? If so, was it bombed by the US(-led Coalition)?
Methodology
The investigation uses only openly available sources on the Internet. In addition, a request to comment on this incident and the accusation was sent to the Coalition’s public affairs office.
Sources were found by, for example, searches on social media using certain keywords of internet. One can think of a search on ‘Red Crescent Idlib’ in Arabic: “الهلال الأحمرادلب‎‎”. These social media include Facebook, YouTube, Twitter, and Instagram. After the exact location of the incident was found, the coordinates have been used to conduct location-based searches in a 2-km radius on those same social networks. Reference photos have been found via geotagged pictures on Panoramio. All YouTube videos have been checked for the date and potential other, earlier uploads using Amnesty International’s YouTube Data Viewer, and images Google’s reverse image search has been used to see if the photos were not posted before the date of the attack. Bellingcat has published several how-to guides to for beginning or advanced open source investigations. All sources but one has geolocated.
Open Sources
These are the main open sources found and used outside of the use of freely available mapping data and satellite imagery via Wikimapia, Google Earth, Google Maps, and Bing Maps. Pro-opposition media, local activists and organisations have shared videos and photos showing the same severely damaged building, and large number of them accuse the Coalition for conducting the airstrikes:
Al-Baladi News also published a video showing SARC employees in a ravaged building [link/archived], and Al Jazeera Arabic also sent a reporter to the scene who said it is not clear who conducted the airstrike [link/archived].
Findings
1. Is there a Carlton Hotel in Idlib city?
Yes, based on open sources it can be confirmed that there is a Carlton Hotel in Idlib city.
First of all, the hotel (Arabic: فندق كارلتون) is listed as a four-star hotel on hotel booking websites (for example, hotelscombined.com [archived]) and Google Maps [archived].
The hotel is located in southern part of Idlib city, at the coordinates 35.9206001, 36.634748 (Wikimapia).
Secondly, the hotel is referred to in media, for example by esyria.net [archived] and dp-news.com [archived]. The articles are an interview with the hotel director and a news report that the hotel closed its restaurant in 2011, respectively.
Thirdly, photos that are geotagged on Panoramio, or that can be geolocated to the area, indeed show a building which bears a sign ‘Carlton Hotel’, including a logo and four stars.
2. Was the Carlton Hotel in Idlib was used by the SARC as a headquarters?
Yes, based on open sources it can be confirmed that the SARC used the Carlton Hotel in Idlib as a headquarters.
First of all, nearly forty SARC members of the Idlib branch posed with four ambulances in front of the Carlton Hotel in a picture uploaded to their Facebook page on Jan. 17, 2017. The same Facebook page also reported the attack which destroyed ‘most of [their] administrative offices but didn’t injure humans’ [link/archived]. Photos posted in Feb. 2016 shows that the Carlton Hotel was already used by SARC at that time [photo 1: link/archived; photo 2: link/archived; photo 3: link/archived; photo 4: link/archived]. Members of the SARC team declared to the Website for the Syrian Revolution that the building does not contain any headquarters of military factions.
Secondly, the official Twitter account of the delegation of the International Committee of the Red Cross in Syria condemned the bombing, “a clear violation of [international humanitarian law”, hours after the attacked reportedly happened. This tweet thus suggests that the ICRC Syria is aware that there is an aid branch operating in the destroyed location, which brings us to the next point: Was the Carlton Hotel in Idlib indeed bombed?
3. Was the Carlton Hotel in Idlib was bombed? If so, was it bombed by the US-led Coalition?
Based on open sources, it can be established that the west wing of the Carlton Hotel has been severely damaged. Below are photos and screengrabs from videos showing that damage.
All footage taken outside of the building can be geolocated to the immediate surrounding of the Carlton Hotel. The map below shows all locations from where photos or videos of the ravaged hotel were taken.
In a news report [archived], MICRO SYRIA claims that the Carlton Hotel, which houses the Red Crescent, was targeted by unknown aircraft at three ‘o clock at night of February 1, 2017. A reconnaissance aircraft was seen in the sky an hour before the airstrike happened, media activist Ammar al-Adalba told Website for the Syrian Revolution. He said that it was most likely an airplane from the Coalition that struck the SARC headquarters, resulting in destruction and fire inside the building.
As there is no (open or commercial) satellite imagery available around the date of the alleged strike, the exact date of the destruction cannot be confirmed.
There are no obvious visual indicators, like weapon remnants, that could identify the perpetrator of the attack, or the cause of the damage. There is, however, footage of an object that appears to have pierced at least four floors of the building.
A rescue worker interviewed by Step News Agency claims that “a missile went from the roof down to the first floor.”
Some have suggested the damage at the Carlton Hotel may be due to a delay-action bomb, which is designed to explode some time after impact (think of a bunker buster, for example). Building upon that claim, it has been suggested that it must have been either the Coalition or the Russian Air Force that conducted the bombing; the Syrian Arab Air Force (SyAAF) is believed to be incapable of delivering such bombs during night time. However, as Tom Cooper detailed in this War Is Boring piece, may well have the capacity to operate at night.
There is currently no visual evidence to confirm the use of a delay-action bomb, so the above remains speculation.
In reply to a request of information, the Coalition’s public affairs office e-mailed Bellingcat that “the Coalition did not conduct airstrikes in Idlib on Feb. 1, 2017.”
With regards to injuries and casualties, the estimates by the local sources range from nine to eleven individuals that were injured.
At least one ambulance has also been severely damaged in the incident, photos and videos show.
Update: In a previous version, the article stated that an e-mail was sent to US Central Command. This is not correct, the e-mail was sent to the US-led Coalition. The reply came from the Coalition’s press desk, which is was an e-mail-address associated with the US military.
Update: An earlier version of this hotel incorrectly mentioned that the Ghreir family was killed in this incident. That is not correct, the Ghreir family was indeed confirmed by SARC as being killed in Idlib, but in a separate incident.
Ad blockers are often painted as the enemy of online publishers, but sometimes things are more complicated.
AdBlock Plus, for example, just announced that they’re working with startup Flattr on a new product that allows readers to pay the publishers who produce the content they read, listen to and watch.
As a result of the partnership, AdBlock Plus said it has also made a small investment of undisclosed size in Flattr .
Together, the two companies have created a new product called Flattr Plus. Like Flattr itself, it allows users to allocate a monthly budget that they want to pay publishers. Unlike Flattr, users don’t have to click a button to “Flattr” a website — instead, it will automatically track their browsing activity and distribute the money based on their engagement.
It sounds somewhat similar to the way a company like Spotify distributes subscription fees to musicians — except it’s not just for artists on a single website or app.
Plus, the question of exactly how to calculate engagement is a tricky one. You probably don’t want to reward a worthless article with a dumb-but-effective clickbait headline. You might also leave an article open for hours without actually reading it.
Ben Williams, who leads communication and operations at AdBlock Plus, told me that the product is still in beta testing (the plan is to do a full launch later this year) partly so the team can experiment with ways to measure engagement — it will involve some combination of factors like time spent and scroll activity.
Publishers will have to sign up with Flattr Plus if they want to get paid, b ut Williams said that if they’re don’t, the money they’re due will be held for them until they join the program . (Update: Williams said that actually, AdBlock Plus won’t hold the money — it’ll just tell them how much money they could have earned.)
The goal, he added, is to earn half a billion dollars in revenue for publishers next year. He doesn’t necessarily expect every AdBlock Plus user to volunteer to pay, but he predicted that many will — and when you’ve got 500 million downloads, just a small percentage of users paying a few dollars a month can add up.
In fact, he said the AdBlock Plus users who opt out of seeing any advertising whatever (even if it’s part of the company’s acceptable ads program), are the ones who “have been the most vocal in asking for solutions like this.” Basically, these users have told the company they don’t like any ads, period, but they still want to support publishers and creators. Now we’ll get a chance to see if they meant it.
Oh, and if you want to hear more about Flattr Plus and AdBlock Plus’ broader vision, I’ll be interviewing CEO Till Faida next week at Disrupt NY.
F ancy cars have always been an important element in rap music. You can find many articles online talking about what cars rappers love the most, but they all lack the research, running their mouth with no backing evidence. Is rappers’ preferred ride Cadillac, Mercedes, Bentley, or Ferrari? Let’s turn to data science to settle this debate once and for all. By analyzing all lyrics on Rap Genius , we’ll see which rides have been celebrated the most. Check the results...
Digging deeper to look at songs released throughout the years, the following is a time series analysis to see which makes have gained popularity and which have lost favor. An interesting correlation here is a dip for some of the most frequently mentioned luxury car makes around 2008, 2009 financial crisis. Shout out to Johnny McNulty at Someecards who have some interesting analysis around the trend’s correlation with economy here .
As seen from the chart, the most frequently mentioned car make is Mercedes Benz. Rappers love to rap about their benzes, but they rarely mention what models. If we get down to the specific model that’s most frequently mentioned, there’s no debate — it’s the Chevy Impala, most specifically the ’64 year model.
The chart above is an overall view. We’ll break it into several categories in the following:
Top 3 Makes
There were more mentions of Cadillac relative to Chevrolet between 1997 to 2006, but the trend inverted since 2007. Benz managed to stay on top most of the time, except between 2004 and 2007 where it was briefly surpassed by Cadillac.
Makes Gaining Popularity Over Time
The following are makes that have been on an upward trend over the years. Worth noting that Porsche, Ferrari, Maybach, and Bugatti peaked around 2012/2013 but have declined since.
Makes Losing Popularity Over Time
The following are makes that have declined or maintain the same level of mentions in rap songs over time. Due to the increase in rap songs released over the years, these are makes that are losing popularity amongst the artists. Jeep and Hummer serve similar functions and have very similar levels of mentions across time. Lexus was extremely popular around 1998, but has since lost steam.
Endorsements
Some major car companies benefit from rappers as endorsers, a task the rappers often take on voluntarily. But who is doing the best job as pitchman? Seeing song mentions as a form of endorsement, the following is an analysis of the top endorsers.
Turns out The Game dominates the car endorsement game. The Compton artist wins the category hands down for seven of the major car makes, having logged more automobile mentions than any other rapper. The following are The Game’s most-endorsed cars, in order of frequency:
Chevrolet Impala
Chevrolet Impala
Chevy mentioned in 93 of The Game’s songs, 82 of them mention Impala. His love for Impala is expressed in his song “In My 64.”
The eight World Cup second round matches are spread over the course of four days. Here’s previews for the first half…
Uruguay v South Korea
Uruguay have been one of the most impressive teams so far – playing for and achieving a draw against France, destroying South Africa and recording a solid 1-0 victory over Mexico.
They started the competition with a 3-5-2 shape, which became more like a 5-3-2 when the wing-backs had to contain France’s wingers. They’ve since switched to a 4-3-1-2 with Diego Forlan playing behind the main two forwards, and they’ll surely play the same formation after their two wins.
South Korea’s first XI is fairly predictable. The only changes they’ve made so far have been at right-back, bringing in Oh Bum-Suk against Argentina – but he was the worst player on the pitch, so Cha Du-Ri has regained his place.
The formation will probably be 4-2-3-1. Playing Park Ji-Sung on the left-hand side might be useful to track the forward runs of Maxi Pereira, although he was fielded in the centre of the three against Argentina.
Picking up Forlan is the obvious task – with two holding midfielders, Korea will have a man tracking him, but must worry this will concede the midfield ground to Uruguay. Korea should look to play down their left-hand-side, because Uruguay’s shape tends to be slightly lopsided. Alvaro Pereira, generally a left wing-back, is playing a more central role but tends to drift back out wide, sometimes meaning Uruguay look like two banks of four minus a right-sided midfielder.
USA v Ghana
The US start as favourites, but this one might suit Ghana tactically; they will be content to sit back and soak up pressure, before hitting the US on the counter-attack. The American full-backs have appeared a little slow in recovering their position after forays forward so far in the tournament, which will be perfect for the pacey Ghana wingers.
The best course of action for the US is to put the Ghana centre-backs under as much pressure as possible early on. Ghana will probably have a 3 v 2 advantage in the centre of midfield, so more direct balls towards the strikers (with Clint Dempsey and Landon Donovan supporting very close by, something they did particularly well against England) might be a better approach than playing through midfield. In particular, 20-year-old Jonathan Mensah has looked slightly nervy so far, and Jozy Altidore should be able to get at him.
The midfield battle will probably be quite reserved. Both central midfields generally sit deep rather than look to make penetrative runs, so they may play in front of each other, and create a slightly static contest.
The US will dominate possession and territory – but finding a way past Ghana’s good defence won’t be easy. This one calls for all of Bob Bradley’s tactical ability – so far he’s generally got his team playing better after half-time, but below-par first halves won’t be acceptable in the knockout stages.
Germany v England
If the two sides perform to the standard as they have so far in the competition, then England are in for a thrashing. First and foremost, their ball retention must be far better.
As with all 4-2-3-1 v 4-4-2 battles, the main task for Fabio Capello is to work out how to deal with Mesut Ozil – the match-winner for Germany against Australia and Ghana. Those two teams both allowed him far too much space between the lines, and it’s likely that Gareth Barry will have the task of tracking him, something he did well against Algeria.
This would mean a numerical disadvantage further forward in midfield, where Frank Lampard would be forced to pick up the runs of both Sami Khedira and Bastian Schweinsteiger, so it’s likely one of England’s two strikers will be given more defensive responsibility when out of possession.
The natural man to do this is Wayne Rooney, who has consistently shown his defensive awareness throughout his career, particularly when playing on the wing for Manchester United. However, Jermain Defoe has done well in recent months in this respect. Capello won’t want Rooney to become overburdened defensively, nor will he want Defoe playing a permanently withdrawn role and negating the threat of his pace in behind the German defence – so it’s likely they’ll take it in turns to pick up the Germans’ deepest holding midfielder – most likely Schweinsteiger, if fit.