Marine Corps Study on Women in Combat ‘Flawed’ and Incomplete 

Last month, the Marine Corps released a year-long study that tested women’s readiness to serve in combat units. At the time of the study’s release, NPRcalled the results “stark,” adding that the “all-male units outperformed mixed-gender units across the board.” That study came months before January 1 deadline to integrate the military by opening all jobs to women. The then Commandant of the Marine Corps, Joseph Dunford, reportedly weighed the study as he prepared “to make a crucial decision regarding the integration of female troops into closed combat roles.” According to reports, Dunford made his decision prior to being promoted to the head of the Joint Chiefs of Staff in September.

Now it looks that the study was not only deeply and fundamentally flawed, but that the Marines didn’t release the entire study.

The Guardian reports:

“[..] critics said the corps’ decision to release only a four-page summary of its study, which focused on negative aspects of women as marines while keeping the bulk of the study under wraps, was ‘hugely problematic.’

One of the report’s released conclusions found that: ‘The integration of females … will add a level of risk in performance/effectiveness and cost … The bottom line is that the physiological differences between males and females will likely always be evident to some extent.’

But other pages from the report, seen by the Guardian, indicate that women were not expected to pose problems for ground-combat units, so long as clear standards, diligent screening of candidates and good training and leadership were in place.”

Other unpublished parts of the study found that mixed-gender units had “superior problem-solving” and better discipline. The four pages released by the Marine Corps emphasized that men were better at long distance hikes with heavy equipment and their lower rates of injury.

The Marine Corps is believed to be the only branch of the military to ask for an exclusion to the gender integration mandate. The recommendations were made to Secretary of Defense Ashton Carter in early October. Carter will announce his decision before the January 1 deadline.

Round Two

Last month, the Marine Corps released a year-long study that tested women’s readiness to serve in combat units. At the time of the study’s release, NPR called the results “stark,” adding that the “all-male units outperformed mixed-gender units across the board.” That study came months before January 1 deadline to integrate the military by opening all jobs to women. The then Commandant of the Marine Corps, Joseph Dunford, reportedly weighed the study as he prepared “to make a crucial decision regarding the integration of female troops into closed combat roles.” According to reports, Dunford made his decision prior to being promoted to the head of the Joint Chiefs of Staff in September.

Read more…

http://ift.tt/1GM5SjE

View original post

Reddit Wants to Combat Its Homepage Problems

‘You can’t code an algorithm and expect it to work forever’

Users of Reddit have bee complaining about the site's algorithm, which is slow to update. (Photo: Reddit)

In an ironic twist, some of the most popular recent threads on Reddit have been about the site itself—specifically, the problems with its front page.

It all started on September 22, when Reddit’s chief technology officer Marty Weiner took part in an AMA with users. The most common complaint was that the homepage content was “static”—for example, one user pointed out that on September 3, the day Kentucky clerk Kim Davis went to jail and Tom Brady won his “Deflategate” case, none of the posts on the front page addressed these topics.

Mr. Weiner and his staff went on the defensive, saying both that the problem may be imaginary and that it may have existed for years:

Capture-19

This answer contented users until last week, when the shooting at Umpqua Community College took over news feeds on every social network—except Reddit. As Motherboard first pointed out, stories about the shooting didn’t reach Reddit’s front page until 12 hours after the event.

Multiple threads popped up, both complaining about the site’s issues and trying to guess why they exist in the first place:

(Photo: Screenshot)

(Photo: Screenshot)

The problem seemingly lies with Reddit’s algorithm. To begin with, the site is structured so that stories that get a lot of downvotes (i.e. those with negative or controversial topics) are ranked lower on the site. No matter how important it is, a story about a school shooting would not do well under this criteria.

Also, Reddit’s popularity has skyrocketed in recent years—there are now 36 million user accounts. Casual users now treat the site like any other social network, skimming the homepage and upvoting stories they like. More upvotes gives a story higher placement on the site—and a funny animal video will always be more popular than news about murdered college students.

Web developer Chris Fohlin told the Observer that Reddit, like any website, has to adapt in order to succeed.

“You can’t code an algorithm and expect it to work forever,” Mr. Fohlin said in an email. “The flow of content and behavior of users changes over time.”

To its credit, Reddit is working on ways to fix this. Mr. Huffman said that his team is revising the in-house simulator which approximates an average day on Reddit. These new figures will hopefully show how to keep the site interesting, and thus keep users coming.

One long term fix that Reddit is considering is personalized homepages—once a user has clicked on a post, it would simply disappear from the homepage.

The fact that it took a while for Reddit to discover these new metrics for user engagement is a function of its size, according to Mr. Fohlin.

“The larger a company grows, the slower it moves,” Mr. Fohlin said. “Ultimately, users have to decide if they’re willing to stick around.”

Reddit did not reply to an Observer request for comment.