Mass surveillance silences minority opinions, according to study • Hacker News Board

Re: Mass surveillance silences minority opinions, according to study

#201

Post by nitrogen » Wed, Mar 30, 2016, 5:47 AM UTC

p4wnc6 wrote: post #198

nitrogen wrote: post #197

Earlier quoted context omitted.

It only takes about ten redundant comments (wild guess number) to derail a thread for thousands of would-be participants. This doesn't indicate community value; at best it indicates accidental nerd sniping, and at worst could be a coordinated attack on what would otherwise be a productive discussion, from which more people could derive far greater value.

But the comments are drawn from the distribution of Hacker News readers. You can't presume that just because you see a thread of ten comments as an off-topic distraction that so must the other thousands of readers, especially if there are healthy replies and up votes. Your comment that it "derail[s] a thread" makes the implicit assumption that the comments are actually off-topic. If they are on-topic, but you happen…

Interesting points. I'll just add that I don't think that enjoyment is the best metric for evaluating a thread, but maximal enjoyment and/or maximal usefulness. Especially if what I'm calling off topic prevents the community from addressing a threat to its values. I could be wrong, but it seems that the HN team want HN to be more than just a fun place to hang out for hackers.

There was another example of derailment recently, with a dozen comments arguing over what a "middleman" is when the original comment was really about how best to support indie makers.

Anyway, thanks for the interesting discussion, and I hope I haven't made this thread less useful for someone else in the process.

Re: Mass surveillance silences minority opinions, according to study

#202

Post by daemin » Wed, Mar 30, 2016, 8:39 AM UTC

p4wnc6 wrote: post #190

daemin wrote: post #189

Earlier quoted context omitted.

But a company generally doesn't want diversity, they want predictability. Predictability as to how fast and when something will be done, predictability on the quality of the work. Sure having thought diversity in solving problems is a good thing, but look at how complicated software project scheduling is already.

This usually applies to commodity work, and there's a long history of Taylorism. It makes a lot less sense in knowledge work, where the problems are not standardized and often solutions are unknown. In these cases, the uniformity desired through these surveillance measures makes no sense. It's fundamentally not the same as a McDonald's burger, no matter how much management wishes so. Worse still, even if you did want…

The whole having to be productive in an otherwise anti-productivity environment seems like some sort of weird hazing ritual.

Though I still think that regardless what job one is doing (knowledge work or factory), that subconsciously the company wants predictability. So maybe putting people in an environment where it is anti-productive puts people of all skills at the same basic predictable level?

Re: Mass surveillance silences minority opinions, according to study

#203

Post by p4wnc6 » Wed, Mar 30, 2016, 3:48 PM UTC

daemin wrote: post #202

p4wnc6 wrote: post #190

Earlier quoted context omitted.

This usually applies to commodity work, and there's a long history of Taylorism. It makes a lot less sense in knowledge work, where the problems are not standardized and often solutions are unknown. In these cases, the uniformity desired through these surveillance measures makes no sense. It's fundamentally not the same as a McDonald's burger, no matter how much management wishes so. Worse still, even if you did want…

The whole having to be productive in an otherwise anti-productivity environment seems like some sort of weird hazing ritual. Though I still think that regardless what job one is doing (knowledge work or factory), that subconsciously the company wants predictability. So maybe putting people in an environment where it is anti-productive puts people of all skills at the same basic predictable level?

That is a good point. I had never considered that they would actually Harrison Bergeron their own people, but it's entirely plausible. If they want predictability more than output, it would make sense, and then they extract value mostly just from your credential and your organizational fealty.

Re: Mass surveillance silences minority opinions, according to study

#204

Post by dreamfactory2 » Fri, Apr 01, 2016, 6:20 PM UTC

p4wnc6 wrote: post #190

daemin wrote: post #189

Earlier quoted context omitted.

But a company generally doesn't want diversity, they want predictability. Predictability as to how fast and when something will be done, predictability on the quality of the work. Sure having thought diversity in solving problems is a good thing, but look at how complicated software project scheduling is already.

This usually applies to commodity work, and there's a long history of Taylorism. It makes a lot less sense in knowledge work, where the problems are not standardized and often solutions are unknown. In these cases, the uniformity desired through these surveillance measures makes no sense. It's fundamentally not the same as a McDonald's burger, no matter how much management wishes so. Worse still, even if you did want…

The entire offshoring software industry is already Taylorised. I'm not saying it's good or efficient but it is the reality of large software projects.

The architects are onshore and have never coded in their lives. They draw boxes on whiteboards and feed UML to offshore dev 'factories'. That UML gets decomposed so that you will have factory lines of developers producing classes (sometimes just methods) in isolation. LOC is frequently used as a productivity measure by their management. It's about as far from software as craft as you can get.