Please create an account to participate in the Slashdot moderation system

typodupeerror

Comment Re:how often we can expect conspiracies fail (Score 1)303

You can't make a conclusion about mean or average _excess_ speed, or prevalence of speeding on highways just by working with the population of all the measured drivers and the caught drivers. For there is a subgroup of drivers who successfully evade such measurements, e.g. by featuring radar-detection tools, obtaining info about speed detector locations, having the connection to erase records, favoring unmetered places or times etc. You can establish lower _bounds_ on speeding behavior but the _real_ numbers can be vastly different (e.g. some large multiple of the lower bounds, might be a small increment, or potentially some order of magnitude).

The GP's point is that, since conspiracies never uncovered weren't (and obviously couldn't) be included in the study, the whole exercise attempts to give the lowest bounds on how long it may take to have a conspiracy unravel. For example, take the climate change example from the study. Its higher range is 26.8 years. If, however, lots of conspiracies are never uncovered, then it means that, were they to be uncovered and included in the study, the number of years would probably become much larger. So we have an already pretty comfy 26.8 years for a conspiracy - a time span enough for the original conspirators to achieve their purpose, go through their productive lives till retirement and be comfortable, not to mention a lot of crimes aren't punishable after some number of years - and it is feasible that it's just a lower bound (again, because the hypothetical inclusion of uncovered conspiracies would imply reduced chance of finding out).

Yes, one can say that 'sure but working with the numbers of proven cases can still be representative'. Not really, because it's a _model_, and nobody proved this model, moreover, there may be systematic factors that separate successful and unsuccessful conspiracies which, in terms of metrics, would have the same input and output numbers in this specific study.

As an example, think of a group of 5 with small children vs. hardened criminals. It's more likely that the former group will not keep a secret for a day and the latter may stick to their wows forever. Same model, huge difference. So unfortunately this study is actually optimistic, because the stronger the reason for maintaining secrecy, the more likely it is that politicians, secret services, military and other, more resilient actors make up the conspiracy, and more likely that good planning went into keeping secrets within a small group (mafia organizations practice this art since forever). So as we move toward the more and more interesting and relevant conspiracies, so decreases the reliability of the proposed model.

I believe we can safely say that a conspiracy that stays secret for 4-27 years based on a small sample size study that obviously couldn't factor in uncovered conspiracies, it's not only believers of conspiracy theories who conclude that 'yeah, numbers support the sentiment that there can be vast conspiracies that go undetected during my lifetime, but for sure there's a large likelihood that a good number of conspiracies on crucial matters must be going on all time'.

Comment Underwater? (Score 1)29

I haven't RTFA but what's underwater specific about it? The haptic gloves stop working above sea level? If you have haptic gloves - basically a rendering device like a computer screen - you can show anything on it, like virtual objects in a 3D scene, or ... surface scenery, or cloud distributions (I mean old fashioned, floating clouds in the air) or even remote galaxies.The very definition of a rendering device is that you can render whatever you want, up to the constraints of the device (resolution etc.) which all have nothing to do with what is rendered.

Submission + - Y Combinator, the X Factor of tech (economist.com)

universe520 writes: Since 2005 YC has taken on batches of promising founders, and this month will celebrate the funding of its 1,000th startup. Though about half of its startups have failed, which is typical of early-stage investing, it has had a head-turning record of success. In addition to Airbnb, YC has had a hand in Dropbox, a cloud-storage firm, and Stripe, a payments company (see table). Eight of its firms have become what Valley folk call “unicorns”, valued at \$1 billion or more. Combined, the companies it has invested in are worth around \$65 billion (based on their most recent funding round), although YC’s share is only a small fraction of that total—perhaps \$1 billion-\$2 billion. It is because of this record that YC has become a juggernaut in Silicon Valley.

Comment Re:/sigh (Score 2)60

This. It's not even more sensational if it fails after 'seconds' rather than in a minute, so this is probably a routine journo hyperbole, the problem is, as you say, that they write without considering the target audience (i.e. they don't give a shit about them).

Comment Re: Pretty standard procedure on a large campus (Score 1)284

This. While most replies assume it's okay for local responders to triage, it's just a waste of time. Even calling security adds a layer of indirection. In case of emergency, call 911 and then security - this way the ambulance will be on the road and security will still have ample time to give guidance to them.

Comment Re: Bad news for them (Score 1)129

The fact that people obviously versed in the art are commenting on Slashdot, rather than excitedly hunching over the algorithm, putting it in whatever they work on, is telltale sign that it's not as revolutionary as the gushing summary states.

Comment Re:Massive Scientific Visualization (Score 1)111

The simulation part is very performance intensive, but the visualizations themselves look like something you could do with WebGL, or often, just some SVG and CSS. What are the thousands of cores used for? Not even a super-high resolution seems warranted, because of the continuity of material properties etc. Apparently the result is some 3D model which can be interactively rotated and zoomed, likely on a single local machine that takes direct input from the user, i.e. the thousands of cores don't even seem to do the real time part of the rendering.

Slashdot Top Deals

Were there fewer fools, knaves would starve. - Anonymous

Working...