Slashdot is powered by your submissions, so send in your scoop

 



Forgot your password?
typodupeerror
Check out the new SourceForge HTML5 internet speed test! No Flash necessary and runs on all devices. Also, Slashdot's Facebook page has a chat bot now. Message it for stories and more. ×

Comment Re:Reproducibility is hard. (Score 5, Insightful) 307

Why are we still using printed journals?

Why is the amount of space a report takes up still an issue?

Details are important. If you want a short version, then make a summary, but don't cut out the detail available to do that.

In terms of ascii/unicode text, we're not going to run out of bytes to explain important scientific details.

Heck - make videos of the processes, mention part numbers, and even show mistakes that you encountered along the way in your notes! Video hosting is free, and shouldn't be going away anytime soon. Making a process replication video should be a normal thing.

If you're spending so much time anyway, so much of your life in these studies, what's the value in holding back important information?

Ryan Fenton

Comment Cool? (Score 5, Insightful) 123

I'm a pretty liberal dude - but this age-information-protection thing is the wrong role for any governance to be playing.

It's an objective, publicly available piece of information. Birth records aren't secret, or in any way protected from public view. Trying to punish websites for listing that among other pertinent details on public figures like actors is just crazy.

That's not to say age discrimination is an unrealistic thing to fear - but this is exactly the wrong way to combat it, akin to punishing kids spreading rumors of an upcoming fight, rather than any of the participants. It's just bad tactics too - objecting to information only spreads that information further (justly called the Streisand effect).

I'm struggling just to wrap my head around how stupid an idea this law was, or who would propose it as a valid way to use law.

Was this some kind of a protest law, or a game of legislative chicken gone wrong?

Ryan Fenton

Comment Useless? That article. (Score 5, Insightful) 722

Here's the thing - basic income CAN theoretically not work out... but some an economist with a stake or two against it working is NOT evidence that this version of it hasn't panned out. Especially when it's posted on fricken Bloomburg news!

That's what the experiment is for. Instead, it's to see if the money spend on THIS style of program is as effective as the several other programs it can replace, and whether that replacement will be practical. It's money that will be spent in any case! You need experimental comparison to judge the merit of the approach.

Again though - until RESULTS are in, hearing some talking head berate the idea of it as not to his liking isn't helpful.

It's like folks who dismiss needle exchange programs to reduce communicable disease, without actually bothering to look at the numbers, and what the studies actually account for.

Ryan Fenton

Comment Science! (Score 1) 268

This is more on the entertainment side, but also feature some of the deeper discussions on important topics than you'd find almost anywhere:

This Week in Science (TWIS) - going for over a decade, and still just as energetic, and honestly hilarious as ever. Just the right mix of solid detail and genuine humor.

Skeptics Guide to the Universe - has also been going for over a decade, and has some of the warmest, funniest folks out there. Lots and lots of science too - since at the heart of modern skepticism is the drive to understand why we can know things more than others.

Data Skeptic - relatively new, but really good, deep dives into what makes meaningful data, in a very entertaining manner.

As with most all podcasts, just make sure whatever you're listening through has a handy 30-second skip to jump past any sponsor bits, they're usually quite well-labelled in those podcasts.

Comment Yeeeees! (Score 1) 262

I'm ALL FOR mouse controls on consoles!

Why? I want the design for mouse to be something at the forefront for developers, alongside controller support.

For the past half-decade or so, UI developers for cross-platform projects really seemed to give mouse/keyboard users a raw deal over interfaces, acting as though everything was just emulating a joystick, with horrible positive/negative acceleration logic, capping allowed movement per second, etc. Really bad controls on their ports to mouse-based systems.

Both mouse and joystick controls have their virtues. But man, when you have a choice, mouse controls with a TINY bit of practice are just amazing for their rapid precision. Controllers are cool for their analog inputs, compactness and portability.

The nice thing is that when you support both of these styles, you make it much easier to future proof your product. Once you have those handled, controls like a Steam-controller trackpad, or a trackball, or a wiimote/virtual reality controller become much easier to pick up, however they appear on later control systems.

As for the guy in TFA, yeah - I can see how he doesn't want to have to ask his funders to support any additional costs in testing. I can empathise, but as a mouse user, I say, the more platforms that get them, the better!

Ryan Fenton

Comment A lot of it is 'frontin' (Score 1) 409

None of this is news. Almost all jobs these days exist more for 'coverage' rather than full-on throughput. On an instant-to-instant basis, 90+% of human 'work' time is waiting/transition/communication rather than raw action. You can often tell a long-time professional by how they spend 'in-between' time as much as traditional knowledge domain stuff, there's a sort of performance art folks pick up that's no longer 'looking busy', but instead putting folks at ease when there's nothing else to actively do.

Sure, anything repeated with predictable variance can be increasingly automated. But the job market we've grown into is based on low-balling everyone possible, then selecting the 'expensive' folks based on a random hodgepodge of subjective expectations (largely self-serving for the hiring folks). Automating lets you hire fewer grunt workers for serialized tasks - but it doesn't free you of the need for 'coverage', and it makes a larger portion of your hiring effort the 'expensive hire' style, which is a VERY mixed bag.

Don't get me wrong - almost everything we count as a 'job' WILL eventually be obviated indirectly by automation assuming we don't find a way to stagnate. There's just too much a reward at large scales to automating supply, even when wasteful, and although we'll keep getting waves of demand, it simply won't make sense to spend 40+ hours a week as a workplace like now. We'll find ways of needing less 'people coverage' and more 'system coverage' over time. Greed for time may start pushing back at greed for stuff in the mix of all that.

Ryan Fenton

Comment Mario not a launch title... (Score 2) 167

The video says "Holiday 2017" for Mario Odyssey, and if history is any indication, there will be delays. Launch lineup is very likely going to be primarily cross-platform ports and Zelda Breath of the Wild.

Still, that will be enough to sell a lot of units, and provide a lot of great entertainment, just don't expect an avalanche of first-party top content out the gate this time. It'll appear in pretty wide intervals, but when they're putting their focus on something, the quality level from Nintendo tends to beat just about anyone outside of Blizzard and a few other top-end developers.

I skipped the WiiU since none of the first-party games appealed to me - no Mario Galaxy/Metroid Prime/Zelda games - Pikmin was cool, but not enough, and I despise time limits on open world games. The switch, however, I'm seriously considering picking up, if only for a nice open world Zelda game and eventually the Xenoblade/Mario games. Here's hoping they bring back the Metroid Prime team, and make Metroid Other M retroactively (pun intented) non-canon.

Ryan Fenton

Comment Re:Why do these journals still exist? (Score 1) 137

Several methods I can think of. First, just leave the printing stage off, let any libraries that find that important just print off their own copies, and remove that excuse for high fees. That's like version 0.01.

After that, you can experiment (which is sort of being done) with proper reputation systems to replace the "we're a big organization with $X, no one else can play" model. Sure - the big organizations would still dominate most of those, and scoring 'points' in such a system would still require money - but that money should hopefully go more towards people doing work, and less towards organization fees, licenses, etc. That would get you to something like version 0.35.

Getting to this point would involve lots of scandals - but proper ones that really should happen. To get further, you'd want replace the flawed "because we're older and got more mentions" system with a proper interactive vetting process, where replications are worth a larger percentage, even if they don't get 'published'. You can start to bring the newer system into the hiring process instead of 'must have published in x or y' process we've got now. That would get you around 0.5.

To get further than that, you'd have to get outside parties interacting with the process better. Imagine a world where not only free access, but journalists would actually use it, because it's mostly as convenient as 'industry sources' info. That, and being able to contact often obscure scientists to ask a question without having to wait for days in administrative limbo as often.

I'm not coming at this from a 'oh, why won't they support my pet topic' perspective - but as someone whose had friends that have had to deal with the system as it has existed, and who is into proper James Randi-style skepticism (not "science skepticism"), who sees flaws in journals and journalists covering topics lead to mass public misunderstandings greater than just a few simple scandals.

Any system is going to have flaws - I just don't see the journals as useful to anything at this point, when expert gatekeeping can be done so much better in other circles.

Comment Why do these journals still exist? (Score 0) 137

They are absurd. They exists purely for the purpose of acting as gateways to science, except they're largely privately owned, and often deeply corrupt.

It's not helpful anymore. All the benefits of such a system can be achieved in far better ways in the modern era - peer review doesn't need a publishing system anymore, nor does statistical analysis, replication studies or metastudies.

The closest thing to a remaining benefit would be reference count - but even that's a dubious statistic, since so many journals exist largely to provide networks of references.

I mean, the whole process has always been somewhat corrupt in the past too - but better options can be built, and better standards should be valuable to enough people to be worth replacing these absurd journals.

I agree with the notion that we need networks to separate science from psuedo-science, but making everything crazy expensive is NOT fulfilling that logical need, nor is it reducing fake science reported as real when you get right down to it.

A real modern science network would inform journalists and laypeople about the best science available as much as the current approach. This is desperately needed, but instead, we still have journals dominating the field, to the point where jobs depend more on the journals than the actual science...again, truly absurd.

Ryan Fenton

Comment Sounds about right... (Score 3, Interesting) 80

Sounds about right, for the circumstances.

I'm working on a project right now using CMU Sphinx (because it's free/open source) to identify word starts/ends for the sake of syncing word display to audio. All the tools available for speech-to-text are going to require human editing:

Comparrison of commonly used speech-to-text tools

...lots of words end up word salad with any tools, even custom-trained, but the tools are nice for being able to at least have the words show up on beat once they are human-corrected.

Syncing video frames of talking without the audio has got to be even more ambiguous, with more reliance on context.

Sounds like a good challenge for a learning system to pick up on. The 5000 hour mark seems almost analogous to what a human child might pick up raised watching TV in a language different from their family.

Ryan Fenton

Comment Re:Better option... (Score 1) 1368

Agreed, actually - I was just saying for the motivations expressed, the better technique towards the same end would not be secession, but equal vote counting. Easier to accomplish, since liberals don't work in that headspace that would allow breaking away from a common governance.

Lots of compelling arguments against that, just this approach suits the desires of those same tech guys.

Ryan Fenton

Comment Better option... (Score 1) 1368

Just put a indentation stamp into every product you sell, stating "replace the electoral college with actual democratic representation".

Provide details on why the Electoral college exists, and why it's a horrible idea. Tell how the founders wanted to prevent democracy, and how that has hurt us, and how we're now stunted by our lack of actual democratic representation through equal vote impact for citizens in each state..

It's more likely to succeed, and if it does, population centers like California would essentially govern the nation from a more progressive stance going forward.

Secession isn't really a hallmark of modern liberals anyway - they want to make government WORK, and won't really fight to break a government the way modern conservatives will. Instead, they'll tend to only really agree to a fight to defeat profound inequities, like ethnic cleansing or terror - but having their vote not count as much as a rural state might just be enough if it's presented strongly enough.

There's plenty of counter-arguments against this line of reasoning - but for what you're asking for, seems that's your best choice.
Ryan Fenton

Comment Ever play an antihero video game? (Score 4, Interesting) 187

You know, where the character is actively rewarded for, and celebrates some 'negative' ethical attribute? Dungeon Keeper, God of War, most action video games, etc.

Well, if they're paying you money to lie, and more money to lie bigger, then seems to me that's the same dynamic. It's a reward loop, with a context. Doesn't seem to me that this would be any more likely to cause an increase in general preference for lying, than the millions who played Dungeon Keeper being more likely to turn to workplace abuse as a first resort.

That said, context adds a lot - the classic Stanford Prison experiment and similar studies showed how far context and roles can push people with very little prodding.

Seems to me, that more thought should be put into what roles we're building for folks, especially with things like the stock market, the legal system, and managerial roles. Unbounded reward loops have a way of being pushed until something really bad breaks, even with 'normal' people.

Comment Stickers... (Score 2) 113

Now available in my "NotExista' store page, I have large stickers featuring the following civic-minded messages:

"Remember kids, look both ways before crossing the street, to prevent accidents!"

and

"Be on the lookout for police brutality! It's all our jobs to record police in order to prevent crime!" ...along with the FINEST of google-translated Swedish-language versions. Yours for only 50kr! Or get 2 for 80kr! Some shipping and taxes may apply.

Transform your old-fashioned 'drone' into the latest in mobile crime-prevention and accident-prevention platforms today!

Ryan Fenton

Slashdot Top Deals

Just about every computer on the market today runs Unix, except the Mac (and nobody cares about it). -- Bill Joy 6/21/85

Working...