I feel like they could've won over a fairly large handful of people by increasing the flexibility of the UI configuration... e.g. win10 internals with win7 GUI. I'd be nearer the threshold of 'deal with it' if their start menu/taskbar menus were actually responsive. When I right-click on a taskbar item, i want a damned menu, not 2-10 seconds of waiting, followed by a flyout transition, just after having right-clicked again because I thought it wasn't working, followed by more of the same...
Pale moon... not sure what you're talking about there. Home page defaults to 'start.palemoon.com' (redirects to palemoon.start.me)... it's easy to change it. It even still has the ability to start with a truly blank page.
It's kind of like the folks who insist that they should not have to fasten their seatbelt when they drive, because they do not understand how this could impact anyone else. All it takes is a comparatively minor bump/hump in the road to displace someone from their seat, leaving them unable to control the vehicle.
I'm going to argue that in the context of training AIs (neural networks, esp.) on data sets that we may very well be imparting biases on them. If the conclusions present in the data were arrived at by biased means (in this context, I'm suggesting historical prolific racism/sexism), those biases should be present in the behavior of the resulting construct.
That aside, attempting to compensate by overriding the output of the AI with some sort of counter-bias indeed seems like a terrible idea.
Probably making my points here less relevant, I did not see any direct references to neural networking; if these are all just human-programmed algorithms (lacking the abstraction of the neural net stuff), I don't have much else to add.
Or, rather, adopt the mindset that an AI is somewhat like a child. A child that grows up in a (racist/sexist/whatever)-ist household is statistically more likely to turn out fairly similar, as is a child whose school curriculum holds such biases. The people implementing/training these things are going to (hopefully subconciously) impart their own biases upon them, or at least the biases present in the training datasets. If you train a parole-bot with all of our (US, but probably most places) historical parole data, of course it's going to be quite racist! I don't know what the 'proper' solution is, but I feel like attempting to manually adjust the AI after the fact is a terrible idea; to me, it makes more sense to manipulate the training data set until you get a reasonable result.
Anyone know if there has ever been an attempt to form a collective/group buy to make that feasible for the average fellow? I can see pros and cons for both consumers and Microsoft...
From an MS point of view, I see these as primary:
Pro: Large purchase is quick money
Pro: As this would appeal to people who whine loudly about Microsoft, it would probably appease some of them.
Con:Less control over the user, and fewer advertising opportunities
From a consumer point of view:
Pro: read MS Con above
Con: Probably cost somewhat more than buying OEM Pro licenses
Poorly thought out so far, but I feel like it could possibly be worth a try.
Fully electric cars with tolerable range (my made up bs threshold: 200+ mi) don't seem to be cheap enough to compete on their own merits yet... will probably be a few more years, but it's coming.
You could have multiple windows in all ancient Windows; 2.0 was the first one that actually let them overlap. Ergo iPhone user interface has some aspects that are inferior to a 30 year old Microsoft product... not sure if that's sad, or outright impressive.
How about instead of a picture-in-picture, just do a perspective-adjusted underlay that allows you to 'see through' the phone? Amazon's firephone with its extra head-position sensing gear might have been a good testbed for this, but I feel like anything with a front+rear camera should be able to do it. Of course, battery life will suffer somewhat, but since "everyone" is already cool with daily recharging their phones, there shouldn't be a major issue there.
Need some data on how much of these various issues with smartphone+person interactions are due to the touch screen interface. I have no doubts (though i haven't actually been arsed to look deeply into it) that if you were to correlate cell phone usage related accidents against various benchmarks of cellphone uptake, the trend would more closely follow the prevalence of touch screens than anything else. Probable reasons include: no useful tactile feedback (a little vibration when you trigger something is insufficient), difficult to hold (still don't understand why people want a slippery glass puck), and wibbly wobbly user interfaces that rearrange themselves at a moment's notice. T9 (and related) text entry systems are, in many ways, ideal for entering text when a full keyboard is not available; touchscreens will only ever be a (low manufacturing cost, aesthetic) compromise that sacrifices overall usability, at least for 'classical' computer usage.
Slashdot Mirror
I want it... I want it...
I feel like they could've won over a fairly large handful of people by increasing the flexibility of the UI configuration... e.g. win10 internals with win7 GUI. I'd be nearer the threshold of 'deal with it' if their start menu/taskbar menus were actually responsive. When I right-click on a taskbar item, i want a damned menu, not 2-10 seconds of waiting, followed by a flyout transition, just after having right-clicked again because I thought it wasn't working, followed by more of the same...
It would probably be sufficient to just nuke existing ratings when changing hands. Should discourage them from *immediately* going fully hostile.
It should also be fairly clear that I do not use Chrome. Minimalism and caution when dealing with plugins is relevant to any browser.
Only thing I throw on top of that is a link re-writer for eliminating search engine tracking links. That and DownThemAll.
Pale moon... not sure what you're talking about there. Home page defaults to 'start.palemoon.com' (redirects to palemoon.start.me) ... it's easy to change it. It even still has the ability to start with a truly blank page.
I think this bro is indeed mad.
Give it to them in writing. Or sing it in a song? Draw some pictures?
Future NDA clause: I hereby certify that I have never acted out the ___________ in a game of charades....
It's kind of like the folks who insist that they should not have to fasten their seatbelt when they drive, because they do not understand how this could impact anyone else. All it takes is a comparatively minor bump/hump in the road to displace someone from their seat, leaving them unable to control the vehicle.
I'm going to argue that in the context of training AIs (neural networks, esp.) on data sets that we may very well be imparting biases on them. If the conclusions present in the data were arrived at by biased means (in this context, I'm suggesting historical prolific racism/sexism), those biases should be present in the behavior of the resulting construct.
That aside, attempting to compensate by overriding the output of the AI with some sort of counter-bias indeed seems like a terrible idea.
Probably making my points here less relevant, I did not see any direct references to neural networking; if these are all just human-programmed algorithms (lacking the abstraction of the neural net stuff), I don't have much else to add.
Or, rather, adopt the mindset that an AI is somewhat like a child. A child that grows up in a (racist/sexist/whatever)-ist household is statistically more likely to turn out fairly similar, as is a child whose school curriculum holds such biases. The people implementing/training these things are going to (hopefully subconciously) impart their own biases upon them, or at least the biases present in the training datasets. If you train a parole-bot with all of our (US, but probably most places) historical parole data, of course it's going to be quite racist! I don't know what the 'proper' solution is, but I feel like attempting to manually adjust the AI after the fact is a terrible idea; to me, it makes more sense to manipulate the training data set until you get a reasonable result.
Think of all the internets they could've provided with that money...
Saw one come through the firehose that described it as twice the size of Luxembourg, and the weight at "trillion-tonne"...
Does it run Windows phone?
Now that's a titanic proposition!
Corporate interests wish to turn the 'web' into a series of apps... it's sad, really.
I read his response as "I'm incredibly sorry ... that someone called me out on bad behavior."
Anyone know if there has ever been an attempt to form a collective/group buy to make that feasible for the average fellow? I can see pros and cons for both consumers and Microsoft...
From an MS point of view, I see these as primary:
Pro: Large purchase is quick money
Pro: As this would appeal to people who whine loudly about Microsoft, it would probably appease some of them.
Con:Less control over the user, and fewer advertising opportunities
From a consumer point of view:
Pro: read MS Con above
Con: Probably cost somewhat more than buying OEM Pro licenses
Poorly thought out so far, but I feel like it could possibly be worth a try.
Fully electric cars with tolerable range (my made up bs threshold: 200+ mi) don't seem to be cheap enough to compete on their own merits yet... will probably be a few more years, but it's coming.
They can, but how far and how long depends on a number of factors.
Software is typically licensed, not sold... so it's hardly untrue. It's nominally on permanent loan to you.
How about getting some more in-house QA going for the Windows 10? It's got more bugs than a (sarcastic comparison to things with many bugs).
You could have multiple windows in all ancient Windows; 2.0 was the first one that actually let them overlap. Ergo iPhone user interface has some aspects that are inferior to a 30 year old Microsoft product... not sure if that's sad, or outright impressive.
How about instead of a picture-in-picture, just do a perspective-adjusted underlay that allows you to 'see through' the phone? Amazon's firephone with its extra head-position sensing gear might have been a good testbed for this, but I feel like anything with a front+rear camera should be able to do it. Of course, battery life will suffer somewhat, but since "everyone" is already cool with daily recharging their phones, there shouldn't be a major issue there.
But what if I'm particle man?
Need some data on how much of these various issues with smartphone+person interactions are due to the touch screen interface. I have no doubts (though i haven't actually been arsed to look deeply into it) that if you were to correlate cell phone usage related accidents against various benchmarks of cellphone uptake, the trend would more closely follow the prevalence of touch screens than anything else. Probable reasons include: no useful tactile feedback (a little vibration when you trigger something is insufficient), difficult to hold (still don't understand why people want a slippery glass puck), and wibbly wobbly user interfaces that rearrange themselves at a moment's notice. T9 (and related) text entry systems are, in many ways, ideal for entering text when a full keyboard is not available; touchscreens will only ever be a (low manufacturing cost, aesthetic) compromise that sacrifices overall usability, at least for 'classical' computer usage.