This is precisely why I lost all interest in Oculus the instant I heard that it had been acquired by Facebook.

> First, there is no reason to believe that we can built robots that can reproduce themselves.

What? This is exactly the technology humans are trying to reach! We're already a significant way down this path!!

> Second, there is no evidence that we or anyone else can build intelligent machines, as the original story seems to presuppose.

Nature did it. We can do it.

> Third, biological organisms are so many orders of magnitude more efficient and flexible than machines that it barely makes sense to put them into the same qualitative category "form of life".

This whole conversation is about extrapolating on the cosmic scale. If you look at the path robotics has taken in the last century it does, as pointed out, actually support the premise of this article.

> Hint: A human consumes only about 2.9 kilowatt hours per day, the equivalent of 1-2 light bulbs ...

Not relevant. Once machines are replicating and repairing themselves they'll do exactly what we do and find other sources of energy.

Frankly I agree with you that it's hard to picture Transformers inhabiting the universe, but OP did make a really good point that extrapolation isn't even in the ballpark of refuting this clown. Honestly I'm shocked he didn't come back with that XKCD cartoon.

Just wanted to thank you for the links. I was especially pleased to find that Python is supported!

Thank you! You've given me reason to sit up and pay attention when 3 rolls around, I appreciate that.

I would recommend against showing the more diehard Photoshop fans that link, though. It won't get you anywhere because what it really needs to be is a list like this:

- GIMP has a plugin/feature for automatically generating normal maps from elevation data.

- GIMP has a perspective correction feature that is superior to Photoshops in that it...

- GIMP's 'save all layers' button saves all of the layers in your file into seperate files.

.. or something like that. In the list you gave me, points 1 through 4, and 7, are irrelevant if somebody already has Photoshop. Given its de-facto marketshare, that is likely.

5 is horribly overrated. Lots of artists can script, but few (if any) can make actual plugins or modify the source code. (Even if they do dig in to the code how do they maintain those features when a new version of GIMP comes along?) I do want to mention, though, that there's another reply to my original post that seems to have covered the scripting point. I haven't checked it out yet but given that scripting is something I do, I'm certainly interested in trying that out.

6 needs an extra line, something like: "its better than Photoshop's Batch feature because...."

10... actually this is a really good one. In fact, just before this thread started, I went and found the portable version and downloaded in. Why? Welp, if the scripting that Culture20 posted a link to turns out to be worthwhile for me, coupling that with a portable version of GIMP is *awesome*. What that means is I will be able to automate certain tasks AND keep a fresh install on my DropBox account so I can even use it off-site. This is 1 out of 9.5 (I gave partial credit to the source-code bit) and, as you can already see from other replies you've gotten, most are refutable.

I'm a little worried you might read my post and think that I'm trying to perpetuate the GIMP vs. Photoshop debate. I'm not, instead I'm trying to explain what needs to happen explanation-wise to get more Photoshop people to try GIMP out. I think there's this mentality that people should switch to GIMP and that's simply not true. If you got the professional Photoshop users to start using GIMP for certain tasks, you may find that some studios may find it worth their time to invest some development time into improving it. Given how Adobe has been dicking around with the licensing, this would be a good time to get that ball rolling. Start touting the unique features it has that shave man-hours off a project. If those features don't exist, then the team needs to start talking to people like me and finding out what else they need.

Care to run off a list of ways that "GIMP doesn't come close"? If it's really so bad, it shouldn't be that difficult to name at least a dozen or so... In actuality, I expect that enumerating the shortcomings of GIMP will not be in quantity, but in terms of a relatively small number of particularly desirable features that many may perceive as critically important in such software.

Hi, professional artist here. Your latter point, at least from my perspective, is correct. I know Photoshop really well, but since I make my living doing this work I am not biased in a way that'd prevent me from using a free tool. Let me be extra clear: It would hurt me to be fanboyishly loyal to be any particular app. I do pick up and mess with GIMP from time to time, but it has two critical omissions from Photoshop that make it unusable in my field. First, it lacks adjustment layers. Second, it lacks Smart Objects.

These are both features intended to do non-destructive editing of imagery. Let's say you have a tree with green leaves. You can create a Hue/Saturation 'adjustment layer' that will turn all the green pixels beneath it blue. If you put a picture of a different tree below that layer, its leaves would turn blue, too. If you took that tree and made it a 'smart object', you'd effectively be snapshotting that image and every operation you do causes it to regenerate itself. In other words, if you shrank a Smart Object down, then scaled it back up again, you'd get all its original detail back.

If you're creating imagery it doesn't take long for these two features to change your workflow in such a way that you gain a HUGE time savings. In fact I have created several templates to speed up the generation of images I do that I just plain cannot do in GIMP. Realistically speaking that is enough man-hours lost that I'd actually make a greater profit paying for Photoshop than I would saving the cost of the license in favor of GIMP.

With that said, I'd be *very* happy if you told me that version 3 would add these features. I'd also be very happy if somebody could tell me what GIMP does that Photoshop doesn't. It's free. if it shaves man-hours off my work, then load me up with the tips. I ain't gonna switch, but I ain't above using both.

Actually this kind of 'disaster' cancelling occurs quite often in other companies as well. Typically it goes something like:

The company reaches a point like: "We made a few games, we know our trade, we have the cash, now let's do something interesting." Then they throw all their best ideas onto a huge pile, and the game-design sanctioned people try to make sense out of it. At this point, a lot of creativity is already out of the door, since of course, the huge undertaking has to play safe ball to ensure success, and who knows better than anyone else how huge games work except game designers, right? In parallel, work starts on pre-production, concept art, prototyping, level design, game play mechanics, effects, you name it. After a while, it turn out that the really fun bits are not fun at all, no matter how much you tweak them, and everything starts to look like a tech-demo, because everyone is focusing on just a small fraction, and there's no coherence whatsoever. How could there be. Of course by then we're 2 year after the project starts, and canning it is starting to sound expensive. In the end, it comes down to a financial gamble: releasing crap can mean the end of the company (ahum: Destiny). You can sell crap once with success and maybe break even or profit, but you shit most of your loyal fans in the face, and usually they tend to not take that lightly. Or you can cancel, and swallow the loss and work on something that holds the promise to bring more grit (of which, of course, there is no proof yet).

If there's one team that has the money and the minds to work on very ambitious projects, it's Blizzard. And apparently the teams values their future productions and fan-base as more important than selling Titan. That said, Titan did look impressive from the setup, so I hope the tech and team survives.

If you're instant messaging someone on the company's IM platform on the company's time why the fuck would you have any expectation of any sort of privacy?

I know my company can see everything I can do when I'm logged on to their computer. This is part of the agreement I signed with them. It's also the reason why I don't do stupid shit on my company's network like look for another job or send out resumes from my company email address.

Oh wait, the outrage is because it's Microsoft. Got it.

That would stop mashup services from working easily, for example embedded maps, which I work on: openspace

You are not being bayesian enough.

You need the probability someone is telling the truth given the evidence of being dumb enough to say "I've got a bomb".

If the Hypothesis is 'got a bomb' and the Evidence is 'said bomb', here are my estimates:

p(E) = 10**-7 proportion of travellers who are dumb enough to say bomb (one in ten million)
p(H) = 10**-13 proportion of travellers who have had a bomb so far (one in ten million million)
p(E|H) = well, ZERO so far, but let's say 10**-3 bombers who mention the word bomb (one in a thousand)

Then p(H|E) = proportion of people who say bomb who have a bomb = P(H) * P(E|H) / P(E) = 10**(-13-3+7) = 10**-9

One in one billion people who mention the word bomb will have a bomb.

So if we close the airport each time for two hours, our losses for an actual bomb need to exceed the damages for closing an airport for two billion hours before it is worth it. That's over 200,000 lifetimes of waiting in the departure lounge.

Just.

Any comments on my maths / approach happily received.