Please create an account to participate in the Slashdot moderation system

 



Forgot your password?
typodupeerror
It's funny.  Laugh.

Journal: On Spam, Faith, and Bullshit

Last Monday we had a front page article asking about the "effectiveness" of ISP-level spam filters. I eventually responded with a rehash of my central thesis on slashdot, pointing out that spam is an economic problem and filters will never resolve it.

My comment was generally well received (as seen by the moderations applied to it), though clearly some people were confused by it. Note for example two anonymous applications of the standard form, neither of which showed good comprehension of my comment. Overall my comment fielded 13 replies, many of which seemed to struggle with my statement in one way or another.

However the one who really failed most dramatically was this comment claiming - based on nothing at all - that everything I said was completely wrong. I eventually challenged his faith in spam filters, which apparently caused him to take off the gloves and make it personal.

In fact, so personal, that he kept bringing the conversation back to himself. Eventually I got tired of trying to bring the discussion back to being about spam, and he apparently got tired of talking about himself.

Even for this crowd, that was an odd discussion. Something like 18 comments from him in ~5 days and possibly not a single fact across the lot of them.
User Journal

Journal: Thanks, Obama! 33 33

Talking to a Caribbean-based business acquaintance this week reminded me that, no matter my feelings for the Obama administration, I will be eternally grateful that he's taken Cuba off the table. Because in the pantheon of stupid american wedge issues, the Cuban embargo is near-lock for the title IMO.

Bonus conspiracy fun: The lifting of the embargo, and its timing could be seen as a nice little spoiler for the only 2 GOP candidates who have any shot at beating Hillary next year, Bush and Rubio. Most of the other candidates can choose a wide range of answers when presented with questions on the topic, whereas the Floridians actually have serious history and ties on this, making finessing the issue that much harder.

Oh, and before someone thinks they're clever by telling me that President Perry or whoever would just go back to the status quo faster than you can say "fuck you liberals!", good luck with that. The GOP only cared that it reliably delivered Florida's electoral votes, and it stopped doing that a good couple of cycles ago. It's dead, Jim.

User Journal

Journal: Obviously, this validates the conspiracy 18 18

Islamic State operative suspected in 2012 Benghazi attack killed in US airstrike

Because certainly they must have known that he had all those deleted emails from Hillary Clinton's email server - particularly the ones where she asked him to initiate the strike - on his person when he was killed in Iraq this week. Hence this airstrike was done only to bring about the coronation of Mrs. Clinton.

Am I getting the conspiracy about right this time? I haven't heard anything from the usual gang here to tell me what to think about this yet.

User Journal

Journal: Leveling up with python 2 2

$ py
Python 3.4.3 (default, Jun 14 2015, 02:11:57) [MSC v.1800 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> dat="http://docs.aws.amazon.com/IAM/latest/UserGuide/Using_ManagingMFA.html"
>>> print(dat.split("/"))
['http:', '', 'docs.aws.amazon.com', 'IAM', 'latest', 'UserGuide', 'Using_ManagingMFA.html']
>>> print(dat.split("/")[1:])
['', 'docs.aws.amazon.com', 'IAM', 'latest', 'UserGuide', 'Using_ManagingMFA.html']
>>> print(dat.split("/")[2:])
['docs.aws.amazon.com', 'IAM', 'latest', 'UserGuide', 'Using_ManagingMFA.html']
>>> print(dat.split("/")[3:])
['IAM', 'latest', 'UserGuide', 'Using_ManagingMFA.html']
>>> print(dat.split("/")[3:-1])
['IAM', 'latest', 'UserGuide']

I hadn't ever used the python REPL to work on a bit of code, but it sure makes exploring the slicing notation easier.

User Journal

Journal: Refuting the Chinese Room Argument

Chinese Room

I find the using of the Chinese Room argument to disprove AI is possible to be completely backwards. Most everyone agrees consciousness is the process that arises from the passing of electrical signals between neurons in an organized fashion. Each individual neuron is merely executing/passing along instructions based on reading instructions (receiving incoming nerve impulses) with no understanding of the overall program. Thus our minds ARE a Chinese Room. Consciousness is the byproduct of the process of message passing between neurons which individually have no understanding of what they are doing.

It may seem alien and ethereal, but should you create a Chinese Room large enough to simulate in sufficient detail the message passing of the neurons of someoneâ(TM)s brain, then that Chinese Room would possess a consciousness (albeit one that runs incredibly slowly). Not just a simulation of one.

It shouldnâ(TM)t matter whether the message passing is pieces of paper with a number representing the weight of an action firing potential or a flood of sodium ions in an actual synapse. Both are message passing between discrete elements. In the case of the Chinese Room simulating a human mind, when the person in the Chinese Room shuffling papers gets bored and quits then the Chinese Room mind dies, just like a human mind dies when its brain is starved of oxygen and its neurons decide to quit their jobs.

Now to get our Chinese Room mind up to speed lets do our message passing with electrical signals between transistors in a computer. This is much closer to how our neurons work anyway (though irrelevant as to where consciousness derives from). Whether it is conscious or not will strictly be determined by its degree of self-awareness. Not what components it is made of.

Note: this is a re-posting of a response I made to Android Authorityâ(TM)s Will the emergence of AI mean the end of the world?
User Journal

Journal: Winduhs

I think the whole mobile operating system thing has screwed up GUI design to a certain degree. Microsoft, Ubuntu, and GNOME have both been brave and tried something new, but what they ended up with ended up being highly unpopular on the desktop. And to be honest, I think only Microsoft ended up with something truly good on a touch interface, though I admit to not using Ubuntu or GNOME in those contexts, just being aware that they've not really encouraged an ecosystem for applications to work well in a tablet environment, leaving users with only the main shell being friendly. So the loss of optimization for the desktop lead to no significant gains elsewhere.

The way I'm seeing it, Windows 10 seems to be genuinely exciting, and a decent modern desktop, that also encourages cross interface design. Microsoft has learned from the mistakes it made with Windows 8, kept the good parts, and put together something truly great and modern.

I don't really want to be stuck with Windows though as my primary OS. I'm hoping Ubuntu et al actually learn from it.

This is something you'll never normally hear from me, but perhaps they need a Miguel type figure to take a lead in either GNOME or Ubuntu. At this point, at least to me, it looks like Microsoft is the one with the good ideas about how a UI should work and the relationship of an application to the UI frameworks of the underlying OS. I don't want anyone to clone Windows, but it would be nice to learn from it, at least.

Back in the 1990s, nerds like me put together our own "desktops", running random window managers, app launchers, and file managers (if that) that seemed to go together. I'm feeling like the FOSS "desktop" is heading back to that era, of stuff that doesn't really go together, being shoehorned to fit, with no real philosophy binding the system together.

Debug is human, de-fix divine.

Working...