SoylentNews

There is now a feed set up in the #github channel on SoylentNews IRC for reporting push, pull and issue events for SoylentNews/slashcode, and push events for some other SN user and staff repos.

The feed works by querying GitHub API URLs every 15 minutes.

--
http://sylnt.us/exec

Seems comment submission in article threads won't work properly. One has to click submit button upto 5 times before it works. Something wrong?

moved nick tracking inside bot process cos i was having all sorts of grief trying to make it work the way i wanted in a separate script. seems to work pretty smoothly now, and i don't need to worry about pausing/unpausing the socket reader.

starting fidgeting with irciv again now that i got some reliable generic authentication features. trying to consolidate all actions into a single alias and get player authentication back up.

event response is a bit more readily available to scripts with event handler registration. the bot keeps track of command/handler pairs in an encoded bucket, and on various events the register is queried and any handlers that exist are executed. this opens up possibility to have the bot trigger a script on an event without using the exec.txt file, and as the register is a bucket it can also be edited from within irc. will have to think about what sort of security risks this might pose, but ability to manually edit buckets is limited to privileged users so not overly worried. irciv will register event handlers on startup to manage player tracking, which was formerly done in cmd.php (eventually want to remove cmd.php, to be replaced with registered event handlers in startup.php). some templates are supported in event handlers; %%nick%%, %%trailing%%, %command%% and %%params%%, which are replaced with actual values when the handled event occurs

a little SN funding feed has been enabled. polls home page slashbox every 15 mins. message appears in #soylent if funding amount increases

comment feed has been changed to use simpler xml feed and only top 20 articles are scraped, as i found the latter 30 articles never got that many comments (assuming probably cos they fall off the SN front page). /join #comments

--
http://sylnt.us/exec
https://github.com/crutchy-/exec-irc-bot

What makes someone rise to the top in music, games, sports, business, or science? This question is the subject of one of psychology’s oldest debates. In the late 1800s, Francis Galton—founder of the scientific study of intelligence and a cousin of Charles Darwin—analyzed the genealogical records of hundreds of scholars, artists, musicians, and other professionals and found that greatness tends to run in families. For example, he counted more than 20 eminent musicians in the Bach family. (Johann Sebastian was just the most famous.) Galton concluded that experts are “born.” Nearly half a century later, the behaviorist John Watson countered that experts are “made” when he famously guaranteed that he could take any infant at random and “train him to become any type of specialist [he] might select—doctor, lawyer, artist, merchant-chief and, yes, even beggar-man and thief, regardless of his talents.”

The experts-are-made view has dominated the discussion in recent decades. To test this idea, Swedish psychologist K. Anders Ericsson and colleagues recruited violinists from an elite Berlin music academy and asked them to estimate the amount of time per week they had devoted to deliberate practice for each year of their musical careers. Based on these findings, Ericsson and colleagues argued that prolonged effort, not innate talent, explained differences between experts and novices. These findings filtered their way into pop culture. They were the inspiration for what Malcolm Gladwell termed the “10,000 Hour Rule” ( http://gladwell.com/outliers/the-10000-hour-rule/ ) in his book Outliers.

However, recent research has demonstrated that deliberate practice, while undeniably important, is only one piece of the expertise puzzle—and not necessarily the biggest piece. In the first study ( http://www.ncbi.nlm.nih.gov/pubmed/17201516 ) to convincingly make this point, the cognitive psychologists Fernand Gobet and Guillermo Campitelli found that chess players differed greatly in the amount of deliberate practice they needed to reach a given skill level in chess. For example, the number of hours of deliberate practice to first reach “master” status (a very high level of skill) ranged from 728 hours to 16,120 hours. This means that one player needed 22 times more deliberate practice than another player to become a master.

In concrete terms, what this evidence means is that racking up a lot of deliberate practice is no guarantee that you’ll become an expert. Other factors matter.

http://www.slate.com/articles/health_and_science/science/2014/09/malcolm_gladwell_s_10_000_hour_rule_for_deliberate_practice_is_wrong_genes.single.html

[Related Abstract]: http://www.ncbi.nlm.nih.gov/pubmed/?term=(Macnamara+and+Hambrick)

Perhaps the wiki could go https just as the main site? considering the opportunity for manipulation and eavesdropping these days, it would be nice.

From this:
http://wiki.soylentnews.org/wiki/Finances

To this:
https://wiki.soylentnews.org/wiki/Finances

When one submits a story to the editors for publication on the site. It would be useful to have a separate input field where one could write comments on the submission itself. That should stay private between the submitter and the editors.

It could be something like "please check the links, good story but in a hurry" etc.

Leaving a comment inside the submission box in parenthesis or similar has the risk of getting unintended publication..

A rigid notion of determinism turns it into a mechanical folly and inverts the importance away from higher functions and towards the lowest detectable causal events, it creates a world in which ultimately the smallest and most remote causal interactions are given disproportionate amounts of importance even though such interactions are constantly changing at speeds far faster than the results they are supposed to have deterministically forced into being. With such an outlook it wouldn't be “turtles all the way down” but instead “turtles all the way up”.

By analogy of a computer program the importance according to such determinism is given to the bits flipping between zero and one rather than the higher structures ruling their behavior.

Such determinism remains technically true but becomes devoid of meaning, comprehension, and value[¹], and thus also without importance.

Instead for any given end result determinism acts as a negative feedback loop in relation to its own importance when given enough complexity: a robot operating its algorithms on the basis of what might as well be an infinite number of ever-changing, causal, and mutually connected variables cannot remain a robot, it is forced into random output and/or the beginnings of intelligence where it chooses which output to give and later also chooses the reason why it is supposed to be the correct output or why a different output is more correct.

Hmm, googly eyes or Einstein afro? It makes sense to me…

(Also ¹ looks like a nice explanation of why “pop” determinism and nihilism so often end up as best friends.)

(And another tangential: if it was possible I wonder what an inverse square type of law would look like for each causal step in determinism, the fact that it rained yesterday has no discernible impact on me writing this journal entry (but now it has and thus two points were made rather than one: one about determinism and one about indirect Wittgensteinian word games).)

The focus of this post was really meant to be the part at the end that I made bold.

To me it makes one or two connections in a way that I haven't seen before (and I know about system complexity and emergence and such). In some way it feels a bit more direct and explanatory tying in a correct understanding/evaluation of determinism as well as (possibly the most basic) evolutionary pressure/fitness challenge. In this way it gets very hands on and mucky (conceptually, and also conceptually reducing the challenge of creating intelligence to that of triggering such a first move and then escalating it). Has anyone seen anything similar elsewhere?

P.S. Yet another tangential: a different kind of amusing folly, more entertaining than determinism but maybe not all that different after all?

exec now has a basic nick tracking script that hooks the join, nick, quit, kick, part and 353 events. this feature is similar but nowhere near as complete as xchat's (or $insert_client_here's) user list.

it has lead to the addition of internal stdout commands for retrieving a space-delimited listing of bucket indexes, and commands for pausing and unpausing the processing of irc data, to prevent corruption of bucket data that might occur due to multiple processes triggered by irc events trying to read/write to the same bucket(s) simultaneously.

http://sylnt.us/exec#Buckets

nick tracking enables scripts to find out what nicks are in a given channel or what channels a given nick are in. currently only channels shared with the bot are tracked, but with additional event hooks (such as a whois 319 numeric) additional channel info can be tracked.

with proposed addition of whois account querying (330 numeric) by the user tracking script, other scripts will be able to authenticate instructions with a simple function call.

the irciv script was originally designed with player authentication using the 330 nickserv account numeric. however it was messy, with irciv-specific code sprinkled throughout the main event handling script (cmd.php). the generalized tracking system will give all scripts access to the same user data via a small set of lib functions.

--
http://sylnt.us/exec
https://github.com/crutchy-/exec-irc-bot