Trott

Host target
  ProxyCommand ssh -q intermediary.example.com nc target.example.com 22

Trott

{
  "camelcase" : true,
  "indent": 2,
  "undef": true,
  "quotmark": "single",
  "maxlen": 80,
  "trailing": true,
  "curly": true
}

Trott

<?php

// Here's our multiline test CSV string:
$csv = <<<EOD
2012,Camel,1
2012,Marlboro,15
2013,Camel,15
2013,Marlboro,47
total,,78
EOD;

Trott

DROP PROCEDURE IF EXISTS sp_create_and_display_exposures_table;

DELIMITER //

CREATE PROCEDURE sp_create_and_display_exposures_table() BEGIN

DROP TABLE IF EXISTS actlzd_vw_total_exposures_all_years;

CREATE TABLE actlzd_vw_total_exposures_all_years AS

Trott

var answers = document.querySelectorAll('.multichoice .answer div');

for (var i=0, l=answers.length; i<l; i++) {
        answers[i].appendChild(document.createTextNode('\u2717'));
        answers[i].addEventListener('click', function( e ) {
                if (e.target===this) {
                        if (this.getAttribute('style') === 'text-decoration: line-through') {
                                this.setAttribute('style', ''); 
                        } else {
                                this.setAttribute('style','text-decoration: line-through');

Trott

grep -i uptodate spu* | grep proxy  | cut -f3 | sort | uniq > uptodate.csv

Trott

// Get date object for the current time.
var now = new Date();

// Now let's get a timestamp for midnight UTC time. 
// Well, one millisecond before midnight so we don't have to deal with incrementing.
var midnight = Date.UTC(now.getUTCFullYear(), now.getUTCMonth(), now.getUTCDate(), 23, 59, 59, 999);

// And the number of minutes until that time.
var diffInMinutes = Math.round((midnight - now.getTime()) / (1000 * 60));

Trott

• Someone asked me what I was working on. My answer was tedious even for me to listen to, so I set out to do something in my spare time that was at least energizing to me. Music Routes Playlist Generator Not sure there's anything groundbreaking here, but it's fun (for me, anyway) and I get to play with technologies that interest me (Web Components, Browserify, etc.)

• Physical Web: Excited to try some stuff out at work but that's as far as I've gotten. I literally know nothing more than what I've seen in a talk or two and what I've read on a few websites.

• Ageism in tech communities. I'll just stop there.

• Making music, but I do that with instruments and not via high tech.

• At work, I created a federated search tool and I gave a talk about it

Trott

Naively Proposed Solution

1. All access goes through a proxy server. Publishers only need to allow the proxy server and nothing else. This simplifies management. You don't have to worry about expanding networks, remote networks for affiliated organizations, whether or not you should allow your VPN, and any other number of issues.
2. The proxy server should use a single sign-on (SSO) service like Shibboleth.
3. If you are on campus, and you authenticate onto your Wi-Fi network or your Active Directory domain or whatever, you are automatically authenticated via SSO as well. This way, you will immediately get access to licensed resources seamlessly.

Remaining Problems

1. What about the situation where a colleague emails you a link or you otherwise have something that is not going through a proxy server?
2. What about public workstations in the library and elsewhere that theoretically should have access based on their geographic location?

Trott

Install solr-proxy

sudo env "PATH=$PATH" npm install -g solr-proxy

Edit /etc/init.d/solr-proxy

There's a bug in the current init script that prevents the proxy from stopping. Add tobacco to the end of the line that stops the proxy. In other words, change this:

runuser -c "$DAEMON stop $SERVER"