Last year Amanda and I went to Montana for a week in early May and visited Yellowstone for a couple days before the official start of the summer season. It was chilly and some roads were closed due to snow, but we avoided the crowds and got to explore the park in relative solitude.
I haven’t shot photos much since graduating from Hillsdale, so I decided to take my camera along. Like any other skill, you need to practice photography regularly so you don’t lose your sense of composition, light, and technical settings like aperture and shutter speed. It was frustrating to discover that I’ve lost some of the edge that I worked so hard to hone in high school and college, but it was nice to put my gear through its paces again.
Here is the best of what I shot over two days in Yellowstone. I took a few landscapes, but I mostly focused on the colors and textures I encountered.
We also drove down into Grand Teton National Park for a little while:
According to the Jekyll docs, the jekyll-paginator plugin does not support categories or tags. That means there is no way to exclude a category as a whole. You can only exclude posts individually by including hidden: true in each post’s YAML front-matter.
I’m hiding all of my TIL posts from the front page of my site, so I did a global find and replace to accomplish it.
Mirror of If statement in Liquid
Liquid has a neat concept that is the inverse of an If statement. It executes if a condition is not met.
{%unlesspost.categoriescontains'TIL'%} Code here to display all the things! {%endunless%}
Inserting via mysqli()
Using the mysqli() function I wrote about yesterday, here is how to insert something into a table:
$connect=newmysqli($server,$db_username,$password,$db);$connect->query("INSERT INTO `table_name` (username) VALUES ('$user');");
Make sure your variables are sanitized first! You don’t want someone doing a SQL injection.
Eric Davis told me that mysqli() is the old way of doing things and I should check out PDO instead.
This checks for a table called scorecard_test and creates it if it doesn’t exist. The SQL parameters for the columns are:
An integer called ID that is the primary key and auto increments
A username that can’t be NULL
A column called counter that has a default value of 1 if there is nothing passed, and the length can’t be longer than one digit
A column that holds the current timestamp.
// SQL syntax$sql="CREATE TABLE IF NOT EXISTS scorecard_test ( id int(11) NOT NULL AUTO_INCREMENT PRIMARY KEY, username varchar(255) NOT NULL, counter int(1) NOT NULL DEFAULT 1, time TIMESTAMP )";// Connecting, sending query, showing errors, and closing connectionif($connect->query($sql)===TRUE){echo"Done!";}else{echo"Error: ".$connect->error;}$connect->close();
Philosophies like Agile Development are great for teams working on large projects, but for someone just getting started with development and working on small projects, they can be a little much. For situations like this, Readme Driven Development strikes the right balance of planning out your project, writing documentation, and actually doing the work.
Write your Readme first! (H/t to Eric Davis for telling me about this.)
WordPress Plugin Basics
I have a food & drink website where I post regularly. Lately I’ve been trying lots of kinds of coffee to dial in my ROK Espresso maker and I’m starting to reach the limits of how many I can accurately remember. I’ve been wanting to make a WordPress plugin for a while, so why not make a plugin that makes a Coffee Reviews custom post type? (The main goal here is to write my own plugin and learn how it is done, so please don’t suggest already made plugins that I could use.)
Hooks allow you to tap into WordPress without editing the core files. (Rule #1 is to never edit the core files.) You pass functions through hooks to extend or change WordPress functionality.
The register_activation_hook() runs when you press “Activate” on a plugin. Since I’m making a custom post type, I’ll want to use the activation hook to refresh permalinks once the post type is created. If your plugin needs a custom database table, you can create that with a function passed to this hook.
The register_deactivation_hook() is basically the reverse of the activation hook. When you deactivate the plugin, the items passed to this hook fire. I’ll want to clear the permalinks again.
I made a barebones plugin tonight that registers the custom post type. I definitely have more work to do, but saving, hitting refresh, and seeing this appear never gets old!
{%highlightliquid%} Put your foo here bar end foo {%endhighlight%}
I need to go back though my previous posts and remove the Pygments highlighting then uninstall the plugin. Rouge supports fewer languages than Pygments, but Rouge supports all of the languages I use, plus Liquid and Markdown, which Pygments doesnât.
Downloading WordPress.com Email Subscribers
I learned today that it is possible to download a CSV of the email subscribers to your WordPress.com blog (or WordPress.org blog powered by WordPress.comâs JetpackArchived Link). They donât make it obvious, but the option is there:
Back in the summer of 2015, I decided to trash my website and rebuild it from scratch on top of Jekyll, a static site generator.
My goal was two-fold:
Build a new, fresh website to house my personal brand.
Learn the Liquid templating language.
I found a starter theme, gutted it, and spent a week digging in to how Jekyll works in order to build the features I wanted. Along the way I built out some tools and templates for myself and decided to share them on Github.
When doing development work, it is best if you have a separate, dedicated area for each of these things:
Development – The area where you do your development work.
Staging – The area where you deploy changes and thoroughly test them. The only thing that should be out of sync with the production environment is the single change you are testing.
Production – The live system for stable, tested code.
I’ve attempted to replicate this best-practice with the development of my Toggl slash command for Slack. My development environment is on a Homestead box on my local machine. The staging area is a folder on my dev server that a /toggldev command posts to. Once the changes here have been confirmed, I commit my code to the repository and then deploy it to production. The production environment is a folder on my main server that the /toggl command posts to. This is the one my team uses and it stays up to date with releases in the repository.
Of course, the entire process is better if you have version control:
A method for committing changes in Git
Eric Davis advises me that it is better form to commit changes after each feature you want to release is completed rather than commit after a full day of work. If you don’t commit after each feature (and develop each feature in a separate branch), you’ll end up with a mass of code changes in each commit that is about as tangled as spaghetti.
Right now, since I’m the only one working on my Toggl project and the changes I’m making aren’t major, I’ve only been working directly on the master branch. This Atlassian tutorial on branches has convinced me that I should be using branches on a daily basis for development, so I’m going to start that this weekend with some new features I plan to work on.
For the Toggl Slack slash command, I’m using it to check if someone’s Toggl API key is in the array in variables.php:
if(key_exists($user_name,$toggl_api_key_mapping)==FALSE){echo":warning: You haven't set up your Toggl API key with this command yet. Get it at the bottom of https://toggl.com/app/profile and send it to @$your_admin_name.";}
Notes:
Case sensitive. If you need it to be insensitive, use strtolower() on the $key
Boolean (returns TRUE or FALSE)
Looking up a key and returning a value in an array
$array['key']
For the Toggl Slack slash command, I’m using it to set someone’s API key from the array in variables.php:
If you are serving something on your webserver that you don’t want anyone else to be able to access, you can restrict access to by adding this snippet to your site’s .htaccess file:
log.csv>orderallow,denydenyfromall
If you need only a certain IP address, you can achieve this by adding allow from 0.0.0.0 (replace that number with your IP address):
I asked Eric Davis about logging for debugging and usage stats on my Toggl Slack slash command and he suggested that I look into writing to a CSV or JSON file. I opted for CSV.
Here I build the array from the information I get back from Slack (plus a datetime stamp) and pass it to a file called log.csv via fputcsv().
A note on fopen() – A lot of tutorials replace that 'a' with a 'w'. According to the docs, w writes at the beginning of the file, which is why my first couple tries overwrote each other. The a starts writing at the end of the file, which is always a new line since fputcsv() always ends with a line break.
Installation is simple: Clone the project to your _plugins folder and add it to the Plugins section of your _config.yaml, then you are ready to go.
Usage is equally as simple: Specify the language at the end of the first line of your code blocks and the plugin will add the proper classes and highlight the code according to the language.
I didn’t do much technical work today besides some front-end debugging. Instead I did a lot of administrative and project management work and ended up thinking a lot about how I work.
Here is what I thought about.
Always provide a solution
As a consultant, you aren’t paid to tell people they have problems. They already know that and chose to hire you. It is your job to provide solutions.
Instead of being negative or saying that a proposal can’t be done, try to get at the heart of the request and offer an alternative solution. You and your clients will be much happier.
As things get frustrating, it is easy to take the, “look how messed up this thing is” route. But that isn’t your job. You need to roll up your sleeves, brew some coffee, and come up with a solution.
Learning
I find time and time again that I retain so much more of what I learn if I’m applying it as I’m learning. So why do I continue to use books with pre-designed activities that I mindlessly copy? Because it is a lot easier than coming up with a project from scratch.
From now on I’m applying what I learn in my own way. No more copying examples. I’m going to figure it out and learn it the hard way.
Culture
Sometimes you need to find software that is a better fit for your culture. But sometimes you need to take a hard look at the culture and its ability to adapt if there doesn’t seem to be any good solution.
Like learning, sometimes cultural change only comes the hard way: Adapting because it is necessary.
Distractions
Second screens are great for productivity when you are in the zone. Unfortunately they are also great for procrastination when you aren’t in the zone. Sometimes you need to turn that second screen off and focus on one thing at a time.
Filter, filter, filter. Reduce the number of things that can distract you. If you don’t need to make a decision on it, don’t let yourself see it. For example, in Excel, if you only need to read items that have a specific attribute, filter your view by that specific attribute and hide the rest.
Idea for a Slack bot: What is distracting you right now?
Regular messages (10, 2, and 4?) via Slack
Respond with comma separated items that are distracting me.
Save those responses in a database for later analysis
I like to learn by giving myself projects. I treat them as if I were releasing them to the public, even if I’m going to move on after I’ve learned what I want to learn. This is one of those projects. I created a roadmap of features that I probably won’t ever implement, but I learned a ton getting what I have so far to work, so I’m calling this a win.
What I learned
How to read from and write to an API
How to structure complex if/then statements to parse input text
Dealing with arrays in PHP
Deeper regex processing
Unicode conversions
Releasing code to a public repo on Github
Designing security measures for a system deployed on the web
How to structure and design a codebase from scratch
How to work within the bounds of the different systems you are working in and finding ways around barriers
Implementing command line options
Using other projects in your own project via dependency managers like Composer
Writing documentation
Features
Show a list of projects and their corresponding IDs in a given workspace
Show a list of tasks associated with a given project
Add time entries to Toggl straight from Slack’s message input box
Screenshots
Setup & Installation
Dependencies
You must have PHP 5.3.2+ installed locally and on an accessible server.
You need Composer installed locally to install the third-party dependencies.
Two third-party libraries are dependencies included via Composer:
Clone or download this repository onto your local machine
Install Composer in this repository if you don’t already have it
Open a command line terminal and navigate to this directory.
Run: php composer.phar install via the command line to install the third-party dependencies
Configure the variables file.
Copy the variables-dist.php file to variables.php
Fill out the Slack token you got while setting up the slash command.
Fill out users’ Slack usernames with their corresponding Toggl API keys in the array. They can get those keys at the bottom of https://toggl.com/app/profile.
Enter a workspace ID for your team. to find yours, log in to Toggl, then go to https://www.toggl.com/api/v8/workspaces. Pick the ID of the workspace you want to use. Don’t wrap it in quotes; it needs to be an integer.
Set your team’s default timezone. Right now it is set to america/new_york.
Save the file as variables.php in the same directory and you are good to go!
Upload to a server with a valid SSL certificate
Once the dependencies are installed, upload the whole directory to a server running PHP 5.3.2+
Ensure you have a valid SSL certificate. Slash command URLs must support HTTPS and serve a valid SSL certificate. Self-signed certificates are not allowed. Check out CloudFlare for an easy way to obtain a valid certificate.
--date [mm/dd/yy] – Adds the time entry to a specific date. If none is passed, it defaults to today’s date.
--task [task id] – Adds the time entry to a task ID. See above to find your task ID for a project.
Logging
I added basic logging to CSV for usage stats and debugging. This happens individually per project, stored in log.csv. Nothing is transmitted back to me.
If you’d prefer to not log usage, simply comment out lines 14-19 in slash_parsing.php.
Since the log includes a Slack token, you’ll want to deny access to log.csv on your webserver. I achieved that via my .htaccess file.
Unicode quotation marks (sometimes called smart quotes) are those curved/slanted single and double quotation marks: “ ” ‘ ’. The ASCII ones look like this: " '. They are usually interchangeable in word processing, but the stylized ones sometimes cause issues on the web (in a parser, for example).
To convert one to the other in PHP, use iconv():
$output=iconv('UTF-8','ASCII//TRANSLIT',$input);
If you use OS X and want to turn off smart quotes, go to System Preferences > Keyboard > Text and uncheck “Use smart quotes and dashes.”
Remote git repositories
Using git locally is one thing, but it is far more powerful (and exciting!) to push that code to a remote repository like Github for display and collaboration.
Assuming you have a local repository that wasn’t cloned from a remote one, you’ll need to hook it up to a remote repository:
$ git remote add origin
If there is code already on the remote repository, you’ll need to merge it in to your local:
$ git pull $ git merge origin/master
Then push your committed files to the remote repository:
In CSS some styles are automatically inherited from their ancestors (i.e. elements which contain that element). For example, if you set body {color: red;}, all text inside the body element, regardless of what other elements the text is in, will be red unless the color style is set on one of those other elements later in the .css file.
This simultaneously makes writing CSS a breeze and debugging it a bit of a headache. There isn’t a great way to tell if a style is inheritable without looking at the docs.
If you encounter an issue where your style isn’t displaying as it should and it is not responding to changes, check the docs to see if the style can be automatically inherited. If so, check the ancestors of the element you are working on.
The issue I ran into today was that my
snippets kept wrapping no matter what overflow and width were set to. It turns out that word-wrap is automatically inherited and I had it set on an ancestor element.
Date/time conversions in PHP
Date conversions are tricky. Thankfully, PHP has some built-in functions to make this easier.
time() returns the current Unix timestamp (see below)
date() converts either the current time or a given timestamp into a specified format.
Example: date("c", time()) will take the current time (provided as a Unix timestamp) and convert it to ISO 8601 format.
Unix timestamps are the number of seconds that have elapsed since the Unix Epoch (00:00:00 Coordinated Universal Time (UTC), Thursday, 1 January 1970). They can be frustrating at first because they are so foreign looking, but they are really useful because:
They are not subject to timezones.
Since they are simply seconds, time math with Unix timestamps is pretty simple: Convert the time you wand to add or subtract into seconds and do the direct calculation. Then if you want to get it back into a recognizable format, use one of the functions above.
I’ve been working on a custom Slack command to input time into Toggl, so I’ve been doing quite a bit with time conversions to get everything right.
I used to be really interested in how people worked. When they get up, what their morning rituals are, what tools they use, how they use them, etc. While these things are interesting and tell a story about a person, they miss the point.
How someone works isn’t nearly as important as the fact that they are doing the work. They made the choice to do what it takes each day to practice their craft. The rest is minutiae.
If using the GTD method or having morning rituals works for you, great. But don’t think that those things are what really matter. If you’ve committed yourself to doing the work, you’ll find a way. If you haven’t committed yourself, none of the rituals or methods will make a difference.
This commitment is very difficult, but necessary.
Flipping the Status Quo
We all suffer from status quo bias. How do we overcome it? We can get 80% of the way there by:
Instead of asking, “Why should this be different?” ask “Why should this stay the same?” – This is especially powerful when confronting clutter. (By the way, the answer, “because it gives me joy” is a valid reason for keeping something. Change for change’s sake isn’t helpful.)
Recognize the difference between acknowledging a feeling and rationalizing that feeling to keep the status quo.
It was starting to annoy me that my tags were not in alphabetical order on my Today I learned page. So I learned how to sort them alphabetically with Liquid and I added anchor links with counts for each tag.
Slack is wonderful for collaborating with remote teams, but if there is a lot of activity at once, it can be easy to lose files and requests.
If IâÂÂve seen the request or file, I acknowledge by marking it with , then I put it on my to-do list of choice, WunderlistArchived Link. (They have a great Slack commandArchived Link to make this easy: /wunderlist add)
If IâÂÂve fulfilled the request or processed the file, I mark it with
If the request/file is no longer needed or IâÂÂve tried to process it, encountered an error, and notified the appropriate parties, I mark it with so that I know in the future that there was an issue with it instead of assuming it was done and I moved on.
Using SlackâÂÂs great search functionality, I can easily visually figure out what IâÂÂve processed and what is still outstanding.
If I get a request when IâÂÂm in the middle of something else, I copy the link and use SlackâÂÂs /remindcommand to remind me about it later in the day.
There are a lot of things that are easy to track that are completely useless to track.
Make sure the things you track relate directly to your goals.
Sometimes the things you want to track aren’t easy (or possible) to track and you need proxies instead. This is okay.
Visible statistics like Facebook likes and Twitter followers, etc, are easy to obsess over because they are public and they are so easy to find. But don’t fall into to the trap of thinking these numbers are what really matter. Getting more followers is unlikely to be your actual goal. Purchases, downloads, or donations and what directly affects those things are what you should really focus on. The closer the things you test are to your actual goals the better.
Instead of Twitter followers, test and track your click-through rate for specific types of content at different times of day.
Instead of website pageviews, track the length of time on page and how far people scroll down through your content.
Instead of likes on a Facebook status track how many people see that status and follow your calls to action.
Scalability
You can always scale with more people or by working more hours. The trick is to design your business to scale the amount of products and services you provide without a 1:1 (or worse) increase in costs, labor or otherwise. In order to scale you want to become more efficient and more automated. When building your product/service, bake in efficiency and automation from the beginning so you can eventually move on to the next thing (or next version) while it sustains itself.
RSI
I’ve been experiencing some RSI lately in my right wrist, just below my thumb. It primarily hits when I use my mouse for an extended period. I read a number of sources and asked some friends today about how to cope with it. Here are some solutions:
Stretching and regularly stopping to do so during the day
Varying input devices, such as switching between a trackpad and a mouse or a Wacom tablet and a mouse
Switching to a trackpad (like the Magic Trackpad 2) to minimize wrist movements.
I decided to buy a Magic Trackpad 2 and try it. The reason I went with that over a tablet or ergonomic mouse is that I can’t stand the idea of using my computer without gestures. I’d have to change all of my workflows. If the trackpad doesn’t help, I can return it within 14 days for a full refund and try one of the other mice. The past hour of use has been promising. We’ll see.
Meteor is a platform for building real-time web apps that sits between your appâÃÂÃÂs database and its user interface and makes sure both are kept in-sync. It is built on Node.js, so it uses javascript on both the client and the server.
To then run that new app at http://localhost:3000/:
$ cd[project directory] $ meteor
To stop the app from running, press ctrl+c
Adding a package like Twitter Bootstrap to a Meteor app is incredibly simple. No files to link up, Meteor takes care of all of that out of the box:
$ meteor add twbs:bootstrap
Structure
Meteor has five types of packages. most can be seen in [project directory]/.meteor/packages:
meteor-base is MeteorâÃÂÃÂs set of core components.
First-party packages that come bundled with Meteor that can be removed, such as mongo (the database) and session (a client-side reactive dictionary).
Local packages specific to your app, which are stored in /packages
Third-party packages available at MeteorâÃÂÃÂs online package repo, AtmosphereArchived Link. These are named in the author:package convention.
NPM packages in Node.js. They arenâÃÂÃÂt listed with other Meteor packages, but can be imported and used by other Meteor packages.
Meteor has some special directories:
/.meteor/ is where Meteor stoes its own code. DonâÃÂÃÂt modify it.
Code in /server only runs on the server
Code in /client only runs on the client
Everything else runs in both places
Static assets should be stored in /public (The exception is CSS. Meteor automatically loads and minifies CSS, so it should be stored in /client)
Meteor loads files in a specific order:
Before anything: Files in /lib
After everything else: Any files named main.*
Everything else is loaded in alphabetical order by file name.
Deployment
To quickly set up a staging server, you can create a Meteor account and deploy to a Meteor subdomain for free:
$ meteor deploy yourappname.meteor.com
For deploying to your own server, check out Meteor Up, a command line utility that automates setup and deployment for you.
404 Error Pages
Making a 404 error page isnâÃÂÃÂt enough, you have to tell Apache where to find it. IâÃÂÃÂm used to working on WordPress, which takes care of that automatically. Turns out IâÃÂÃÂve been running this site for a few months without my 404 page working. Whoops!
To do so, add ErrorDocument 404 /404.html to your .htaccess file. Replace /404.html with the path to your 404 page.
Secure Hash Algorithms
What are they and what is the difference between SHA1 and SHA2?
SHA stands for Secure Hash Algorithm. The short story about how is works is that a mathematical operation is run on a given input and a unique output, or hash, is generated. By comparing the output to an expected output, you can verify the dataâÃÂÃÂs integrity. The theory is that no two different input values should result in the same hash output (called a collision).
SHA1 algorithms produce a 160-bit hash value, while SHA2 algorithms can produce 224, 256, 384 or 512 bit hash values depending on the function used.
The very short explanation is that SHA2 hashes larger and are theoretically much less likely to have a collision than SHA1 hashes due to the underlying algorithmic changes.
What is it used for?
Generally, SHA is used for verification. If the hash you calculate matches the expected result, your data has most likely not been tampered with or corrupted.
Git uses SHA1 to verify data has not changed due to accidental corruption.
SHA is used to sign SSL certificates.
Bitcoin uses SHA2 to verify transactions
SHA2 is used in the DKIM message signing standard (i.e. what checks to make sure someone isnâÃÂÃÂt spoofing your email account)
Some software vendors are adopting SHA2 for password hashing.
Why should I care?
SHA1 is on its way out. Chrome is showing errors on sites with SHA1 SSL certificates that expire past Jan 1, 2016. All major browsers will stop accepting SHA1 SSL certificates by 2017.
Every site using SSL signed with SHA1 needs to update their certificates.
I might start using this for my code snippets in the future. Right now I have some CSS styles that conflict with it, so I need to sort those out first.
VPN Connection issues
I had VPN connection issues today. The solution was to stop using the IPSec client and start using the AnyConnect client.
Iâve been having a lot of discussions about privacy and data collection recently, both at work and with friends. This episode of Whatâs the Point on privacy with Kashmir Hill cleared up some common misconceptions. The main points:
If a companyâs fundamental business model is collecting data and figuring out things about you (i.e. Google and Facebook), they arenât selling your data, but rather selling access to your attention to advertisers. They donât want to give up your data to other people because it is such a valuable resource to them.
Some companies who sell you physical products have a side-business selling data on what you bought and when to others.
Privacy has become a genuine concern for companies recently, as evidenced by the rise in end-to-end encryption in consumer products like iMessage. This is worrying to some government agencies.
If you read privacy policies, they actually tell you all the ways a company is going to violate your privacy. As long as a company doesnât do anything they didnât say they were going to do, they are protected.
I’ve been making some patterns for the header images on these TIL posts in my down time the past few days. There are two ways to tile patterns in Photoshop. One is manual and time consuming, the other is fast, easy, and less prone to error.
The manual, time consuming way is to open a pattern in one layer, duplicate the layer multiple times, and move the layers so that they line up in a tiling pattern. While PS is pretty great at snapping items to a grid, sometimes they don’t always line up and you only find out after you export the image.
The fast and easy way is to open the pattern in PS, go to Edit > Define Pattern…, then use the Pattern Stamp Tool to paint the pattern you just created across a large canvas.
This would have saved me a lot of time on the first nine headers, but c’est la vie. This will definitely save me time going forward.
Homestead Provisioning
I use Laravel Homestead for my local PHP environment. Provisioning a new site is something I’ve done multiple times, but I do it so infrequently that I have to look it up each time. So here are my quick notes:
Create the site folder in ~/Projects/
Map the folder and domain in ~/.homestead/Homestead.yaml
Map the domain to 192.168.10.10 in /etc/hosts/ (reminder: $ sudo nano /etc/hosts/)
Run $ cd ~/Homestead
Run $ vagrant up --provision or $ vagrant reload --provision if vagrant is already running.
Site should now be accessible locally.
If you need a new database for this site and load it from a SQL dump:
$ vagrant ssh $ mysql -u homestead -p-e'create database DBNAME;'$ mysql -u homestead -p DBNAME < path/to/file.sql
Composer Basics
Composer is a dependency manager for PHP. I’ve used Composer, but like Homestead provisioning above, I’m a little rusty. Writing it down helps me remember.
After you’ve installed it (I have it installed globally), the basic usage is:
Set up composer.json in the directory you are working on. A basic structure looks something like this:
When writing these TIL posts, I want to eliminate as much resistance as possible so I can get straight to writing. The more steps I have to take, the less likely I am to follow through. So I took some time today to automate the one of the tedious parts: Setting up the markdown file I write these posts in with most of the specs already filled out.
I created 3 TextExpander snippets to help with this process:
One with today’s date string and extension separated by dashes for the file name: %Y-%m-%d-.md (I tried a fill-in for the title, but the filename doesn’t stay selected in Coda when another window launches)
One with Javascript randomly picking one of the 7 default header images I use (this is used in the next snippet):
Besides for the normal scope, budget, platform, dependencies, timeline, and underlying technologies, here are a few things you need to consider when architecting and planning out a software development project:
After delivery, what resources will we have available to maintain the system and handle emergencies?
What impact will maintenance and support have on our ability to sell and develop new projects in the future?
What is the opportunity cost of using one developer over another?
Use the jQuery parents() selector to return the appropriate parent element, then use the css() method to set the style you want. For example, ("a i").parents("li").css({"margin-right": "5px"}); finds list items who are the ancestors of icons wrapped in links and makes their right margin 5px.
Change the structure of your code and add specific classes that you can apply your desired styles to. For example, instead of applying a right margin to all list items and then a different right margin (via parent selector) for all list items that contain icons, make two classes with the proper margins and apply one to the plain list items and the other to list items that contain icons.
I opted for #2. It makes your CSS cleaner, keeps you from having to load new dependencies like jQuery, and keeps the styles in one place so you don’t have to search 6 months from now for where that extra margin is coming from. I’d only use the first option in cases where you have complex rules that can’t be accomplished by restructuring your code and/or adding classes to make your desired change possible.
Project-wide Find and Replace in Coda
I do the vast majority of my code-writing in Coda. I used to have to open up TextWrangler to do find and replace across multiple files, but apparently Panic added this feature into Coda sometime between v1 and v2.5. I got used to having to switch and didn’t look for it again until today.
It is buried in the sidebar under “Find In” instead of in the regular find and replace bar, so I never saw it. You can search across open files, the entire site, or in a specific directory on your local machine.
The upgrade itself is simple: In Terminal, run gem update jekyll. Use sudo if you run into permissions issues like I did.
If you use pagination in any of your templates, you’ll now have to add gems: [jekyll-paginate] to the # Plugins section of your _config.yaml or else your site won’t compile. This wasn’t necessary in 2.x.
Making an index for my TIL posts in Liquid
I wanted to have an index for my TIL posts that was organized in two ways: reverse chronologically and by tag. Here is how I’m handling that with Liquid:
The general structure of the tags method comes from Joe Kampschmidt. I started searching when it didn’t work the same way as the categories method I wrote earlier.
This method for displaying tags works because I’m only tagging my TIL posts. Everything else goes into categories. If I were to tag other kinds of posts, I’d need to first limit by posts in site.categories['TIL']
Category names are case sensitive.
You can display raw Liquid markup without it rendering by wrapping it in {%raw%} and {%endraw%}.
Eric DavisArchived Link and I ran into a strange CSS issue where an image was scaling in a funky way when we resized the browser. The height was staying fixed while the width was changing, but there was nothing in the CSS setting a specific height.
It turns out that one of the parent
s had display: flex; flex-direction: column; specified for layout order purposes, and when we turned that off the problem went away. So then we went searching and the quick-and-dirty fix is wrapping the image in a vanilla
I love the idea of the Five Minute Journal, but I know that I won’t use yet another notebook or iPhone app to fill it out. I took a look at the tools I use on a daily basis and came up with a solution: A TextExpander snippet that I can put into the journaling app I already use, Day One.
All I have to do is type ;5am in the morning and ;5pm in the evening to fill out the journal. It automatically adds the date and time then pre-populates the template. Presto. It works on TextExpander’s iOS app, too.
I formatted the snippets with Markdown so they can be easily used in a wide variety of apps. If you want to change the formatting, feel free to make it suit your specific needs.
For the past two weeks I’ve been using a single iPhone home screen configuration a la CGP Grey.
Most of the apps live in folders on the top row, divided into four main buckets: Media, Work, Life, and Other. The other three rows of apps are the ones I use most frequently and want quick access to.
The rest? I search for them. This is just as convenient for me as swiping through multiple screens, but the added benefit is that my screen is much less cluttered.
In the process of reorganization, which I completed during a 30 minute train ride, I took a moment to delete applications I no longer use, enable me to waste too much time, or provide too little value. For the distracting/time wasting apps I still want to use (Instagram, Twitter), I made the intentional decision to move them into folders so I am less likely to use mindlessly tap them when I have a free moment. I filled their would-be spots on my home screen with apps I want to use more: Day One to journal and Pocket to read some of my recently saved articles.
The Results
My screen is less cluttered, so it is easier to look at.
I’m spending less time mindlessly tapping on an application and scrolling through stuff I didn’t really look at anyway.
I’m spending more time using my phone for productive purposes like journaling and reading articles I’ve saved.
My default method of finding an application is now using the search functionality (swiping down and typing) instead of swiping through screens, which is very useful when using a different iOS device.
I really like this layout and I think it will prove to be my long-term configuration.
