
As companies had to adjust to remote working last year, many were open to hiring outside their location. When deciding what to pay a software engineer, most do a quick search for market rates in that area and then offer a salary around that rate.

I have been working remotely from Mexico for years and have also hired freelance software engineers and designers. I have had no shortage of work at multiples of the market rate (a few thousand dollars a month) in my area, Guadalajara, for years. If you make me an offer at market rates, I will never accept. Neither would many other engineers I have met in Guadalajara who also earn significantly more than the market rate.

Companies that are known for adjusting heavily for the cost of living, such as GitLab, tend to offer low salaries even considering market rates. Whenever I encounter this situation, I can guarantee the company is not hiring the best engineers in the area.

I recently received an email for a job that pays in the range of 110k - 170k USD. If I actively search for a job, I know I can find more offers of a similar, or better, salary range. I was 22 when I was no longer "employable" in Mexico because of the salary I demanded.

I have learned that there are two types of companies that hire remotely. The first type looks to hire engineers at a cheaper rate. The second type wants to hire the best engineers. Be the latter.


What Companies Get Wrong about Remote Salaries


Ever since deciding to permanently disable images on all the emails I receive (even from people I trust), I occasionally run into the inconvenience of viewing an email with a bunch of blank squares where I suppose a huge image is supposed to be. Just an image, and almost no text to explain what the email is about. This is a shame because I do sometimes subscribe to newsletters on purpose because I want to stay up to do date with a service that I am using.

Now, I am definitely _not_ going to enable images just for these emails. I disabled images as a matter of principal against all of the tracking that happens when an email is opened. But _you_ can write a [multipart email](https://www.w3.org/Protocols/rfc1341/7_2_Multipart.html).

Basically, you can send a rich HTML email, with a plain text alternative for those of us who prefer plain text. Rails already supports multipart emails and I imagine many other web frameworks as well. Services such as Mailchimp also support multipart emails so there really isn't an excuse not to include a plain text version in your next newsletter.


Please Write Multipart Emails for your Newsletters


I was around 15 when I purchased my first domain. Of course I used GoDaddy. Since then I have tried many other registrars, including (in no particular order):

- Dynadot
- namecheap
- iwantmyname
- Gandi
- DNSimple
- a couple others that I can't remember

In my experience, most people can't tell the difference between all the registrars. What could one registrar possibly offer to make it worth paying an extra $2-5 a year? Turns out, a lot.

First, the one registrar you want to avoid at all costs is GoDaddy. They have been known to buying domains that you have searched for in the past and then try to sell it to you at a higher cost. They also try to upsell you on extra domains or features that you probably don't need. If you are new to buy domains, avoid being scammed by using a better registrar (maybe from the ones I will talk about next).

If you just want to buy a domain without being bothered about any other features I highly recommend [iwantmyname](https://iwantmyname.com/). They are a no-frills domain registrar, so they _only_ sell domains. If you need email, hosting, SSL, or anything else, you need to purchase them somewhere else. They have a clean UI and is, in my opinion, the easiest to navigate since they don't have endless menus to go through. They also automate setting up your DNS records with many services such as Fastmail, GitHub Pages, and Tumblr.

## Domains for Businesses

This is where it gets interesting. If you are a large company, stop fooling around and just go with [MarkMonitor](https://markmonitor.com/). MarkMonitor only works with large companies (or anyone who is willing to pay their fees I imagine) who have business critical domain names. Registrars for businesses should mainly offer the following services:

1. High account security
2. Good Support
3. Ability to share administration of domains
4. Invoice payments

MarkMonitor can pretty much check all of those conditions. It is amazing how few registrars check points 1 and 2 which should just be standard. Points 3 and 4 might not seem obvious at first so let me explain. Point 3 basically means that someone else can create an account at the registrar and be able to handle billing, DNS changes, etc. themselves as well. A lot of registrars don't allow this so you have to share credentials for the one account that can manage those domains, which is not ideal especially if you want to use 2-factor authentication. Point 4 helps prevent the issue where you card details are out of date or for some other reason the registrar is not able to charge you for a renewal. This can cause headaches later on when you find out that the renewal did not go through because your site, email, and services relying on your domains are down. They will renew the domains, then invoice you (and escalate if they don't get paid).

Now, what if you want some of the features MarkMonitor provides but aren't willing to pay the thousands of dollars needed for them to take you on as a client? This is where I believe [Gandi](https://www.gandi.net/en) comes in. They have a service called [Gandi Corporate Services](https://www.gandi.net/en/corporate) which is for companies that want more help managing their domain portfolio. This includes the important invoice payments that few registrars provide. The service costs $65 a month and you can cancel at any time.

Gandi also offers the largest selection of TLDs I have seen anywhere, although some of them require you to subscribe to their Corporate Services to purchase, supposedly because it requires more work on their end to register.

Other features Gandi has include:

1. Ability to manage personal and business domains with separate billing details using just one account
2. Strong 2-factor authentication support including U2F
3. You can become a reseller to manage the domains of clients
4. They are always on the bleeding edge of new TLDs

These are the reasons why I use Gandi for personal and business use.

## Summary

So my registrar recommendation is as follows:

- No-frills: iwantmyname
- Large business: MarkMonitor
- Medium business: Gandi with Corporate Services
- Not sure: Just go with Gandi, they are good with scaling from small project to a large business.


Thoughts on Domain Registrars


It’s no secret that Slack has mixed reviews. On one hand, people talk about how it has improved the experience of chatting inside of an organization, while others say that it is slow (Electron) and is moving us away from more open alternatives such as IRC.

As a freelancer, I don’t believe I would be able to switch all of my clients to use alternatives such as [Twist](https://twist.com) which has native applications. Instead, I tried to find different ways to access my Slack messages.

I recalled that Slack had an IRC gateway, but I later found out that it was removed. I thought that ruled out using IRC to communicate with Slack, but I then found out that [IRCCloud](https://www.irccloud.com) has a [feature that connects to Slack](https://blog.irccloud.com/slack-integration/) and makes the messages available through IRC. _This even works when you use IRCCloud as a bouncer_. This allowed me to access my Slack messages with _a native client_. Yes, _native_, not a glorified web app installed on your computer.

Unfortunately, you do have to pay $5 a month to access this feature and it’s still in beta. You also do not have access to your previous messages, only those going forward. I also had some issues reconnecting using the IRC client I was using (Textual 7), so I switched the IRCCloud web app for now. However, it is _much_ faster than the standard Slack app, even when using the IRCCloud web app.

We are probably stuck with Slack for a while longer. This workflow is not ideal and your data is still trapped within Slack. Perhaps open standards will improve enough so that people who are not tech savvy are able to use them in the future.


Using Slack Through an IRC Client


## Background

I am a freelance software developer. I mostly work on Ruby on Rails and Ember.js projects. Prior to this week, I would work almost exclusively in the terminal with NeoVim (_never_ a GUI Vim) and tmux. I still use tmux and [tmuxinator](https://github.com/tmuxinator/tmuxinator) to set up the dev environment for the current project I am working on (changing to the correct directory, switching to the correct ruby version, starting servers, etc.). I am not scared of the terminal and continue to use it for Git, the Rails console, running tests, among other things.

## Why I am leaving Vim

I have been using Vim for over 4 years. I am not the best Vim user, but I know it enough that any editor that forces me to use the mouse would make me less productive. I have slowly written a `.vimrc` that fits my needs and use a few plugins. I really liked that I could use it in the terminal where I could easily switch to do other things on the terminal.

What I didn't like about Vim was that a good number of plugins always felt like a hack. Vim has built-in functionality to be extended, but every plugin seemed to make it gradually slower, more than on other editors.

One example is CtrlP. It is currently the most recommended way of opening files in Vim. The default setup can be improved by using [The Silver Searcher](https://github.com/ggreer/the_silver_searcher) to find the files and using [ctrl-py-matcher](https://github.com/FelikZ/ctrlp-py-matcher) to match files with your search query. I _always_ had issues with it. From having to refresh every time I made a new file, to folders such as `node_modules` being included in the results unless I refreshed again (although this could be an issue with The Silver Searcher). I have tried for a long time to get it to work as well as it works on Sublime Text, but I was never able to get it even _close_ to the performance and quality of the results of Sublime Text.

Another popular plugin is Syntastic. I absolutely hate the way it displays the errors. It is not a smooth transition from "no error" to "error". Not much to say here, it just looks awful, yet it is highly recommended.

I don't use many plugins, most of them are just to work with newer languages such as Rust and Elixir. For all other plugins, you need to really consider whether you need it or not. If you don't stick to the bare essentials, it is likely that Vim will start to slowdown.

One of the settings that really slowed my Vim down, was the `relativenumber` setting. If you are not familiar with this setting, what it does is instead of displaying the actual line numbers in the file, it displays the how many lines away a line is relative to your current line. If that sounds confusing, just type `:set relativenumber` and you will see what I mean. Apparently this can make your Vim really slow when moving up and down. I had this setting for years and it never occurred to me that this could be affecting my Vim speed. It was only through some random web browsing that I found out that this made Vim much slower.

Other parts of Vim I just never really understood. Swap files have always annoyed me and I would disable them. When I would edit a file, Vim wouldn't automatically reload it without me doing it manually (and only realizing I need to do this after editing a file). In order to automatically reload files you need to add a line or two in your `.vimrc` to look for watch for file changes then reload. It's not a simple one line setting. I never tried it because I was worried about slowing down the editor.

Also, few people actually use Vim to its potential. "With Vim you never have to use a mouse" is not a good argument. Any popular editor usually has a "Vim mode" that you can install and you can achieve native Vim productivity. If you want to convince someone to specifically use the actual editor Vim, being mouse-less is not going to work if they can already emulate Vim commands in their editor.

And probably the most superficial and shallow reason I switched: I want prettier font rendering. I didn't spend money on a retina MacBook Pro to be limited to 256 colors.

## Why Sublime Text

My issues with Vim were small, but still annoying. I wanted to find an editor that was just as fast but wouldn't slow down to a crawl once I started modifying it to my needs. So I went back to Sublime Text which I haven't used since 2012. Back then, Vintage mode (the name for the "Vim" mode in Sublime Text) had a few missing commands that I regularly used in Vim. Now those commands are available in Sublime Text so I am not missing out on anything from switching away from Vim, at least for my workflow. It's also **much** easier to extend than Vim and does not slow down as much when customized. Sublime Text just feels faster and smoother for me than Vim ever did. If I had added all the Vim plugins needed to match Sublime Text's functionality, it would have been even slower.

I admit it, I never paid for Sublime Text when I was using it. So I decided to pay for it this time and support the project. Now, I know some of you will say: "But Atom and Visual Studio Code are free and open source!" I don't give a damn. Open source is an awesome thing. However, open source, with its potential to bring hundreds (if not thousands) of developers together to help create amazing software, has given us: editors that run slower than [what we had available in the 90s](https://en.wikipedia.org/wiki/Gedit). More and more "desktop" software will be built like this. I'd rather do my part to ensure good desktop software will continue to exist, even if it's proprietary. I will not use shitty software just because it's open source.

Also, why not Emacs? I wanted to change my editor, not my operating system :P.

## You should still learn how to use Vim

I hope I have not discouraged you to not learn how to use Vim. Everything I have learned in using Vim has carried over to Sublime Text. Almost every popular editor and IDE has a Vim plugin. Being forced to use only the keyboard in an editor has made me more productive. Also, if you will be accessing servers through SSH frequently, the best editor available is usually Vi or Vim.


Switching to Sublime Text from Vim in 2017


So you want to create an API only Phoenix app and don't want to include any of the frontend assets? Just run the following command when creating a new app:

`mix phoenix.new your_new_app --no-brunch --no-html`


Excluding Frontend Assets on Phoenix Apps


## TL;DR

`git config --global help.autoCorrect -1`

## The Details

Ah yes. Once again you have typed `git stauts` and have received the infamous response:

```
git: 'stauts' is not a git command. See 'git --help'.

Did you mean this?
	status
```

Never fear, you can configure git to automatically correct your typos using the `help.autoCorrect` configuration option. Here is the description in the documentation:

> Automatically correct and execute mistyped commands after waiting for the given number of deci-seconds (0.1 sec). If more than one command can be deduced from the entered text, nothing will be executed. If the value of this option is negative, the corrected command will be executed immediately. If the value is 0 - the command will be just shown but not executed. This is the default.

So basically, you use it like this:

`git config [--global | --local | --system ] help.autoCorrect [number of deciseconds (0.1 sec) to wait before executing or a negative number for to execute immediately]`

Here is how I use it:

`git config --global help.autoCorrect -1`

Now you will get this message:

> WARNING: You called a Git command named 'stauts', which does not exist.
> Continuing under the assumption that you meant 'status'

**Edit**: I updated the "TL;DR" command to run immediately instead of waiting 0.1 seconds.


Automatically correct typos in Git


I have been using vim for almost two years now and one thing that has always bothered me has been the escape key. According to the [Vim Tips Wiki](http://vim.wikia.com/wiki/Avoid_the_escape_key):

> The Vi editor was originally written on an ADM-3A terminal, which had the Escape key positioned where the Tab key occurs on most modern keyboards.

Well, I guess it would be a good idea to use something other than the escape key to return to normal mode. Luckily, `Ctrl-[` does the exact same thing as the escape key in vim. Although it does require you to use two fingers, I have felt that it doesn't cause as much of a disturbance to your vim workflow as the escape key does.

Another alternative would be `Ctrl-c`, which lets you return to normal mode. One of the things that I like most about `Ctrl-c` is that it doesn't show any lag in returning to normal mode like `Ctrl-[` and the escape key do. The downsides to using `Ctrl-c` is that it does not expand any abbreviations or trigger the **InsertLeave** event, so it could cause some issues with plugins or with your .vimrc config.

Of course, you can always create your own custom mapping (e.g. `:inoremap jj <Esc>`) for the escape key if you don't like any of the alternatives. I am probably going to play around with the alternatives and see which one I like most for a bit longer though.

**Edit:** Updated `:imap jj <Esc>` to `:inoremap jj <Esc>`. Thanks to [bilalq's comment on HN](https://news.ycombinator.com/item?id=13101691).


Escape key alternatives in Vim


QWERTY, the most common keyboard layout out there, was designed to prevent you from typing too fast! If you want the long story (and part of how Dvorak was created) you can read [The Dvorak Zine](http://www.dvzine.org/zine/index.html). If you want the short story, QWERTY was designed to place the most common letters away from each other because, in typewriters, it was common for keys to jam when typing too fast. So why are we still using this layout when we don't have to worry about this anymore? It's time for everyone to switch to a new layout.

About a month ago, I decided to switch to a different layout. I only knew about Dvorak and Colemak at the time and, because Dvorak was more popular, I decided to go with Dvorak. After two days, Dvorak wasn't for me. The 'L' key was in an awkward position where I had to use my pinky a lot even though it is a common letter. I also didn't like how it rearranged the punctuation marks. It was kind of disappointing that my first attempt of switching layouts was a failure, but I kept at it with Colemak.

Colemak is a much newer layout. It improves on some of the flaws that Dvorak has and doesn't change as many keys as Dvorak from the QWERTY layout, only 17 are moved. I almost have a month with Colemak and I don't plan on ever going back to QWERTY (well, it will remain on my phone). I still haven't reached my QWERTY speeds (I was able to type at around 100 wpm) but I am slowly improving (currently at 45 wpm).

Switching layouts is not easy, you have to unlearn QWERTY and will be hitting keys thinking you are still in the QWERTY layout. But if you begin the journey, don't give up! Here is a list of benefits, listed on the [Colemak FAQ](http://colemak.com/wiki/index.php?title=FAQ) that Colemak has over other layouts to get/keep you motivated:

> - There's an active and growing community of users.
> - Colemak has been designed to be easy to learn and to allow easy transition from QWERTY.
> - Colemak keeps the Ctrl+Z/X/C/V in the same place.
> - Colemak doesn't overwork the pinkies.
> - Colemak's design is considered stable.
> - Colemak is available for many operating systems.
> - Colemak has typing lessons.

I hope you will join me in leaving QWERTY.


QWERTY is slow and inefficient. Try a more modern keyboard layout instead.

Switching from QWERTY to Colemak


I am using the [Ghost hosted platform](https://ghost.org/pricing/). They support custom domains and I wanted to use my root domain (eduardobautista.com) instead of a subdomain (www.eduardobautista.com). Ghost advises against doing this but I was not sure why; they are not that clear about the reason.

So despite their warning, I went ahead and did it anyway. That's how I found out that when you add a CNAME record on the root domain, it will no longer use your DNS records. I found out about this when I was expecting an email and it never arrived (I use MX records on my domain because I use Google apps).

So that is why my blog is located at www.eduardobautista.com instead of just eduardobautista.com.

~~**Update:** I am no longer using Ghost and have switched to Jekyll, but this is still useful for anyone that might run into the same problem.~~

~~**Update #2:** I switched back to Ghost. They now offer a desktop app which I believe is awesome.~~

~~**Update #3 (2017-04-17):** I switched back to Jekyll. A static site handles heavy loads with ease.~~

~~**Update #4 (2019-01-11):** I switched to [Hakyll](https://jaspervdj.be/hakyll/) for the faster compile time.~~

**Update #5 (2020-06-08):** I switched to [Gatsby](https://www.gatsbyjs.org/) because I gave in to our TypeScript overlords.
