My Experience with MailSpring on Linux


Email was a significant improvement over its predecessors. Instead of physical messages passing hands several times, a message could be sent at the speed of light as a series of packets to anyone attached to the same network. Email changed the world.

Now email is old tech. But things that are old have a way of becoming new again. IRC is pretty old tech, but its latest evolution via Slack, RocketChat, and Discord seem to have given new life to the ideas behind IRC.

In the space of email, many companies have put effort into creating a clean interface to help people manage their inboxes. My least favorite of all of these is Microsoft Outlook, which I’m required to use at work. Between its lack of theming and limited functionality (work turned off the RSS feed reader) it epitomizes critical mass stagnation.

Microsoft isn’t the only one who’s tried to solve the email problem. Google entered the space with Gmail, which has gone through a couple of interface redesigns. Mobile reduced our inboxes to tiny screens, where each pixel represented a valuable piece of real estate. This pixel scarcity furthered the design of email interfaces. Google now has an app called Gmail and another called Inbox. The latter enables snoozing of messages to be rescheduled at later times.

On the Linux Desktop, there are quite a few choices for email applications. Each of these has their own pros and cons which should be weighed depending on one’s needs. Some clients will have MS Exchange support. Others do not. In general, because email is reasonably close to free (and yes, we can thank Hotmail for that) it has been a difficult place to make money. Without a cash flow to encourage developers, development has trickled at best.

Thunderbird and Nylas Mail have both been discontinued by their original sponsors. Thunderbird still works, comes preinstalled in many distributions, and has an interface reminiscent of earlier days. For all intents and purposes, Nylas solved the interface problem and open sourced the code they did it with. So last summer when development was suspended the project got forked.

Welcome to MailSpring!

MailSpring offers the clean interface of Nylas and an easy install via a snap. Unlike its predecessor MailSpring will let you adjust the text size to accommodate screen scaling and the age of one’s eyes. There’s a limited theming function. Snoozing emails appears to be part of the original design and it’s a well done feature of the app. The app handles signatures quite well. It uses industry standards for URL tracking with your email. And you can’t help but feel like the whole app is designed to encourage you to use the paid version of their service. They do keep these opportunities to remind you pretty low key, but they’re still baked into the app.

I’ve been successfully running MailSpring on Ubuntu 16.04 for a month now and have enjoyed the experience. Notifications are working well and the app looks like it belongs in this era.

What else can I say? Email clients aren’t exactly new territory. There’s not much in the space to get excited about. It works. It’s clean. And email is still a necessary evil. If you’re going to have to use it, it might as well be using an app that lets you use it on your own terms. MailSpring hits the mark for me.

Does Adobe Hate Linux?


As the press prepares to cover the release of Ubuntu 17.04, it should be clear in the tech industry just how big of a player Ubuntu is to the ecosystem. While a good bit of reviews will focus on what’s new in the release and what’s headed down the pipeline, I’d like to comment on what’s still missing and better yet, what can be done about it.

What’s missing is a graphics suite and there’s really no excuse for not having one. Yes, we have graphics applications, but there are advantages to having a suite, not just a one-off application that can do something in 12 steps when its competitor can do it in three. The industry leader in this market is Adobe, whose Creative Cloud suite is leaps and bounds away from its competitors in terms of market share.

Of all companies, Canonical should understand why market share matters. Market share translates to a community conversation that reduces the burden of support from the vendor. Need to learn how to do something on Ubuntu and with a quick Google search there are hundreds of helpful responses to the same question from the community. Want to know how to change someone’s eye color on a photo and your search will return with YouTube videos explaining just that. (an online courseware website) has precisely 1 series on Gimp, but 23,932 on Photoshop. These numbers are a good indicator of market demand for a product. The real problem here isn’t recognizing demand; it’s charting a course to satisfy that demand.

From what I can see, Adobe’s not going to be bullied into porting their software. If bullying and requests worked then the numerous forum requests for Photoshop on Linux would have made their mark years ago. They haven’t been the catalyst people had hoped for, but we do have a potential catalyst in the community, Mark Shuttleworth.

Canonical’s recent past provides a template that can lead us to a solution. While exploring the mobile side of Linux, Canonical has developed SNAPs. They’ve also learned as an organization to work with outside partners. I can run Ubuntu inside of Windows! That would have been a blasphemous statement ten years ago and now it’s very much a reality. It happened because Canonical is uniquely positioned and willing to partner in the market.

Another aspect that gives Canonical a unique advantage is the ability of the ecosystem to adopt its choices. While more innovation may occur on bleeding edge distros, more adoption occurs through decisions at Canonical. Ubuntu is the base for not only its flavors, but a lot of derivatives. Where it goes a large part of the market goes. If Adobe wanted to move into Linux, Ubuntu would be a key part of that strategy.

Mark Shuttleworth took the bold idea of convergence and went and built the most popular indiegogo campaign of 2014. The campaign didn’t reach its goal, but it didn’t fail to garner the attention that moved the tech community forward. I think it’s time for Mark to try another round on indiegogo, and this time being the attention back on the desktop. Run a campaign for funding to buy the first license for Photoshop on Ubuntu. Call Adobe and ask for the price tag and let us help you pay for it. Once they’ve got one customer, they’ll be able to have more. It’ll shift the burden of production costs from Adobe to the market and prove at the same time that there’s market demand to justify that shift. This campaign needs a face, and there’s no one better than Mark.

Mark, Ask for the full Creative Cloud Suite, but get started with Photoshop and let the momentum build from there. Start soon and you can have the campaign coincide with Photoshop being delivered in time for 18.04, the next long term release.

When I first used Ubuntu it had the motto of Linux for human beings. It was a great tag line to tell the world how your software choices were going to match the needs of the market. I’d like to see that tagline return and have the needs of market extend beyond the nerds who know and understand why ZFS is significant. I’d like to see the tagline apply to folks who truly want to get things done and need a suite that will elegantly allow us to do just that. If I get more choices when I run open source software why can’t I get to choose to run the best software for my work flow?

Mark, give Adobe a call and I’ll gladly give from my wallet to help make it a reality.

LibreOffice 5.3 – Freedom Meets Functionality


Freedom to create with code is not the same as the freedom to create a specific product. Sometimes the freedom offered in the open source community makes it easier for me to be more productive. Other times, not so much. The biggest excuses I have to grab one of my machines with a closed source operating system consists of the following photo editing (Adobe CC), video editing (Final Cut Pro), and Civilization IV. Yes, I’m still playing Civ IV. It’s my favorite. I don’t need to upgrade. I’d love to find a tutorial that worked to get it working under Neon, but sadly the community that would write such a post appears to have moved on.

I used to think that I couldn’t create documents under Linux but LibreOffice 5.3 has really been a game changer. Everyone else beat me to the flashy reviews, so this isn’t a review that exposes the new features. This is a commentary of my experience.

The ribbon (Notebook Bar) feature they’ve added is perfect because although it costs me a bit of screen real estate, I get to have both the menu and the ribbon. It’s not intrusive, it’s complimentary. I get quick access to the things I need using two familiar ways to find them. Talk about flexibility designed for meeting user expectations! At work, I have to use the MS Office Suite and it always bothers me that I can’t quite tweak the interface to what I’d prefer. For LibreOffice, I didn’t have to tweak much once I enabled the ribbon.


The application does have a limit with larger files, but this is common amongst most text & document editors. When I was writing one of my books, I hated that it would take Word so long to load my file. The book wasn’t even that big but there was a huge amount of lag while I would wait for it to load. I threw a 1.7+GB text file at LibreOffice the other day and on a machine with 24GB of RAM. It decided it wasn’t going to play nicely. It’s not hard to understand why. That’s a lot of text to render. The 508 page book I opened didn’t have an issue. Sure, it took just a bit for the pages to load so I could scroll down, but it’s still rather impressive how responsive it was. I’ve never had MS Word feel this responsive. Ever!

My understanding is that building up to 5.3 involved a lot of code cleaning and making this a solid product underneath. It seems to have worked for ‘em. I’ve now composed several articles and most of my homework for this semester in the application suite. It’s held up for every single one. It doesn’t take much to get me to switch to a new application but it takes a lot to get me to stay there. So far, I have no reason to leave. I can easily switch from .odt to .docx files for things I have to share with folks at work. My midterm is due this week and it’ll be done with LibreOffice. For the first time in 19 years, I’m looking for my next job. The resume is getting worked and reworked in this excellent application. That’s how confident I am that it can deliver.

I know lots of other people already covered the features, but it felt not enough of us talked about how solid this thing is. Talk about reliable. When I’ve had my computer lose power without saving, it recovered every last character of my work. The only part that really annoys me about the application is the default fonts. The Libre fonts are nice, but nice isn’t the way I want to dress up my text. I usually want wow. Thankfully, it’s not hard to change the defaults.

While there are some parts of my work flow that keep me working in Windows and OS-X, document production is no longer one of them. When it comes to this software freedom has given me the freedom to produce efficiently.

The Novelty of KDE Neon

KDE Neon

The good folks at KDE managed to engage a market of Linux desktop users underserved by other distribution models. Or, maybe it’s just me.

KDE has a long history in the desktop ecosystem. It was the first Linux desktop I was exposed to back in 2006. Back then, it was on OpenSUSE and it was clean and functional. For some reason after that, installing KDE had never really appealed to me. I’ve tested it out briefly when poking around at what the OpenSUSE guys were doing and I’ve run Kubuntu for brief snippets. For years, I’ve been trying to find out what type of desktop user I am and which distro fits my needs.

I’ll admit that I’m a moving target. What I like changes depending on workflow expectations and machine capabilities. I’ve started to notice that I expect a desktop to have modern features while still maintaining a familiarity that doesn’t cost me days to adjust to the new environment. In this space, there are a lot of really neat contenders. Gnome has got a great desktop, but it renders slow on X. Budgie on Solus is a cool project with a lot of momentum, but isn’t ready for what I’d like to do (yet). Ubuntu MATE got me back to running Linux full time and it’s so familiar it’s hard to not use it. In addition, the software boutique and MATE tweak features gave me some really great access to modern features and software. Why the software boutique isn’t installed by default on every distro is just beyond my comprehension. That thing is just amazing!

MATE failed me when I got a new Hi-DPI machine with a backlit keyboard. In some cases, there were solutions I could apply to the issues arising from this new machine, but over time it became too cumbersome to keep searching to find them. I had to get work done. I knew the team was working to fix them, but the gap between the team fixing the issues and me needing to be functional was too long for my patience. The sour onion in my sandwich was when I had trouble grabbing the edge of a window to resize it and realizing how many attempts it took to perform this simple task. Sure, I scaled up my fonts, but that didn’t scale up the window edges.

I poked around at other distros and eventually landed on KDE Neon.

I know that to gongoozle is a verb that means to stare idly at a canal or watercourse. It’s an oddly specific verb and it’s even more odd that I know what it means. That being said, I’ve never bothered to learn how the community labels some things as distributions and some things as not distributions.

Here’s what I can gather:

  • Neon is not a distribution. It’s a desktop that sits on top of Ubuntu LTS (currently 16.04).
  • It’s not Kubuntu. Kubuntu is a Ubuntu flavor.
  • The word flavor reminds me of ice cream. So I guess when you’re trying a flavor you’re licking it?

So I guess the way to look at this is that Neon is an open-faced sandwich where the bread was made by the Ubuntu bakery. It’s good bread.

What I get on Neon is a desktop that’s updating and becoming more refined while still maintaining the underpinnings of what makes Ubuntu so marketable. This is exactly what’s missing from the Ubuntu ecosystem. In that ecosystem, you can run dated a dated desktop for several years and watch its wrinkles become more frustrating over time. Or you can run the nightlies as your OS and watch things break and get fixed. You’ll have the latest desktop the good folks have selected, but it may not work the way you’d expect. I did this for several months and it was unpolished but quite enjoyable.

So then there’s Neon. The desktop updates as needed and with the underpinnings of 16.04 still get you SNAPs, ZFS, and a great repository of software. Since I’m human, I interface with the machine through the desktop (and the occasional command line). I don’t directly interface with the code underneath. I want clean lines and elegant functional design. I want to be able to resize my windows on the first try. In Neon, scaling for Hi-DPI is easy, font management is excellent, alt+space launcher is awesome, super key search is flawless (even works when I misspell things), and not only are there elegant lines there’s an amazing amount of design thought into the way everything works and works together. I get that, and all the familiarity of the Ubuntu stack underneath.

You might pick on me for touting font management, but it’s a serious indicator of a polished desktop. If the font management is good, it’s likely because the design team had people on it who understand fonts. So you’re only likely to see this on a more polished desktops. Font management is also never the priority. So if the developers got around to getting it done, then it means they’ve worked through quite a large stack of issues to get fonts going. So yeah, for me you can tell the quality of the desktop by the way it manages fonts.

It might just be me, but I believe these KDE guys are on to something. They’ve been able to federate the effort of one of the most popular Linux distros (Ubuntu) and marry that to their effort on the desktop. They’ve created a wonderful balance for a symbiotic relationship that benefits both parties. Canonical should be advertising this solution while their users are waiting for Unity 8.

Groke is another old fashioned word with Scottish origins. It’s a verb that means to gaze at someone while they’re eating in the hopes that they will share their food with you. Folks who have been working with Ubuntu’s deployment scheduled releases have certainly benefited from their professionalism over the years, but I’ve heard of many who’ve been groking at those with rolling release desktops. Now finally it seems they can have both. KDE Neon is where the rolling release desktop meets a stable foundation and it’s where a great team hit the moving target of what I want in a distro. Thank you! Now I have more time to gongoozle.

2016 was the Best Year for Linux

2016 was unequivocally Linux’s best year yet. It’s on more devices than ever before and more secure than ever before. Were there embarrassing moments along the way? Yes, I kept reasonably close to the news and watched a few of these evolve and get patched as quickly as they were found.

I’d also like to predict that 2017 will end being Linux’s best year yet. And I’ll even go one year further than other folks making predictions and say that 2018 will top them all.

For those of you who are glass half empty folks, let’s talk about a couple of the flaws found in 2016. LUKS looked pretty bad and Dirty Cow caused a few headaches, but the latter had a patch available within hours. And because it’s worth being redundant, let’s remember there was a patch within hours. While some would argue that the potential attack time for dirty cow was nine years, the published attack time was only a matter of hours. If you want to be a glass half empty type of person go ahead and set your clock for nine years. I still think that it was Linux’s best year ever and next year will be even better.

Why can I say this? Because Linux is honest. Honesty doesn’t mean perfection. It means openness. Linux’s faults are out there and ready for the world to see. Sometimes they’re caught early and sometimes they’re caught later.

2016 did burst the bubble on the narrative that in the land of a thousand eyeballs all bugs are shallow. It’s sounds nice, and I’m sure some projects run that way, but that’s not the way things are done anymore. I hope in 2017 we can make a better argument for open source security, and we can do it by talking about our talent management.

Outside of the honesty in the open source ecosystem, the open source talent management is our second greatest asset and every project lead knows how to leverage it. Jim Collins’ book Good to Great highlights business practices that if followed drastically improve a company’s performance in the long term. One of the most core principles is hiring the right talent, even if you have to wait for that talent to emerge. Linux’s talent management is unsurpassed because the power of that talent is published.

Want to know how good someone is? Read their code. Want to know how passionate they are? Read their posts. Bryan Lunduke has a full time job for being loud and passionate and remarkably entertaining while he flirts with a bit of rudeness. I know two project leads that recruit hires specifically from their volunteer pools. I’ve heard of Redhat and others doing the same. From what I can see, the researchers finding the bugs in the code aren’t locked behind ivory towers of corporate influence, they’re emancipated. They get hired to work on what they love and what they’re good at. They find the flaws and responsibly disclose what they’ve found. Because of their paycheck, they have the ability to research the technology that often doesn’t get looked at.

Yes these flaws get press. Shouldn’t they? Isn’t press good? Doesn’t it encourage us to audit more and improve? We’re doing that. Are the other guys? I’m confident that Redmond and Cupertino have areas that don’t gather much attention. I have a hiDPI screen and sometimes run Windows 10. I can see the areas they didn’t think anyone would notice. Not everything in Windows 10 has a hiDPI icon. I noticed. That same machine is a Lenovo. For the first time since 2009, it’s not a MacBook. Why? Because when I look at the MacBook, it’s easy to see there’s a whole division at Cupertino that isn’t getting much attention.

While our efforts may be more ad-hoc, our talent management program is better in the long term. My current employer hires only on the basis of certifications and I can see how that affects our workforce. Since they started enforcing certifications, we stopped innovating and instead merely executed. While I’m not knocking certifications, organizations who rely solely on certifications for hires often miss out on the right talent to take them from good to great. While the open source community has its share of certifications, it more importantly has an open repository of talent information.

How do we combat the glass is half empty people in the blogosphere and the occasional pessimistic podcast? Talk about our talent library. Our talent library created some of the most inventive and functional desktop interfaces ever dreamed of and only for 2% of the desktop market share. Imagine how good things will get as that percentage grows! We’re talking a lot about this year about Solus, but with a larger market share how many more Soluses are we going to see ahead? Our talent library brought us a great 2016, and it’s destined to bring us an even better 2017.

Overthinking Font Installation

Over Thinking Installing Fonts Title

If you overthink a problem, you’re bound to make it harder than it needs to be. My experience with Linux has taught me to respect the simplicity and elegance of the command line, easy access to system files and the beauty of collaboration that creates such functional tools. Recently, I’ve been impassioned by what is likely another very odd hobby. I’ve fallen in love with fonts.

My passion with this newfound arena has grown in part because it feels like I’ve stumbled on a world that’s always been there, but I’ve never noticed it before. It’s like when I took the math class in college on contemporary mathematical topics and learned about mobius strips, gerrymandering, gps satellite calculations, and later hexaflexagons. It’s like when Luke discovers the force or you find there’s multiple ways to end a level in Super Mario World.

Over Thinking Installing Fonts

Those uninitiated to the font world who might need an introduction could start with this quick rundown of resources:

For those of you already aware of this amazing world of letter dressing you’ll find the above list remarkably refreshing and easy to share.

Now, let’s talk about trying to figure out how to manage fonts on Linux. Because my other operating systems (Max OS-X and Windows 10) make managing fonts with font managers I expected the same on Linux. While I was playing around with Kubuntu over the weekend I discovered a very elegant interface to a font manager. The plasma desktop’s clean lines and thoughtful functionality asked me if I wanted to mass install my collection of 534 fonts to the user or to the system. It accepted open type fonts (OTF) and truetype fonts (TTF) without any issue. Wonderful.

But after testing Kubuntu for a weekend, the plasma desktop didn’t feel like it was my cup of tea. I’ve traditionally been a Ubuntu MATE guy, not just because I like its project lead but because the familiarity made me more productive. I made managing fonts on MATE harder than it needed to be. I kept searching for a font manager only to find orphaned posts about font managers whose latest updates were several software generations old.

I googled harder, to no avail.

Finally, I just picked a file and clicked on the file for Trajan Pro. Such an elegant font! It has the most wonderful descending j that balances delightfully with the rest of the characters. I was surprised that after I clicked it I was greeted by a dialogue with a button to install the font. BOOM! I was in business. Individually, this solution would work to manage fonts. I was disappointed that the character preview wasn’t working, but I generally preview my fonts elsewhere before I install them on the system, so I could overlook this issue.

So if I were writing a guide for installing a single font it would look like:

1. Close the application you want to use the font in
2. Download your font.
3. Find it in the folder you downloaded
4. Double click on it.
5. Press the Install Font Button.
6. Open your application you want to use the font in
7. Donate to the project of your choice (optional)

Over Thinking Installing Fonts2

As I mentioned before, I didn’t have just one or two fonts to import. I had 534! While I did contemplate clicking the install font button 534 times I reminded myself that I was probably over-thinking it. There must be an easier way. And here’s how easy it is:

1. Close the application you want to use the font in
2. Download your font.
3. Open a File Manager Window to your home directory and view hidden files (CTRL+H)
4. Open another File Manager Window and find the fonts you downloaded
5. Copy them into the .fonts folder
6. Open your application you want to use the font in
7. Donate to the project of your choice (optional)

After discovering how easy it was to adopt my font library I was once again reminded by my love for the simple elegance that Linux offers to my weekly workflow. It’s the same love for simplicity and elegance that had me enthralled when I discovered that fonts were a thing.