Rob Fahrni

nnw.ranchero.com

NetNewsWire 6.0.1 for iOS

Congratulations to the NNW team! 🥳

Y’all are an inspiration!

If you’re a recruiter from Apple, Microsoft, Amazon, Google, or Facebook please save yourself some time and don’t pursue me as a candidate.

I’ve been around for a long time as a developer but I’m not someone you’d want to hire.

I’m not that smart. More than likely I won’t be able to get through your interview process. I do ok as a developer but I’m not an algorithm guy. I try my best to write easy to read, maintainable, and stable code.

I’m not a ninja, rock star, or 10x developer. I’m kind of slow paced, iterative, and discerning. Definitely not a code factory.

If you’ve ever seen the movie Bull Durham I’d compare myself to Crash Davis – played by Kevin Costner. My best days are behind me. I had my time in The Show as part of Visio – which became part of Microsoft – but that was over 20-years ago. I’m in that stage of my career where I’m trying to help younger developers learn the business so they can get to The Show.

I still like to write code but I’m not a good fit for Big COs. I don’t have the energy or desire to work 80-hours a week for months on end. Living for the company.

I’m beginning my twilight season and I’m certain you’ll find your perfect candidate.

Take care, and remember, we only get one shot at this beautiful life. Make the most of it.

P.S. - If you’re a Facebook recruiter, this is the page I want you to see. You should really question why you work for a company like Facebook. If the answer is “because money” you’re doing it wrong. Facebook is a vile company.

Mr. Fredricksen

Quartz: “The investment in Redmond also assumes workers will be be using offices in the future, when it’s clear many won’t. When given the choice of working remotely, many workers opt to stay home (or find a nearby coffeeshop), to the extent that Automattic, maker of Wordpress, shut down its San Francisco office because no one was using it."

This is from a late 2017 article. Clearly folks have wanted to work remotely for quite a while. COVID-19 has definitely driven that point home, no pun intended.

I’ll continue to advocate for it.

Well, hello there, beautiful.

Bug holding a cicada. Either it’s a dead one or an exoskeleton of one.

Either way, they’re cool looking.

Kim found these this afternoon while watering our butterfly bush.

Found this little fella while clearing out some underbrush and trimming up trees today. Nifty little critter.

So far my – for personal use only, Cocoa inspired – framework is coming along. It’s not like I haven’t done some of these types of things in the past.

Cocoa Inspired Classes

• HSString
• HSArray
• HSDictionary
• HSThread

Of the classes above HSThread is not yet complete. It may become something like HSOperation. I’m not sure yet. Grand Central Dispatch is so good I could see using it under the hood. I could also go with modern C++ thread support. Which might be the easiest way to get my threading for any platform. We’ll see.

The String, Array, and Dictionary classes are working just fine. Complete with unit tests.

For the String class I’d like to add sub string finding and a split method. I’d imagine it will expand as I continue using it.

Not Cocoa

• HSDatabase
• HSResultSet
• HSRef

The database and result set classes were inspired by the great FMDB.

HSRef is a smart pointer implementation. Used throughout the library.

These classes are mostly complete and what has been completed is fully unit tested.

Interfaces

Remember in C++ they’re interfaces, not protocols.

• Enumerator
• Database
• Collection

Enumerator and Collection are used for HSArray and HSDictionary. Which, until a change is necessary, both use the same base class that uses a linked list for storage. If/when it becomes too inefficient to use I’ll change to something else. For my needs, it’s fine.

What’s Missing

Still need networking and parsing for HTML, XML, JSON. Which will lead to RSS, OPML, and JSON feed parsing.

Then Stream C++ models and view models will layer on top of those.

The framework code is currently building and unit tested on macOS. I’ll get everything running on Windows next.

Closing Thought

Remember, this framework is purpose built to serve the needs of the underpinnings necessary to build a cross platform Stream. So, yes, it is definitely going to be missing stuff. A lot of stuff.

In the end I’m a bit undecided on what direction to go for the UI. On Apple platforms will I go UIKit and AppKit or SwiftUI? On Windows is it going to be all C++ or will I use C#? On Windows I am definitely building for Windows 11 using WinUI 3.

I figure this is going to take quite a while to complete. I need to get Stream for Mac out the door and do some more features, across the board, for both.

I think the one little thing that bothers me about the whole 1Password using Electron kerfuffle is a statement made by Michael Fey, that I can’t seem to find now, so I may be completely off here. Even if it wasn’t Mr. Fey who said it, it still bothers me.

Me paraphrasing.

“The Electron UI is a thin layer over the Rust engine.”

That is a really great design, right? All the shared code jammed into this newly built Rust engine. Rust is fast and built so folks avoid lower level pitfalls. Great!

What I want to understand better is, why not use your existing Mac and Windows desktop client code and build feature parity on top of that new engine? Mr. Fey had mentioned the Mac and Windows clients had diverged so just work toward getting them in sync?

On the flip side how can you not consider using Electron when what you’re selling is a service? Yes, 1Password is a service that works across multiple platforms. I tip my hat to the entire team for that. They’re going to provide an easy to use password management system across iOS, Mac, Android, Windows, and Linux. Wow! That’s quite an accomplishment.

Would I love a native client? Of course I would. Will I stop using 1Password? Nope.

Oh, something to keep in mind my fellow Mac psychofants. An M1 Mac can run the iOS version of 1Password, it’s a native application.

Flynn living his best life. Chillin on two pillows.

It’s Twitter pile on time for the fine folks behind 1Password. I, myself, thought about joining the hoard screaming how horrible they are to abandon native apps in favor of Electron. I think I’d be justified as a Mac loving sycophant.

After my initial disappointment I thought I’d think on it for a bit before writing my screed to Agilebits.

I don’t think I’m gonna do that, I’m gonna give the app a chance. Still love the app. Given their history and reputation I’ll wait and see how it turns out.

From the Horses Mouth

Michael Fey, their V.P. of Client Applications, and the decision maker on the matter has a nice writeup on their weblog about their challenges around rewriting their client code – which is, BTW, something you should avoid because it’s costly.

Ultimately we made the painful decision to stop work on the SwiftUI Mac app and focus our SwiftUI efforts on iOS, allowing the Electron app to cover all of our supported Mac operating systems.

As Mr. Fey points out. It was a simple business decision. They’re no longer the scrappy little startup. They’re big, and growing, and now they’ve taken VC money, which I’m certain brings its own set of expectations.

The Pundit

Over at Six Colors – one of my personal faves – Jason Snell is having none of it.

1Password, originally a Mac-forward software developer, has simply decided that the Mac isn’t important enough.

A lot of folks will think that’s really harsh of Snell. The entire article is worth a read. You can tell he’s torn. But damnit, this is, after all, the Mac! We Mac zealots are very picky.

The Softie Perspective

I follow a lot of Apple and Microsoft folks, pundits and employees. I follow a Softie named Kyle Pflug and he’s disappointed by backlash to Mr. Fey’s blog post.

Great blog post. Horrifyingly toxic replies. Sigh.

Take a minute to read Mr. Pflug’s Twitter thread. It’s worth it.

Apple’s Involvement

Apple is a very slow moving company. Folks may disagree and that’s ok. They’ve been heavily criticized for their slowness in adopting web standards in Safari.

In a similar vein, Safari has consistently lagged behind competing browsers in supporting modern web APIs and features, presenting considerable challenges for developers wanting to create products that work consistently across all the major browsers (Chrome, Edge, Firefox, and Safari).

Mr. Sun points out Apple’s slowness, or reluctance, to support modern web standards that allow web apps to be more like desktop apps.

A more standards based Safari would allow folks to make better web apps and give users more choice on their platform. That would mean we wouldn’t have to install another browser to work with certain web apps and web sites. They would just work.

But Native Apps

We come back around to the native apps argument. Apple really wants developers to use their frameworks to create the best applications your money can buy for the Mac. As a developer, fan, and everyday user of a Mac I want the best possible experience I can get from my computer.

As noted earlier, Apple could, and should, embrace the open web 100%. Why? Good question. Because developers making these choices are not going to suddenly change their minds when they’ve decided the lowest common denominator experience is good enough for their users. 1Password isn’t some super sexy illustration or photo editing software. It’s a boring utility that does a job and does it really well I might add – yes, I’m a user and recommend it all the time.

If Apple would invest more time and resources to making Safari the best possible platform for creating apps and, at the same time, work on Electron so it would be a more performant, better citizen on the Mac, wouldn’t that be good for everyone? Imagine if you would an Electron for Mac package that includes Safari as it’s engine instead of Chrome. That would be amazing and would hopefully make the Electron experience suck less on the Mac!

The Problem

<img border=“0” src=“http://static.crabapples.net/apple-rotten.jpg" align=“right” alt=“Rotten Apple”/>In the end the big issue with that line of thinking is 15-30%. Apple likes it’s cut of each and every app sold in its App Stores. Granted we’re talking about Mac Apps here which, as of this writing, can still be distributed outside of the Mac App Store.

Apple also loves its platform and honestly believes its tools are the best to use when creating desktop applications. They have a long history of delivering on that belief; Objective-C, Cocoa, AppKit, UIKit, Swift, Catalyst, and finally SwiftUI. The last few years they’ve moved extremely quickly to deliver new developer tools. They really want us to take advantage of all that effort to turn out the best apps possible.

But, but…

I got nothing in conclusion. I love making native apps but I also use a few Electron apps I really enjoy, like Slack and Notion to name a couple. Sure, at times, the hog memory and make my computer fans start roaring, but restarting them usually fixes it for a while. The only native app that ever does that on my computer is Xcode and it gets a pass because it a dev tool.

Everyone is going to have to make their own choice about Electron based apps going forward because they’re here to stay no matter what we zealots and Apple say.

WordPress is an awesome platform for building all kinds of amazing websites, no doubt. According to WordPress “42% of the web is built on WordPress.” That’s a crazy big number.

There are times I wish I had the chops and the time to turn WordPress into a headless API that generated static HTML. I know, I know, folks think I’m nuts for my constant harping on this idea. But I like static sites. That’s why the rebirth of this site is running on Micro.blog, it spits out raw HTML.

WordPress already exposes two API’s that could serve as the headless part I’m after. A lot of tools use those API’s today. But how about a simple, get out of the way, editor? I mean one devoid of all the administrative stuff needed to properly operate a WordPress site? An editor purpose built for writing. Yes, yes, keep the current admin site, we’ve already established it’s needed.

Next create a layer on top of the entire WordPress engine that publishes content when it’s published to the backend. This is where quite a bit of time could be spent to make my perfect system.

Heh, in the end what would you get? You’d get Micro.blog with a WordPress backend.

Hey, WordPress, how about you buy Micro.blog and get Manton to build this perfect blogging tool on top of WordPress as a backend?

I really miss San Luis Obispo.

Should’ve never left.

Kim and I

Bubblegum Alley, San Luis Obispo 2013

Remember when Instagram was fun and all about the filters?

When I moved my blog to iam.fahrni.me I never really liked the subdomain name I picked but it worked, so I carried on with it, for quite a number of years; 2010 - 2021.

I’m back to blogging at home, here, where I started on January 21, 2001. It’s been over 20 years.

Here’s hoping I can give it at least 20 more, 40 if life allows it.

Let them die 💀

Flynn doing what Flynn does.

We lost a family member today. Ms. Sophie was 19-years old. She was a constant companion while I watched TV at night after everyone went to bed. She was super sweet and would purr when you’d touch her, even to the end. ❤️

Windows 11

Hope and Worry

Automattic ❤️ Day One

Windows 11: Sun Valley