inessential by Brent Simmons

I can’t help but wonder if — given equivalent knowledge of UIKit and AppKit — iOS development isn’t harder now than Mac app development. At least for some kinds of apps.

Mac apps don’t have to deal with size classes, safe area insets, two very different classes of devices, getting killed by the system, a split view controller that isn’t suitable for some common purposes, presentation controllers, user activities, and — toughest of all — background app refreshing.

And some things that were Mac-only, such as multiple windows and contextual menus, are now iOS features too.

Even if I’m wrong, I can’t help but notice, as we work on NetNewsWire for iOS, that iOS development is starting to approach Mac-app-like complexity, and is already more complex in some areas.

Why I Listen to Podcasts at 1x Speed

On my microblog I mentioned that I always listen to podcasts at 1x speed.

Here’s why:

We’re in danger, I think, of treating everything as if it’s some measure of our productivity. Number of steps taken, emails replied-to, articles read, podcasts listened-to.

While accomplishing things — or just plain getting our work done — is important, it’s also important that not everything go in that bucket. The life where everything is measured is not really a full life: we need room for the un-measured, the not-obsessed-about, the casual, the fun-for-fun’s sake.

So I’m in no hurry. I will never, ever be caught up on all the podcasts I’d like to listen to. So, instead, I just play whatever I feel like whenever I feel like listening.

I’ll miss things, and that’s totally fine. But, in the meantime, I get to listen to the human voice somewhat close to realistically, with its the natural human pauses, with its rhythms and flows relatively unmediated and natural. Its warmth and music means so much more to me than being caught up.

But, again — I’m not saying this is right for you. But I would remind people that we have choices about what falls under productivity and what doesn’t.

The Apple Curtain

I’m always happy for a friend when they start a job at Apple — but I’m also sad when it means they have to stop their community activities: no more podcasting and blogging, developer meetup organizing, presenting at conferences, writing side-project apps, contributing to open source things.

Another friend of mine at Apple, who worked in an area relevant to some trouble we were having with NetNewsWire, wanted to look at the source code – and they had to go ask permission before they could even look.

I understand! I understand why Apple PR and legal departments are the way they are. But I still feel a loss to the community every time somebody I know goes to work at Apple.

There’s a curtain between us and them. Colorful, well-designed, made by lasers — but still a curtain.

PS This would even prevent me from ever working there. I have a great job, and I intend to stay at this job till I retire, but if I were unemployed and saw the openings in Apple developer publications based here in Seattle, I totally would have applied. Except that working there would have meant the end of NetNewsWire and, effectively, the end of this blog. I would have had to give up the two biggest pillars of my career. It’s flat-out not worth it.

PPS I don’t write this to make any of my friends feel guilty! Working at Apple is a wonderful thing. But it’s bittersweet, and I do wish Apple would do more to take out the bitter part.

NetNewsWire Status: Progress Toward Shipping the iOS App

The team — we have a team! which you can join! — is continuing to work on several different things: NetNewsWire 5.0.4 for Mac, NetNewsWire 5.1 for Mac, and NetNewsWire 5.0 for iOS. We’re also figuring out what will go in subsequent releases.

The most attention has been going to the iOS app. We’ll be doing a public TestFlight beta as soon as it’s ready, which I’ll announce here and on the NetNewsWire blog. It’s getting close! A bunch of us have been using it every day for a while now.

In second place for attention is syncing with Feedly, which will appear in the iOS app and in NetNewsWire 5.1 for Mac. (We’re working on adding other services too, but it’s yet to be decided which ones will appear in which versions.)

No ETAs, of course — we’ll announce dates only when we know them for sure.

* * *

The design of the iOS app is what you’d expect, I hope. As NetNewsWire for Mac is so very much a Mac app, the iOS app is very much an iOS app.

Our design goals are the same as with the Mac app: make it stable, fast, clean, modern, and — most importantly — as obvious as possible, which helps make it easy to learn and use.

In other words, we did what we always do, which is to do the best job we can by using classic user interface design principles.

It definitely looks like an iOS 13 app. You can see its bones.

While I think it’s awesome, people who prefer something richer have other really great apps to choose from. We know NetNewsWire is not the only RSS reader, and we’re not trying to please everyone. Which is very freeing — it means we can please the people who like apps like this. :)

A Naming Challenge

In NetNewsWire, we have a concrete Feed type that’s what you expect: it describes an actual web-based feed.

But there are other things that are feed-like in some important ways. Smart feeds, script feeds (in the future), search results, and folders.

They have some things in common: an icon, the ability to fetch articles, the ability to provide an unread count, etc. These common abilities each have a separate protocol: UnreadCountProvider, for instance.

But this is so common that we should have a single protocol that bundles these up into one.

The problem is — what should we call it?

* * *

Maybe the right for that single protocol is Feed. If so, then what should we call the concrete type for web-based feeds?

* * *

In the original design for NetNewsWire, five years ago, Feed was going to be a protocol. Then I started working in Swift, found that I wanted to use Set<Feed> and couldn’t, so I made it a concrete type.

I love Swift, but this limitation keeps coming up for me. I use both protocols and sets a lot, because they’re often the best choice, but in Swift they just don’t play together well.

* * *

Update a couple hours later, at 1 pm: We’re going with a protocol named Feed and the concrete type name WebFeed for, well, web feeds. This means a bunch of renaming in the app, but I think the result will be worth it.

Remember that it’s an open source app. The marketing team is not counting on a specific deadline. Our goals are 1) app quality, and 2) making it easy for people to work on the app (which really supports the first goal).

PS The marketing team is me. :)

For the entire 20 years of inessential.com, I’ve been in the same house, at the same desk, with even the same keyboard.

Happy 20th to This Blog!

Today is my blog’s 20th birthday!

It started like this: at the time I was working at UserLand Software on a blogging app called Manila, and this was my own personal Manila blog. It’s gone through a few other engines since then. (These days it’s rendered — as a static site — by some Ruby scripts.)

* * *

It‘s tempting to think that The Thing of my career has been NetNewsWire. And that’s kinda true. But the thing I’ve done the longest, love the most, and am most proud of is this blog.

* * *

The original tagline was “You don’t need to be here.”

The name inessential sounds like a low self-esteem thing. It’s not. It was named for the idea that literature is inessential. Without it, people would still go on fucking and fighting and grocery shopping.

In other words, the name comes from huge and unwarranted self-esteem, as if my blog would be the first to be literature. Sheesh! Now I laugh at myself. What arrogance!

I hope I’ve matured, at least a little, since then. But I keep the name, because nowadays that’s just its name.

* * *

Old proverb: “The best time to start a blog is 20 years ago. The second-best time is today.” :)

This Feature Should Be Easy

From time to time people note, in feature requests, that a particular feature should be easy, and they’re surprised it hasn’t been done yet. Sometimes they even say that it’s so elementary that they really consider it a bug that it hasn’t been done yet. Especially given how easy it is.

I’ve written about this before. See Bug Guilt Trips from 2003 and Anatomy of a Feature from 2009.

A feature is pretty much never as easy as it seems. The main function of a feature — whether that’s making a https call to a read-it-later service or adding some fairly simple new view — is often the easiest part.

It’s everything around the feature that makes it harder: UI design, localization, refactoring, accessibility, state restoration, getting new artwork (for a toolbar button, for instance), dealing with errors, testing, updating the documentation, etc.

And then you still have to schedule it: as important as a feature request is, there might be others with a higher priority, and resources are always limited. Doesn’t matter who you are.

* * *

When you get developers alone, and you make a joke like “Why don’t you just…” or “I bet you could do this in an afternoon…”, they all laugh, because they’ve all heard this.

I most definitely do not want to discourage people from making feature requests, whether for my app or anybody else’s app. Please make feature requests! And report bugs!

But I would ask that you try not to call something easy. The developer of that app will have a pretty good idea of the requirements and the level of effort, and it’s never as simple as it looks on the surface. And they also have a good idea of what needs to happen, for various reasons, before work starts on that feature.

Amateurs

One thing I’m weirdly proud of is my position as an amateur programmer.

When I point that out, people say, “Well, but…” — and I know where they’re going, that after 25 years of professional experience I’m not what you think of when you think of “amateur.”

And yet, it’s still true. It’s just that I’ve come out the other side, and now I get to work on exactly what I want to, the way I want to, without any thoughts of trying to make money at it.

I can take risks! I can work with anybody who shows up! It’s a pure thrill. It’s like writing single-malt apps.

And I would wish for more people to find themselves in this position — eventually, anyway — because I want to see what they would make.

PS The Dictionary app on my Mac says of the origin of the word “amateur”:

late 18th century: from French, from Italian amatore, from Latin amator ‘lover’, from amare ‘to love’.

Spot-on.

You Choose: Follow-Up

It came to my attention after writing my blog post about how we choose the web we want that the pessimism is about not being able to make a living from blogging.

Here’s my followup: I don’t care. Bite me.

ETAs: Follow-Up

Some people took my post No ETAs as if I were arguing against doing software estimates of any kind, ever.

I didn’t actually mean that. If your boss, project manager, or person you’re contracting with asks for an estimate, do your best to come up with something accurate. If you’re writing enterprise software, you may even be contractually bound to provide estimates for when features will ship.

There are ways to get pretty good at this. Pay attention to history and avoid wishful thinking. Don’t assume perfect productivity. Allow for the unexpected, because there’s always something.

What I’m talking about is the case where you’re writing a consumer-facing app — something that would get published on an app store, for instance — and customers or potential customers ask about an ETA for a given feature. Don’t do it! (For the reasons stated in the article.)

You Choose

On Twitter, Molly Lambert writes:

RIP blogging we all tried real hard to make the internet good and then corporations and rich idiots destroyed everything a generation of writers tried to build

Charlie Warzel re-tweeted her with this comment:

there’s almost no space for writing anymore that’s joyful or an attempt to be creative. hardly anyone is playing around with form or even just trying to entertain. so much of the joy has been sucked out of the internet unless its crowdsourced by platforms from ppl who aren’t paid

They’re both right.

But I’d also note: fuck that shit.

Here’s the thing: there are good blogs to read. Some old ones are gone, but new good ones are created all the time.

And there are good RSS readers which you can use instead of (or in addition) to Twitter and Facebook.

And — most importantly — nothing is stopping you from writing joyfully and creatively for the web! You can entertain, you can have fun, you can push the boundaries of the form, if you want to. Or you can just write about cats as you develop your voice. Whatever you want!

There are plenty of great places for it. (I quite like Micro.blog, personally.)

* * *

You choose the web you want. But you have to do the work.

A lot of people are doing the work. You could keep telling them, discouragingly, that what they’re doing is dead. Or you could join in the fun.

Again: you choose.

No ETAs

People often ask me about ETAs. When will the feature they’re waiting for ship? If you’re a software developer, they probably ask you too.

I totally get it! Though I write an app, I’m mostly a user of apps, and I too want to know when the features I’m waiting for will ship.

The Problem

But here’s the thing: ETAs are very hard to estimate with any amount of accuracy. Even if you plan well.

There are just too many variables. You have some very rough idea of the work to do, and how much time it ought to take, yes. To start with: those ideas are wrong.

You don’t know what you forgot to take into account, or what pieces you underestimated or overestimated. You never know what weird bugs will trip you up.

You don’t know what OS release between now and then will make you spend extra time on something. (For instance: iOS 13.2 just came out today. How is it different? Does it affect the app? Every OS release is a moment of anxiety for developers.)

You don’t know if you’ll need to do some serious, time-consuming refactoring or rearchitecting. You don’t know if the person working on the feature might get pulled away for some reason (vacation, emergency, reassignment, etc.).

You don’t know if the fallout from some other feature in progress will require more time, which ends up pushing back this feature. You don’t know if some other feature or bug fix will suddenly get a higher priority based on any number of good reasons.

You don’t know if your testers will be pulled away to work on something else. You don’t know if your tools will be updated in a way that affects productivity. (Hopefully for the better! Though, if there’s a learning curve, maybe it will hurt productivity at first.)

There’s all that and plenty more — and none of the above are extraordinary conditions. They’re all quite normal. So any ETA will almost surely be quite wrong. Even if you take all of the above into account.

On Not Setting Expectations

The problem with stating an ETA then, is that it sets up expectations. When you don’t meet them — and you won’t, most of the time — people sometimes get upset. Not everybody. But some people take ETA to mean “I promise to ship that day or damn close to it.”

So the better thing to do is plan which features go into which releases, and have a date for internal use for just the next release — and refrain from making public ETAs.

Even if you the software developer think you can make accurate ETAs — and maybe you did, once or twice — you’re not going to be that lucky most of the time.

PS This is all just to say that app-making is nothing like building a house. It’s more like building the first house ever in the history of houses, with a pile of rusty nails and your bare hands, in a non-stop tornado. It’s different every time, and it’s astonishingly complex, non-linear, and unpredictable. We all do our best to mitigate this, to make it more regular, but the industry just hasn’t gotten very far with that yet. The only reason anything ever ships is because people just keep working until it’s ready.

Update 30 Oct. 2019: See the follow-up, which explains that I’m not talking about all ETAs.

NetNewsWire 5.0.3 for Mac Released

NetNewsWire for Mac icon: globe with a satellite in the foreground.

The main things in this release are 1) enhanced performance and 2) importing subscriptions from NetNewsWire 3 (since it won‘t run on Catalina).

There are also a bunch of bug fixes — including a fix for the space bar behavior on Catalina — and there’s a new feature: you can type the s key to star and unstar an article.

For more details, read the change notes on the NetNewsWire blog.

SwiftUI Is Still the Future

We’ve been using SwiftUI in NetNewsWire for iOS, for its settings screen and related pieces (such as adding an account).

But now we’re redoing that code in classic UIKit. Maurice Parker, NetNewsWire for iOS lead developer, lists the limitations we ran into with SwiftUI:

1) Unable to customize cell selection color for our vibrant cell selection requirement. I was able to hack around it, but it made cells behave strangely. You couldn’t tap and hold the cell. You also couldn’t scroll using the modified cells.

2) No pop back to root for NavigationView. Longer scene flows weren’t possible.

3) ActionSheet flat out just crashes if you use it on an iPad app. https://forums.developer.apple.com/thread/124530

4) TextField will crash the app if it has the cursor and is hidden. We wanted this for the show/hide password functionality.

5) There is no way to display attributed text. You have to bridge to UIKit and use a UITextField that has a modified intrinsicContentSize calculation. We had to take input from GeometryReader and pass it into the UITextField to calculate intrinsicContentSize.

There are more, some of which I have probably forgotten about.

We very much want to use SwiftUI, and we believe it’s the future of Mac and iOS development — but emphasis should be on future, because it’s not quite ready in the present.

Which should surprise nobody, given that it’s so new. But I thought it might be interesting to know exactly what issues we ran into when using it.

We look forward to the day when we can use it everywhere, but we suspect it will be incremental. Hopefully next year we’ll be able to do the Settings screen using SwiftUI. And maybe some parts of the Mac app, too.

PS Once we’re ready for a public beta of NetNewsWire for iOS, we’ll announce it here and on the NetNewsWire blog, and you’ll be able to sign up to get it via TestFlight.

NetNewsWire 5.0.3b2 is faster — and it adds the ability to import NetNewsWire 3 subscriptions.

If you’re a NetNewsWire 3 user who needs to upgrade because you’re on Catalina (where NetNewsWire 3 won’t run), then I recommend either trying this beta or waiting a few days till the final 5.0.3 release.

Archive