Hello Weather

What should come as no surprise to anyone roaming the App Store as much as I do, I’ve been through using nearly a dozen different weather apps during the last years. Supply is plentiful on the App Store, and there’s always a new hotness to discover and check out.

Some weather apps provide good and accurate data but are hard to navigate and it is made unnecessary hard to get answers for the essential questions when it comes to weather:

  1. What are the conditions right now?
  2. What’s it going to look like in the next couple of hours?
  3. What is the forecast for, say, the next week or so?

Enter Hello Weather. This is a minimalist weather app that answers the three questions above in three sections of just one screen, one section below the other. Each section can be tapped to reveal further information information.

The essential point with regard to the questions mentioned above, Hello Weather answers the questions in a truly beautiful design that further underlines the simplicity of the user experience.

By default, Hello Weather comes with weather data from Dark Sky which (YMMV) I would personally not consider the most accurate source of weather forecast in central Europe.

On the other hand, Hello Weather’s in-app purchase1 unlocks the ability to use Weather Underground data. This data is what I personally prefer and which has delivered rather accurate forecast in my experience.

I have made Hello Weather my default weather app for the last four months. Yes, I keep other apps around as a reference. Some are a match in terms of quality of data, none is a match in terms of user experience.


  1. Which also gets you a night mode and a “secret experimental mode” (with no explanation of further detail). 

Burned

Interesting read1 in The Economist about the “the death of the internal combustion engine“.

The technology for combustion engines is a remarkable engineering feat. It took more than a century of constant iteration to get to the currently existing state-of-the-art in power-train technology. I wonder how long it will take until, once it became obsolete by the wide-spread adoption of electric cars, the knowledge2 to build a combustion engine on the current technological level gets lost forever.

Compared with existing vehicles, electric cars are much simpler and have fewer parts; they are more like computers on wheels. That means they need fewer people to assemble them and fewer subsidiary systems from specialist suppliers

This is the key observation and herein lies one of the main challenges in the transformation to electric power-trains on a broad basis. The article paints a dark picture for the industry in general:

While today’s carmakers grapple with their costly legacy of old factories and swollen workforces, new entrants will be unencumbered. Premium brands may be able to stand out through styling and handling, but low-margin, mass-market carmakers will have to compete chiefly on cost.

You could argue that premium brands are already distinguishing themselves in terms of styling and handling and that low-margin OEMs already chiefly compete on cost. But still:

On the most extreme estimates, that could shrink the industry by as much as 90%.

This sparks a comparison to other industrial disruptions on a similar level, for example the closing of coal mines in the Ruhr Area that started several decades ago: the painful implications still haven’t been overcome to this day.

On the other hand, parts will still have to be made on an industrial scale and especially the software branch may find plenty of opportunities. For example, the creation and maintenance of a reliable charging infrastructure with high availability will take a lot of resources and this should mean business for existing companies that are willing and able to take the challenge.

Therefore, I sincerely hope that the consequences for workers and engineers will not become as terrible as painted by the article.

I don’t know whether central power stations – as the article suggests – will really play a central (sorry) role in that future as they do today. My hope is that decentralized power generation from renewable sources will eventually become the dominant supplier of the power grid.

Electric cars will come and Otto- and Diesel-cars will shrink in market share over the next decades. This development will be driven by two factors, the fact that fossil oil deposit is limited in quantity and the fact that the technological improvements (once starting to gain pace) will make electric cars more compelling and eventually even more economically appealing than traditional fuel burners.

Driverless electric cars in the 21st century are likely to improve the world in profound and unexpected ways, just as vehicles powered by internal combustion engines did in the 20th. But it will be a bumpy road.

Indeed.


  1. Via Instapaper
  2. And the industrial processes, these are admittedly equally crucial. 

Oh No, Another on Goes Subscription Only …

This time it is Ulysses, my favorite app for writing.

Whether it makes sense or not, subscriptions feel more expensive than one-time payments. You could pay for an app and only occasionally use it over the course of several years. You could ignore paid updates released on the meantime, and still enjoy the app that you bought back in 2015, at no additional charge.

The most prominent example of this category of apps, off the top of my head, is Acorn. I don’t need the app on a regular basis, but when I do I’m happy to have a powerful tool at the ready.

Sure, it would be possible to sign up for short-period subscriptions, but that sounds like at least a mild pain in the hind quarters.

With more and more apps going subscription only, I feel less and less comfortable buying yet another subscription on top of the pile of existing ones1.

Last year, paying for a subscription of Bear was a no-brainer. The app is such a joy to work with. Despite the obvious overlap in functionality I would have no problem using Ulysses and Bear side by side going forward. But this will have to change.

The thing is: I can use Ulysses for what I use Bear for2, but I simply can’t use Bear to replace Ulysses functionally, not even close. And if I have to chose between the two to keep the costs low it is obvious that I’ll be terminating my subscription to Bear.


  1. To give an example, TextExpander did not make the cut. 
  2. Which is mostly note-taking and being a bucket for text shared from other apps. 

Things 3

It’s been a couple of weeks since the release of Things 3. I have had some time to use the app and explore its various new features.

Admittedly, my first reaction after downloading was mixed. Like many others, I had the initial impression that the changes between the new version and Things 2 were mainly limited to visual refresh1. However, this impression went away pretty fast.

Here are some of the highlights of the new Things according to my observation:

  • I’d like to start with an aspect of Things that I liked before the launch and continue to like after the fact: the distinction between areas and projects makes a lot of sense to me. Projects are, by all common understanding, defined by a temporal limit. In other words, a project has to end at some point in time. Areas (e.g. “Work”, “home”, etc.), on the other hand, can be taken as perpetually existing buckets for items that may or may not belong to a project and that are related to the meaning of the area.
  • The concept of a project has been massively enhanced by the ability to define sections inside the project as well as the ability to add a description to the project itself. Projects become a lot easier to manage by this means.
  • Subtasks have more utility than expected. For example, I use subtasks within the definition of a task to create the expense statement within a project created to organize a business trip. In other words, I create the task to fill in my expense statement in advance of the trip and then create a subtask for each expense to be included in the final statement.
  • “Upcoming” is nice to have, but I use it less than I initially though I would.
  • I’ve successfully reduced the number of reminders and still haven’t lost track of anything that needed to be done.

Here are some aspects that I sort of wish were supported:

  • There is no way to display a list of everything. I don’t know, maybe sometimes you want to get an overview of all the unresolved tasks, no?
  • Dark theme. I’m a sucker for dark themes.

Overall I really dig the new version of Things. It came with a lot of stuff that I never knew I wanted. But now that I’ve made myself comfortable with Things 3 it is hard to imagine using anything else for the foreseeable future.


  1. Which does not mean that the visual refresh wasn’t appreciated. Even with the visual refresh the launch of Things 3 would have been impressive, in my opinion. 

E-Mail on iOS

E-mail clients are plentiful now on iOS, although there have been times where this category of apps were not available on the App Store at all.

But since third-party clients have been made available, I have taken some of them for a spin to see whether I could find the one that works best for me.

I wanted (but, in all honesty did not fully expect) to see whether it is really possible for me to point to one of the apps and declare it the best one overall, period.

The Sweet Setup thinks the best e-mail app on iOS is Outlook. I have tried Outlook but (as much as I respect The Sweet Setup) it did not even make it into the three apps that I have finally spent some time with:

Spark

Spark is a really solid option on iOS with some minor drawbacks. It is very well designed and visually the most appealing apps of the three.

One of the most prominent features is the “smart inbox” that does a pre-sorting of incoming e-mails and presents the user with a structured list of e-mails separated in different categories (e.g. newsletters, notifications, and personal)

Spark supports snoozing and also the opposite of it, i.e. pinning an e-mail to the inbox.

The feature that I like most is that Spark is able to make a proposal for the target folder of a move operation. The proposals are surprisingly accurate.

And Spark is fast. Switching between e-mails as well as access to attachments is almost instantaneous.

Spark supports push-notifications for IMAP accounts, but in my experience the notifications have a tendency to stop coming in after an initial period where everything works well.

Until yesterday, search used to be more or less a desaster. I’ve heard a lot of praise about Spark‘s search capabilities, but that must have come from gmail-users, I assume. On IMAP, search just did not work. It was unreliable and utterly slow.

Today saw the release of version 1.8.0 that fixes most of the annoyances. Search now works as I want it to, at least if I have opened the folder in which potential hits are located previously. Spark will just ignore folders that have not been opened in the app for search.

One minor issue that I have with Spark is that it does not have the ability to make any changes in the folder structure, i.e. renaming, adding or removing folders is not supported.

Airmail

Spark’s greatest weakness (in the past) is where Airmail shines like a beacon in the night. It has the best search engine for IMAP accounts that have ever found on any e-mail app on iOS.

It even runs reliable search queries when offline, i.e. it must have the ability to create a search index on device. Also, it is very easy to take na e-mail address and find every single mail received from this specific sender in the entire IMAP account.

Airmail shows the name of the folder as part of the response´use to a search query, a very nice detail and an information that requires an additional tap in Spark.

Airmail is the only app of the three that supports adding and deleting of folders in an IMAP account. Neither Spark nor Newton deliver any amount of control over the folder structure. Airmail also comes with option to color-code the folders, which is very nice.

Airmail has the best and most elaborate documentation of the three, despite the fact that the documentation is hidden behind a rather misleading topic named “give feedback” in the preferences.

Despite its undeniable technological advances, Airmail has a certain reputation of being buggy. I can at least confirm that I have been hit by the occasional glitch, especially when composing a message.

Newton

Newton is the most expensive option of the three by a large margin. The business model is based on subscriptions that amount for 50€ per year in the German app store. 50€ is a lot of money and that naturally raises expectations.

But frankly, Newton was by far the biggest disappointment of the three e-mail apps. Newton is ridiculously slow, to the point sometimes I had the impression that it was just a wrapper around a web mail view.

When switching between individual messages there’s plenty of opportunity to stare at a spinner. Scrolling is the worst. The app scrolls a few lines and the takes some time to add a further chunk of messages to the bottom of the list. When scrolling beyond this point, the procedure repeats.

And the same thing happens even after I scrolled on particular folder from top to bottom. If I tried to scroll the same folder some time later the app behaves as if the scrolling were done the first time.

Search is even slower than scrolling. I cancelled all my queries in a state of frustration, every single time.

And it gets worse: I had cases where an e-mail vanished from my inbox, only to reappear after a couple of hours. However, as other clients confirmed, the mail was positively in the inbox all of the time.

Another aspect where Newton failed all my expectations is support, which is limited to a rather limited FAQ page on the website.

Conclusion

If you are interested in further candidates I’d recommend listening to Canvas episode 10 where Federico Viticci and Fraser Speirs explore the world of e-mail clients on iOS on a broader basis.

Myself, I will use Spark as my main mail app, now that such big improvements have been made with the latest release. Airmail, on the other hand, is more than a solid alternative despite its occasional tendency to drop the ball.

Things 3 is coming

From the Things Blog:

We’re very excited to report that Things 3 is almost ready for release. After a final round of beta testing, we expect to ship in May.

I have been using Things on and off over the years. I have tried other apps of that category, starting with Remember the Milk, then OmniFocus, Todoist, purely textual apps, and 2Do.

I guess the reason why I sooner or later return for another stint with Things is because the app hits a spot between an obsessively simplistic and a demanding and complex GTD approach.

The approach implemented in Things really stands out from the crowd. Things is not, and has never been, like any other generic task manager. There’s a concept behind it and I like that a lot even if I ended up writing entire blog posts to help myself figure it out.

However, or alas, that spot occupied by Things is no stable equilibrium. It works fine for some time, but frustration slowly but inevitably builds up until it is time to switch.

For explanation, one aspect that used to make my relation to Things complicated is about a glaring shortcoming in the workflow. In the current version, the creation of a repeating task is only possible in the Scheduled section and it is not possible to convert a non-repeating task into a repeating task.

I really hope there was time enough during the development of the new version to fix this.

On the other hand, there is so much to like. Let’s face it, Things‘ synchronization of data between devices is lightyears ahead of any competition. Nothing else comes even close. It took time to get there, people had to be patient, but the result is truly outstanding.

So I welcome a next iteration of the Things saga and I personally can’t wait to give it a try.

Joel Spolsky’s new Computer

Joel Spolsky made a splash on Twitter:

Sorry Apple. After 10 years loyalty, this latest MBPro with useless touchbar and unreliable keyboard was last straw. Switched to Dell XPS13👋

I am using a Dell Precision 5510 (which is basically a rebranded XPS15) at work for more than a year now.

My only minor complaint is that Dell doesn’t seem to care about fine-tuning the hinge such that it is possible to open the lid with one hand while the machine is lying flat on the table.

Apart from that, it’s a solid hardware with modern components that has a Skylake processor and that supports being loaded with up to 32 GB of RAM1.

Even as a Mac user for more than 20 years, I wouldn’t hesitate giving money to Dell for an XPS 15 (or 13) to run my my private stuff if only there was an officially supported way to run macOS on it.

Joel, for what it is worth, does not seem to care much about the software aspect. While many will agree with Joel on the hardware choice, I bet that the question of the operating system is not as unequivocal.

Windows has gotten better with Windows 10. But in my personal experience, it is still way behind macOS in many measurable ways. Plus, for people heavily invested into software based firmly in the Apple ecosystem the idea of switching to Windows is still a scary idea2.


  1. I have personally no experience with this option or the implications on battery life. 
  2. Especially the prospect of having to keep a Windows installation that is not maintained by a corporate IT department free of malware would give me nightmares. 

App Store Adventures

It all started with a mistake, that’s fully admitted. I was browsing a web site that had some app recommendations. Tapping one app icon on the site opened the App Store app that in turn told me that the item was not available in the German Store. Oh, and whether I wished to switch to the U.S. store.

Thanks, no. Somehow (and here comes the mistake part) my tap was a bit too much off-center and registered with the “yes, switch me over” button.

Welcome to the U.S. store.

At least, that’s what it looked like. Instead of naming the price of an app in € the labels all had a $ on it. At this point, it was totally safe for me to assume that the switch actually happened.

I struggled with a faint memory of a similar incident a couple of years ago: there had to be a way to switch back immediately, and got confirmation by the fast-reacting @AppleSupport Twitter account. Kudos, I was really impressed by the prompt reaction.

Unfortunately, I was still unable to follow the advice: In order to proceed into the “Country/Region” section of my Apple Id properties dialog, I was told that first I would have to cancel my ongoing Apple Music subscription and then wait until it expired. Which would not happen before (you guessed it) end of February.

Great!

As an aside, I really don’t understand why tapping on “Country/Region” results in the information that I can’t switch without cancelling Apple Music instead of just telling me what my current region is. If this were the case and Settings.app had informed me about my current region the whole issue would have been over by then (I have filed a radar).

Frustrated, I called Apple support on the phone and was quickly connected to a support person who, after listening to my story, confirmed to my surprise that the switch had actually never happened. According to the support person, I was positively still on the German Store.

And yet, this information was somehow at odds with the fact that the App Store app on my phone insisted on app prices in $ instead of €.

So I did what I was supposed to do in this situation: reboot the phone. Still: $.

I went to my iPad, launched the App Store app and watched all the little € appear in the place where they are supposed to appear. This was another reassuring indication that the switch really never happened. But how on earth would I be able to get my phone back to my native currency.

After some time, I wondered what would happen if I tried to purchase an app on (what still seemed to me) the U.S. App Store.

Here’s the thing: after confirming my intention to buy a random app I was informed that this app was not available on the U.S. App Store and, hallelujah, whether I wanted to switch to the German App Store.

Yes, sure. Please do it. Can’t wait.

Now that everything is back to where it belongs I’m still struggling to understand what was happening. It seemed that the App Store app initiated the switch which, however, was immediately deflected by my existing Apple Music subscription.

But somehow the App Store app had convinced itself that the switch went through and it would only be able to get back on track when an actual purchase transaction was attempted.

Whatever, I’m pleased that this adventure turned out well. Apple support in Ireland (yes, you apparently get connected to Ireland if you call from Germany) deserves an honorary mention for the crucial confirmation that my account was never actually moved.

I can’t help ending with the clever conclusion that if you want your iOS devices to prevent your account from inadvertently switching to a different App Store: buy an Apple Music subscription!

Pocket Casts Notifications, reloaded

I had this struggle with double notifications in Pocket Casts. First, I seemed to have found a solution by signing out and into my account again, but after some time the pair-wise notification were back.

Tired of this issue, I eventually deactivated notifications within Pocket Casts. After that, you’d expect that notifications would stop popping up altogether going forward, right?

Wrong.

Here’s the thing: I still got notifications, only the notifications din’t come in pairs any longer. Weird.

OK, time for more radical measures. I deleted the app, and reloaded it from the App Store. Unfortunately, I did not even make it to the first notification when I realized that the audio effects were simply not available.

The buttons were just inactive, and it was not possible to use these features. This was no good.

After taking a deep breath, I was willing to give it one additional shot. If that worked I would stay with Pocket Casts, otherwise it would finally be time to leave.

Again, delete the app. Reload. Launch. Try audio effects. Wait for incoming notifications. Phew.

The tentative conclusion is that the last install seem to work according to my expectation. However, I have to say that this job stretched my loyalty to the product dangerously close to the snapping point.