Also in computing, less can be more (Sunday musings / discussion)

Hi everyone,

after having spent some time with a Surface Go 2, a thought crossed my mind that I wanted to bounce off of you: also in computing, less can be more.

There’s a high likelihood I am pointing out the obvious here. Then again, there are so many instances in which the more powerful solution gets the thumbs-up, the shoulder-clapping, and an respectful grunt while the more well-balanced and well-integrated solution gets the “meh” that I think the idea - while not necessarily very innovative - might still be interesting to rehash, to ponder and discuss.

Couple of observations:

Screen size
The past year, I tried working with a dual-27-sans-bezel screen setup (DELL U4919W). While some things were great (immersion while gaming), work simply felt overwhelming with many open windows staring back at me and the GUI elements being crammed way to the right and left, way out of my field of view. It felt like the screen worked me instead of me working the screen. So I got rid of it and got a 32 inch (DELL U3219Q) instead which fits perfectly into my primary field of view. Still, doing “thought-focused” work on the 10,5 inch Surface Go feels even more manageable. It simply gets more out of the way and is less distracting than anything bigger. It turns out that the larger share of my projects require thinking than pushing pixels from A to B (even though pushing pixels feels like I am working). Since it’s obvious that screen size is more or less useful with the type of job to be done, I am contrasting “thought-focused” work with “visually demanding” here. In the former, I suppose the tool rather assists a thought process and in the latter, you need to work with the data (usually coming from multiple sources) on the screen. When trying to do thought-focused work, a large screen can be distracting.

Simple mail software (Windows Mail) vs. full groupware (Outlook)
While I’ve been happy with working on Outlook for the past years, the web version of Outlook seems to have surpassed its desktop client counterpart in terms of GUI and ease of use. At the very least, it’s more modern. The desktop client is gradually following suit, but the online client felt more useful since it made clever decisions about dropping functions and buttons I could really do without. Enter Windows Mail: while I did my best to sneer at Windows Mail for five years, the Surface Go made me rediscover that little piece of software that goes even more into that direction: less buttons, more focus, less stress, more getting-things-done. I have to say that this may in part be the case due to it working well with Microsoft 365 (duh, but I tried other services with it before and it did not work well for me - particularly the integration of mail and calendar). Once again, rediscovered something simpler and turned out more happy with it.

i7, i9, 3950x, 24 cores vs. ARM, Apple chips, m3
One thing that the Surface made me aware off is that the speed of a reasonably modern CPU does not matter as much as its cooling and throttling. I am using a Surface Go 2 with an m3 chip. It can boost to 3,4 GHz, generally works around 2 GHz - but when it gets a heat stroke, it can go as low as 390 MHz. Whenever it has to do much of a prolonged amount of time, going far beyond 1,1 GHz is impossible. This leads to the CPU being busy and delaying responses often running into the “this app has ceased to work”-thresholds of the Windows GUI. Consequently, the computer feels inoperable - not, because its CPU would be too slow or its RAM would be to small, but because it cannot manage the exhaust of its temperature very well. In the past, throttling has been the source of a few great product innovation bruhahas (specifically the first introduction of i9s in MacBook Pros). If you are, like me, interested to go into rather smaller and more efficient technology solutions, the problem really occurs at the fanless balance of power and portability so somewhere around the m3 or Apple processors. If that balance is well done, a well-integrated smaller CPU implemented with a set of challenging restrictions in mind, will “outshine” (in terms of “It could do that, too? Wow!”) a more power CPU any day. I suppose that more challenging context simply will lead to more advanced products. If you can draw 250 Watts and throw 100 of that into your cooling fins, not much is won for most people without gigantic work loads.

More RAM, less RAM - "a thousand tabs in Chrome"
While I am writing this on a computer with 32 GB or RAM (not because I really need that but because RAM was cheap recently and “more is better”), I was excited to find that working with an 8 GB little tablet machine really is more than I’d most likely need. I would not consider myself entirely opposed of challenging computer tasks, but it’s simply impressive to see that the smaller, maybe even more frugal solution would work just as well for you, 95% of the times. While being cheaper and less wasteful (I know, there’s a philosophical debate to be avoided) 100% of the time. For an amusing twist: I just remembered that I have connected the Surface Go to my screen and am typing this on it instead of the 32-GB-desktop. Surprise, surprise: does not take more than 8 GB of RAM to browse the web or write a forum post. Indeed, the task manager says something about 4GB in use. Certainly, computing life is more complex than that but: it’s useful to reflect on how much you really need. That “Chrome is a memory hog” when reflecting that you have dozens of tabs open may not be a call to order new chips on amazon but maybe a reminder to close some tabs.

Integration
All of those observations, we will agree, depend on the nature and subject of your work and work flow - as well as preference, of course. I pointed to the proposed idea of the “thought-focused” and the “visually demanding” above.

However, there is one substantial influence on all of us: advertising. Most established brands are advertising and “making cool” the most powerful and most expensive solution with any derivative of that being a lesser trade-off to buy into innovation while spending a bit less. Very few brands on the marked advertise toward a less-is-more, focus-your-mind, you-don’t-need-that approach. The only one coming to my mind right now is The Light Phone (https://www.thelightphone.com/).

Given that the market is geared towards the bigger, the brighter, and the more powerful, there is a point to be made in the other direction: how much tech becomes too much and simply stands in the way between you and your project. No more “If you’re a big guy, you need a big computer, too, of course!”. Too many times, these tropes are being introduced by tech brands and echo back through tech journalists, in my opinion. Small solutions are always “cute”, big solutions are “wow”. I’d wager that often times, it’s worth to explore if the smaller solution is not the one that you can thoroughly put to use and that does not require so much attention that you spend more time administering than working.

As a final devil’s advocate: “get with the times, software will only become more demanding - of course you need to go more powerful than what you basically need today to future-proof” - sure. Don’t buy hardware or software that was obsolete yesterday. The idea is only to give a strong (!) counterbalance to feeling the urge toward buying high-performance solutions you don’t need. They might, counter-intuitively, stand in your way, afterwards.

Ok, after having written that, I feel considerably like I’ve been preaching to the choir. Still, I’d be interested to learn your perspectives. :slight_smile:

4 Likes

I think software is the problem, not screen size. UI elements’ becoming far-flung is solvable by better design meant for that physical size, or even customization options for where to dock which palettes and so forth. Physical screen size only works against you when the screen itself’s physicality is a factor, which is why so many people love the iPad (I prefer Apple so my analogies will be to them). Screen space itself isn’t the issue and is critical where it is intelligently utilized such as for programming code, media production, and art (gaming, as well, as you note). The only issue with screen space as a quantity is the human’s field of view which curved screens help address (though I’m not a fan of those). Accept your own cognitive limitations and place proper responsibility for design on the UI, and the issue of screen space largely evaporates.

I’m not in MS ecosystem so can’t comment on Windows Mail but I will mention the forthcoming Hey.com from the founders of Basecamp which aims to tame email. You may also be interested in their book It Doesn’t Have to be Crazy at Work tackling many of the corporate overhead issues to which you allude in proffering “single-user” software.

You are clearly among the class of white-collar professional not relying upon heavy computing capability, and for those individuals a “thin client” such as an iPad is perfect. Safari now has automatic close-tab settings.

I can understand why from your perspective there’s little distinction to draw between hardware performance and software performance, but I think it has led you to malformation of your critique: software design’s impact upon UI is the software author’s responsibility rather than giving yourself less screen space excusing poor utilization of more screen space. Similarly, toxic office social dynamics afflict software to the extent software reflects those, rather than software being superior by stripping away the ability to readily collaborate. Hardware heft is indispensable for those who actually do use it, rather than focusing on the long tail who are perfectly happy with something not innovative (I fully agree against the culture of glorifying bigger and better for the sake of marketing hype and profitable over-consumption, though!).

As someone who grew up on PCs then switched to Macs and now can only afford an iPad, your complicity in complacency with middling specs and minimal functionality is infuriating as that’s why iPad OS remains incompetent in empowering users for even the most basic control over our data such as file management and reliable storage/transfer (which I’ll spare you the nitty-gritty of here, since you’re a Windows guy). I’m all for simplicity which reveals complexity for those who seek it out, but it can never come at the expense of core user empowerment whether or not the user is aware of how disempowered they are to be without it.

1 Like

I’m working on two 22-inch monitors right now. I think that’s a sweet spot for me currently. I prefer the two monitor set up to ultra-wide monitor as I can more easily snap an application full screen on one monitor. I tend to do that - have a couple of apps full screen. I will switch between apps using the task bar or use the virtual desktops in Windows. I find that works for me much better than hunting down windows.

2 Likes

Largely agree with your points. For the mainstream, usability is king. That’s why iOS was successful. People always tout the app store ecosystem as the reason, but remember in the very beginning there was no app store, and then for a time afterwards it was very sparse. I’ve always believed iOS succeeded because of the responsive nature of the UX, something that had never before been seen in computing. And at the core of that ultra-responsive UX is the simplicity that lets the layperson pick it up and use it. The lacking feature set shaved precious milliseconds off every interaction which made all the difference in the world.

Our challenge these days is no longer hardware. We have the hardware design ability to do incredible things. It’s the software side that sorely needs catching up. There’s so much inefficient, poorly thought-out, cookie-cutter code strewn everywhere throughout our lives.

2 Likes

It’s so funny because to me iOS’ UX is garbage because it so sloppily maps to actual functionality. I mean, just in the Photos app alone, to get something into Pixelmator, I first tap Share, then scroll to Open In… then a nearly identical looking menu comes up over that where I then choose an icon in the top row labeled Send to Pixelmator which I had to first edit into being in that list! Or to duplicate a photo, the Share stuff is in the top-right usually, unless you’ve selected multiple photos it’s in the top-left! Drives me NUTS. Pixelmator’s own functionality is another case in point: since iOS refuses to allow persistent functional UI elements (like the Mac menu bar) there’s now a “Create New +” button parked in every folder view! Absurd waste of space. There are sooo many examples of the stilted, lobotomized, divorced relationship between UI and actual user-function interactivity loops in iOS it makes me sick. And it’s all seemingly because mobile designers got it into their heads that interface should lead functionality, when that’s not why good UI succeeds. The original Mac interface was much truer to this despite having a mouse: its simplicity was 1:1 physical space, an icon was a file was a printed page. I hate those limitations, but that doesn’t mean that having functions play hop-scotch across my screen constantly is acceptable in the name of refusing to resemble something just because it’s already proven successful! So, I take issue with your describing iOS as having a “responsive nature…that had never before been seen in computing” aside from touch-screen, which has nothing to do with what I’m talking about in the didactic between user interface design and functionality.

On the hardware front, iPads are still decapitated by absurdly limited RAM, excruciatingly low thermal tolerance, and dongle city (I know, you’re mostly just talking for kiddie-pool uses, but that’s part of my point: those who don’t know better and shouldn’t have to for themselves are nonetheless drowning out those who insist there’s more to know, and in the meantime the user-empowering fundamentals of user control over their own computing have been and continue to be eroded).

I suppose it depends on what you need. I find there is a lot of missing functionality or it is hard to access with Outlook online. Plus I have multiple accounts, so it is easier than using Outlook.com and the web interface for my domain provider’s email service. At work we are Exchange and Outlook is streets ahead of OWA.

Very much agreed. Having a powerful processor is a waste of time if it is constantly being throttles. The fist Lenovo Yoga Pros had that problem. They used Y-series processors, but were generally 20%-30% slower than other manufacturers devices, because they had compromised the cooling.

That said, a well cooled, low power processor is fine for simple “on-the-go” work, as long as you don’t expect to be doing video editing etc. on it. It is like a Twingo isn’t a Ferrari, but you can pack a lot more shopping in a Twingo and you don’t worry about parking it in the supermarket car park and getting scratched or bumped.

Again, it is the workload that defines the horse. Sometimes a donkey is more practical than a Shire horse. My home PC is 32GB and 16 threads and I use it to experiment with virtual machines in Hyper-V. It is great. My work laptop has 8GB and is used for RDP, Outlook, Word, Excel, Teams, TeamViewer and a couple of other business tools and even running everything at the same time, 8GB is more than enough.

Heck, most of our users are still on 4GB RAM in their new (2019/2020) PCs and don’t have any major problems with performance.

My home PC is a Ryzen 7 with 32GB RAM and 3 SSDs and a 2TB spinning rust drive. Because I bought it to be a photo processing and VM testbed machine, it fulfills that role admirably. I also have a 2016 HP Spectre X360, which replaced my Surface Pro 3, when I left a previous employer (I bought it privately, but my employer bought it off me, so I could use it at work). I had worked out that I used it as a tablet, without keyboard, for less than 1% of the time. That laptop has an i5 processor, because an i7 would be more than I need and would run hotter and louder and reduce the battery life. The i5 was the sweet spot.

The same for my work laptop, a ThinkPad T480 with an i5. An i7 would have been nice, but I don’t need the extra grung 95% of the time.

When Windows 8.1 came out, I had a Samsung ATIV Atom based tablet. That was a great little machine, more than powerful enough for what it was designed to do. But it got a bad reputation as being underpowered, because people were buying this 400€ tablet and expecting it to compete with 1000€+ laptops with Core processors and lots of RAM. Used as a lightweight tablet, it was fine, it could even run Outlook and Word or Excel (not the “or”) and web browsing was “quick enough” and it could play Store games well enough. Trying to load up all of Office at one time or load Photoshop onto it and you quickly found its limits, but that is the wrong machine for those applications.

And that is the key, I can think of several scenarios, where the Surface Go 2 would be an ideal, relatively cheap, tablet for some verticals - we use a Windows tablet that uses 1 custom app to talk to our industrial fork-lift scales, which is linked to our ERP software and used weigh chemicals into the reactors on the production line. The Go 2 would be perfect for that, running the single app, a Core based tablet would be overkill.

1 Like