Archive for April 26, 2016

Tuesday, April 26, 2016

Dropbox’s Project Infinite

Yesterday, I wrote about the superiority of BitTorrent Sync’s Selective Sync over Dropbox’s. Today, Dropbox gave a technology preview (via Mitchel Broussard, Hacker News):

Project Infinite will enable users to seamlessly and securely access all their Dropbox files from the desktop, regardless of how much space they have available on their hard drives. Everything in the company’s Dropbox that you’re given access to, whether it’s stored locally or in the cloud, will show up in Dropbox on your desktop. If it’s synced locally, you’ll see the familiar green checkmark, while everything else will have a new cloud icon.

It’s not clear whether this will be limited to business customers.


This feature is implemented at a low level, and works on the command line.

For example if you have a directory that is all stored in the cloud you can cd to it without any network delay, you can do ls -lh and see a list with real sizes without a delay (e.g., see that an ISO is 650 MB), and you can do du -sh and see that all the files are taking up zero space.

If you open a file in that directory, it will open, even from command line, then do du -sh and see that that file is now taking up space, while all the others in the directory are not.

Dan Luu (in 2015):

Something that’s occasionally overlooked is that hardware performance also has profound implications for system design and architecture. […] Consider the latency of a disk seek (10ms) vs. the latency of a round-trip within the same datacenter (.5ms). The round-trip latency is so much lower than the seek time of a disk that we can disaggregate storage and distribute it anywhere in the datacenter without noticeable performance degradation, giving applications the appearance of having infinite disk space without any appreciable change in performance. This fact was behind the rise of distributed filesystems within the datacenter ten years ago, and various networked attached storage schemes long before.


However, while it’s easy to say that we should use disaggregated disk because the ratio of network latency to disk latency has changed, it’s not as easy as just taking any old system and throwing it on a fast network. If we take a 2005-era distributed filesystem or distributed database and throw it on top of a fast network, it won’t really take advantage of the network. That 2005 system is going to have assumptions like the idea that it’s fine for an operation to take 500ns, because how much can 500ns matter? But it matters a lot when your round-trip network latency is only few times more than that. The caching and other obvious wins if you have 1ms latency may not buy you much at 10us latency, and it may even cost you something.

Latency hasn’t just gone down in the datacenter. Today, I get about 2ms to 3ms latency to Youtube. Youtube, Netflix, and a lot of other services put a very large number of boxes close to consumers to provide high-bandwidth low-latency connections. A side effect of this is that any company that owns one of these services has the capability of providing consumers with infinite disk that’s only slightly slower than normal disk.

Dan Luu:

We’ll see how well the implementation works, but when Dropbox was first released people said, “who cares, it’s just rsync”.

Modular Frameworks and Module Maps

Omar Abdelhafith (via Ole Begemann):

If you are like me, then you banged your head trying to solve “Include of non-modular header inside framework module ..”. But what is a modular header.

A modular header is a header that is included in the module map. This header can be either imported in the umbrella header in case you are using umbrella header "header.h", or it can be explicitly imported in the module file using header "header.

Any header that is not included in the module map is not a modular header and hence cannot be imported in any of the modular headers (modular headers can only include modular headers).

A Flop Unlike Any Other

M.G. Siegler (via John Gruber):

Last year, Rolex did $4.5 billion in sales. A solid year for the premium watchmaker. Of course, it was no Apple Watch. That business did roughly $6 billion in sales, if industry estimates are accurate.

John Gruber:

It’s a misconception that what Apple does best is unveil mind-blowing new products. What Apple does best is iterate year after year after year — exactly what Apple Watch needs.

Craig Mod:

I sold mine after a month and have never — not for even one microsecond — reconsidered or desired the Apple Watch

Casey Chan:

I bought the Apple Watch a year ago. I stopped wearing it two months ago, and I’m not sure if I’ll ever wear it again. That’s because it doesn’t really do anything that anyone needs, and even when it does, it doesn’t always work like it’s supposed to.


Wearing the Apple Watch for nearly a year did change something in me though, but it’s the opposite effect that was probably intended: the Watch’s constant low-level notifications made me realize that there’s nothing really worth being notified about. Being able to feel every text, email, and whatever else, made me see how useless they mostly were. I used to joke that wearing a watch is handcuffing yourself to time. Wearing an Apple Watch (or any smartwatch, really), doubles down on that by locking you in a barricade of notifications too. So I’ve taken the Apple Watch off and don’t know when I’ll put it back on. The Watch isn’t at all worth it, but I’m not sure it’s even possible to make a smartwatch that I, or any reasonable non-tech nerd, would need. The more ambitious a smartwatch gets, the more complicated it is to use.

John Gruber:

“When do I use which button and what do the buttons do?” needs to be obvious for the Apple Watch to truly feel Apple-y. And it fails. The longer I own mine the more obvious it is that Apple dropped the ball on the buttons[…] My hope is that Apple does more than just make the second generation watch faster/thinner/longer-lasting, and takes a step back and reconsiders some of the fundamental aspects to the conceptual design.

David Smith:

Yesterday was the one year anniversary of the launch of the Apple Watch. It is a device that has had a profound impact on my life both personally and professionally. The Apple Watch I received on launch day is still firmly on my wrist each day (notably with barely a scratch).


I remember being rather skeptical of Apple’s original marketing of the Apple Watch as “our most personal device ever”, but a year later I must say that it would be a hard case to make that something that has been physically attached to me for 83% of my life is anything other than personal.

Kirk McElhearn:

I wouldn’t say that the Apple Watch is essential, and I think Apple greatly oversold its abilities, which has led many people to be underwhelmed. I also think it was priced too high, which has led Apple to drop the price, and led to sales in a number of retail outlets, something that isn’t common with Apple products. But I choose to put mine on every day.

Nick Heer:

Calgary is in the middle of a recession, but I have also seen a lot more Apple Watches lately while on my commute, on the pathway system, or on the streets. There’s a lot that I’m hoping for, both in software and in hardware, but as far as “failures” or “flops” are concerned, the Watch is a long way off, and the potential is huge.

Update (2016-04-28): Joe Cieplinski:

Apple obviously thought that sending little hand-drawn pictures was a big enough feature to warrant a dedicated hardware button. Perhaps now, armed with a year of data, they can reassess that decision in the next update to the OS. If you never put a product into the hands of real people, you can’t learn anything about how people will use it and what they’ll want from it.


The Watch is flawed, no doubt. Just like the original iPad, the iPhone, the iPod, and the Mac were before it. But there’s no question in my mind, one year into wearing this thing every day, that it’s a device that has a bright future. The vast majority of Apple customers will never witness these growing pains. They will buy version four or five. Don’t let your early adopter frustrations cloud your perspective on how most people view these products in the long run.

Update (2016-04-29): John Gruber:

Did you know there are games for Apple Watch? My favorite: launching any app and seeing if anything other than a spinner appears on screen.

Update (2016-05-03): Ole Begemann:

The one thing that’s holding the watch back the most at the moment is speed. I’m not sure the next version will make a huge performance leap (remember, the iPhone 3G wasn’t faster than the original iPhone), but performance will surely improve dramatically over time. It will be interesting to see where the watch stands on that front in two or three years.

I think it’s safe to say that I wouldn’t wear my Apple Watch anymore if it weren’t for the activity tracking. But that feature alone has hooked me to an extent that I will probably buy the next model. I may be better off with a cheaper fitness tracker, but now that I’m kind of locked into Apple’s fitness metrics I’ll probably stick with it.

Update (2016-05-09): See also: The Talk Show. John Gruber speculates that the Apple Watch Edition will not be updated and that it existed at all because “Tim Cook wanted to make Jony happy.”

Update (2016-05-20): John Gruber:

Posit: The things on Apple Watch that people actually like and use are the things that aren’t slow (notifications, activity tracking and goals, Apple Pay, complications, maybe Glances) and the things that are slow are the things people don’t use (apps, especially). Apple should have either cut the slow features from the original product, or waited to launch the product until all the features were fast.

Lightroom $4K iMac vs. $4K PC Performance Test

Pye Jirsa (via Hacker News):

So, I wanted to do another Mac vs PC test to see if we could purchase pre-built Apple computers that can compete in terms of performance and price with our custom PCs which have been built by our own internal IT guru Joseph Wu.


While we really appreciate Apple’s approach to their hardware quality and design, we can’t justify the price to performance difference at this time.

The custom water-cooled PC was between 26% and 114% faster than the top-of-the-line iMac. The most obvious objection to these results is that the iMac has a Retina display and so was pushing around a lot more pixels:

We were using a 2.5K Eizo vs the 5K Mac, however, to compensate we lowered the Smart Preview resolutions to 2048px on both machines, so all tests/previews were running at the same sizes on both sides.

John Gruber:

To me it just shows how many hoops you have to jump through to make iMac look bad.

Marco Arment:

There’s a lot of bullshitting in the parts choices, but the conclusion stands.

Cabel Sasser:

To me it’s a broader “Mac performance has lagged behind PC performance more than I can remember”


Of course a purpose-built PC will outperform an iMac, dollar for dollar. That’s not exactly news. The iMac also has a severe price-to-performance disadvantage here by including an expensive 5K panel, while the PC is tested with a much cheaper 2560x1440 monitor.

However, the disappointing thing to me here is that you can’t reach performance parity with the iMac by throwing more money at it. The iMac tested here is a completely maxed out machine.


The Skylake processor in the iMac is something the end user cannot overclock (without doing something extreme). And none of the Mac Pro cores are as fast. And Apple doesn’t water cool their machines. They simply don’t offer robust overclocking options to us. So, for a single-core, the 4.0 GHz Skylake in the iMac is the fastest thing Apple offers.

If there is an underlying, valid core criticism in this article, it’s the same one everybody already knows about: Apple doesn’t offer their end users a truly competitive array of system configurations. Instead, they force their users into carefully defined and limited product tiers, each one costing significantly more than the last.

Antitrust and Aggregation

Ben Thompson:

The Play Store is the one part of the Google Mobile Services suite that is irreplaceable and thus the leverage enforcing the various requirements the [European] Commission objected to, like making Google Search and Chrome defaults, and forbidding AOSP forks.


This monopoly, though, is a lot different than the monopolies of yesteryear: aggregators aren’t limiting consumer choice by controlling supply (like oil) or distribution (like railroads) or infrastructure (like telephone wires); rather, consumers are self-selecting onto the Aggregator’s platform because it’s a better experience. This has completely neutered U.S. antitrust law, which is based on whether or not there has been clear harm to the consumer (primarily through higher prices, but also decreased competition), and it’s why the FTC has declined to sue Google for questionable search practices.


One more implication of aggregation-based monopolies is that once competitors die the aggregators become monopsonies — i.e. the only buyer for modularized suppliers. And this, by extension, turns the virtuous cycle on its head: instead of more consumers leading to more suppliers, a dominant hold over suppliers means that consumers can never leave, rendering a superior user experience less important than a monopoly that looks an awful lot like the ones our antitrust laws were designed to eliminate.