Dropbox Broke My Heart

For The Last Time: Tech’s Unsatisfied Need For Cloud Storage

There is a major failure in the design of personal cloud storage services, this is why and how we act on it

Google Trend for Cloud Storage interest over time (based on News Headlines, search frequency, and more)

For at least four years now, Cloud Storage has been an eminent category of data sharing and consumption. Unlike Cloud Computing, Cloud Hosting, and other various brother and sister terms—which refer more broadly to the medium and technicality of “the Cloud” itself—Cloud Storage is most commonly for end-users.

Personal Cloud Storage services, like Dropbox, Copy, and box, appeared as early as eight years ago, but their spike in popularity occurred only within the past four years. This was based on a fundamental shift in the goal of these services—to provide a sort of “grand central station” for your data. Because the technology of uploading and downloading files was vastly improved with the adoption of new, multi-core and multi-thread, block-by-block protocols (reminiscent of P2P-based clients like BitTorrent), and broadband speeds continued to skyrocket, even the least ambitious hosting companies could offer barebones file synchronization.

In this new model, the primary advantage of using Dropbox, or an app with a Dropbox core—like Carousel—is offloading. No longer do you have to “worry” about managing your own space.

Secondary is the ability to access your files universally, with the assurance of prime usability, regardless of medium.

The third and final main objective seems to be security. Even services like iCloud Drive, which offer minimal file syncing for Yosemite Macs along with a limited app-folder syncing for iOS 8+ devices, prioritize your file’s safety and boast of failsafe encryption.

Now, these reasons sound all well and good, but are they fulfilled using the fundamental principles of design? A product which requires harmonious integration with your day-to-day requires good design. If there is no overlap between the projected profitable purpose and the benefits of good design, can we trust any software to fulfill our needs? Or even be enjoyable to use?

Much less, do these cloud-business objectives resemble the pillars that made us truly desire an offloading service in the first place? Our interest in “offloading” is not simply a physical or virtual removal; we want to cleanly remove the responsibility of files. Dropbox is designed to keep you content and sandboxed into both a role and a virtual location on your hard-drive—fulfilling the very opposite.

Now that the culture of technology realizes the capabilities of high-speed internet and reliable LTE network service, have we forgotten to take a step back and make sure we are using them to the best of their abilities?

Maybe I’m just infuriated that the Dropbox sync client—remember, Dropbox is today’s most prominent and seemingly profitable cloud storage service—is the most unstable, dysfunctional piece of software on my computer—even if other services like MEGA and Copy suffer from similar afflictions. But, even if it wasn’t, how could I possibly use up all of the 1TB? The original offloading goal is impossible using the client’s somewhat arbitrary restriction to the Dropbox folder, and the lack of configurability to sync, say, more than one folder. And, if you lack high-speed internet access (or something else on your computer is hogging it), or something goes wonky with the client, there is no way to deduce the issue or manually perform any interactions with the system. Or even see a system log or diagnostic data. And if you jump ship? You are certainly no better off, without a simple way to migrate data or intelligently recover changes or thorough status history (unlike the incredibly lightweight git system, which has been around for ten years).

Error when managing several files in the Dropbox web interface. These files are not movable to the Dropbox folder, so this instruction is impossible to follow. It’s insulting that Dropbox would assume everyone has an active hard-drive with their entire Cloud—it contradicts the entire point of my 1TB plan!

Offering to free up photo space on an iPhone is great, but offering up almost every megabyte of storage on a Mac could be so much more.

Not only is that not a consideration, it’s a literal impossibility.

Not one popular cloud storage service obeys the basic principles of problem-solving, human-based design

Dropbox has a robust user experience, browser-based (complete sync-requiring) linear versioning, social integrations, and the most widespread 3rd party developer integration of any cloud-based service in the app community. All of these aspects, however, ultimately take away from the core user experience, either by bulking up your Dropbox — Where do I put all of these shared folders? How do I make sure they’re always up-to-date? — or taking away from your personal storage — I want to share my giant family photo album, with my maximum storage plan, but if anyone has the starter 2GB plan, there’s no way for them to get a glimpse.

What problem is Dropbox trying to solve?

It’s not even a feasible replacement for a USB Drive, especially when drives are currently cheaper per unit of data, don’t provide the false illusion of instantaneous synchronization, allow for prioritization of data transfer (speeds and methodology), and have all of their contents completely usable once connection is established (without need of extra software, a mobile phone to 2nd-factor authenticate, or access to a password manager). Dropbox can’t handle large uploads and downloads, even now (unless you play within your browser for ages). Even if you want to remove files from your Dropbox, you’re restricted to the minimum arbitrary scratch space Dropbox requires on your hard drive. In addition, as Dropbox has increased their prevalence into photo centralization, I’ve found on iPhone it provides a nearly identical service to the current iCloud Photo Library, and on Mac ends up clogging the CPU, extra storage, and network as it migrates your entire Photos directory tree into a new folder (the location of which they don’t let you choose), at which point enabling Selective Sync will take all day, if not all week. Even though I was really crossing my fingers on this one, Dropbox is also far from capable of hosting music and video files for saving, streaming, or sharing (which is just on Dropbox’s end, since cloud players like Tunebox and Eddy already exist).

kernel[0]: process Dropbox109[26506] thread 391845 caught burning CPU! It used more than 50% CPU (Actual recent usage: 83%) over 180 seconds. thread lifetime cpu usage 178.025760 seconds, (139.961418 user, 38.064342 system) ledger info: balance: 90002563068 credit: 176312147119 debit: 86309584051 limit: 90000000000 (50%) period: 180000000000 time since last refill (ns): 108204299426 (pastie.org)

Business-based cloud storage is just not a reliable solution to backup and centralize anything.

“I’ve been amazed at how often those outside the discipline of design assume that what designers do is decoration — likely because so much bad design simply is decoration. Good design isn’t. Good design is problem solving.” — The Art & Science of Web Design by Jeffrey Veen

A truly good service is honest, and spurs innovation. Dropbox is not honest, is not thorough, and does not capitalize on the restrictions of its software model. Instead, it blatantly gives up and punishes the end-user for slow network speeds and complex legwork and migrations. If I have a new idea for how to use my Dropbox space — like storing my iTunes Library in a nested Music folder — an inevitable struggle will ensue, that is, until I concede defeat. Life becomes restrictive fast if original thoughts, ideas, or as some say, innovation, is suffocated. In this case, my Mac client became stuck “Indexing…” for ten straight days, the web client refused to be responsive when more than 100 files were uploading at a time, and the speeds of either were miserable more often than not. There is no middle ground. Either you’ve succeeded or failed, and Dropbox—the service, not the people—won’t help you regardless.

Using Dropbox makes my days worse, more than any other service I use regularly.

That’s why I’ve written this article, and why I’m certain we need a reflection and redesign. Even the worst-designed products shouldn’t negatively impact our lives, especially on a personal, life-changing matter. Dropbox does, and has, but there’s nothing we can do without rethinking cloud storage from the get-go. Essentially, we need to take a step back. Dropbox might be one of the most obvious examples of business turned left and right and twisted all about. From its shortcomings, we can learn a whole lot. I think this market is primed for an actual competitor.

Someone needs to analyze and transform the stubborn ideology of these services, as well as propose something of a solution, or at least a way to take care of the oppressive software restrictions. Here’s my take.

Taking Care of the Cloud Storage Need

“…today’s applications will naturally move towards a cloud model as they become more pervasively available through the web, require more data processing, and span the boundaries of multiple devices.” — Julian Friedman, quoted in ComputerWeekly, 2009

Cloud Storage has championed social recognition as the most reliable, extensible, and sensible model of data storage. Cloud Storage services, on the other hand, couldn’t be more confusing and somewhat infuriating to choose from and ultimately use as an end-user. Besides clear developmental and design-based problems, which plague all parts of the app-based software industry, this undue confusion is rooted in an ignorance of need.

This need in particular—which a majority of users seemingly notice very little—is a composite of all the inevitable problems with limited, physical, vulnerable types of data storage. And, since virtually only Chromebooks are off this model, every traditional personal computer user is risking a whole lot by not realizing and fulfilling the need which Cloud Storage has been crammed into—sort of by chance.

Having terminology like “cloud” become mainstream and trendy alongside a technical push to switch to the smarter, recently feasible style of data transactions, made everyone jump to use it. But, how do they use it? How can cloud storage be sold to the consumer who doesn’t understand the words in the first place? Here’s where the “sort of by chance” comes in: cloud storage marketed a few possible uses, competitors fought, and what resulted was an incredible possibility cornered into a niche, although still wonderful, use-case—kind of like Twitter.

So, what are the main problems with basic personal computer storage? What is cloud storage in the position to solve? Common disk storage is:

  1. Subject to physicality. Hardware damage and technical usability with working parts and machinery, time insensitivity and inevitable wear, general insecurity considering it can be robbed like a piece of jewelry, subject to the limitations of time and space. Data only exists once—arbitrarily, considering modern standards—and can only be viewed using a specific type application, sometimes one-at-a-time.
  2. Inconsistent to use. Sometimes this means, well, it’s damn hard to organize and get through file structures on your computer. When I pick up a new computer, god knows I have to shift and re-evaluate my expectations for hierarchy and organization. In certains models, you’d expect certain usability, but can’t have it. A file in your Mac’s Documents isn’t on your iPhone. You restored from a Mavericks backup on Yosemite, and a few applications went missing. Your memory doesn’t serve you right, and there’s no way to get a list of all the movies saved on your laptop, desktop, external hard drives, and backups disks or services.
  3. Frustrating to manage. Manual sorting, cleaning, checking, renaming, and generally managing a structure of files and folders on a computer (and most certainly in real life too) is a pain in the ass. Although many accept this, and sort of use most computers as a big pile of data—sometimes right on the desktop—that we put in a closet when it gets too obsolete to use. We’d hardly consider hiring a maid to virtually sweep the inside of our laptops—which is a problem in responsibility and ownership of items, not just basic personal computer storage.

Dropbox and existing personal cloud storage sync services do not alleviate any of these. If a file in your local sync folder hasn’t finished uploading, you’ll never get it from your laptop back home while you’re on a trip. If iTunes Match fails to upload a few songs, you’ll never play them from anywhere but your computer and (iTunes Match-disabled) phone. If the Copy app isn’t working well to steam albums, you can’t use YouTube for every track without a network connection, SoundCloud won’t have all the tracks, Grooveshark will suffer from all of the above, Spotify only works if you download local songs via Spotify on the computer, and god damnit Dropbox and iTunes Match are still having the issues from before. Why can’t I backup my whole computer onto Dropbox if I’ve purchased a 1TB plan? Why do great services like CloudApp and Copy not get any attention from 3rd party developers, lag on internal software development, and are thus cornered into following the same lagging model as their competitors who have already gained fame?

Enough! We need a service that works on all platforms, and is unbounded by influence from existing companies and platform dynamics. One that has an open API and is easy to hook into from basic storage-requiring products, anywhere from Pixelmator to Clear to iCloud Backups to iTunes Libraries to arbitrary picture files for tweets to a shared browser cache. One that has source control dynamics, allowing for a timeline perspective over files as well as the ability to visit and restore from revisions. One that isn’t bound to a single server—so, cloud-based—but also not a single account—sharing, like Copy, but also gifting of storage, like partitioning a hard drive, for other users or teams. One that has modes for instant sharing (images, pictures, videos, documents), internet hosting (websites, packages, repositories), and internet storage (backups, libraries, temporary dumps of files). One that has synchronization which is transparent, progressive, and interruptible (like one would expect from the protocols which are presumably utilized). One that can be paid for monthly, but is able to have the data itself transferred to any other system, be it local storage or an enterprise server infrastructure.

Where is it? In our hands, apparently. Waiting to be forced into the vision of existing services or waiting picked up by a careful team of developers and designers. Either way, only with the foresight and understanding presented in this article can one expect their heart not to be broken again, and again.


Researching some background for this article, I found this succinct line about universal design: “Universal design is made up of four elements: accessibility, adaptability, aesthetics, and affordability.” Business-backed cloud storage services are able to be at the forefront of this model of design, because both the product and the impact of the service can obey these principles. Although the former has been improving and improving—Dropbox is available on nearly all devices, usable with a clean, beautiful interface for syncing and inter-app usage, and is affordable enough for the average end-user to afford the most ambitious plans—the latter couldn’t be in a worse position.


Show your support

Clapping shows how much you appreciated Julian Weiss’s story.