Learning by Shipping

What’s in store for 2015 when it comes to technology advances in the workplace?

Originally appeared on <re/code> December 18, 2014.

This next year will see these technologies broadly deployed, but with that deployment will come challenges and choices to make. This sets up 2015 to be a year of intense activity and important choices — how far forward to leap, and how to transition from a world we all know and are working in comfortably. In today’s context of the primacy of smartphone and tablet devices, robust cross-organization cloud services and the changing nature of productivity — all combined with the acute needs of enterprise security — lead to dramatic change in the definition of the enterprise computing platform, starting this year.

Amazing 2014

This past year has seen an incredible — and exponential — diffusion of technologies. Who would have thought at the start of the year we would end the year surrounded by:

• Smartphone/supercomputers, some costing less than $50 contract-free, in the hands of almost two billion people
• Free (essentially) or unlimited cloud storage for individuals and businesses
• Tablets outselling laptops
• 4G LTE speeds from a single worldwide device in most of the developed world
• Amazing pixel densities on large-screen displays, introduced without a premium price
• Streaming 4K video
• Apple’s iPhone 6 Plus “phablet” sold very well (we think) and is now perfectly normal to use
• SaaS/cloud services scaling to tens of millions of business subscribers
• Major cloud platforms putting millions of servers in their data centers
• Shared transportation is on a path to substitute for traditional taxis, and in many cases, private car ownership
• Mobile payments finally arrived at scale in the U.S. and are routine in some of the world’s least developed economies

These and many more advances went from introduction to deployment, especially among technology leaders and early adopters, thus creating a “new normal.” In terms of Geoffrey Moore’s seminal work from 1991, “Crossing the Chasm,” these technologies have been adopted by technical visionaries and are now crossing the chasm to the broader population.

In the real™ world, technology diffusion takes time (deployment, change, etc.), so we have not yet seen the full impact of any of the above. Moving forward to that future — not just making changes for the sake of change — requires a point of view and making trade-offs. This post has in mind the pragmatists (in Moore’s terminology) who want to accelerate and get the benefits from technology transition. Early visionary adopters have already made their moves. Pragmatists often face the real work in bringing the technology to the next stage of adoption, but often also face their own tendency toward skepticism of step-function changes, along with trade-offs in how to move forward.

Viewpoint 2015

Even with many hard choices and challenges, for me, the coming year is a year of extreme optimism for what will be accomplished and how big a difference a year will make. Looking at the directions firmly seeded in 2014, the following represent strategies and choices for 2015 that demand an execution-oriented point of view:

• Enterprise cloud comes to everyone
• Email isn’t dead, just wounded, but kill off attachments with prejudice
• Productivity breaks from legacy work products and workflows
• Tablets make a “surprise comeback”
• Mobile device management aims to get it right
• Hybrid cloud ROI isn’t there, and the complexity is huge
• Cross-platform really (still) won’t work
• Massive security breaches challenge the enterprise platform

Enterprise cloud comes to everyone.

When it comes to cloud services for typical information workflows, bottom-up adoption, enterprise pilots and trials defined 2014. The debate over on-premise versus cloud will mostly fade as the pragmatists see that legacy “on-prem” or hosted on-prem software can no longer innovate fast enough or connect to the wide array of services available. Cloud architecture is different, and new software is required to benefit from moving to the cloud. The defensibility of holding an enterprise back or attempts to find plug-replacements for existing legacy systems proved weak, and the demand from business unit leaders and employees for mobile access, cross-product integration, enterprise-spanning collaboration and the inherent flexibility of cloud architecture is too great.

The most substantial development in 2015 will be enterprises defaulting to multi-tenant, public-cloud solutions recognizing that the perceived risks or performance and scale challenges are far less than any existing on-prem or hosted solution or upgrade of the same. The biggest drivers will prove to be the need for primarily mobile access, cross-enterprise collaboration and even security. The biggest risk will be enterprises that continue to shut off or regulate access to solutions, especially by preventing use of enterprise email credentials or devices.

The biggest enterprise opportunity will be integrating leading offerings with enterprise sign-on and namespace to permit easy bottom-up usage across the enterprise, with minimal friction. Because of the rapid switch to cloud, we will see legacy on-prem providers relabel or rebrand hosted legacy solutions as cloud. The attributes of cloud “native” will be key purchase criteria, more than legacy compatibility.

Email isn’t dead — just wounded — but kill off attachments with prejudice.

So much has been said and written about the negatives of email and the need for it to go away. Yet it keeps coming back. The truth is, it never went away, but it is changing dramatically in how it is used. Anyone that interacts with millennials knows that email is viewed the way Gen-Xers might view a written letter, as an overly formal means of communication. Long threads, attachments and elaborate formatting are archaic, confusing and counter to collaboration. Messaging services and apps trump email for all but the most formal or regulated communication, with no single service dominant, as context matters. In emerging markets, email will never attain the same status as developed markets. Today, receiving links to documents is still suboptimal, with gaps to be closed and features to be created, but that should not slow progress this year.

Using cloud-based documents supports an organization knowing where the single, true copy resides, without concern that the asset will proliferate. Mobile devices can use more secure viewers to see, print and annotate documents, without making copies unnecessarily. The idea of having a local copy of attachments (or mail), or even just an inbox of attachments, is proving to be a security nightmare. Out of that reality, many startups are providing incredibly innovative scaleable solutions that can be deployed now based on using cloud solutions,.

Services like DocSend can track usage of high-value documents. Textio can analyze cloud-based documents without having to extract them from a mail store, or try to locate them on file shares. Quip edits documents and basic spreadsheets, and integrates contextual messaging avoiding both mail and attachments while safely spanning org boundaries.

This year, casting technologies will allow links to be sent to displays via cloud services for documents, as video is today. The leading enterprises will rapidly move away from managing a sea of attachments and collaborating in endless email threads. The cultural change is significant and not to be underestimated, but the benefits are now tangible and needed, and solutions exist. The opportunity for new solutions from startups continues this year, with deployments going big. Save email for introduction, announcements and other one-to-many communications.

Productivity breaks from legacy work products and workflows.

The gold standard for creating business work products is not going anywhere this year, or for 10 more years. The gold standard for business work products, however, is rapidly changing. Nothing will ever be better than Office at creating Office work products. What has significantly changed, in part driven by mobile and in part driven by a generational change in communication approach, is the very definition of work products that matter the most. Gone are the days where the enterprise productivity ninja was the person who could make the richest document or presentation. The workflow of static information, in large, report-based documents making endless rounds as attachments, is looking more and more like a Selectric-created report stuffed in an interoffice envelope.

Today’s enterprise productivity ninja is someone who can get answers on their tablet while on a conference call from an offsite. They focus their energy on the cloud-based tools that have the most up-to-date data, and they get the answers and don’t fret about presentation. They share quickly knowing that content matters more than presentation because of the ephemeral nature of business information. The opportunity for the enterprise is on the back end, and moving to real-time, cloud-based solutions that forgo the traditional delays and laborious ETL efforts of dragging massive amounts of data onto client PCs for analysis. The risk is in seeing cloud solutions as anything but the definitive source of data and as workgroup or side solutions, so integrating with the primary sources of transaction data will provide a great opportunity to the organization.

Tablets make a “surprise comeback”

Some thought 2014 was the year tablets faded. Many debated the long replacement cycle or weak competitive position of tablet between phablet and laptop. The reality is that tablets will outsell laptops this year. Some discount all the cheap Android tablets barely used at home, but then one must discount the laptops that go unused in analogous scenarios. Regardless, one thing distinguished 2014 with respect to tablets, defined as iPads: You see them in the hands of business people everywhere, from the coffee shop to the airport to the conference to the boardroom. On those iPads, there are enterprise apps, email and browsing (and now Office), doing enterprise work.

The big change in 2015 will be (and I am guessing like everyone else) the introduction of a new iPad, and likely first-party keyboard attachments and/or (at least) iOS software enhancements for improved “productivity.” A tablet properly defined is not just a form factor, but is a hardware platform (ARM) and a modern/mobile operating system (iOS, Android, Windows Phone/Windows on ARM). Those characteristics, being a big phone, come with the attributes of security, reliability, performance, connectivity, robustness, app store, thinness, light weight; and above all, those attributes remain constant over time.

Laptops will have their place for another decade or more, but they will become stationary desktop tools used for profession-defining tools (Excel in finance, Photoshop in design, AutoCAD in architecture, and many more). Work will happen first on mobile platforms, for both team agility and organizational security. The scenario that will resonate will be a larger-screened modern-OS tablet with a keyboard and a phone/phablet as a second screen used in concert, as shown by Apple’s Continuity. The most significant opportunity for those making apps will be to design tablet- and phablet-optimized experiences and assume the app is the primary use case.

Mobile device management aims to get it right.

From the enterprise IT perspective, the transition from managing PCs to managing mobile devices (phones and tablets) is both a blessing and a curse. The faster that IT can get out of managing PCs, the better. The core challenge is that in the modern threat environment, it has become essentially impossible to maintain the integrity of a PC over time. Technical challenges, or even impossibility, mean that 2015 could literally see pressure to reduce PCs in use.

If you doubt this, consider the Sony breach and the potential impact it will have on the view of traditionally architected computing. The rise of tablets for productivity is, therefore, a blessing. Over time, any device in widespread use is eventually a target. Therefore, mobile presents the same risk as the bad actors find new techniques to exploit mobile. The curse, and therefore the opportunity, is that our industry has not yet created the right model for mobile device management. We have MDM, sandboxing and user profiles. All of these are so far not entirely well-received by users, and most IT feels they are not yet there, but for the wrong reasons. IT should not feel the need to reintroduce the PC approach to device security (stateful, log-on scripts, arbitrary code inserted all over the device, etc.).

This leads to a lot of opportunity in a critical area for 2015. First, a golden rule is required: Do not impact the performance (battery life, connectivity) or usability of the device. It isn’t more secure for the company to issue two phones — one the person wants to use, and the other they have to use. Like any such solution, people will simply work around the limitation or postpone work as long as possible. This dynamic is what causes people to travel with iPads and leave the laptop at home (along with weight, chargers, two-factor readers and more).

The best bet is to avoid using or emphasizing management solutions that work better on Android, simply because Android allows more hooks and invasive software in the OS. That’s quite typical in the broad MDM/security space right now and is quite counterintuitive. The existence of this level of flexibility enabling more control is itself a potential for security challenges, and the invasive approach to management will almost certainly impact performance, compatibility and usability just as such solutions have on PCs. As tempting as it is, it is neither viable nor more secure long term. Many are frustrated by the lack of iOS “management,” yet at the same time one would be hard-pressed to argue that the full Android stack is more secure. There will be an explosion in enterprise-managed mobile devices this year, especially as tablets are deployed to replace PCs in scenarios, and with that, a big opportunity for startups to get mobile management right.

Hybrid cloud ROI isn’t there, and the complexity is huge.

In times of great change, pragmatists eager to adopt technologies crossing the chasm may choose to seek solutions that bridge the old and new ways of doing things. For cloud computing, the two methods seeing a lot of attention are to virtualize an existing data center, or to architect what is known as a hybrid cloud or hybrid public/private (some mixture of data center and cloud).

History clearly shows that betting on bridge solutions is the fastest way to slow down your efforts and build technical debt that reduces ROI in both the short- and long-term. The reason should be apparent, which is that the architecture that represents the new solution is substantially different — so different, in fact, that to connect old and new means your architectural and engineering efforts will be spent on the seam rather than the functionality. There’s an incredibly strong desire to get more out of existing investments or to find rationale for requiring use of existing implementations, but practically speaking, efforts in that direction will feel good for a short while, and then will leave the product or organization further behind.

As an enterprise, the pragmatic thing to do is go public cloud and operate existing infrastructure as legacy, without trying to sprinkle cloud on it or spend energy trying to deeply integrate with a cloud solution. The transition to client-server, GUI or Web all provide ample evidence in failed bridge solutions, a long tail of “wish we hadn’t done that” and few successes worth the effort. As a startup, it will be tempting to work to land customers who will pay you to be a bridge, but that will only serve to keep you behind your competitors who are skipping a hybrid solution. This is a big bet to make in 2015, and one that will be the subject of many debates.

Cross-platform really (still) won’t work.

It has been quite a year for those who had to decide whether to build for iOS first or Android first. At the start of 2014, the conventional wisdom shifted to “Android First,” though this never got beyond a discussion with most startups. With the release of Android “L” and iOS 8, the divergence in platform strategy is clear, and that reinforced my view of the downsides of cross-platform. My view was, and remains, that cross-platform is a losing proposition. It has really never worked in our industry except as an objection-handler. Even today, almost no software is a reasonable combination of cross-platform, consistent with the native platform, and equally “good” across platforms.

As we start 2015 it is abundantly clear that the right approach is to focus on platform optimized/exploitive apps, leading with iOS and with a parallel and synchronized team on Android. Android fragmentation is technically real, but lost in the debate is the reality that the highly fragmented low-end phones also almost never acquire apps nor do they represent the full Google stack of platform services. So the strategy is to focus on flagship Android, such as Nexus, Samsung and Moto (though one must note that the delay there of “L” was more than a month even on Moto) or to focus on a distribution of Android from a specific OEM that has some critical mass, and is aimed at customers who will actively acquire apps.

To be clear, we are in a fully sustainable two-ecosystem world. But given the current state of engagement, platform readiness and devices, 2015 will see innovation first and best on iOS. If you’re building your app and working on core code to share, one should be cautious how that goal ends up defining your engineering strategy. Typically, once core code is in place, it selects for tools and languages as well as overall abstractions, and what system services are used. These have a tendency to block platform-native innovation, or to constrain where code goes. Those prove to be limitations, as platforms further evolve and as your feature set expands. The strategy for cross-platform apps also applies to cross-platform cloud. Trying to abstract yourself away from a cloud platform will further complicate your cloud strategy, not simplify it. The proof points and experience are exactly the same as on the client.

Massive security breaches challenge the enterprise platform.

2014 will go down as the “year of the massive security breach.” Target, eBay, J.P. Morgan, Home Depot, Nieman Marcus, P.F. Chang’s, Michaels, Goodwill and, finally, Sony were just some of the major breaches this year. This next year will be defined by how enterprises respond to the breach.

First, the biggest risks are endpoints. Endpoints as defined by today’s technology are likely vulnerable in just about all circumstances, and show no signs of abating. Second, the on-prem data-center infrastructure suffers this same limitation. Together, the two make for a very challenging situation. The reason is not because today’s infrastructure is poorly designed or managed, but because of the combination of an architecture designed for another era and a sophistication level of nation-state opponents that exceeds IT’s ability to detect, isolate and remediate. As fatalistic as it sounds, this is a new world. Former DHS Secretary Tom Ridge said in an interview, “[T]here are two types of companies: Those that know they have been hacked by a foreign government and those that have been hacked and don’t know it yet.”

The challenge for 2015 in this year of adapting to new technologies is managing through the change. The good news is that there are tools and approaches that can make a huge difference. This post picked many trends that taken together are about this theme of securing a modern enterprise. If you use public cloud services on next-generation platforms you aren’t guaranteed security, but it is highly likely that the team has assembled more talent and has an existential focus on security that is very difficult for most enterprises to duplicate. If you use cloud services rather than local or LAN storage for documents, not only do you gain many features, but you gain a level of security you otherwise lack. Not only is this counterintuitive, it is challenging to internalize on many dimensions. It is also the only line of sight to a solution.

As endpoints, the combination of a modern mobile OS and apps is a new level of security and quality. The most innovative and forward-looking solutions in security will be found in startups taking new approaches to these challenges. Even looking at basics, deploying enterprise-wide single-sign-on with mobile-phone-based two-factor would be a substantial and immediate win that accrues to both legacy solutions and cloud solutions.

Technologies to watch in 2015

Above represents some challenges in the extreme, but also a huge opportunity to cross the chasm into a mobile and cloud-centric company or enterprise. Even with all that is going on to get that work done, this will also be a year where some new technologies will make their appearance or begin to wind their way through early adopters. The following are just some technologies I will be watching for (particularly at the Consumer Electronics Show in January):

Beacon. To some, beacon is still a solution searching for a problem, but I think we are on the cusp of some incredibly innovative solutions. I have been playing with beacons and encourage startups that have any potential to use location to do the same. In terms of enterprise productivity, beacons plus a conference room or auditorium is one area where some incredibly innovative tools can be developed.

4K and beyond. Moore’s law applied to pixels has been incredible. Apple’s 5K iMac topped off a year where we saw 4K displays for hundreds of dollars. In mobile, pixel density will increase (to the degree that battery life, OS and hardware can keep up) and for desktop and wall, screen size will continue to increase. Wall-sized displays, wireless transmission and hopefully touch will introduce a whole new range of potential solutions for collaboration, signage and education.

Tablet keyboards. I am definitely biased in this regard, but I am looking forward to seeing a strong combination of tablets, keyboards and mobile OS enhancements. If you’re developing tablet apps, I’d make sure you’re testing them out with keyboards, as well. The idea that a laptop clamshell form factor can be a mobile OS is going to be normal by the end of the year. The need to convert between “tablet mode” and “laptop mode” isn’t a critical feature for productivity, especially for large screen size. Physical keys will define a clamshell, and make converting to a “tablet” awkward. Innovative touch-based covers could make a resurgence for smaller tablet form factors.

The following is a set of “everyday” things you can do, starting immediately. They are easy. They almost certainly require a behavior change. They will make a difference.

Payments. Apple Pay arrived in 2014 and will have a huge impact on how we view payments. Yet the feature set and usage are still maturing. The transformation of payments will take a long time but happen much faster than many think or hope. I am optimistic about traditional bank accounts, credit cards, currencies all being transformed by the block chain and mobile. Because of the immense infrastructure in the developed world, it is likely the developing world will be leaders in payment and banking.

APIs. One of the most interesting differentiators of cloud services is the way APIs are offered and consumed. Every cloud service offers APIs that are easily consumed at the right abstraction levels. In the old days, a client-server API would look like SQL tables. Today, this same API works the way you think about developing custom apps, time to solution is greatly reduced, and integration with other services is straightforward. I’ll be on the lookout for services with cool APIs and services that take advantage of APIs used by other services.

Machine-learning services. Artificial intelligence has always been five years away. I can safely say that has been the case at least for my entire programming lifetime, starting with, “Would you like to play a game?” Things have changed dramatically over the past year. We now see ML as a service, even from IBM. The ability to easily get to large corpora and to efficiently compute training data in cloud-scale servers is a gift. While it is likely that everything will be marketed using ML terms, the real win will be for those building products to just use the services and deliver customer benefit from them. I’m keeping an eye on opportunities for machine learning to improve products.

On-demand. On-demand is redefining our economy. In many places, a few people still view on-demand as a “spoiled San Francisco” thing. As you think about it, on-demand and same-day delivery bring a new level of efficiency, reduction in traffic, pollution, congestion, infrastructure and more. It is one of those things that is totally counterintuitive until you experience it, and until you start to think about the true costs of consumer-facing storefronts and supply chain. On-demand will be viewed as a macro-efficient necessity, not a super-luxury convenience.

From the coffee shop to the boardroom, 2015 will be a year of big leaps for everyone, as we tap into the new normal and execute on a foundation of new services, new paradigms and new platforms.

—Steven Sinofsky (@stevesi)