The Situational Cloud

They’re all residing on my laptop, smartphone and tablet now. Like four quarreling sisters, SkyDrive, iCloud, Dropbox and Google Drive are ostentatiously competing for attention. As a consumerized IT user – in my own micro space – I am confronted with exactly the same question that enterprises face more and more: what shall I put in whose cloud?

As always, a simple answer would be the best. It would be great to just choose one superior supplier – say Acme SuperCloud 9 – and stick to it for all cloud business. Unfortunately, reality may turn out to be more delicate and a situational approach will often need to be taken.

Much like the way I find myself currently dealing with various, personal cloud storage options. As most of my media is still brought to me through Apple channels and devices, I turn to iCloud to hold it for me. For all the materials I often need mobile access to (I am a Windows Phone 7 user), I prefer SkyDrive. For open documents that I want to collaboratively work on with others, Google Drive from now on is my natural choice. DropBox has the slightly unrewarding, but in fact glorious role of being a general backup. Then – of course – there is a limited, but crucial collection of confidential data assets that in no way should be stored anywhere outside my own premises. I keep them on my own machine in an encrypted data vault and have them backed up by a Time Capsule.

Clearly, reputation is everything here as the various financial plans for cloud storage tend to gradually resolve around one and the same benchmark. But the track record in real life is even more determining: if one of the solution installs improperly, crashes just a few times, is too often unavailable or – the horror – loses data, it will mercilessly be replaced by one of the other cloud sisters in the house. That set aside – assuming that robustness and trust are being dealt with sufficiently by providers – I can explore and switch my options freely, increasing my benefits from the cloud over time, while learning more and more.

For enterprises, at their own scale of reality, it is a very similar play. There is no specific point in selecting one cloud provider to deal with all purposes. There is also no urge to determine right now what computing, storage and application loads should be moved to the cloud and what should stay on premise. Possibly with the exception of start-ups and certain small and medium businesses, the hybrid scenario will be the de facto standard in the forthcoming years.

We will need to move forward step by step, opportunistically weighing our options, considering new paths as they emerge, leaving the unreliable and dusty ones, improving while we learn. Data sovereignty considerations may keep enterprises from moving their applications to a public cloud that is run from another continent: understood, run the delicate ones on a regional cloud or keep the crucial data on premise and process it on a public cloud. Two years from now, regulations may have been changed and options with better economies of scale become feasible. You may not be ready to fully migrate these vital, but ill performing client / server analytical applications to the cloud: fine, just replace the data crunching parts and have them run by a superfast, specialized Big Data cloud.

The options are limitless and smart navigation is key. Come to think of it, we may want to coin the concept of the Situational Cloud right here, right now as a flexible way of benefiting from various, evolving cloud options, supported by a set of well-defined design patterns or ‘cloud stereotypes’.

Providers need to understand that the ability for enterprises to effortlessly navigate the evolving cloud landscape is instrumental to broader adoption. I can easily switch between my personal storage clouds because they all have been integrated in the same way in my file system. Similarly, cloud providers need to move to joint, open standards around how to store and compute in the cloud. This may temporarily feel to them like cannibalism, but soon the increased usage of the cloud by enterprises will bring them the desired additional revenues.

In the end, we want to be able to have different cloud deployment scenarios at our fingertips and literally play with them: seamlessly changing as insights, alternatives, rules and regulations, technologies and business cases evolve. Some enterprises will prefer the details of such a situational approach be hidden, for example through automated, policy-based tools or by partnering with a specialized provider that takes care of the orchestration activities. Other enterprises may choose to stay in control for now and will strive for maximized transparency around the attributes of different cloud options. They accept that choices will need to be made and frequently re-assessed.

Then again, making choices can sometimes be very simple. Aspiring and already quite successful startup GreenButton enables applications even literally with a green button: if you push it, external cloud power – running on Windows Azure, Amazon or vCloud – will be activated on the fly to speed up calculations, analytics and graphics rendering. If you want to know what pushing the button additionally will cost (or save, think about it), all the information is there to make the decision.

Knowing what buttons to push when, and have all the insight for support: it’s really what the Situational Cloud is about. Let’s go and push a few of these buttons.


Posted in English | Leave a comment

What would Amazon do?

It’s still an established benchmark for any application developer building a web shop: before starting to discuss structure, activity flow and lay-out, you have a look at the world’s leading example and see what’s hot. What would Amazon do?

Well, they are doing it again. With the launch of the AWS Marketplace, Amazon shows any IT department what they are up against in the forthcoming years: a neatly organized, easy accessible catalogue of open, highly standardized IT services, ready to deploy in seconds, paid per use, all on one invoice. And – of course – at incredibly competitive prices.

I have often discussed with our clients how quick – and through what steps – they could be benefiting from the public cloud. And the same advice would be coming back over and over again: “We’re not saying your entire IT landscape should be on Amazon next year. But for sure, they are quickly defining a new normal in terms of how fast, easy and cost-effective you should be able to deploy new solutions”.

That benchmark just became more tangible and solid than ever before. Go at the AWS marketplace yourself and browse around a bit. Will your IT department be able to provide the same catalogue, with the same self-service, usage-based pricing and deployment in minutes? And even so important: are your prices more or less on par with what Amazon is offering?

More improvements should be expected. It seems like a missed opportunity not to present the AWS marketplace in exactly the same way as the Amazon web shop (that would have made the point even more). And for now, you can only imagine what will happen when more and more business applications will become available (did anybody say recommendation engine?) through the very same marketplace.

For now, the bar has been raised. We are not saying the internal IT market place of your organization should be just as good as Amazon’s. Not yet, that is. But a new normal has been defined and we’d all better have a good look. Happy shopping.


Posted in English | 1 Comment

7 Signs an Enterprise is getting the post-PC thing

Guess we are all in agreement by now? As Tim Cook confirmed this week when launching the new iPad (not ‘iPad 3’ not ‘iPad HD’ not iPad Maxiumus Optimus Prime GT 7.3.1 X, just iPad, sounds simple, makes even more sense) we have clearly entered the Post-PC era. The App Effect changes our everyday lives and – in the wake of that – the way enterprises connect to their clients, employees and partners. Enterprise Mobility thus tops the priority list of CEO’s and CIO’s the like. Strategies are not only being made, but are executed on as well.

It is a quickly emerging area that will change our industry. So it is only logical for a company like ours to launch a specialized, global service line that builds on Enterprise Mobility best practices and accelerators that we acquired throughout the globe. Much to discuss about the topic and we invite you to do so, together with us in the forthcoming months. As an appetizer, here are 7 signs that we have come to appreciate as indicators for enterprises that actually understand the power of mobility and the post-PC era:

1. Not Another Channel. Enterprises need to understand the transformative power of mobility. A smartphone or tablet is not a trimmed-down version of a laptop: it inspires us to think far beyond the boundaries of current processes, organizational structures and established management theories. There is nothing against a mobile version of a website or an internal application. Good start. But paving the cow path just won’t bring the potential benefits that a more disruptive approach will.

2. On Par Experience. We are all spoiled consumers of apps that excite and engage us us with compelling user experiences. The thing is, when it comes to Enterprise mobile apps, organizations can’t really afford to produce anything less than that. An app with a clumsy and suboptimal user experience will annoy us, then have us quickly turn straight against it (and we might even share that with the rest of the world). You don’t necessarily have to win design prices with your new mobile apps. Just remember to never, ever go below what is considered on par at the App Store.

3. Inside & (even more) Outside. For sure, the mobility revolution brings excellent opportunities for enterprises to redefine the way their employees work and how they collaborate with their partners. So by all means, define and execute your B2B and B2E mobility strategies. But don’t forget that the real magic may be happening outside, with your customers. One single, well-chosen consumer app may create many more benefits than all of your internally focused apps together.

4. Trains versus Scooters. Mobile apps have a different lifecycle: they often need to be created quickly and should be updated frequently with useful additions. These dynamics are different from what the organization is typically used to, both at the IT department but just as much at the business side. It is a matter of understanding the difference between a Train and a Scooter (here is our manifesto, in case you haven’t read it yet) and creating the governance within the enterprise to deal with it.

5. BYO Reality. Face it: where you still may keep your employees satisfied – sort of – for a year or two with the company-issued laptop, you won’t cut it with mobile devices. They will bring their own, no matter what. Obviously, this is also the case with your customers who may change their preferences overnight in this whimsical, highly consumerized market. Ask RIM. Ask Microsoft and Nokia. Ask Apple (even Cook couldn’t believe his own success). Ask Amazon. So better make sure you have the development platform, the management tools and the agility to deal with the next craze in mobile devices.

6. Supersize. The most successful enterprise mobile apps seem to bring together everything that currently drives innovation and growth. Apps are supersized with real-time analytics and intelligence, business process management, rules-driven decision support, web services, composite workflows, personalization and – of course – social interaction. To top it off, mobile apps are particularly suitable to explore a (public) cloud strategy as they may require extremely scalable computing resources with a minimum of upfront investment.

7. Let Them Build. There will be many places inside and outside the organization with great ideas – and needs – for new mobile applications. It is tempting, particularly from the perspective of the central IT department, to carefully gather all requirements into a prioritized long list of mobile initiatives and then start to build the apps, one by one. But that will not do justice to the potential of the crowd, nor does it provide a short enough time-to-market. Focus on building a ‘hub’ platform instead, with a catalogue of enterprise-level services to catapult new mobile apps that are secure, well-integrated, manageable and consistent. Then actively support and grow the ecosystem around you in building the actual apps. True mobilization indeed.

Posted in English | Leave a comment

CTO blog item: 7 Enterprise IT areas to watch in 2012

You didn’t think we would let you go into the New Year without at least one top 7 list from our side, now would you? After all, it may be a time of economic pressure in some parts of the world, only very rarely we are on the threshold of exciting, promising developments in Enterprise IT like we are now. So without further ado – before they already start celebrating in New Zealand – here are 7 areas that you really should watch in 2012.

1. Success Factors. It’s not about the specific HRM cloud-based solution, really. And also not about SAP making strategic acquisitions. We could have mentioned Oracle and RightNow. Or ‘just’ Or NetSuite. The thing is, with SaaS now very rapidly entering mainstream – including ERP and other core applications – we are seeing new benchmarks for how quickly, simple and cost-effective new solutions can be deployed. It will change organizations and their IT departments just as much as it will change the business models of the major, incumbent software vendors.

2. OpenStack. It may be an oxymoron, this ‘private cloud’, but rest assured that many enterprises and governmental organizations will be building their own cloud platforms next year. It will be in an attempt to start enjoying some of the obvious benefits of the cloud without full exposure to the (alleged) uncertainties of the public scenario. With this, it’s time for the evolution of open standards – always an indication of a maturing industry – for cloud-based infrastructure and we’d say that OpenStack has a pretty good position.

3. Windows 8. The upcoming year might very well be the year of Microsoft, as many of its new, promising developments will take centre stage (illustrated by this list). Windows 8 particularly is interesting, because it might be the very last version of a major desktop operating system or one of the first of a next generation of operating systems that target the new reality of tablets and other smart devices. In practice, it is likely to be both. Hang on for the METRO user interface: it’s truly different and people will love it or hate it. Always a good sign.

4. Oracle NoSQL. It’s not about Oracle, really. Or about NoSQL, for that matter, as  Hadoop and Exalytics fit just as much into the picture. Among other things. We could have mentioned SAP with HANA – and many others did, rest assured – or HP with its risky, expensive bet on Autonomy. But yes: we should take notice when the world’s most influential database company starts to move in the world of Big D and we are actually seeing new, unexplored ways of doing something useful with All That Data.

5. Cloud Foundry. Ready to develop your first real cloud applications in 2012? You may want to have a good look into the world of Platforms as a Service in order to leverage all the potential of scalability, flexibility and productivity. For sure, created the archetypical cloud development platform against which others are measured. But obviously, you need to be into the Gospel according to Benioff. If open source is more of your comfort zone, you may be very interested in the ‘upstream’ move of VMWare (ostensibly endorsed by HP) or Red Hat’s entry with OpenShift. Windows Azure is a safe, surprisingly mature bet. The dark horse of 2012 in this space might be Amazon. Oh, and if you are considering new development for the cloud anyway, you may as well want to consider applying more simple, more elegant programming languages that will save you time and frustration. Ruby, for example.

6. Sybase Unwired Platform. It’s not about Sybase, really. Or about SAP putting its strategic acquisition to very good use. We could have mentioned Antenna, Appcelerator, Kony or Adobe’s PhoneGap. Or simply point to Microsoft, still a potential leader in this space. But in any case, next year will see a strongly increased demand for enterprise-level mobile applications. And with that, new challenges need to be met around security, manageability, lifecycle management and integration. But having said that, your enterprise mobile apps better look just as good as their frivolous cousins from the Apple and Android app stores. After all, we are all so consumerized by now.

7. Kinect for Business. For real. Just have a look.

Posted in English | Tagged , , , , | Leave a comment

Slow IT

Ik mocht nu alweer enkele jaren geleden een praatje houden op JavaZone: een congres in Oslo dat jaarlijks 2400 software engineers trekt. Wat u zegt, niet helemaal een gelukkig geval van typecasting. Ik heb me in het verleden nog wel eens licht kritisch uitgelaten over Java, hoewel ik moet toegeven dat het altijd een kwestie van spijkers op laag water zoeken was. Met uitzondering misschien van wat ik zei over de bizarre complexiteit van de taal, de lage productiviteit en de semi-criminele toeren die ontwikkelaars elke dag weer uithalen om hun dolgedraaide object-georiënteerde ideetjes af te beelden op een database. Maar dat zijn details, laten we eerlijk zijn.

En gelukkig viel er heel wat meer te genieten dan geklets over een programmeertaal. Er waren zoveel sessies dat de organisatie zes grote schermen in de centrale hal had geplaatst. Zo kon je tegelijkertijd alle spreekbeurten in de zaaltjes volgen. En met speciale Overflow Headphones kon je steeds wisselen van audiokanaal. Het is veelzeggend dat veel congresbezoekers er de voorkeur aan gaven de hele dag te blijven zitten in de hal, rusteloos zappend op zoek naar interessante fragmenten in de sessies.

Heel passend eigenlijk. Veel voordrachten stonden in het teken van agile systeemontwikkeling en zo’n beetje alle beroemdheden uit het vakgebied waren er. Dat leidde tot het ondertussen overbekende gepreek voor eigen parochie in een gelijkhebberig ons-kent-ons sfeertje. Op een gegeven ogenblik merk je dat het gaat tegenstaan, al die intergalactisch gecertificeerde, 7e dan Scrum masters. Allemaal de mond vol over hoe het nog sneller, interactiever en in nog kortere cycli kan. En hoe de eindgebruikers zich steeds weer blij verrast zullen werpen op de zoveelste daily build.

Hoe weten ze dat eigenlijk zo goed? Zonder uitzondering lieve, vriendelijke mensen, die agile guru’s. Maar de meesten maken toch de indruk dodelijk verlegen te zijn en op zijn minst te lijden aan pleinvrees, dagdromerij en wereldvreemdheid. Ik vraag me af hoe vaak ze in het wild met echte gebruikers te maken hebben.

Al die flexibiliteit en interactie, het wil nog wel eens een tikje kortademig aanvoelen. Op het Internet werken we ons googlend en klikkend door een reeks van nieuwsbronnen heen. Een filmpje hier, een blog-item daar: als het meer dan twee minuten kost, raken we afgeleid. Die hijgerigheid slaat ook neer op onze aanpak van systeemontwikkeling. Niet te lang met je gebruikers praten over specificaties, want die zijn toch nog niet helder. Maak eerst maar eens iets werkends, dan komen we er later wel weer op terug. In een workshop of zo. Niet te diep in je ontwerp opgaan, je kunt in een volgend stadium altijd weer terugvallen op refactoring.

Ik zie nog steeds grote voordelen in een iteratieve benadering en intensieve gebruikersparticipatie. Maar het ruikt onmiskenbaar naar politieke correctheid in agile kringen en de balans is zoek. Tijd voor een tegenbeweging. Laten we weer eens wat vaker bewust de tijd nemen om de essentie van het beoogde systeem tot ons te laten doordringen en te begrijpen wat onze gebruikers nodig hebben. We kunnen veel leren van de principes van Slow Food: een onder koks populaire stroming die ervan uitgaat dat goed eten een kwestie is van geduld, rijping en toewijding. Gewoon drie dagen rundvlees op 57 graden laten sudderen in een Grieks herderspotje. Dat is nog eens wat anders dan een zootje ingrediënten door elkaar mikken in een wok.

Van een echt goed, doortimmerd databaseontwerp heb je decennia lang plezier. En het getuigt van respect naar de gebruikersorganisatie om zorgvuldig aandacht te besteden aan het doorvoelen van de vraagstelling voordat er specificaties worden gekrast.

Slow IT. Het vakgebied is er gereed voor. Nu nog een keer een uur vinden om het concept verder uit te werken.

Posted in Dutch | Tagged , | Leave a comment

Schizophrenic Tester

Ah yes, testers and me. We go back a long, long time. It’s much like love, really. We had our high a few years ago, when I wrote an article series for an IT magazine. In this series I used practical observations and some basic anthropology to describe the psychological mind set of various practitioners in IT.

Experiencing many different IT organizations across the world, I had noticed that there is a strong correlation between the personality of an individual and his or her role in the IT profession.

It is for example quite easy to distinguish between Java and .Net programmers. The first category has strong analytical tendencies, typically hates practical solutions and prefers a painfully slow process of thinking and rethinking before producing anything (if you’re lucky). Their ADHD brothers and sisters in the Microsoft camp on the other hand, are particularly interested in short-term results. This is reflected in flimsy prototypes, a trial-and-error style of developing and an aversion to anything that tends towards structure or documentation.

As you by now start to realize, the article series generalized just a tiny, little bit.

Some truths however were uncovered that turned out to be difficult to deny. About the megalomania of some ‘enterprise architects’ for example that won’t mind a bit to invent their own Oath of Hippocrates to emphasize the presumed importance of their activities. Or about the self-assertion of project managers who all – almost without exception – have suffered in their childhood from a more successful brother, a particularly critical father or both.

I must admit that within the series, testers were easy targets. From my own experience, I had learned to know testers as rigid, over-serious people. Never a smile, even if errors were found. On the contrary: instead of being happy about detecting bugs, testers would tread the instigators (quite often .Net programmers, obviously) with cold disdain. Clearly, this did not add to the popularity of testers, who often could be found in clusters in the company cantina, isolated from all other people.

But that was then. Much has changed in the meantime. The role of testing in the entire applications lifecycle is much better understood nowadays. It is part of the earliest stages – often right at the core of requirements management –so that developers are not unpleasantly surprised at the end of a project by suddenly appearing wall of testing. Also, testers nowadays immerse themselves much better in the context of a project, which has – step by step – adjusted their initial perspective on the world (“testing is the one and only purpose in life”).

And that context is changing quickly, putting extra demands on testers. A powerful wave of ‘business technology’ solutions is enabling an entirely new generation of applications that are developed quicker, have a shorter lifecycle and typically are implemented in close alignment with the business side of the organization. Often, easy-to-use tools and platforms are applied (think BPM and business rules suites, model-driven development, cloud services, mobile and social platforms, open data and app markets, mash-up tools, self-service BI) that are much better understood by business people as well. It creates an ‘outside-in’ perspective, in which many of the newer innovative solutions are developed far outside the central IT department, or even outside the organization.

It means requirements management is in for some changes – if there is still any requirements management left – with an obvious impact on testing as well. Agile thinking and acting will be the default. And more than ever, both insight in the needs of the business and the power of the next generation of tools will determine success.

It will make the role of the tester even more situational. One day, you may be testing a robust, mission-critical ‘Train’ application (check our introduction of the concept in this white paper) that is built for multiple decades of uninterrupted use. Predictability and controllability will be the virtues. The other day, you may be handling an opportunistic ‘Scooter’ application and it may all be about speed, agility and the willingness to accept a certain level of risk.

So in the world of 2012 in which mobile, social and the cloud impose a new rhythm on top of the established, core applications landscape, schizophrenia is not a prerequisite to be a good tester. It would certainly help, though.

Posted in English | Tagged , , , , , , | Leave a comment

Black (pardoned) Turkey

Want to have something nice to read on your iPad this weekend? Download Nassim Taleb’s very instructive, still very relevant ‘The Black Swan’. Taleb describes the phenomenon of the Black Swan as a metaphor for unpredictability. You see, until the end of the 17th century the entire inhabited western world presumed that there were only white swans. This insight was based on observation and a bit of extrapolation: as far as the eye could see, only white swans were present. Furthermore, this always had been the case. Therefore, everything indicated that the next morning the situation would be the same. Until 1697, when Dutch explorer Willem de Vlamingh sailed into the still unknown continent of Australia and bumped into flocks of black swans on the very first river he took.

Black swans occur when we least expect them and they cannot be predicted based on what we know and see.

I had to think about the concept when I recently spoke on a conference for IT service managers. The focus was on the future of service management and some existential discussions were going on. I thought, let’s add some food for thought. Infrastructure from the cloud, rigid standardization and simplification of services, a tweeting and facebooking helpdesk, Bring Your Own, SaaS applications: service management will change its face considerably in the forthcoming years. Come to think of it, it may even disappear a bit.

You could almost feel the atmosphere change in the conference hall when I brought up this last topic. An unmistakably, cold wind blew across the stage. I thought I heard a muffled curse.

Only natural. In the IT profession, we just love groundbreaking paradigm shifts. Unless they pertain to our own ways. Apparently, it is difficult to stay neutral: you are the turkey that is being asked about Thanksgiving.

A good point in time to introduce a new concept that describes this inclination. I propose to use the Black Turkey from now on as a model for the – often involuntary – tendency of IT professionals not to radically renew their profession because this could lead to a loss of work.

And don’t think for a moment that we only encounter Black Turkeys at conferences for service managers. In applications development and maintenance, we are almost trampled by them. Let’s face it: we have some though questions of conscience to answer.

Are we really ready to consider standard, packaged solutions from the cloud as a replacement for bespoke software from the past? Even if we have to stay close to the original, ‘vanilla’ version and cannot elaborately customize and extend?

Do we truly want to get rid of that scattered, best-of-breed applications landscape that can only be held together through stacks of middleware? Even if we have to accept the ‘limited’ functionality of the much simpler applications of that single vendor?

Are we genuinely willing to retire applications? Also if they are the absolutely unique ones that we once built ourselves with blood, sweat and tears?

In applications management, do we really favour the idea of continuous simplification and optimization? Even if it means it will lead to a strong decrease of trouble tickets to solve?

Are we indeed ready to consider new, more productive programming languages to substitute the hideous, overrated complexity of Java and C#? Also if it will take us only a fraction of the time to build the same applications?

Are we still in love with the idea of having a distributed development team that brings in the best-educated, -motivated and most productive capabilities from all over the world? Even if our team members from India, Malaysia, Argentina and Romania cover a bigger and bigger part of the applications lifecycle?

Just a few interesting dilemmas to chew on. Don’t take it too literally though (thanks on behalf of the turkey).

Posted in English | Leave a comment