It's raining Apps and Devices

Recently I had a discussion with some fellow  from CIOs, both the Public sector and Universities, about this summers news on the "End of the PC era" and "Bring your on device". Some are calling this Consumerization of IT.

Naturally this is very interesting and a potential ground shaker in many IT-organizations around the globe. Control is a key word in many enterprises, not only to regulate information security but also to better provide support to end users. At many universities this control has never truly existed, in many "standard platforms" the end user is Administrator (or should be) and can alter programs and settings at will. This should mean that universities are well equipped to handle this shift if/when it truly arrives.

So will this be realized and if so when? I guess there are a few things that point in this direction. My first example is the news from Nvida CEO Jen-Hsun Huang  who claims that Windows 8 will be able to run Windows Phone 7 Apps (http://www.winbeta.org/?q=news/nvidia-windows-phone-7-apps-will-run-windows-8). If this is true I would assume these apps will be bought through the App Store that will be in Windows 8.

There also seem to be devices that tries to close the gap between "tablet" and "pc". An example is Samsungs Slate PC as presented at the IFA in Berlin (http://www.techradar.com/news/mobile-computing/tablets/samsung-slate-pc-series-7-is-a-tablet-and-a-notebook-1007398). Some argue that the entire release of Windows 8 is about removing this gap. And then we have iPads, where the gap has already been removed. 

Both these examples points in the same direction. The end user will loose control over the operating system (OS) of their device. The OS will be deployed over the internet and controlled by the manufacturer of the OS. Much like Apple do in the case of iPads. I think many end users will welcome this development, especially if this mean the computing device will be more stable and fast.

This means that the end user will provision apps on the device, and these will naturally vary from individual to individual and be a combination of personal and business apps.  But in order to help the end user it would be great if the business can provision (and pay for) apps and also perhaps even restrict access to some apps or documents.

An interesting video on this topic is from Microsoft, "How consumerization is changing the role of IT " http://www.youtube.com/watch?v=3nNe8BbXDj8 (take a few minutes to watch that video).

So in the potential future where the lines between personal and professional life has been blurred or  removed and the devices comes with a prepackaged OS; It will rain Apps and Devices, it will virtually pour down. We can only embrace the risk and the following change and help usher both the end users and our vendors in a direction that are somewhat safer and perhaps less chaotic. 

Battle of giants

This last week and even month has been quite interesting from an IT perspective. We have been able to see the announcements at Apples WWDC, the new user interface (UI) of Windows 8, Google+ and Microsoft Office 365.

Google vs Facebook

Looking at the soft launch of Google+ the big issue would be, is this another Buzz or Wave? Two services that simply failed but perhaps for different reasons. Buzz because it tried to copy Twitter and Wave because it either was to innovative that nobody understood it or because it simply wasn't needed.

Well, Google+ takes another approach than Facebook in several areas. But perhaps the biggest problem will be to get a large enough user base so that the service will gain a self momentum. Facebook is a lot of things but it is now the largest Social media player (even if some has started to leave the service).

In any case it was quite interesting to read that Google+ is supposedly Google's attempt in "killing" Facebook. I guess that will not happen and  that Google will integrate more or less with Facebook. Isn't that Google's core business: Aggregation of information across different services?

Edit: I found this good blog entry about Google+ from SiliconFilter. 

Microsoft Office 365 vs Google Apps

The announcement from Steve Ballmer regarding Office 365 was also presented in the press as an attempt to "kill" Google Apps. Having used Google Apps since 2007, LiU launched Apps for our students back in May 2007, I welcome Microsoft's challenge in this area. It will spur both solutions. For students Google Apps still seems better, there is no need for integration at a desktop level and there are more support for various clients (Smartphones, Pads/Slates etc). For faculty and staff where Microsoft already have a big installed user base Office365 makes more sense and definitely rival on-premise solutions.

Apple vs Microsoft in Touch interfaces

When looking at the new UIs from Apple and Microsoft it is clear that they envision a future where you are more tactile with your computer. It is fascinating to see how similar they seem to be at first glance.

I have to say I have more faith in touch than voice. I can't really see how we could use voice as a primary UI, even though I know Google et al has ideas in this area. Just imagine the scenario: You are in the train writing an e-mail to someone, or searching for information in the subway. Even sitting alone in ones office talking to the computer feels strange (perhaps that is a generation issue and todays youth can't see the problem with that).

However I can definitely see that I would touch the screen or use gestures to control the computer. Even better would be to replace the keyboard and mouse with something that combine touch and gestures. Even eye tracking might work well.

My dream would be large screens in conference rooms with a "Minority report" style of UI. I often catch myself pointing at a projection of my presentation or word document wanting to interact with the computer.

In all the cases above we as customers and end users benefit from the competition between the giants. 

A humble wish regarding calendaring

With the launch of Microsoft Office 365 and the ease of use of Google Apps I have a humble wish regarding calendaring.  It is simply: Integration!

I attend a lot of meetings both within my organization and outside. Within the organization this works quite easily, my calendar is accessible for everyone in our Exchange environment. However it is not easily available to people outside the organization. That means I get meeting invitations from Doodle.org, Tungle.me, Foodle or simply an email where I am supposed to respond what times are available. This process takes a bit of time and is cumbersome. For me it is  also impossible to reserve timeslots in my calendar for potential meetings, I would simply have no time slots left.

Now for the request. I would like to see a super simple solution of a standardized way to expose my calendar (or the one I choose in Exchange or in Google Apps) to different services. Naturally it should only be the free/busy and for X months in advance. Naturally this should be able to be exposed to various services on the web using a web interface.

It should be as simple as providing a URL, x.x.x.x/username/freebusy/calendar with some type of OAuth style of authentication.

If this is implemented throughout many services one could also imagine ways of exposing different organizations free/busy to all end users in an Exchange  or a Google Apps environment. Or even a federated approach where all universities expose the information with in the federation.

In the end it is these types of quite "simple" integrations that makes life easier (or not) when using different services.

The university storage challenge

I challenge all online, hardware and storage vendors in particular  too attend the university storage challenge!

At many universities around the globe there is need for massive amounts of storage. In general the storage solution has to have the following properties:

1. Be fairly  safe - some sort of redundancy is needed
2. Be fairly fast -we are talking about massive amount of storage so access time is not the priority, but the end user must not fall asleep. Ie transfer speed (read / write) should be equivalent to  USB 3 external hard discs. 
3. Have a life expectancy of at least 3 years. Some discs might need to be swapped but not so the storage integrity is damaged.
4. Be fairly scalable - we are basically talking about a huge amount of storage, that can be expanded when need arise
5. Be accessible by any computer via standard file access protocols (like SMB).
6. Have an extremely low price, at a maximum, equivalent to a USB 3 external hard disc.

The target price formula is simple! The price for 1 TB in the solution must be the same price as an external USB 3, 1 TB disc bought at your local hardware store, per year. Today that is around 800 SEK, €85 or $120. This means the price for 1 TB in the solution can cost no more than $360 when the system is bought i.e. a depreciation period of 3 years.

The reason for the price is simple, when talking to a professor it is hard to explain that our super duper good storage solution costs 10 or even 100 times more when they can go out and by one 1 TB disc at any local hardware store (or actually any larger supermarket). 

The expected use case is storage of end user files or other files (like massive data sets collected in a research project) where high speed or redundancy is not top priority. Basically it is a solution that rival an USB 3 external hard disc.

Are there any viable solutions out there? Either something on premise or in the cloud.

The cloud based university - Day 2

Day 2 kicked of with some reflections from day 1. Among them the trust issue.  

Microsoft

Microsoft then  presented their vision and strategy for productivity in the future. Different aspects of that future are unified communication, Business Intelligence, Content delivery, Collaboration and Enterprise Search. These aspects can be delivered either on premise or online.

The presentation then focused on Office 365 which includes Office Professional Plus, Exchange Online, SharePoint Online, and Lync Online.

In conjunction with Office 365 Microsoft's adopts a new type of release cycle. It will have a 90 day cyckle for the Microsoft Online offerings.  But Microsoft  keeps a larger buzz (launch etc) when a new wave of products arrives. An example will be Office 2013 (or under which name it will be sold).

The audience had a huge discussion about security. First talking about physical security. But the physical security seems not to be the issue (most cloud vendors are likely secure enough) but rather the discussion started to revolve around logical security. How can we as customer be certain that there is no leakage of data? Or how can we be certain that support personnel in the US (or where ever) access our data without a support request. It boils down to a trust issue. Which naturally can be regulated in a contract. But even with a contract eventually you have to trust the vendor.

In Microsoft's offering for Office 365 there will be a special education version that will include various templates. Unfortunately we didn't see any examples of these templates.

Before a short demo the Swedish pricing was presented. The most interesting is  A3 which is offered at 94 SEK per user, i.e. around $14 per month per user. And that lies in line what was presented earlier this year.

Cisco

Our last speaker for the day was from Cisco and he also started with the statement "Trust is the key to cloud adoption". According to Cisco the trust is built on four pillars: Security, Control, Compliance and Service Level Management. 

The presenter claimed that todays infrastructure is trusted, controlled, reliable, secure and that the cloud offering are flexible, dynamic, on-demand and efficient. Cisco's solution is  trusted clouds that try to bridge the gap between internal IT and cloud services.

A Cisco slogan: "The network is the Cloud Experience! " and that the cloud computing is just a journey that has started.  Foundations for the journey ahead is consolidation and virtualization. After that the crux is  automation and self service.

The presenter then outlined different types of clouds: Private cloud, Community cloud, Public cloud and hybrid clouds. Most interesting is the term community cloud which makes sense in a university and public sector perspective.  

Cisco identify candidate application for cloud by three characteristics: Non-Core, Standardized and Dynamic. Examples are Multimedia services, Grid computing, virtual desktops, e-mail, storage & backup.

Cisco can see that different NRENs (national research and education networks) offering off-site backup for entire datacenters.

Cisco's concept of the trusted cloud infrastructure is built on three building blocks: Platform (network, compute, storage), Services (security, virtualization, optimization), Architecture (Secure, Multi-Tenancy Architecture).

Summary

The two days has been intense with lots of interaction and discussions. A conclusion that can be drawn is the huge trust issue, how to establish trust and how to make sure trust is there. As a fun comment Google has employed a Trust manager, which is to the point! I guess we will and must see more trust managers out there.

When it comes to technology the simple answer seem to be that everything is possible and given the rate of innovation the journey towards the cloud (or what ever it will be called in the future) has just begun. 

The cloud based university - Day 1

This is my first attempt to blog from a conference so it might contain various errors and so on. But I have tried to capture some key points and observations but naturally I haven't tried to capture everything.

The discussions kicked off with a heated debate on different clouds, IaaS, PaaS and SaaS. I have blogged about this earlier, see The cloud - some definitions.

A first observation some in the audience made is that the internal IT organizations often are outpaced by vendors and customers. The internal IT organization is simply too slow to provide and adopt new services. Especially when it comes to new services like Dropbox etc.

Per Sedihn from Proact and SNIA talked about storage in the cloud or perhaps it should be called DaaS - Data Storage as a Service.

A key point was that the user must be able to do self provisioning of storage and virtual servers and more. And naturally only pay for the resources the user uses. The service ordered must be available almost instantly. The user doesn't accept a waiting time of days or even hours. An acceptable time is perhaps only minutes.

Obstacles against cloud services are emotions, legal issues, SLAs and perhaps more important control.

"Internal clouds is built by internal IT organization that has been around for a while". The argument is that it is more likely that a traditional (older) IT organization builds an internal cloud rather than use a "public"  cloud. As an observation most seem to define a public cloud as something provided by someone else.

A good idea is likely to outsource (move to the cloud) things you do well and not things you do bad. Infrastructure or highly standardized services is perhaps a good place to start.

What is the acceptable price for storage? Is it the same price as can be obtained by buying an external sata disk?

• http://aws.amazon.com/s3/pricing/
• http://www.microsoft.com/windowsazure/pricing/
• http://www.rackspacecloud.com/cloud_hosting_products/files/pricing/

But the long term largest cost for cloud services is integration! If you have a lot of different islands of clouds you likely need integrate them. The users probably expect the services to integrate to some extent.

The old file protocols (like CIFS, NFS) should perhaps be replaced by things like HTTP (REST etc..) or some sort of hybrid web protocols.

Some very interesting numbers for storage:

• A few % was only written to disk
• 7-8 % was read only once after creation
• 1 % is read frequently

Microsoft

Microsoft has several different offers. Amongst them are

• Microsoft CRM online
• Intune
• Windows Azure Platform
• Office 365

One of the things that unify different clouds and solutions are Identity management. So a good identity management is key for any organization.

An interesting option to the cloud is the Azure appliance. It is basically the Azure platform but sold as a physical container that is installed locally.

But perhaps the cloud should be called:

IT as a Service - Helps you deliver! 

WMware

WMwares primary idea has been to decrease cost and increase agility when it comes to servers.

Reason why users round internal IT and go to public cloud services:

• Pooling
• Elasticity
• Automation
• Self-service

Reason for internal IT to not allow public clouds:

• Control
• Security
• Compliance

WMware has created a set of APIs in order to create a hybrid cloud. Ie the possibility ensure things like compliance, control and security but still use a public cloud.

End User computing provides a user centric perspective. Basically pulls applications, data and settings to the (a) cloud. Wmware calls their product Wmware view.

Summary

To summarize the day there has been a lot of discussions and ideas. I hope some of the universities can join forces and test one or more solutions in the future.

The legal, contractual and security aspects are naturally questions  that pop up all the time. So it is good that we have a seminar planned to cover just those issues. 

The cloud based university

I have set up a program for a small lunch to lunch conference in Stockholm with the theme The cloud based university. The venue is Stockholm university and the target audience is CIOs and IT architects from Swedish universities.

We will have a program that combines a more internal discussion with different presentations and perspectives from different vendors.

Amongst the speaker we have Per Sedihn from Proact. He will talk about SNIA (www.snia.org)  and Cloud Data Management Interface (CDMI, http://www.snia-europe.org/en/technology-topics/cloud-storage/index.cfm). I also hope Per will have time to cover some of the new interesting products coming to the market, like Cloud storage gateways.

Microsoft will also present both Windows Azure and Office 365.

Wmware will talk about their experiences in the cloud area. I guess we can have an interesting discussion about private vs public clouds.

We will also talk about the cloud ramifications for the network infrastructure and listen to Cisco talking about the cloud and their talk on that.

These two days will focus on technical possibilities. A future seminar will investigate legal and policy questions that arise when using cloud services.