The concept of Functional Boundaries

http://docandersen.podbean.com

https://docandersen.wordpress.com

http://scottoandersen.wordpress.com

My Amazon author page!!!!

http://lukeoandersen.wordpress.com

http://chuckandersen.wordpress.com/

http://NickOandersen.wordpress.com

Standing in line at a coffee shop the other day I was checking my email and a friend sent me a spreadsheet with a bunch of numbers to be reviewed. I opened it on my phone realizing quickly that I couldn’t use or leverage the information in the format I had it. It was more information than I could cope with on the small screen. Now I could have broken out my traveling glasses (basically makes my cellular phone a 50 inch screen) but that really isn’t cool in a coffee shop. Additionally no matter how big I make the screen the actual input isn’t easy.

The problem is three fold, what you are using, how you are using it and frankly how the application was built. The first two are serious considerations in the world of how we interact with cloud based solutions in the future. Information you can’t access or information that arrives after its critical path isn’t good.

A functional boundary is the edge of both how something is used but also the requirements for that use. It has two aspects that must be considered Human and Machine. The human functional boundary can be “physical” such as physical limitations that make usage harder. The Machine boundaries can be as complex as overall maximum processor capacity or as simple as the screen simply can’t display the content.

Functional boundaries don’t impact you or your solution until they actually impact you and your solution. From having to hold your phone at arm’s length to read information or having to wear glasses because the print is so small you can’t read it there are human boundaries to solutions. The machines continue to get smaller, faster and more capable. But the screen size is a significant impact. Watching a movie on your cellular phone is one thing, working your way through a complex spreadsheet or presentation with lots of builds is quite another thing.

Machine boundaries

The machine boundary can be broken into capabilities and capacity. Simply put some devices has a set of capabilities that support the “Syncverse” effectively. Other devices do not have those capabilities and will limit the effectiveness of the solution overall.

Capabilities

· What is the devices video capacity

· Local storage capacity for cache

· Permanent data storage that can be leveraged

· Processor capable of rendering requirements

Capacity

· Network bandwidth (right now)

· Network bandwidth (overall monthly)

· Software interaction that allows for soft keyboard and other functional “interactions” such as camera (video conferencing) and voice.

Human boundaries

Human boundaries are harder to discuss because all of us have them. It is the willingness to try new things, the ability to understand and use technology and a willingness to try new ways of doing things.

Ability

· Can you use an on-screen keyboard?

· Can you see what is on the screen clearly?

· What do you want your device to do?

Access

· Is the environment you are in conducive to the task you are trying to complete (is it loud? To bright? Etc…)

· What are the physical or asset security requirements for the solution you are using, can you easily interact with the security requirements on the device you are using?

o Solutions requiring fingerprints don’t work on devices that don’t have finger print readers

o Optical interaction (facial or retinal) recognition requires a camera capable of producing the required quality of image.

From ability as is often the case we open the door. If you have the ability to do something that is frequently the first step. There is a factor that is limiting your ability and that is simply what do you want the device to do. I bought my first Swiss army knife when I was 11 years old. I used that knife to open wine, carve balsa wood and as a screw driver and all around quick to use device. Would I however use a Swiss army knife to cut down a tree? Only if it was the only device I had for that purpose other than my hands, yes. Otherwise I would never use a Swiss army knife to cut down a tree.

Machine boundaries are often more clearly defined and less variable. They do appear frequently as you move to the edges of what you are trying to accomplish. You see people all the time who struggle with the performance of their device as they attempt to complete a task. This represents the functional boundary of that device for them. It could be as simple as lack of bandwidth or as complex as trying to work on a spreadsheet on a phone screen.

Application Development and Functional Boundaries

The next component of functional boundaries lies within the solution itself. I mentioned already the concept of human and machine boundaries. I also talked briefly about the limits imposed by security and the then physical hardware requirements. You can’t require fingerprint access for a solution if in fact the devices (blackberries) don’t support a fingerprint reader.

There is another side to this that is even more significant. The concept of what and how applications are leveraged on the device becomes a huge issue.

1. Can I get to the solution remotely?

2. Does it require local data?

3. Does it require great security than my device supports?

4. Is the application smart enough to understand what I am using and begin an automated switching solution for the overall processing?

.doc

Developing the future of tomorrow

http://docandersen.podbean.com

https://docandersen.wordpress.com

http://scottoandersen.wordpress.com

My Amazon author page!!!!

http://lukeoandersen.wordpress.com

http://chuckandersen.wordpress.com/

http://NickOandersen.wordpress.com

The developers edge of functionality.

Do applications sense that they are running out of bandwidth?

When does the last time an application said to you, I am switching into light mode to reduce the strain on your network connection. Or I am switching into small screen mode, where a smaller portion of the screen real estate is magnified for easier use.

Probably never.

That is the developers change over the next two to three years. Making applications that work with you and for you.

If you think about cloud computing (I find I spend a lot of time doing just that) in the end you realize that the reality of the future has nothing to do with the cost savings cloud computing offers today. It has to do with the future evolution of solutions to meet the ever growing needs of the users and organizations that leverage them.

Cloud is a platform for the future. As we move down the path of cloud computing its important to remember that. We aren’t building the computing models of today, we are building the computing models for the next five years.

Develop the solution of tomorrow that can grow with its own evolution.

.doc

Limpid pools of data so green they make your eyes look orange…

http://docandersen.podbean.com

https://docandersen.wordpress.com

http://scottoandersen.wordpress.com

My Amazon author page!!!!

http://lukeoandersen.wordpress.com

http://chuckandersen.wordpress.com/

http://NickOandersen.wordpress.com

Heading further down the conceptual path of functional boundaries. In planning applications for the future there are several components that have to be considered. I listed a few of them right before thanksgiving.

Do developers think about these issues? Do infrastructure guys plan for them?

The reality is we don’t. We don’t plan for elastic solutions that can on the fly switch from remote to local processing and are “bandwidth” aware knowing what to send and when to send it. We don’t have the ability today to work from pools of data when seeking a result (see the Syncverse Essays of a year ago for more information in that space).

That last piece about data is actually a new term I am going to call elastic data. Like a rubber band you can stretch your data collection beyond one central store. Data pools allow for great flexibility (I only have to capture the delta or changed information or for that matter the subset of new data that is relevant to what I am doing.

I should probably republish a bunch of the Syncverse stuff eventually. I was trying to pull it into a book but that isn’t going so well.

.doc

Let’s change four-square to a new cloud application…

http://docandersen.podbean.com

https://docandersen.wordpress.com

http://scottoandersen.wordpress.com

My Amazon author page!!!!

http://lukeoandersen.wordpress.com

http://chuckandersen.wordpress.com/

http://NickOandersen.wordpress.com

Do you remember those days of September? Sorry the song was stuck in my head – has been for more than 30 years. How do you remove a song that won’t go away?

Cloud Application Design

As we think about the architecture of cloud applications going forward there is of course the many changes I’ve mentioned here over the past four years.

Applications that are aware of the device being used

  • bandwidth
  • screen size
  • processor and video capability
  • usage pattern
  • storage pattern

What scares me is that we really don’t have good usage and storage patterns today. We have generic ones that we can leverage in the short run, but nothing that allows us to build the application awareness that we will need to solve the functional boundary issues.

How can we build an application that is a partnership a symbiosis with the user rather than an interaction.

Let’s pick on Four Square for a second. Great application, great idea, but it is stupid. Stupid in that it finds out where I am when I launch it, then expects me check in. I should be able to have auto-check in so that when I am out and about and hit a new place (perhaps stores and places could have four-square beacons) it can store that information. Then later you can approve them for publication (rather than as it is now). I suspect you could use four square as a stalking tool, and it probably is, without the end of day report process.

I can of course change the application (say I am exploring on a rainy day – it it starts to rain really hard my wife can pick me up if she knows where I am).

If you think about just the three ideas above – it represents  huge change for the foursquare application team.

.doc

The functional boundary of the cloud…

http://docandersen.podbean.com

https://docandersen.wordpress.com

http://scottoandersen.wordpress.com

My Amazon author page!!!!

http://lukeoandersen.wordpress.com

http://chuckandersen.wordpress.com/

http://NickOandersen.wordpress.com

Device capacity – the functional boundary expanded. I’ve talked a lot about the concept of the functional boundary- and in fact about its impact on the cloud solutions of the future. The next thing or consideration for me at least is the raw concept of capacity of devices.

I have a number of different mobile and not mobile devices. Each of the not mobile devices has a certain capacity or function. For example I use a box at home as an ORB server (don’t ask me why – I started doing it more than 5 years ago and just never really stopped). I have another box serving as a windows home server. It does the automated backups for computers on my home network.

Each box has a physical function and is on most of the time. It so happens that I can also use the processing power of those boxes via my iPad to reduce the stress on the iPad as far as processing locally.

The thing is, most applications today aren’t smart enough to say “hey you are using a portable device – do you want me to switch to terminal service mode – where we leave some of the processing on the host system.

Problem 1: smart applications that are aware of what you need, how you need it and the best possible delivery method.

Problem 2: solution awareness of my needs. IE where I am, what device and I using and in a tracking fashion what device might I use next to access the solution. It’s the old batch jobs from the mainframe days. They were set to run each night and you had information/data in your inbox when you got in the morning.What if applications and solutions were smart enough for me to kick off a job on my smaller smart phone device and produce it for me, the next time I log into a device with a larger screen?

There are many more problems and I will noodle through a few of them over the next few weeks.

.doc

The day the Internet came, but didn’t stay.

http://docandersen.podbean.com

https://docandersen.wordpress.com

http://scottoandersen.wordpress.com

My Amazon author page!!!!

http://lukeoandersen.wordpress.com

http://chuckandersen.wordpress.com/

http://NickOandersen.wordpress.com

A non-cloud topic today.

I’ve been blogging for more than 5 year now –regular posting for most than 3 years now. More than 2100 posts overall. During that time I have noticed an anomaly that doesn’t lend itself to easy theories.

The peaks and valleys that are traffic on the Internet.

I notice that frequently I have days that end up with 2 and 3 times the traffic of other days. WordPress nicely graphs your daily views. They vary from 15-60 views per day per blog.

The thing that honestly has me baffled is that there are spike days. They come along roughly four times per month. They normally double or even triple the average number of views (between 75 and 160 per blog). I can’t explain these changes over time. Not sure what causes that either. It is a curious thing, and throws off your stats for the month. If you remove those blips (which I have in the average above) you tend to go down in average views quickly.

So what are those blips? What causes the number of page views to soar.

My initial theory was cool blog titles and catching the right Buzz phrase for search (200 total views on both blogs the day I posted my notes about Steve Jobs after his passing). That isn’t always the case however, sometimes those trends don’t hit my blog.

Perhaps as my blog used to say “last friendly house” I am to far on the fringe of blogging to be relevant so that when the main stream finds me the blips occur. I have a few more theories that I will share someday. For now – its time to relax and finish up some documentation.

,doc

brave new cloud world

http://docandersen.podbean.com

https://docandersen.wordpress.com

http://scottoandersen.wordpress.com

My Amazon author page!!!!

http://lukeoandersen.wordpress.com

http://chuckandersen.wordpress.com/

http://NickOandersen.wordpress.com

Happy Thanksgiving to those reading in the US.

I am thinking about a new iteration of cloud computing and the issues raised by the concept of private cloud computing. One of the issues I see with the explosion of private cloud is the overall management of that new set of solutions.

The public cloud offers organizations a chance to have two data stores they need to worry about – public and the existing infrastructure required to get people to the Internet. What they buy in the long run is the ability to reduce the infrastructure they have installed and deployed because less and less is needed for their users to access and use the public cloud facing applications.

Of course there are a number of considerations that are not as simple as the reduction of bandwidth. Applications will have to be built differently. The first is the functional boundary that I have talked about again and again (boring, I know). Applications will have to change – they will need to be smart enough to know what device is accessing the data and resources. From there they will need to be able to:

  • scale up or down on the server based processing (more server for a mobile application)
  • scale up or down based on bandwidth (what gets returned to the device)
  • scale up or down on what is presented to the user (based on screen size)
  • prioritize packets based on type of application requesting the data as well as the nature of the request

From there of course there are a number of other things that will have to change over time. As more bandwidth moves to the Internet (according to current stats nearly 34% of all traffic on the Internet is Netflix today – that will have to change).

Ah brave new world – what will we do with you?

.doc