Mobile and “Offline versus Online”

For fun, I just typed in www.sap.com on my Samsung Galaxy S4. You can watch the browser translate that request to the m.sap.com version of the URL so that you get a mobile specific version of the page. Was it instantaneous…not really, earth shattering…definitely not but did it work…yes. The scenario we are talking about here is an example of a connected mobile web scenario, a scenario targeted at folks who would like to download the latest content from the www.sap.com or from any other site. In this example, the scenario is clearly illustrative of the online use case and is the browser based connected scenario of the past 10-15 years.

As I cross the country as SAP Canada’s National Architect for Mobility, a common area of discussion is connected versus disconnected. To most customers, the question implies how will their application react when there is no internet connectivity (no LTE, no WIFI, no hotspot…say it isn’t so). Some customers will say things like “we are just going with a mobile web site” or “our employees/customers will always have a connection” or “we do not have to worry about offline”. Every customer is different and customers across Canada are starting to get their ideas and strategies around mobile organized. However, prior to these ideas getting fully organized and cemented let me suggest the following or as I tell my customers, I’m going to throw a few ideas into the air and you grab the ones you like.

  1. 1) Customers/partners tackling mobility tend to start with the technical requirements and not the user experience. Before you can make a decision on whether or not to support a truly offline experience, you need to decide on the type of user experience you want to deliver. The instant you decide you will not support offline you become limited in terms of your design options. If you have kids, you have probably heard the phrase “Dad, why is the internet so slow”. The internet and its responsiveness or lack thereof is a crucial element of design that must be incorporated into the end goal. Building a mobile application with a world class UX is not very valuable if each screen transition takes 3,4 or 5 seconds. Start with the “What” and finish with the “How”…not vice versa.
  2. 2) Going offline does not only imply you have customers/employees that will be in an area of no coverage. In my mind, having an application that can operate without a network connection is one way that mobile applications can become one step closer to a superior customer experience.
  3. 3) The internet is great, except for when it isn’t. Does everyone really find the request/response nature of the internet amazing…3-4 seconds for a basic web page as evidenced above is not really what I have in mind when I think about a truly amazing, consumer grade application that everyone wants to use…think Angry Birds…do you really think every time you push a button in that game it does a call over the internet….of course it doesn’t. I’m sure behind the covers, it is doing the occasional update, logging of high score but the core functionality is available offline which is why I can play it on the plane Majority of functions work offine, minority work online…now that’s a model to build upon.
  4. 4) The decision for offline versus online needs to go hand in hand for the type of process being mobilized. Every time you perform a local operation and avoid network traffic, I am a happy user because you will be delivering sub-second responses. Think about the following processes and how you would classify them compared to me:
    1. Stock Market Ticker -> Online
    2. Grocery Shopping List -> Offline
    3. Work Order Management -> Offline
    4. Priority Work Orders -> Online
    5. eBanking -> Online
    6. Workflow Scenarios -> Offline/Online
    7. Inventory Management -> Offline/Online

Having made these broad based statements, there are a bunch of other factors and tradeoffs to be considered that are a huge component of any enterprise wide mobility strategy. TCO, time to value, user adoption, ease of use, flexibility, speed of change, etc… All of these factors need to be considered and have an impact on how a particular application could and should be delivered. At the end of the day, certain industries and segments will benefit more from offline than others and this will be the subject of my next blog.

Mobile is one of my passions as I have been doing it on and off for a long time and would love to hear from you. Comment, email, text me, tweet me or come see me at GTEC this year in Ottawa on October 9th as I speak about the opportunities for mobile in government.

Cheers,

Mobile Guy at SAP Canada

To report this post you need to login first.

6 Comments

You must be Logged on to comment or reply to a post.

  1. Michael Appleby

    Hi Scott,

    I guess I would agree with all the ones you have outlined even the order of f. Workflow Scenarios as Offline/Online (in that order).  I have much less familiarity with many of the mobile technologies than I have with the standard SAP Agentry (formerly Syclo SMART suite) applications for which there are apps for c. Work Order Management  and g. Inventory Management.  Not sure how familiar you are with Agentry, but these are robust and long released products designed to work in an off line mode. 

    Regards, Mike

    Rapid Innovation Group – RIG

    (0) 
  2. Midhun VP

    You have given the right points. While the user is using the application, if he don’t want to bother about the availability of network means it’s a good experience for end users.

    SUP (Sybase Unwired Platform) offers APIs that can work in this manner. Where the user work while he is in an occasionally connected environment.

    -Midhun VP

    (0) 
  3. Scott Dillon Post author

    Thanks for the feedback and comments. My intention is to get into some deeper conversations in the coming articles around how we plan to do offline and some of the pros/cons of the different technologies and get more feedback. I’m not an expert in the Agentry application but I certainly know enough to be dangerous and how we plan to model our future offline applications using their approach 🙂 I’m also very familiar with SUP and how it can do offline for both Native and Hybrid type applications.

    oData and offline will go hand and hand in the future 🙂

    Thanks again for the comments.

    (0) 
  4. Joao Sousa

    I think you are raising two points at the same time. True offline and an application making minimal use of network (read heavy Javascript usage).

    In my mind the second is no longer an option if you want to provide a good mobile experience, people don’t expect to wait for server round trip for the PHP to refresh. That is a big No.

    True offline is still a very complex topic and in my mind in depends heavily on the usage of the app and the level of development of the target country. For african countries where networks are slow, offline is usually a must. In Europe not so much.

    (0) 
  5. Scott Dillon Post author

    Hi Joao…Thanks for the comments.I guess I would say they are two tightly connected points 🙂 The only way I know of to eliminate network traffic is to cache data and refresh screens via that offline cache. The alternative is to make calls over the network which even on the best networks has the potential for lag and everyone’s tolerance for lag is different…mine just happens to be very low. In the relevant cases, I would rather endure a 20-30second “sync” and then have everything function seamlessly without lag as opposed to waiting even 2 seconds for each screen call….but again, everyone is different.

    Have a great weekend, I look forward to your comments going forward.

    (0) 

Leave a Reply