Srinivas Reddy’s Weblog

My Tech Rants

Archive for the ‘Technology’ Category

Technology trends and Need for simple software

leave a comment »

 

Web technologies are increasingly moving away from App server (3 tier) to mirco services architecture More details here : http://de.slideshare.net/ewolff/java-application-servers-are-dead and http://martinfowler.com/articles/microservices.html

Balancing IT budgets between specific application needs to generic services/solutions supporting E2E optimization (generic tool adaption in agile DevOps world not be practical). This blog post highlights this dilemma for the SAP aware folks http://scn.sap.com/community/hana-in-memory/blog/2014/10/18/hana-and-the-solman-syndrome. It is helpful to think in terms of Jobs to be done and tools needed to support this, a concept presented well by Harvard Prof. Clayton M. Christensen at a tech event here: http://events.sap.com/sapphirenow/en/session/9602 The trend is towards open platforms where knowledge sharing in network increases value of product/service to unlock trapped value, see link for details http://srobbins.wordpress.com/2014/10/10/a-different-kind-of-trapped-value/

What are the major vendors doing in this context? From the recent Oracle keynotes https://www.oracle.com/openworld/live/on-demand/index.html and SAP TechEd keynotes http://events.sap.com/teched/en/session/13493 the common thread is in bridging the SaaS apps with the on premise apps using PaaS technologies. While the legacy vendors try to simplify their offering using “cloud” cover its really startup and open source solutions that are pushing the innovation envelope.

Multiple layers of abstraction in people and technology stacks is the root of all legacy driven complexity. Simplification needs decision making / smarts to be handled in micro nodes for scalable processing in human and technology contexts. Paradigms: Top down vs Bottom up  – Centralized vs decentralized – Big data / Network optimization / New Biz processes / next gen resource management – Cloud standardization vs custom extension needs (core and context). Cloud apps need fast/productive turnover times for build/deploy cycles which is pretty disruptive for legacy Dev/deploy models (on premise and early j2ee cloud) to compete with. Key part of TCO is TCD in Dev/Ops cycle including maintenance/support/upgrade with extension support.

Change is about people and mindsets. Simplicity starts in the mind. Given individual mindset changing behavior needs transformational leadership.  Fear of change, job loss, uncertainty, insecurity can make people do things which are anything but simple and smart and technology can only accelerate this trendSad smile

Written by srinivasreddy

November 11, 2014 at 7:30 am

Posted in opendata, SAP, Technology

What’s all the “(Big) Data Hype” all about anyway?

leave a comment »

Technorati Tags:
Technorati Tags: ,,,

Its been a long time I blogged, it feels good to get back to it again. With all the digital data around it sure is difficult to take some time out to think what does it all mean anyway. This blog post is an attempt in this direction. We have been hearing the statement “Data is the next intel inside” for some time now. A recent statement by IBM CEO “Data is the next natural resource” is really a much stronger statement and needs some introspection into.What’s all the (Big) Data hype all about anyway?

As predicted here a decade ago, data has grown explosively in the 3-Dimensions of Volume, Variety and Velocity. The database landscape graph here is good portrayal of how the tech community is trying to address this problem in the different dimensions.   As of 2012, about 2.5 exabytes of data are created each day. The info graphic below shows some sources for this data.

 

Given that there is already a huge amount of digital data existing and much more being created everyday why should be care? I was recently at the CeBIT tech exhibition show with the theme “Shareconomy” and how (Big) Data driven “Smart” applications are and will change the world was a common theme at different exhibits. While cloud computing (be it IaaS, PaaS & SaaS) has mostly focused on reduced infrastructure costs its really value is in ability to enable companies learn from the data usage and help them focus on things that really matter for their customers. New business models are enabled by big data. While some companies like Salesforce see the data deluge as an opportunity to build “facebook like” user interface for enterprise end users to ease consumption. Other companies like KhanAcademy leverage the data to create a “flipped classroom” education platform where “learning” is emphasized than “teaching”. Data enables “feedback loops” to “validate learning” which is the key principle for lean startups to enable sustainable business. Advances in IT enable leveraging the unstructured (big) data with the structured (enterprise) data to engage with customers in new ways in real time. Products and services are being built much faster and customer acceptance measured to improve or kill the new offering based on such “experiments” in real time. Such “Agile” enterprises are the one setting the “speed of innovation” for their industries. Facebook’s recently announced “Home” interface for mobile highlights the importance about owning the data interface to end users. If you can “engage” the users you can have a greater impact on your companies bottom line. This quality is best highlighted by Amazon and its letter to its shareholders is very insightful in this regard.

Given that (Big) data is important how can we best profit from it? Given that we live in ever increasingly digital (thanks to our mobile devices) world we need to be smart as to where we are engaging in the “digital world” and why. “Time” is the critical resource and we need to use it in the optimal way. (Big) data needs “Big filters” to help us not be drowned in the data. One needs to develop skills to do “Data driven experiments” to learn from the “consumption” patterns to create sustainable business value. While “Open Data” access to different data sets is partly driving the current (Big) data hype. Its important to ensure “openness” of critical data sets at the same time taking care of the privacy and intellectual property rights concerns.

We will have much more data in the future, thanks to devices like goolge glass. Given a lot of IT innovations are still needed to make Big data vision a reality we will do well to enable this data driven disruption than be surprised by it.

Written by srinivasreddy

May 3, 2013 at 10:00 am

Posted in Mobile, opendata, SAP, Technology

Open Aid data reflections

leave a comment »

I would like to share my idea that I had submitted to “Open Data Challenge” in this blog. Even if the idea has not made it to the winners, I think the idea is worth reflecting upon and hopefully implement it as well. The idea aims to use the “Open Knowledge Foundation” project IATI registry that maintains references to standardized XML file released by aid recipients, donors and intermediaries.

The idea is to create Datasets on grassroots agencies (NGO’s) working in development projects in a standard format (IATA format) by enabling ‘Self Service’ entry of such data by these organizations, including building useful visualizations on the datasets. This idea can take a lot inspiration from the World Bank aid visualizations that are currently available and AidData portal that tracks development finance. What I miss in the current solutions is a “Self Service” interface where recipients, donors and intermediaries can maintain the data for their grassroots development projects in a standardized format.

Who will benefit from this idea?

  • This will provide transparent information access to (potential) grassroots aid donors on current funding agencies, causes and recipients
  • Help (potential) receivers transparently access information on current funding agencies and causes
  • Help funding intermediaries better match the interests of donors and aid receivers
  • Enhances current aid data sets with grassroots aid data and makes the data more representative

What will be needed to implement this idea?

  • Need user studies with grassroots agencies on “open aid data” needs and willingness to contribute to “Self Service” data entry
  • Need to explore ways to aggregate grassroots aid data (both automated and manual)
  • Build infrastructure / tools to help ‘Self Service’ data entry of aid data in IATA format
  • Leverage related efforts regarding map features, API & Technology tools to help build visualizations. While CKAN IATI provides the basis needed for building grassroots aid data format and AidData platform can be extended with grassroots datasets it needs to be seen if data format needs changes/simplifications and to what extent current open visualization platforms can be reused

I would expect getting grassroots agencies to contribute to Open Aid data (be it manual or automated means) would be the biggest challenge. Given the potential transparency benefits for the Aid process it is well worth the effort to give it a try :)  Would be interesting to hear the views of others on it…

Written by srinivasreddy

June 19, 2011 at 6:35 am

Posted in CSR, opendata, Technology

Emerging Business models

leave a comment »

Watching the replays from the recent SAP Sapphire user conference two things that stuck out to me are – need for business model innovation and enabling micro consumption. Be it in the Enterprise IT industry or any other industry for that matter, Consumers demand  “Instant value based on Fresh content”. Most industry business models are not prepared for such agility.  This nice blog post @dahowlett highlights this for the Enterprise IT industry. While adopting to the bazaar business models pioneered by Open source software development become mainstream finding the right Content monetization model is the challenge. The music industry trends show case the current state of the art. The fact that Apple Appstore crossed the 500,000 mark recently shows that the power of micro consumption in the consumer IT has arrived. While Salesforce is trying to replicate this in the Enterprise space with the AppExchange and SAP with its EcoHub, Enterprise consumers can only benefit from the competition.

What is really preventing the traditional Enterprises from adapting the Bazaar type of Business model? The question if this is the right model for all Business to adopt is something I personally feel is not up for debate anymore. At the heart of adoption would be creating learning organizations that empower people to collaborate in the enterprise and across the Business network in a transparent way. In this context I like the thought on “social objects in the enterprise” by JP Rangaswami.  Making people across organisational boundaries to collaborate is difficult enough than the task of collaborating with business ecosystem is a major challenge to conventional organizational setups. The genY style leadership being promoted by Vineet Nayar from HCL is a good example on how we can get there. I also like the following statement from SAP Chief Sustainability Officer Peter Graf  “we need to manage entirely new type of resources – social impacts, environ impacts, network life cycle sustainability metrics (energy, water, ..) not just finance and enterprise assets (people, machinery,…)”.

I liked a statement from an expert at SAP Customer event Sapphire “Future is in harnessing diversity (be it people, technology, domains, systems) and ensuring interoperability for unlocking new value”. While the emphasis is mine, I think they highlight the key aspects of a Bazaar business model. Given the current smart customers who keep themselves informed in real time the next gen biz models begin with empowering the employees.  While providing “self service” access to content inside the enterprise has been mainstream for some time how this can be enabled for the business network will be key for the emerging business models. Enabling Amazon mechanical turk like efficiency to extended business process can make the “Bazaar” like business models possible.

Written by srinivasreddy

May 28, 2011 at 10:37 am

Posted in SAP, Technology

Tagged with

Mobile user notes of a Tourist

leave a comment »

Having recently come back from a short vacation in London (UK) I would like to share some of my thoughts on Mobile usage status and needs as a Tourist in this blog.

– Before our trip we were doing some research on places to visit in London so as to optimize time and cost. Having consulted with some London resident friends, visitors and online travel sites we had come up with a initial short list on paper. I wanted a digital version of this list (kind of Tour itinerary)  with map integration to show location details. Basically during research stage I was navigating different websites (travel sites + Google maps) to arrive at a list of places to see. Ideally I would have liked to consolidate this list, get it pushed to my mobile device and provide offline access. Kind of briefing book interface of this content would be great. In reality it was a pretty inefficient process (add stress factor) to actually visit planned places and my expensive mobile was under utilized due to the expensive roaming charges. Considering the number of people who have a similar need there may be some business opportunity for a techie here …

-  I understand Google Maps offers some way to mark (star) locations that you can reuse, I was trying to star locations on the map (based on where I am) from Mobile device but looks this feature is not enable. I know they have some feature to help you track your movements remotely but that is not what I wanted (think of the roaming costs!)

– I tried to use the Google Maps street view but found it as a good idea but still practically useless, at least the current implementation on the mobile device (am using Android HTC Desire).  I found the need a few times that I need some visual assistance (kind of augmented reality support) to go a place from where I am.  I know there is a app for that but possible trying to startup a different app for each need is something even a techie like me had not done naturally

– I was trying to use Foursquare to check-in at some locations and get some tips. The data quality and interaction speed was so poor that I gave up after a few tries.  I was actually looking for some kind of location history (details for a tourist spot I am currently at) and info on other interesting sites nearby. Yes my travel guide provides this information but my “smart phone” can make this process more enriching and fun

– We found out it can be difficult to locate and meet people even knowing they are pretty close by. There should be easy way to share location information and guide the user to the shared location. I know given the accuracy of our GPS systems this is asking for too much but the mobile networks know enough to provide a good enough approximation.

– The biggest hurdle to mobile data usage is the roaming charges. This is still far too expensive currently for tourists. Given the promotion budgets of Governments’ & local businesses to attract tourists, its amazing that the only innovative business model for free Mobile internet usage is to go to Starbucks. At some locations open Mobile Wi-Fi networks were available but did not work at least for me.

– There were lot of promotional offers  with complex pricing models (be it for using the public transport, visiting tourist spots, hotel stay,..). It would be interesting to have a Groupon like service which not only makes last minute offers but also helps optimize standard tourist needs

– It was helpful to use Mobile to locate nearest food joints (the user interface can be better). Use of mobile search and usage of user reviews has arrived

– SMS based messaging is still the killer app for mobile. Given the noise level in London subways it was the best way to communicate. Expect this will be the case for most big city tourists but can imagine instant messaging may also be helpful (if only we can get most users to one system, Facebook?)

– I was impressed by the wealth on knowledge at the Natural History Museum. It is a bit sad that you have to be physically present there to experience this. Given the current E-Learning technologies creating virtual tours for the remotest of schools in the world should be feasible. Actually it was notable how understaffed the museum was, considering the looming British budget cuts this can only get worse. Mobile technologies may help to augment visitor experience in a efficient way. Every object in the museum had a great story to tell (be it the Dinosaur skeletons or the many rock samples…) if only we can make them smart enough to tell their own story and enable the visitors to engage with them in a whole new way…

– Last but not least I had to visit the Apple store to experience it first hand. It was amazing to see people still waiting in queues for hours to buy the iPad. I am convinced mobile internet is here and hopefully we will see mass market data charges sooner than later. The apps to make Tourist experience more rewarding will come Smile

Written by srinivasreddy

April 16, 2011 at 4:47 am

Posted in Mobile, Technology

Tagged with

Cloud computing Trends and Market needs

with 2 comments

Year 2010 may well be the year of Cloud computing (probably with some close competition from Mobile and Sustainability themes) for the Tech folks. In the recent weeks I took some time to understand in more detail if Cloud computing has really arrived or is one more hype theme for IT vendors. I had actually blogged as far as in 2007 on ‘Everything as a Service’ and in 2008 on ‘IT as a Service’. Interestingly most of what I had written then is still valid just that the plot got more complex and audience more interested! So it’s time to take another shot at the clouds.

Without getting into what Cloud computing means (I am sure you have read enough about it!), I would like to think in terms of the “End-user” and “Application developer” be it in Enterprise or Consumer domain. What are their key challenges and to what extend can the current Cloud offering help? It’s after all “market consumption” that decides the success. The Key challenge in Enterprise IT is “Total Cost of Ownership (TCO)”. Given the huge “legacy” of applications that IT needs to maintain at considerable operational costs, making any changes (how about integrating with some cool “social” apps?) is considerable effort. The maxim “Do not touch a running system” still holds. While “packaged” applications come with huge “support” costs and related upgrade requirements, very little room exists in Corporate Budget for new “cloud solutions”. If we look at the increasing revenues of cloud vendors (Amazon, Salesforce.com, Google…) obviously market demand exists…  So why is on-premise computing so expensive and can it really be compared to the cloud offering?  From my SAP development experience I can say supporting all product standards (check slide 15 in link) comes at a considerable cost but with related value add for the customers. Given that not all solutions/customers need such standard compliance it would be fair to say that the “cloud offerings” with all their limitations still have enough niche markets that can be addressed. The real challenge for cloud computing would be to build “complex” business applications and support them over time for a large installed base. While ‘Software as a Service (SaaS)’ is only one aspect of cloud computing (maybe the most important for end-users / app developers), we do see good traction in ‘Infrastructure as a Service (IaaS)’ (aka hosted hardware) while the market for ‘Platform as a Service (PaaS)’ (aka hosted hardware with infrastructure software) is still being defined.

What are the expectations from “Application Developers” from a cloud computing platform? As for any computing platform they would need a ‘productive’ development environment (be it programming language, app framework, IDE support, Source code management…) and runtime environment (be it on-device mobile container, web/app servers…). Offering a productive ‘end2end’ environment can be challenging considering no one vendor may/can control all the ends…The rapid pace at which new cloud programming languages/app frameworks are emerging (be it JRuby, JPython, Scala, Dojo,….) together with new NoSQL cloud databases (Cassandra, CouchDB…) makes Application development more difficult. Till there are some clear cloud platform winners making a technology choice for application development is risky. It could also explain the new found love for open source software of most IT vendors as a risk mitigation strategy ;o).

Given the market needs what are the major IT vendors doing? Having attended a recent IBM cloud computing for developers session, watched recent Google Atmosphere youtube webcasts on cloud trends, attended Deutsche Telekom Developer garden session on Cloud computing and recent SAP DKOM session related to Cloud computing among others, can share some key trends I see.  While some key design issues (Multi-tenancy, In-memory databases, App Lifecycle Management…) need more research, Cloud computing is accepted as a viable option for some use cases. Cloud computing has a lot to do with the many recent/continuing IT acquisitions we see. While the IT stack is at a inflection point, the platform vendor (hardware / system software / applications) has to win in the “Cloud Data Center” to survive. The stakes are huge, one can only hope that open systems win. I like in particular IBM view to cloud as ‘Cloud = Virtualization + Standardization + automation) and SAP focus on ‘on-premise, on-demand and on-device’. It would be interesting to see how ‘best of breed’ solutions migrate to “cloud appliances”. I personally do not think one need’s to own the whole stack (Oracle strategy) to win. Having the right eco-system partnerships and adapting to customer needs will be key.  So ‘End-user’ may well be the winner after all even if  he has to wait longer to harness the clouds.

Written by srinivasreddy

May 14, 2010 at 10:42 am

Posted in Mobile, SAP, Technology

Tagged with , ,

Enterprise 2.0

leave a comment »

Key features of Enterprise applications include being highly controlled (for legal compliance and other valid reasons) and oriented towards individual users (be it transactional or analytical applications).  In contrast Web 2.0applications are highly uncontrolled (anybody can access it) and collaborative (be it social networks, online games, wikis…). As such Web 2.0 applications have found a natural home in consumer web applications.  Off late Enterprises see value of having web 2.0 like applications in the enterprise. This is what Enterprise 2.0 is about. In some sense the current state of Enterprise 2.0 is comparable to ‘walled garden’ approach to mobile content practiced by telecom providers currently. 

The major disruptions Enterprise applications are currently seeing include Software as a Service (SaaS), Virtualization (leading to server consolidation), Shared Services (centralize common/standard processes), Outsourcing (be it as BPO, BPU…), Distributed development teams (be it in the same enterprise or by outsourcing partners) and Service Orientation of Applications (legacy applications are being replaced/updated with SOA capabilities).

In this context, Enterprise applications are already in a major flux and adding Web 2.0 like features is not the highest of priority for the enterprises.  Extending Enterprise portals with collaboration features (including access for suppliers and customers) is already practiced in innovative enterprises. In order for real value of collaboration and web 2.0 technologies to be delivered in enterprises,  its important to provide tools,  re-engineer business processes and  underlying service infrastructure to support collaboration.  In web 2.0 environment any one can create content (like Wikipedia, Second life,…), end users determine content value ( like user ratings in Amazon), end users select content to use (as in Yahoo widgets) and users can easily search all content (be it Google like web search or desktop search feature) .  If such features where to be extended to enterprises it would be possible to design adaptive enterprises.  While Self-service applications are common in enterprises we can expect Collaborative self services in real time in the future.

While Enterprises try to learn from Web 2.0 technologies, Consumer web can also learn from the process orientation and governance discipline in enterprises.   While in the near future we can expect the ‘digital’ consumer to be ahead of the enterprise user, as the boundaries between both become increasingly blurred it will be win-win for both.

Written by srinivasreddy

September 30, 2007 at 8:24 pm

Posted in Technology

Tagged with