About Samir

This author has not yet filled in any details.
So far Samir has created 14 blog entries.
9 Mar 2015

Docker Adoption in the Enterprise

2017-05-16T17:19:42+00:00 March 9th, 2015|Continuous Delivery, DevOps, Docker, PaaS|Comments Off on Docker Adoption in the Enterprise

Docker is hot!

Docker was the #2 “best overall open source cloud project” according to a survey by The Linux Foundation and The New Stack in July 2014.  Google Trends sheds some light, though, on the relatively new entrance and rapid acceleration of Docker versus OpenStack and “Virtualization”.

Docker trends

Google Trends - March 5, 2015

So, why is Docker so hot?

Software used to have a major delivery drag – it had to be delivered on some physical medium like tape, disk or download and then installed. Since software could only be consumed so quickly, the impetus of independent software vendors (ISVs) to build and deliver new software was perhaps at an annual or semi-annual basis. SaaS has changed all of that. Now, there’s no reason why a bug cannot be fixed and delivered…instantly, or at least as “fast as possible.”

Web scale SaaS companies (a la Facebook, Netflix, and a big slew of smaller SaaS companies) have been striving to optimize their software release stream to do just that. These modern ISVs have given rise to “DevOps” teams that focus on this task of “Continuous Delivery”. This has given rise to scripting technologies like Puppet and Chef for automating release cycles.

The problem with these approaches is they are still prone to error. Creating a test environment via scripts is not 100% guaranteed to build an exact replica of the development environment, for example. So when QA finds a bug that is not reproducible in the development environment, valuable time is wasted determining if the bug is in the software or the environments.

Docker changes this paradigm. Instead of pulling your hair out to recreate “identical” environments, Docker gets much closer to actually just moving the actual environment around. This eliminates the errors due to release environment differences and the need to build and maintain lots of complex scripts.

Docker provides additional benefits to SaaS ISVs as well. Docker is very light weight, and fast, which makes it easy to scale, resource efficient, and for new cloud ISVs that may not be invested in technology like VMware, Docker providers many of the classic benefits of virtualization.

Enterprises need Docker too

Understandably, large scale SaaS ISVs lead the way in innovation, including Docker implementation. Enterprise developers learn from these best practices and are starting to adopt Docker. Enterprises, too, benefit from the speed and efficiency of continuous delivery (CD), especially those enterprises building customer-facing scaled SaaS applications. But there’s an even larger number of internal applications that perhaps don’t need the daily updates, but certainly need greater efficiency delivered by Docker on internal teams with limited resources.

What enterprise IT may not realize is that if they do not provide Docker for their developers, their developers have numerous external choices now for public hosted Docker services. In many cases, IT does not want their apps, workloads and data running on unsanctioned services, and the real problem is they probably won’t know it’s happening until well after the fact.

Forward thinking IT are providing Docker for their developers, and those furthest along are discovering additional benefits of Docker ranging from lower costs via greater resource utilization or reduced VMware licensing costs to workload portability for moving apps to/from public and private cloud infrastructures.

But it ain’t quite that simple

But what enterprise IT departments are also realizing is that Docker is not so simple to implement, especially given the unique needs of an enterprise. Docker is not an out of the box solution. To implement Docker, you must not only manage containers and orchestration, but ensure resource isolation and access control for security; streamline diverse stack support and upgrades; optimize data snapshotting, backups and recovery; implement monitoring at machine, instance, container and workload levels; and integrate with existing systems like LDAP or Active Directory.

SaaS ISVs need to scale a single application, on a single app stack, on a single infrastructure, and they have engineering resources dedicated to implementing Docker and all requisite components. Enterprises, on the other hand, have diverse workloads, varied app stacks, heterogeneous infrastructures, limited resources, and additional needs for security, control, visibility, governance and compliance, so implementing all of the Docker related technologies for these permutations can be daunting.

That’s were PaaS comes in. A Docker-optimized PaaS for enterprise should take care of implementing Docker and required components, while meeting additional needs like role based access control based on existing IDM with visibility and monitoring. A well designed PaaS will also leverage the unique advantages of Docker to deliver features like idle workload hibernation for further infrastructure cost savings.

Docker is young in the timeline of enterprise technology, but we’ve rarely seen one grow and rise so quickly. Enterprises will lag slightly behind SaaS ISVs in Docker adoption, but the smart IT organizations will not only provide Docker to their developers before they wander elsewhere, but will also discover the other powerful advantages of Docker unique to enterprises without spending an arm and a leg.

24 Oct 2012

Social Fills Gap In Business Process Spectrum

2017-05-16T12:25:56+00:00 October 24th, 2012|Collaboration, Email, Social|Comments Off on Social Fills Gap In Business Process Spectrum

Process Long Tail

Within a business, the collective processes can be viewed as a spectrum.  The more repetitive, team-oriented and generic, the more likely there will be a packaged line of business (LOB) application, like CRM, ERP, etc.  More proprietary processes, those linked to the business’s competitive advantage, will likely need BPM solutions or custom applications.  However, if a proprietary process is not highly repeatable, or does not involve many people, individuals tend to choose the tools used – email, spreadsheets, chat, etc.

Processes Matrix

Social Collaboration fills the gap where processes are too proprietary and ad hoc (different enough each time) to warrant a custom or BPM solution, but also too collaborative or team-oriented for email to be an efficient solution.

By leveraging Social Collaboration, these important “long-tail” processes can be also tamed by capturing and sharing learnings, providing transparency for broader contributions, enabling faster responses via self-subscription activity notifications, etc.  These benefits are lost when relying on email, chat, phone and meetings to fill this gap alone.

15 Apr 2012

McKinsey’s MECE

2017-05-16T17:42:31+00:00 April 15th, 2012|Best Practices, Problem Solving|Comments Off on McKinsey’s MECE

Forgive me if I’m stating something obvious.  I’m just surprised how often I see people (in all kinds of situations) make recommendations without providing any rationale behind their recommendations.  IMHO, this makes recommendations, and all the hard work and thought that went into them, virtually meaningless.  Always frame the problem first and start with the primary objective.

Marquee consulting firm McKinsey has an approach to problem framing called “MECE“.  This refers to “Mutually Exclusive and Collectively Exhaustive.”  The value is more in the philosophy than a specific approach.  Implementation can be done in many ways (outlines, graphics, mindmaps, etc.).  The key is achieving the MECE framing objective.

23 Mar 2012

Collaboration Knowledge and Big Data

2017-05-16T12:28:05+00:00 March 23rd, 2012|Collaboration, Knowledge Management|Comments Off on Collaboration Knowledge and Big Data

Gutenberg’s printing press initiated mass knowledge capture via books.  The Internet is enabling another monumental knowledge boon, but this is not limited to just web pages.  In fact, rapidly increasing pace, more multi-tasking, greater agility, and even shortening attention spans are moving us away from the era of large, heavy books or documents to an increasing amount of micro-content.  Major decisions and valuable knowledge can be found in micro-content including email, microblogging, and SMS mobile texting.  The question is: how much of it is useful knowledge and how much creates noise that just makes finding the useful stuff more challenging?

The above chart illustrates concepts and is not intended to indicate relative magnitudes in any dimension.  However, it may be interesting to think about which collaborative interactions should be prioritized for knowledge retention (i.e., most valuable knowledge).  Does the interaction medium effect the likelihood of useful information?  or the number of people involved in the interaction?  For example, do more real-time media lend to less pre-thought posts, perhaps resulting in less useful info early in a chat conversation, but more valuable conclusions worth saving at the end?  In contrast, an initial blog post has considerable thought (well, some do at least) up front, followed by less formal comments by readers.  The bottom line is we don’t know, so all if it is kept for future retroactive analysis, which leaves us in the burgeoning age of big data, with snowballing growth of lots of little data (micro-content) instead of traditional heavy data (big books and documents).

20 Feb 2012

Convergence of Messaging Systems

2017-05-16T12:31:11+00:00 February 20th, 2012|Collaboration, Email, Social|Comments Off on Convergence of Messaging Systems

Email Legacy

It is becoming increasingly more difficult to imagine life before the Internet, a time when networked Email was a breakthrough, providing electronic messaging to the growing masses of PC adopters.  But Email, at least Networked Email (vs. its predecessor, Mainframe Email), required a store-and-forward architecture to enable wide-area transportation across disjoint computers, often over periodic slow phone line connections.

Converging Delivery Times of Messaging Platforms

The connections are faster now, but the physical networks making up the Internet are still heterogeneous resources, so Internet Email uses effectively the same store-and-forward architecture (via SMTP).  This legacy architecture carries additional weaknesses (see also “Seven Reasons Why Email Sucks“) for Email:

  • Security Risk: Messages are transported over unknown networks.  And most people do not encrypt their email.
  • Storage Waste: Email creates physical duplicates for each recipient (excluding some intranet email systems).  No wonder email quotas are so common.
  • Unreliable Delivery: While Email delivery can be fast, sometimes Emails are delayed or never delivered.  Worse, there’s no reliable way for the sender to be sure.

Over time, new messaging mechanisms were created to fill Email shortcomings:

  • IM/Chat: Provide faster real-time communications.  Enabled by Internet connectivity, and peer-to-peer technology.  Typically, not persistent (conversations usually are not saved)
  • SMS “texting”: Required by new mobile phones and devices which initially did not support email.


But, today, do we still need all of these different messaging systems?  Email delivery times have decreased over time as connectivity and computing capacities have improved.  Fortunately, mobile devices (“smartphones”) now support Email, and Email does provide more persistent information than SMS or Chat.

So, can’t we get the best of both worlds?  1) Real-time (or near-real-time) chat plus 2) Persistent data retention?  Unfortunately, Email still has its shortcomings.

Google Wave was an attempt to create a new messaging application (“from scratch”) that combined multiple messaging needs such as: real-time and message persistence (storing messages for future reference) minus Email failings.  But…Google Wave was just too different.  Only the most revolutionary innovators could take the time to actually adopt Google Wave.

Instead, Social Collaboration, for the first time, offers the best of all worlds, including but not limited to Chat, Persistence, Mobility, Security, Reliable Delivery, Easy opt-in/out, Separation of alert and content (see “Delete Dilemma“), Efficient storage, all in a familiar & easy to use interface (a la Facebook and Twitter).

19 Jan 2012

The Real Cost of Email

2017-05-16T12:33:27+00:00 January 19th, 2012|Collaboration, Email, Knowledge Management, Social|Comments Off on The Real Cost of Email

The biggest cost of Email is the cost of lost opportunity.  Using email for much of our collaboration, idea sharing and decision making is costing organizations dearly in lost knowledge capture.

Imagine we had to buy everything directly from manufacturers – that we had no stores from which to buy items we wanted, not even online stores.  How would that work?  We would have to figure out if the product we wanted even existed and then who made it.  We’d have to contact manufacturers asking if they made the product we wanted or if they could refer us to the right manufacturer.  We’d then have to keep track of who made what and hopefully, they were still the right manufacturers later on.  Finding the best product or price would be even more daunting.

Crazy, right?  Guess what.  That’s pretty much how knowledge is distributed and managed via email today.  When Joe needs some info, he has to figure out…often guess who has it, or interrupt others to find out who might.  He may remember next time or have to go through the same process again next time.  A great deal (most?) of business discussions, debates, background information and ultimately decisions happen in email today.  And most of it is lost to the organization…

Non Discoverable

When sending an email, the sender must know who needs that information and who needs it right now.  Maybe the sender has no idea that someone else could really benefit from inclusion.  And tomorrow, if someone else could benefit from it, they will never find it.  Unless, of course, they start asking various people, often interrupting people from their own work who cannot help anyway.


To ensure anyone who might benefit from an email, the sender may over distribute it.  In which case, recipients not wanting it may feel “spammed”.  After all, we’re all overflowing in email already, right?  When you do receive clearly useful or targeted info, great.  But what about when it might be useful later?  You keep a copy in case you need it [see “Delete Dilemma”]…as does everyone who received it.  Now you’ve got numerous copies of the same thing floating around.  If one person updates it (this can be a document or just a discussion update), how can you be sure they’ll  a) send the updated version to all people who got the previous version?  and b) those recipients will save that updated copy (and ideally delete the older one).  This gets complex when you think about it!

Lost Context

Another shortcoming that we just seem accept and live with is the “context bottleneck” effect of email.  When I send someone an email, I already have some context about the email – e.g., subject, categories, pertinent accounts, orders or in fact, any associated business process or object.  But the poor recipient has to figure out for herself all of the relevant context.  Granted, her context may be different than mine, but more than likely most of the time, much of the context would be relevant and useful, but it’s lost unless I somehow recreate it.

Solution: Social Collaboration

Social collaboration provides a familiar experience (a la Facebook) that, if done well, integrates well with email and solves many of these knowledge management problems.  Once social collaboration liberates knowledge from email silos the growth of the organization’s collective knowledge will compound rapidly.  Good search and proper contextual social collaboration (like Qontext.com) actually makes finding relevant information easier (see my post on “Context Collaboration”) than in email.  And this collective knowledge benefits individual workers with productivity gains and the organization with greater creativity (see my post on “ROI of Social Collaboration”).

It’s surprising what shortcomings we get used to living with.  That’s why often important changes can only occur with generational change.  I sincerely hope, for all of our sakes, that it won’t take that long for people to wake up to the various shortcomings of email (see my post “Email Sucks”).

9 Jun 2011

Seven Reasons Why Email Sucks

2017-05-16T17:44:41+00:00 June 9th, 2011|Collaboration, Email, Knowledge Management, Social|8 Comments

Ok, maybe that’s a bit harsh…or is it?  We can’t live without email, just like we couldn’t live without telephones even when all we had was rotary dials.  Who wouldn’t say “rotary dial sucks!” today?  We’re dependent on email and drowning in email at the same time.

Look, email, in some form, will not get replaced completely.  Just like TV didn’t completely replace radio.  But following are just some of the reasons why we are way overdue for a change.

1. Knowledge Silos

Email is a poor (and that’s being polite) information sharing repository.  When sending an email, the sender must know who needs the info now.  A colleague who could benefit from the emailed information tomorrow, will never know the info even exists.

Since each email inbox is completely private, stored content can not be shared without explicitly and actively forwarding the content.  This a) requires time to forward those requesting it, and b) creates duplicate copies of the content.

2. Delete Dilemma

deletedilemmaI’ve dedicated another entire blog post to email’s “Delete Dilemma” because I believe it is so important and I don’t recall ever seeing anyone previously identify this issue.  Basically, because an email includes both content and the notification of that content, recipients can not delete the notification if they feel they might need the content again later.  And the expectation of others is that if they emailed you some info, you’re responsible for remembering it.  The result is keeping “everything”, relying on unread marks, manual flagging and labeling.

3. No Priority Hierarchy

The email inbox is more or less a flat list of stuff.  Sure, the sender can set a low/medium/high priority, but who does?  And is the sender’s high priority necessarily yours?  You can also try building some crude rules in Outlook to change colors of emails based on sender.  Gmail even tries to do this for you.  But is every email from your boss or mother really your top priority?

4. Lack of Context

What’s been the latest big innovation in email?  Threading!  Ooooh.  Email threading is nothing more than a rough attempt to automatically group emails by their explicit Subject.  Yes, this is helpful, like…spitting on a fire to put it out.  How often do email threads take tangents that have nothing to do with the original subject?  And how many people just reply to an old email (even keeping the old subject) as an easier way to address a new email? (It’s easier than trying to figure out which is current of the multiple emails addresses suggested by your email program.)

When you think about it, the email sender already has the context of the message.  Instead, we leave it to the recipient to figure out the context and organize (tag, label, file in folder, etc.).  With the increasing volume, organizing received emails usually just does not happen any more.  Instead, we rely solely on sort and search…or probably more prevalent than we’d like to admit, we just ask people again for answers we already have buried in our email inboxes.

5. Distributed Mess

Email’s store-and-forward basis made sense in a world with proprietary dial-up connections and latencies.  Today, this cobbled-together public infrastructure has vulnerabilities.  E.g.,:

  • According to Microsoft, “more than 97% of all e-mails sent over the net are unwanted.
  • An email sender can never be 100% sure when or even if  all intended recipients will receive their email.
  • And even with “standards”, many deal with incompatibilities on a regular basis.  Anyone seen “winmail.dat” attached to an empty message?  Or incompatible formatting?  Or ever try to send meeting invitations including different recipients using MS Outlook, GMail, and Lotus Notes?

6. No Data Life Cycle Management

OldEmail data management has long been a concern for IT and more recently for Compliance.  IT sets crude storage quotas to manage costs (email proliferates unnecessary duplicate copies of files and data).  But data archiving, regulatory compliance, legal holds, etc. are a nightmare.  With email’s store & forward architecture and increasing mobile device adoption and diversity, IT demands security features like remote wipe so an employee can’t just walk away with critical corporate data.

7. Ambiguous Etiquette

Some people ignore emails if they’re only included in the “cc” or “bcc” list.  For example, when introductions are made, often the match-maker is stuck left in the thread, cc’d in all the subsequent emails.  So does cc equate with fyi?  When multiple people are included in SendTo, who’s responsible for responding?  With email lists and listserves, this gets compounded.  Do you reply to the group or just the sender?  Do you then cc the group?  ReplyAll as become ubiquitous with internal spamming.  Inevitably, someone seems compelled to ReplyAll to explain why ReplyAll should not be used.

Let’s face it.  Most of us have a Love/Hate relationship with email.  When all you have is a hammer, everything looks like a nail.  We use email for “everything” collaborative.  In email, we share and co-author documents, discuss, announce, invite, schedule, flame, update, etc.  The “Least Common Denominator” functionality of email provides flexibility but also creates inefficiencies for certain tasks where a more appropriate collaboration tool would be better.

But we’ve lived with the pains of email so long we don’t even notice them.  Today, imagine getting rid of your cell phones to carry coins around for pay phones.  Ridiculous now, right?  No one thought so before cell phones.

Shouldn’t we be able to: Subscribe to what we want? Decide how/when we get notified?  Discover useful info we didn’t know existed?  That’s where “Social Collaboration” comes in…for the first time, Social Collaboration offers a supplement, if not a real alternative to email.

18 Apr 2011

Email’s “Delete Dilemma”

2017-05-16T17:48:23+00:00 April 18th, 2011|Collaboration, Email, Knowledge Management, Social|1 Comment

Email Overload

Sure, we can blame email overload on spam, the exploding information age, or our increasing life complexity and responsibilities.  But volume of emails received is only part of the problem, in fact, arguably, not the primary problem.  Even a trickle of running water into a tub that does not drain will overflow.  The unspoken culprit to our email overload, then, is our inability to delete emails.

We keep so many emails because these emails include both content (e.g., message, answers, files) and the notification of this content.  Since we have no other guaranteed way to access an emailed content, we don’t dare delete the email on even the slightest chance that we may need the content again in the future.  This inevitable email hording creates a build up of emails in mail files, whether we organize email into folders or leave them all in our inboxes.  To compound things, our emails include a mix of different content types from various sources (e.g., personal and work) and varying priorities.  This makes finding information complex and inefficient at best.

For example, we may receive multiple versions of the same document over weeks, but how do we remember to delete older versions when we receive an updated one?  This results in multiple versions of the same document, which just adds to the noise when searching for information within my Email.


The good news is that there is a solution, and a seemingly simple one: Separate email content from the notification.  Storing the content in a central repository and sending a notification of the content plus a link to the content allows the recipient to read the notice and delete it without worrying about losing the content.

Content management systems have existed for years.  What’s really enabling this separation of notification from the content for the first time is Social Collaboration.  That’s right, from the roots of Facebook, come real, tangible, game-changing ROI.  By separating this notice from the content, users can literally cut email in half.  This is not just a shift of email from one place to another.  This separation simply allows us to eliminate the temporary notifications while getting the persistent information out of our email inboxes where they do not belong…finally freeing us from the email “Delete Dilemma”.

26 Jan 2011

Years in the realizing…Social Contextual Collaboration will deliver major productivity gains.

2011-01-26T14:35:36+00:00 January 26th, 2011|Collaboration, Social|Comments Off on Years in the realizing…Social Contextual Collaboration will deliver major productivity gains.

The culmination of cloud, social and contextual collaboration is creating a perfect storm.

When all you have is a hammer, everything looks like a nail.  And so email has been used for decades as the de facto collaboration tool.  Since IBM Profs in ’80’s and Lotus Notes in ’90’s, companies have tried to offer better alternatives.  However, these alternatives fell short due to complexity, inflexibility and cost.  Web solutions cropped up, but they just duplicated an antiquated paradigm.

For the first time, the decreasing costs and increasing capabilities and acceptance of the cloud, coupled with the familiarity of social technology and ubiquity of thin client, contextual collaboration is poised to deliver transformative value to business, no less than the size of the email industry.

Social promises to make information more accessible by turning information sharing upside down.  Rather than forcing people to fit into information structures like folders and subfolders, social is making the info structures fit the social structures.