IT Management and Cloud Blog

« | Home | »

Does ITIL Really Matter – The Debate on the Tivoli Mailing List

By John | November 25, 2007

I actually posted this question a few weeks ago on the Tivoli mailing list. Here is a clip of some of the responses from that thread.

bar-b-q.jpg

If there’s one thing that’s annoyed me the most in environments where I’ve worked, it’s been the ITIL-like processes. Sure, the purpose is to make sure you don’t bring the system down and miss SLAs, but sometimes management’s religious fervor can be absurd. Even when I’m annoyed, I still believe ITIL or ITIL-like processes should be here to stay, but management should be more educated on what constitutes a serious change to the environment. There have been times when I’ve developed custom automations that required minor changes such as a path to a file or directory which caused management to want a whole battery of tests to be performed. In those scenarios, I could’ve made money betting with management that the tools would’ve passed 100% of the tests.

——————————————————————————————————————————————————

I couldn’t care less about ITIL … to me it’s merely another $6 buzzword /
management fad … it will be replaced in a few years just like preceding
management fads that have been superseded by ITIL.

Being a technical implementer and not a strategy
consultant/management/software sales person, ITIL has minimal impact on my
day to day work style, it simply adds work — bureaucracy for the most part,
but not nearly so much as ISO 9K or SOX ;-) . From my perspective, ITIL has
merely changed the shape of the hoops I jump through, and not the fact that
there are hoops. The hoops have always been there, they probably will always
be there.

I am curious, however, about many things alluded to in your blog as well as
in that Gartner report linked from your blog, John. The Gartner article was
very … um, fluffy … no hard facts or technical details. This Googlish &
Amazonian style of IT … does this ‘cheap hardware, free software, damn the
torpedoes’ methodology encompass those companies’
back-office systems as well? I mean, okay, I can grok the warehouse full of
linux boxes running a massive transactional application & web service, but
does that translate to, for example, corporate financial processing? I
mean, even in a big company, you don’t need that much sheer front-end IO
capabilities to do your accounting … seems to me on the back end you need
reliability and security more than anything else. HR systems? R&D?
Manufacturers need design programs, they need business resource management
programs. Distributors need logistics planning & management programs …
warehouse management, inventory manageme! nt … retailers need supply
ordering, stock replenishment tools. On and on the list goes. But all this
back office stuff — very important to have a very sound Information
Technology solution, yes? At least that’s the way I see it …

So … how does this ‘new’ IT (or IT-less) paradigm fit in that picture? How
does it work?
Can it work? I’m not really getting it … and that Gartner article did not
provide any details (I presume details are reserved for the paying
customers, of which I am not one ;-) .

Anyhow, John, keep up the good work — I like rabble-rousers ;-)

—————————————————————————–

John G.  you hit some good points.  But ITIL is not a fad.  It is the
formalization of a lot of standards/practices. It will be of greater
importance from a competitive standpoint as time goes on. ITIL has been
around for many years and is widely adopted in Europe where it was hatched.
Ford has strict requirements on bids for it's contracts with suppliers.
Those requirements call for ISO certifications. Yes, I know these are
manufacturing certifications, but how long will it take for companies to
require it's IT providers/outsources provide ISO type credentials that state
with certainty how data is processed, secured, backed-up, audited etc  let
alone, how incidents are handled. Basically the time will come (5-10 years)
when those lacking ISO for IT / ITIL like certification need not bid/apply.
Most shops already perform many of the processes and don't recognize how
close they are to being ITIL compliant already. It is often a case of
refining a few things.
These refinements will generally make SOX, ISO 9k etc much easier to comply
with as well.
Regards,

————————————————————————————————————————–

I raised the question because of what I see going on in a small part of IT right now (i.e., Google, Amazon, …). There is a game a foot that is turning what we have been doing for xx years into a true commodity. Will this scale in the enterprise as we know it (not likely for quite a few years). Jason is exactly right you need process. Will it be ITIL or something else, who knows…

However, if centers like Amazon and Google start proving commodity like savings for computing services I believe the large Enterprise CTO’s will be forced to look at this model. I was talking to a data center eng for a local Atlanta company a few nights ago and he told me they built a data-center in Austin that was suppose to have enough capacity till 2012. They are going to run out of power in about 1.5 years from now. One of their options is to re-build a new data center.

Just my 2 cents

John

———————————————————————————————————————

Gartner 1996……”by 2002 80% of desktop devices will be replaced with thin client technologies” (probability 80%).

There is the Gartner world, then there is the real world.

ITIL…..ISO9000…..ITSM……the fads come and go. Enterprise Management challenges stay the same…….

Regards, Len

————————————————————————————————————————

 :-D   Beautiful, Len!!!
>
> I bet it would be a hoot to go through archives of Gartner's (and others)
> predictions from a decade or so ago ... count up how many came to fruition
> and how many were total bombs.
>
> regards,

————————————————————————————————————————

ITIL is likely another fad, but the underlying concept of change
control, monitoring, etc were there before and will continue to be
around tomorrow.

What I like about the Google model isn't that it does away with
monitoring, but rather that it turns all outages into routine
maintenance.  And rather than doing this with expensive hardware, it's
done as a api framework that all the software is designed to reuse.

Tivoli showed us both the power and problem of a solid framework.
Mainly that there's many advantages of reusing authentication, gui's,
tasks, a database, etc across multiple products, but no one solved the
problem of how to expand beyond the original corba design.  They never
really handled fault tolerance like Google has.  But then Google built
their api's to handle infrequent writes to a distributed db and lots of
distributed reads, not a design that you see in most corporate
environments.

Of course, at a high level, we still have a lot to gain by designing our
software to expect and continue seamlessly on an outage.  That is best
done with some kind of api that allows lots of vendors to make use of an
easily managed farm, think utility computing if it were done right.  In
the mean time, we have virtualization and vm farms to get us halfway
there.

————————————————————————————————————————-

A less known fact about Gartner predictions is that the methodology is based on
extrapolation of purchasing intentions by early adopters (50% probability).
Since there is no downside to saying 'our architecture will be 80% SOA in 5 years', I
wouldn't blame analyst firms too much. Anyone feels the heat of mortgage lending meltdown
yet? I predict a new wave of cost-cutting, cost-saving and budget-squeezing acronyms in 2008
(50% probability). I also predict that SAP will buy its way into systems management space
(50% probability). Finally, I predict that ITM6 EIB will be replaced by an SQL-compliant
embedded database system by 2010 (50% probability).

—————————————————————————————————————————-

How about IBM buys another monitoring company to replace the EIB by 2010 (50% probability)

—————————————————————————————————————————

I think that probability should be a little higher given past indicators :)

————————————————————————————————————————–

About 10 years ago Larry Ellison gave a presentation at some software executive briefing/conference and he told the audience that his cost of development went from 39% to 17% by off shoring his developers. Software executives from other companies ran like gazelles to the phones to instruct their managers to start off shoring development to India.

The net-net is that once someone finally leaks how much it is costing Amazon and Google to run 1 million core data centers the CIO’s will be running like gazelles to managers to get answers. Worse than that, picture that mad money idiot on MSNBC doing a report on why stockholders should buy shares in company x vs. company y due to infrastructure costs.

One last story, if you have not heard of the story of iLike.com. iLike.com decided to port their interface to Facebook and in one 24 hour period they had 50k new registered users. Can any of the brick and mortar fortune 500 company boast of a story like that? IMHO, the answer is no. Would a large bank like to get 50k new registered users in one day, IMHO, the answer is yes. Why can’t they? – Because their infrastructure may not be agile. These are the kind of things that are going to change our lives over the next 10 years, IMHO…

John Willis

Compnay : Zabovo.com

Blog : johnmwillis.com

Training :training.xtrain.com

Facebook: botchagalupe

Twitter : botchagalupe

Dopplr : botchagalupe

———————————————————————————————————————————

You just hit on an article read today that said RedHat and Amazon are teaming up to host applications on Amazon’s infrastructure to recoup some of that cost, co-location goes full circle I guess – I put 50% probability that eBay will do the same thing. These two companies have a dynamic infrastructure – the biggest difference between them and someone like a large bank = no legacy systems, they were built from the ground up mostly on Linux based servers running Apache/Tomcat/MySQL.

———————————————————————————————————————-

Rob,

You and Jason definitely make sense, and I perhaps oversimplified things a bit when
I called ITIL a fad. The word "ITIL" is itself still a buzzword though. I work for the
biggest buzzword using organization on this planet, I know how to recognize them ;-) .  

When I refer to management fads, I'm thinking about all of the various business or IT
processes or methodologies over the years that have been touted by (specifically)
'management' or 'consultants' as being the 'way of the future'.  Potentially, all
of those processes or methodologies could have been perfectly sound and might have
resulted in improved efficiencies or increased ROI or some other benefit if they had
been implemented appropriately.  But the problem is that appropriate implementation
requires a sound understanding of the process/methodology in theory as well as
>>in practice<<, and in my experience any complex information theory will usually be
misunderstood and/or misapplied in most business environments.

I've seen quite a lot of instances of people making references to ITIL just to show
their street cred and get props ... kinda like somebody using 'l33t', 'h4x0r', 'pwnd'
and 'woot' to show how kewl they are in whatever technical pseudo-underground community
they have appended themselves onto [wanna-be hackers, online gamers, and linux
enthusiasts come to mind ;-)  ].  If you want to talk process improvement and standards,
I'm hip to that jive ;-)  ... I like process improvement and I like standards and I
dislike chaos. It just irks me that if somebody is trying to implement some processes
for a perfectly valid business or technical reason, they can't seem to get their heads
around the concept that process improvement and improving efficiency or productivity can
be done without constantly banging the buzzword drums. A couple/few years back,
I was on a team that went through an ISO9K implementation ... the corporate drones
who were foisting ISO on us (and we were an IT department, not a manufacturing shop)
had no viable understanding of IT ... they just said "here's the ISO rule, you will conform"
regardless of the facts in the real world that would get in the way. High-level people who
had no idea how IT worked at the front line set some high-level mandates that were
interpreted and re-interpreted at every level of management until it got down to the
front line and at that point the requirements made absolutely no sense at all. I just think
that management and politicians should not be involved in the creating and setting of
technical policies. Just to touch on the subject of documentation that Jason
brought up ... the flawed logic and inappropriate requirements mandated by nabobs
is the biggest bureaucratic boondoggle pain in my side.  I'll be blunt ... I don't need
to document how to delete an endpoint because it's in the stinkin framework manuals.
Over and over I have had to fight with the mandaters on the subject of "work instructions"
... I don't have or use any "WI" ... I manage a complex IT system from the standpoint that
I use my inherent intelligence to analyze and understand what I'm doing. If I need to learn
how to do something I don't already know, I find a resource to learn from, be that a manual
or a coworker or this listserve. Those same resources are available to anybody that wants
my job, so I am not irreplaceable. The biggest problem with WI and other documented processes
or procedures is that these things quickly become religion -- you get some bureaucrat that
"owns" documentation and it becomes a Scripture that cannot be changed or deviated from for
fear of eternal damnation, even though there are a million perfectly viable, sound, and feasible
ways to do any given task.  I don't do what I do the same way every time ... changing how I
do any given task from time to time helps keep my mind flexible -- I mean, how are we supposed
to get any process improvement or gain any efficiencies or increase in productivity if we are
bogged down by mandated 'single solution' work instructions and never try different approaches
and methods?  The whole notion that a technical job should be boiled down to a lockstep
procedure "step 1, step 2, step 3" written in 4th grade English is abhorrent to me.  

But that's just my opinion ... ;-) .

John Guad.
-----------------------------------------------------------------------------------
 I think in our LatinAmerican World, its very important to follow standards or to have a mandatory
 that you must have to follow a rule o process, because we have a lot of CAOS, and at this time I
think that ITIL is just and interesting way to do something, the problem is that no all but the a
lot that I know are ITIL Consultant that are certified in ITIL but they never work in real life,
they only know what the ITIL books said, but they dont have the understanding of but have to do
to make something to work fine. Thats is my point of view of a problem that we have with standard
here they become a business that are full of pirates.-
--------------------------------------------------------------------------------------------
I believe IT standards are important and if you truly understand ITIL and have practiced ITIL
in a real life scenario you also understand the importance change has to the IT process.
Without a process like ITIL you may find yourself resisting to change without noticing the
business urgency. ITIL is not a step1, step2, etc process it supports IT changes needed by the
business. 

My 2 cents...
--------------------------------------------------------------------------------------------
I must quote Mick Jagger here:    "Let it bleed!"
--------------------------------------------------------------------------------------------
The Helpdesk is NOT an ITIL process, it is merely a function.
The CAB (Change Advisory Board) may well be considered the 'god'
that relies on the data in the CMDB (or IBM's CCMDB) when making decisions
about changes, their impact etc.   http://en.wikipedia.org/wiki/Information_Technology_Infrastructure_Library 

	http://www.itsmwatch.com/itil/article.php/3707341 

	ITIL v2’s Service Support book identified the service  desk (HelpDesk) as the lone function (group of people
	and the tools they use to carry out one or more processes  or activities).   ITIL v3 now identifies three other  functions in addition  to the service desk: Technology  Management,  ITOperations Management, and Application  Management.	

ITIL processes incorporate these functions into a consistent
approach to providing IT services.
Regards,

———————————————————————————————————————————————-

I agree with this sentiment. ITIL is like religion - open to interpretation,
with ordinary followers, fanatics as well as detractors. Rather than
meaningless terms such as "ITIL compliant", as you say, we need agreed upon
"protocols" between various layers in the ESM stack. Vendor products that
don't support the protocols won't work in an Enterprise and those that do,
integrate with minimal effort.

——————————————————————————————————————————————

Topics: itil, tivoli | 7 Comments »

7 Responses to “Does ITIL Really Matter – The Debate on the Tivoli Mailing List”

  1. People Over Process » links for 2007-11-29 Says:
    November 29th, 2007 at 2:19 am

    [...] Does ITIL Really Matter – The Debate on the Tivoli Mailing List Whole lotta email-list talk on ITIL. (tags: itmanagement itil email) [...]

  2. Berkay Says:
    November 30th, 2007 at 8:43 am

    Hey John,
    the new blog looks good but you broke all the old links :)
    I’ve updated the link to this one.

    Take care,

  3. Danielle Says:
    December 7th, 2007 at 2:47 pm

    You guys are all missing the point and furthermore so does most of the IT industry. Unfortunately, over the last few years, some of your criticisms of ITIL based Service Management have been proven correct. Although the ITIL Library has been around for over 20 years, it has become a buzz word, commodity and a fad. This is especially true in the U.S. The concepts, initially developed from the best examples taken from practice, have been polluted by those who know little or nothing about management consulting. This has made it extremely difficult to distinguish between the chatter surrounding the real value of adopting ITIL and that around ITIL for the sake of ITIL.

    What I’ve noticed is that many organizations opt for the second (ITIL for the sake of ITIL)forgetting that ITIL is not an end in and of itself. Rather, it is a means to an end. Therefore, any organization that attempts to utilize the best practice guidance outlined in the library will ultimately be unsuccessful in delivering any real value. ITIL initiatives must be linked to clearly defined organizational objectives and all measurements must reflect the progress, or lack thereof, toward meeting these stated goals. This is a point that is lost on both the critics and cheerleaders of Service Management.

    Additionally, it cannot be rolled out in a vacuum. Any undertaking related to changing the way work is done within an organization (ITIL fits into this category) has to have the participation from multiple departments and divisions including, HR, Business Units, IT Operations, Support Oranizations, Procurement, Suppliers, Development and User groups.

    In short, (and I could go on), achieving the benefits of implementing IT Service Management requires a level of planning and investment that most organizations either don’t understand or are not seriously interested in making. And why should they when it is so much easier to just publicly say that your organization has “implemented ITIL?”

    The danger in this is reflected in the comments posted on this site and is being reverberated in IT organizations across the US.

  4. People Over Process » IT Management Podcast #001 - barcampESM, Monitoring, The Cloud, 2008 Predictions, and more Says:
    January 15th, 2008 at 2:38 pm

    [...] in the new OMC around the topic, I ask John to tell us about the continued discussion around his does ITIL matter? discussion in the Tivoli mailing list and [...]

  5. Don Schueler Says:
    July 19th, 2008 at 9:13 am

    Having ITIL processes in place is better than not having them in place. That being said, ITIL is really targeted to the internal data center support world. It is not really robust enough to address the world of “vendor support” since it is really a general framework and does not bring the real world best practices into the equation. (I will admit to having a tainted viewpoint on this since I do work with Service Strategies, owner of the SCP standard that was built by the major software and hardware vendors out there…but I have also run major service organizations and have been involved in ISO, Baldridge etc. etc. and they all had this fundamental flaw that makes it harder to implement and of a lesser real value. But again, something is better than nothing!)

  6. John Says:
    July 19th, 2008 at 10:51 am

    Don,

    I agree having something as oppressed to nothing is a good thing. I originally posted this question on a very popular IBM IT management forum called the Tivoli TME. These were the responses I received from the question. Most of the companies that responded are fortune 1k.

    Thanks
    John

  7. Matt Says:
    June 10th, 2009 at 12:16 pm

    Good process management is realistically the only way to manage IT well or any company doing anything. ITIL is one method, there are many others and ITIL has huge flaws which make me question who the hell is feeding into its development and has done over the last 30 years. I believe its lack of perfection is down to 2 areas. Firstly its a money making setup, the courses are way too expensive for what is basic common sense. Secondly all the people I’ve met in ITSMF forums etc are just schmoozy management and the whole thing is a ‘club’ this results in little change and poor links to the real world practices. All this said it annoys me just as much as those opposed to it. Its just a framework, processes aren’t bureaucratic, they’re how you operate. If you find your companies processes bureaucratic then they’re badly designed. At least ITIL gives us a common framework and terminology. If you understand ITIL and IT and Processes. getting ISO9000 implemented which covers the ITIL processes is a far better and more meaningful way to go and you’ll reap far more benefits, that said it is far harder and more work.

Comments