Archive for January, 2007
The Symantec, Altiris deal is the first recognition of the fact that traditional anti-virus, anti-phisihing etc is a dying market. And this is not because Microsoft (MSFT) has decided to put some of the features in its newly launched Vista release.
Many corporations have realized that traditional anti-***** products stop the “known bad” stuff from entering their infrastructure. But it is much easier to keep track of “known good” stuff which provisioning systems like Altiris do. They make sure that only “known good” where good is defined by the corporate policy or by Dell for consumers is kept on the machines.
Solicore Systems is another example of this, where enterprise change control policy is used to ensure the known good state of the machine. If it came in via enterprise change control it is good, otherwise it is not.
If you can ensure that only known good stuff is on the machine, traditional anti-virus, anti-spam etc is dead.
January 30th, 2007
This is a great site for enterpreneurs and CEO’s who like to read.
800ceoread.com
January 27th, 2007
Dennis Drogseth is a vice president with Enterprise Management Associates, a leading analyst, market research and IT consulting firm based in Boulder, Colorado, focusing exclusively on all aspects of enterprise management.
This is an awesome article about startup mistakes (Link Here).
January 23rd, 2007
What Steve Jobs did at Macworld is a classic example of what CEO’s should not do. He raised expectations astronomically. (Mike Egan echoed the same sentiment in his article @ computer world). As a CEO there is immense pressure to sell the future. Steve Jobs has to figure out a way to sustain his growth rate for apple; while ipods are selling … the growth is slowing down. So he sold the future … before it happened. The future
almost always never happens the way you had expected.
Maybe this is just a show for the world and Steve has correctly set expectations at his board. But for us mere mortals, setting expectation is where you are seen as successful.
So why do CEO’s feel compelled to raise expectations:
- for most startup’s its because they want to increase the spend.
- feel like something will happen, so it wont be an issue
- over-confidence, in-experience, ego
- the board wants to believe it also, things will be great in the future
Are you setting the right expectations?
January 22nd, 2007
Peter Armstrong, Corporate Strategist BMC Software, talks about IT horor stories in IT World.
Here is Peter’s podcast.
January 21st, 2007
How to Blag an Interview
Every now and then as a technology journalist, you’ll find yourself agreeing to participate in what is referred to a “Press Tour”. Vendors do these once or twice a year, when they have a shiny new product to talk about.
This means you sit in a room with them for 30 minutes to an hour and talk.
Link to the post.
January 19th, 2007
… based on a true story at a Fortune 1000 company
From the wikipedia:
… a tax shelter is any organized program in which many individuals, rich or poor, participate to reduce their taxes due. However, a few individuals stretch the limits of legal interpretation of the income tax laws. While these actions may be within the boundary of legally accepted practice in physical form, these actions could be deemed to be conducted in bad faith. Tax shelters were intended to induce good behaviors from the masses, but at the same time caused a handful to act in the opposite manner. Tax shelters have therefore often shared an unsavory association with fraud.
In most organizations when you want to make a change you need to fill out a change request form. The change request form states what needs to be accomplished, but does not concern itself with how the change is to be carried out. It usually contains when the change should be made and to which machines in the infrastructure. The change requests then go to the change
Again from the wikipedia
Change requests typically originate from one of five sources: (i) problem reports that identify bugs that must be fixed, which forms the most common source, (ii) system enhancement requests from users, (iii) events in the development of other systems, (iv) changes in underlying structure and or standards (e.g. in software development this could be a new operating system) and (v) demands from senior management
Almost every IT administrator I have met hates to fill out requests for change. Several of them see this as a TAX that they have to fill in addition to working 24×7 to get the work done. Now most administrators are smart people and they either created or found their own tax shelter: the Emergency Change.
When a change is labeled as an emergency change most organizations allow all the procedures of filling out a change request and approval to be bypassed, hence avoiding the change management tax. Just like IRS’s tax shelter some of which are perfectly legal and have legitimate uses, the emergency change is required for the proper functioning of the infrastructure; but like several illegal tax shelters it can be used and abused.
Recently a large retail company had outsourced their e-commerce website to an managed service provider (like IBM SO, EDS, Verizon etc). The outsourced service provider had very strict change management procedure, which they had developed over the past several years to ensure highest availability and uptime on the revenue generating sites. But ofcourse there was the change management tax shelter: the emergency change. After about an year of signing the contract, one of the outsourcer’s executives reviewing the retailers account found that 70-80% of all changes presented by the team at the retailer were emergency changes with no documentation or approvals. This was clearly and abuse of the tax shelter.
He called up his counterpart at the retailer and explained his analysis and why their performance fell short of the SLA. The VP of e-commerce operations was a reasonable person and admitted that he was between a rock and a hard place. His development team had revolted that if they had to follow the process and pay the change management tax they don’t have enough time or resources to get he site up before the Christmas season and he had let his operations team use this tax shelter in the contract.
What should we do? They could not come up with a solution at that very moment. Now the person at the outsourcer wanted to help his customer the VP at the retailer. So he got his team together and said we have got to figure out a way to this problem, because we look bad in front of the customer by not meeting our SLA and we are loosing money on the account also.
His team worked hard and came up with requirements: What they really wanted was something that could let the customer just do the change and then pick up all the information for the change and create the necessary documentation. Thus from the customers perspective they got rid of the tax but from the outsourcers perspective they still got the documentation needed and if there were exceptions that would be flagged. If someone could figure out a way of doing that with the IRS we would all save tax and the government won’t go into deficit.
Once they knew what they wanted they began looking into the market for solutions. The first one they encountered was from Mercury (Kintana), now part of HP. This solution would fit well with the development system that the developers were using. But it only took care of changes from the development system; if someone used some other system it would not work.
Next they looked at Bladelogic & Opsware, both tools used for pushing out the changes. Now these tools could detect if a server had been changed and what the changes but they had no idea who made the emergency change, when it was made or how it was made. The integration with their change management system was also not easy. Both these systems could provide some of the documentation required for the process, but not all of it. Then they came across Tripwire. Tripwire could also be run periodically and it would detect what had changed, but it could not provide who made the emergency change (which was a problem, because it was important to show that the change had been made by the retailer) , also there was no SLA on when the change was made and when the documentation would be complete. The other problem they found with the above systems was that when they created tickets from change the volume of change was so high that the tickets created were meaningless.
Finally they looked at Solidcore. Solidcore could tell them what changes had been made, who (user) made them, what application was used to make the change, when the change was made. It could also connect back to the change ticketing system. Before the changes were put into the ticketing system, Solidcore clustered them to find units of change. This dramatically reduced the number of tickets created and also there was more meaning to each ticket.
The next morning the VP @ the retailer received an email: we have discovered a legal tax shelter for you: Solidcore and would like to discuss and mutually agree to discontinue the use of the illegal tax shelter: the emergency change.
January 18th, 2007
There is a battle brewing out there: google versus citrix. Sounds absurd, its not. Google wants to become the place where people store thier data, run applications from and do all their computing. That is exactly what citrix does for the enterprise today.
A number of engineering people I have talked to have mentioned that Google’s strength is the distributed application environment they have built which allows people to develop scalable new applications very quickly. The question is whether newer technologies erode the advantages that google has and almost make them like yesterday’s technology when it comes to applications over the web.
The generally held consensus is that Citrix’s current architecture would never scale to something serving millions of people around the world. But that is of their current architecture. The advantage that citrix like approach is that you can take existing applications and just plop them in and they work.
The first question is whether being able to deploy current applications for use over the web be a strong differentiator. I think it will be. The next question is whether it can give Yahoo, Google or Microsoft (or Akamai) a lead over the other. I again think the answer is yes. Microsoft’s live initiative is heading in that direction although it will run into resistance from the cash-cow products of the company.The other angle this competition could come from is from the people who are putting a box in your living room …
That brings us to the question whether a virtualization based approach will get us there. I recently came across a company called Atlantis Computing which has the technology working. They signed up for the Amazon’s grid and have signed up over 5000 users all over the world for their beta. So while Citrix may not get there … the approach will definitely and it will turn the advantages of today dis-advantages of tomorrow for companies like Google.
January 17th, 2007
Is there a simple ITIL philosophy that can be used as the guiding light for implementing ITIL. The general consensus about six-sigma was that if the top leaders in the organization understood that six-sigma was about “removing variability from a repeatable process“, and then they empowered their people to go achieve this the implementations were enormously successful. This was because the statement was actionable for people, it tied very closely to some metric they were evaluated on and it made intuitive sense. In organizations where they created a group (inside or consultants) to come up with the six-sigma plan, almost always resulted in failure.
Now as IT organizations look at ITIL, how could we describe it in a simple phrase like “removing variability from repeatable processes? What is the equivalent statement describing the ITIL philosophy?
When I searched on the web, here are some interesting possibilities:
- separate administrative tasks and technical tasks
- develop a common-lingo for communication about process and solutions
- automate repeatable process
- get budget from the CFO, wait for the next wave
- provide a service (like a resturant), not technology (cooked food).
What do you think?
January 17th, 2007
Vijay Anand (the force behind Proto) posted this note from a Demo Conference Veteran to help people have the maximum impact during their product presentations ( You can get the original posting from here).
———- Forwarded message ———-
From: Shel Israel
Date: Jan 14, 2007 9:43 PM
Subject: RE: Proto.in Presentation Tips
To: Vijay Anand
1. Be humble but proud. Remember the star of your presentation is the product not you. Your audience wants to see the technology. Spend as much time showing it as possible.
2. It is far more memorable to make one point very well, then to make several points.
3. The objective of a good presentation is to get the people who matter most in the audience to want to come up to you after your talk and say, “tell me more.”
4. Remember to always play nice. Your worst competitor in today’s conference is tomorrow’s business partner, employer or employee.
5. Speaking without hype, in a style that is authentic and in the tones you would use when meeting a new business partner.
6. If there are potential investors in the audience ignore them. They are going to follow the buzz you make with the technologists in the audience. Focus on the technologists.
7. If you are a star of the conference, remember that you just did well at one event and on one day. If you do poorly remember that you just had a bad day and the world has not collapsed. In either case, you will have learned a very valuable lesson to take with you moving forward.
8. You, your product and your company are being judged at every minute during the conference, not just when you are speaking from the dais. The conversations and networking are the most important part of any business networking event.
9. Do not exaggerate what your product can do. Do not be overly optimistic about when it will be ready.
10. Make it clear that you want as many partners as you can get.
–~–~———~–~—-~————~——-~–~—-~
January 14th, 2007
Previous Posts
