As a follow on to Cloud Camp Seattle,
I met two particularly distasteful people, Mr. Snake Oil Salesperson
and Mr. Know it ALL. Both are doing a disservice to everyone, and here
is why these people should not be involved in your cloud computing
project.
Mr. Snake Oil Sales Person
This is a huge problem because companies are looking for products
that work, do what they are designed to do and solve a problem.
Unfortunately, there are a number of products out there that are not
suited or suitable for the cloud computing environment. They claim to
solve a problem, but in reality will cause more problems than they
solve. The process involves a sales person who knows all the right
buzzwords but does not understand the fundamental concepts behind them,
nor can show where their product solved a similar or same problem for
another company.
Mark Suster has a great piece about this: “Danger of Crocodile Sales”.
Cloud computing is rife with companies in Alpha, Limited Beta, Beta, or
selling product that simply is not up to the tasks in cloud computing
that companies are looking for to manage and scale systems. The
immature products that are out there right now will solve some
problems, but not some of the bigger problems out there for audit,
control, CIA (Confidentiality, Integrity, and Accessibility) policy
management, or even simple system administration. Right now we have a
pile of old data center model software that is being cobbled together
to putatively support the cloud. This is a bad thing for companies who
are looking to purchase software to manage their cloud environments.
These people are more of a danger to the cloud computing project for a company than a help. These people are a danger to any project being undertaken by a company.
The reason for being dangerous is that they are going to sell a product
or service that is not suited for what the company is doing. When
things start to fall apart the sales person will make wild offers to
“customize it for you” when really what they are doing is using you as
a learning experience to make their product better, or even just make a
product for the cloud environment. You are paying them to develop a
product for you that they are going to turn around and revise to sell
to other folks. There are better ways to do product development, and
most companies are not “software development test beds” for software
manufacturers.
Mr. Know it ALL
These folks are even more dangerous than Snake Oil Sales People
because no one knows it ALL, and certainly not in cloud computing. They
tend to try to dominate the discussion and tell people all sorts of
arcane and useless knowledge to show that they know it ALL. (The caps
are intentional here, I subscribe to the belief that no one ever knows
everything or that anyone is ever 100% right all the time, it is
statistically improbable that anyone knows it ALL or is right 100% of
the time). There are some very smart people out there, but Mr. Know it
ALL is busy dominating the discussion, talking about useless bits of
information that do not apply to the project, or otherwise obfuscating
what should really be going on with a cloud computing project in that
everyone in the organization will have a stake in this.
The reason for the danger is that they really do not know it ALL,
rather they rely on bluster and volume to make a point that could cost
the project a consensus on a truly important point. Like the workflow
for the Cloud Environment, the way to store crypto keys in a cloud
environment, how to ensure CIA, and other important things to discuss
and reach consensus on. Rather they are busy discussing how long it
took them (a year and a half) to discover the actual VM use in a cloud
computing environment and the pain and suffering it took to discover
this. How they slaved day and night because knowing the VM is an
important fact, it is something that we all should know and how dare
the cloud computing service company obfuscate that knowledge. (This is
an actual event, which was painful to sit through). They are not
helping; rather they are seeking to manipulate the needed and necessary
group consensus for a cloud computing project to their own ends, or to
inflate their own self-esteem and/or ego. This can come at a steep cost
to a project, as they will cause delays in implementation spending a
year and a half seeking to find the VM used by a service provider. No
project can afford this kind of delay or obfuscation of the truly
important part of getting a service or product into the hands of users
to make the company more competitive.
If you see any of these people in your project then the company or
project manager needs to nip this process in the bud. Unfortunately,
they are hard to spot as there is a value to them generally, but the
risks they bring to the project far outweigh any derived benefit that
they bring to the project. Companies still deal with a phenomenal rate of project failure,
these might be two of the causes, which are somewhat easy to eliminate
from a project. This is more of a cautionary note based on my
experience, but one that is well worth noting because in the end, a
company cannot afford the cost or time overruns that these two people
bring to the table.
(Cross-posted @ IT Toolbox)
Right on Spot Dan. But it is very difficult to eliminate these two causes. Especially when they are tuned up a lot. and realizing that client knows the minimum about the cloud.