Building an Effective Information Security Policy Architecture 70
Ben Rothke writes "Security
policies are like fiber, that is, the kind you
eat. Everyone agrees that
fiber is good for you, but no one really wants to eat
it. So too with information
security policies. They are
sorely needed, but most users don't go out of their way to comply with
them. And in many firms, they
are not even trained in what they have to do.
But
failure to
have adequate information security policies can lead to myriad risks for an
organization." Keep reading for the rest of Ben's review.
For
the sake of a basic definition, a policy is a formal, brief, and high-level
statement or plan that embraces an organization's general beliefs, goals,
objectives, and acceptable procedures for a specified subject
area. The purpose of
information security is to protect an organization's resources. The
cornerstone of any information security strategy is a robust set of policies,
procedures, standards and guidelines.
| Building an Effective Information Security Policy Architecture | |
| author | Sandy Bacik |
| pages | 340 |
| publisher | CRC |
| rating | 8 |
| reviewer | Ben Rothke |
| ISBN | 978-1420059052 |
| summary | Good book for information security policy development |
There are many reasons what information security policies are needed. Some of the most imperative reasons are:
- To inform users of their information protection duties
- Advise them what they can and cannot do with respect to sensitive information.
- Define how users are permitted to represent the organization, what they may disclose publicly, and how they may use organizational computer resources for personal purposes.
- To clearly define protective measures for these special information assets. The existence of a policy may be a decisive factor in a court of law, showing that the organization took steps to protect its intellectual property.
- Define both acceptable and unacceptable behavior. For example, spending a lot of time surfing the web and downloading videos off the net are both generally unacceptable.
- Policies are needed to establish the basis for disciplinary action, up to and including termination.
Building an Effective Information Security Policy Architecture does a good job of showing the reader how to start from scratch and build their security policy infrastructure. The book starts off at a high-level about the need for policies, and then goes into details on how to develop, write and sell these policies to management.
The book is a good guide to the entire policy lifecycle, and how to use various means to get to the ultimate goal. At 340 pages, the first ten chapters comprise 155 pages and deal with creating the policy infrastructure, communicating with management, and putting the entire policy puzzle together. The final 185 pages comprise 21 appendices of various examples of different policies.
A most significant downside and frustrating part to the book is that there is no CD-ROM with it, or companion website in which to download and use the numerous policy and process examples. At $80.00, such an option should be de rigueur. The lack of electronic versions of the policies in a book such as this is senseless.
Also, this is the first technology book that I have ever seen that did not cite a single reference. It is hard to imagine writing a book on this topic without using some sort of external reference. While the author may not want to quote sources, she should at least point the reader to other sources of information about security policies. Two notable and essential sources in the information security policy space are the SANS Institute — SANS Security Policy Project, which is free, and Information Security Policies Made Easy from Information Shield, Inc., which is $795.00, but worth every penny for a serious security policy effort. Full disclosure: I am on the Information Shield Expert Panel, but get no financial incentives or compensation.
Overall, Building an Effective Information Security Policy Architecture is a good resource to use if you are tasked to create or modify your organizations set of information security policies. The book will likely find itself on the desk of many information security professionals.
While it is frustrating that the book makes you reinvent the wheel by not having electronic versions of the polices, its value still can't be underestimated. Let's hope future versions of the book will fix that anomaly.
Ben Rothke is the author of Computer Security: 20 Things Every Employee Should Know.
You can purchase Building an Effective Information Security Policy Architecture from amazon.com. Slashdot welcomes readers' book reviews -- to see your own review here, read the book review guidelines, then visit the submission page.
No references? (Score:2)
Re: (Score:1, Offtopic)
Re: (Score:1, Offtopic)
Re: (Score:1)
Preparation:
Mix everything except the instant coffee together in a blender. Serve topped with a sprinkle of coffee crystals.
I don't usuall post off-topic, but this one was too good to pass up. Courtesy of About.com [about.com].
Re: (Score:1)
Re: (Score:1, Offtopic)
Re: (Score:1, Offtopic)
Re: (Score:1, Offtopic)
Tight, but effective.... (Score:3, Interesting)
2. Deny everything else.
Comment removed (Score:4, Insightful)
Re:Tight, but effective.... (Score:5, Insightful)
The problem with the "block known bad things" approach is that there are a lot of unknown bad things. It is far easier to profile for, and allow "known good things."
Watching all traffic for anomalies is a joke. No one has figured out how to do it yet and they have been chasing that goal for a decade at least. I have seen countless demos of "network anomaly detectors" that have all failed. Anomaly detection probably requires AI to work.
Given the technology available today the only effective technical controls we have to enforce an IT security policy is a default deny policy.
Re: (Score:2, Insightful)
that is sooooo stupid.
Not every policy violation deserves the worker to be fired.
What's next? Kill the jaywalkers?
Re: (Score:2)
That's great, until you take into account that slowing down a senior technical person by an hour on some project costs the company hundreds of dollars every time it happens.
Assuming that you're talking about network security for a business, security policy is a business decision. It's a business decision that requires extensive technical expertise to properly make, but it's really a tradeoff between various scenarios where the company loses money. A security breach isn't simply "we lose, game over" - it's
Re: (Score:2)
That's oversimplifying in the other direction.
If there are real security risks, the company is going to lose money either way. The question is very simple: What policy will lose the company the *least* amount of money. That probably means a security policy that occasionally inconveniences people, but not enough to be more costly than the security risks it reduces.
Re: (Score:1)
Re:Tight ... I was going to reply to Bishop .... (Score:3, Insightful)
THE MISSION IS ALL! Security that prevents mission/CoreBiz+ performance is more harmful than valuable to the mission. However, probable mission success without some respectable and reasonable degree of security can be problematic.
Don't let security stop reality. Keep security in perspective and segment the critical (plans, G2, accounts
Re: (Score:1)
yeah, but they stay in that state for about a week
And IT data leakage shows there is too much (Score:1)
Big dirty mess at the end, butt no.body wants to clean up the mess...
Now, IT are looking at "data retention" policies to wipe up the mess before it sticks around too long..., and avoiding legal anu-tubal ligat... umm, litigation...
(see InformationWeek June 9, 2008, p 27...)
First and most important rule (Score:5, Insightful)
Trust me on this one, my company tried to hot dog this ourselves twice and we failed both times. It wasn't until we brought someone else in that we ended up with a good working policy that really worked.
Some people will get their egos dinged and feelings hurt in the process (including some near the top), but a VP's indigestion is far more manageable than a massive level I breach. This is especially true if your company handles anyone else's financial or personal data for a living.
Comment removed (Score:5, Insightful)
Re: (Score:2)
Outsiders are an untouchable extension (Score:1)
I second this, only because outsiders are fairly immune to inner-office politics and squabbles,
Which is usually why they are brought in.
so their recommendations are usually much more "pure" than from people already in the company that don't want to piss off certain people or don't want to anger others.
Wrong. Outsiders are an untouchable extension of the party or interest that called them in. They're more often than not used as a tool or weapon against opponents or to rent loyalty. As long as the outsider is fully dependent, they'll be loyal. So stocking a project or team with rented outsiders is a way of buying control.
Case in point: I've seen several occasions where teams spent a few months investigating and making detailed recommendations, only to be
Re: (Score:1)
works all the time.
Re: (Score:1)
it ain't that simply.
in fact, that is the worst approach.
Colon Blow? (Score:1, Offtopic)
shouldn't a policy most importantly.... (Score:1)
Slashvertisement (Score:3, Insightful)
You can't identify sensitive information assets because there's just too much data and no one can agree on what's sensitive and what's not. You shouldn't bother telling uses what they can and can't do because they aren't paying attention and even if they are, when a situation comes up where they should actually be using that info, they've forgotten it. And users who can't figure out on their own that surfing the net for pr0n on company time is unacceptable will probably do it anyway.
The only thing he got right is "Policies are needed to establish the basis for disciplinary action, up to and including termination." But it's an excuse for firing someone you probably didn't like anyway. If they're actually a valuable employee, you'll probably have to overlook whatever they did.
Re: (Score:1)
Re: (Score:2)
This statement really goes to the heart of the matter. If it's encrypted at all times, no one can every even read that data. Obviously it MUST be decrypted at some point for use, but that doesn't stop many companies from having some vague and bonehead statement like "data must remain encrypted at all times." Are they just stupid? Maybe, but sometimes I bet such as a statement is intent
Re: (Score:2)
Re: (Score:2)
One of the big points is that if no one adheres to your security policy, it is useless.
Hence, you need to design a security policy that users will respect and obey.
Re: (Score:1)
Re: (Score:2)
Security policies (Score:4, Insightful)
Written security policies are just plain indispensable if you're covered by PCI/DSS or HIPAA, since both standards require them. They also give you a way to do knowledge transfer: before a written policy, the technogeeks know not to download free toolbars, after a written policy everyone does.
Anything good has a policy underpinning it. Are the backup tapes encrypted? If so, it's because there was a policy decision to encrypt them, even if that decision was made by an empowered IT person rather than a suit or a consultant.
>You can't identify sensitive information assets because there's just too much data and no one can agree on what's sensitive and what's not.
You can identify enough to be useful. Customer credit card information and health records are things people can agree on, especially when external forces require them to. Protect the things you know are sensitive, and you can reduce the risk of something damaging or embarrassing happening, and reducing the risk is all you can hope for anyway.
Iso 27001 and 27002 (Score:1)
Standards have been made after a long process of elaboration to be able to cope with every possible situations.
More informations on wikipedia : http://en.wikipedia.org/wiki/ISO_27001 [wikipedia.org]
And here : http://www.27000.org/index.htm [27000.org]
NB : allas, the documention don't seem free
Re: (Score:1)
very different.
CIO and CSO (Score:1, Insightful)
Re: (Score:2)
Maybe a stock investment rule can be found surveying corporations' C*O positions and quantity (CEO, CIO, CTO, CSO
Re: (Score:1)
Not having a function responsible for information security separate from the CIO is, unfortunately, generally a conflict of interest. Too many CIOs focus only on the availability aspect of security.
Re: (Score:2)
When I saw CSO, I chuckled, reminds me of legacy viewgraph [powerpoint] engineers
So many can screw-up, all their lives, and get very well paid. We all know that worker-bees, pack-mules, liberal-public-heathens, and unexpected (totally surprising) economic conditions are the most believable cause of business failures, a
Re: (Score:1)
what's really needed .. (Score:2)
No amount of security policies is going to protect the 'computer', unless and until they can come up with a design that don't get 'infected' merely by clicking on a URL or opening an email attachment.
Unfair punishment by Federal Trade Commission (Score:2, Insightful)
please read my book .. (Score:2)
* Phishing and spyware
* Identity theft
* Workplace access
* Passwords
* Viruses and malware
* Remote access
* E-mail
* Web surfing and Internet use
* Instant messaging
* Personal firewalls and patches
* Hand-held devices
* Data backup
* Management of sensitive information
* Social engineering tactics
* Use of corporate resources
- unquote -
If they let their own IT staff get on with the job, instead of ordeing in the latest innovative fad, then we wouldn't even need a security poli
Re: (Score:1)
but then again, what is the post of your posting?
supposed to mean what??>>>
Re: please read my book .. (Score:2)
My point is that none of these would be a problem if they innovated a computer thaat was secure, by default, without the poor sufering end user having to: Take steps to protect against phishing and spyware, identity theft, viruses and malware, e-mail, web surfing and Internet use, instant messaging. Add a 'personal firewall' which is next to useless, endlessly downoad and install patches that break something
C [amazon.com]
Re: (Score:1)
Re: (Score:2)
No, you have to make it so as the end user don't have to do anything, like verifying a color coded URL or verifying an email is really from the sender, without opening it first
Re: (Score:1)
but, this will neva eva happen.
no one can build such a system that is secure and the world will accept.
imaging a real secure Operating system. u think the nascar idiots who use PC's could figure out how to use security?
So Dilbert.com is a security risk? (Score:2, Insightful)
"Define both acceptable and unacceptable behavior. For example, spending a lot of time surfing the web and downloading videos off the net are both generally unacceptable."
Acceptable behavior in the workplace should certainly be codified but by and large it is not a security issue. Gawking at YouTube videos all day is counter productive and probably not what your employers had in mind when they hired you. But it is not a security issue, it's a peformance issue and should be dealt with acco
Re: (Score:2)
Been there done that (Score:1)
When you leave that company do a replace on the company name and make it the official policy of the new company.
In case you get an audit match the auditors requirements with your security policy and enhance it where it lacks using the format of the template you brought along.
During the time of the audit have signs up in the office, revoke the CEO's and any other big shots/pain in the ass user's s
Re: (Score:2)
CRC v. Mathworld (Score:2)
I wouldn't buy a book from CRC personally: CRC Lawsuit Frequently Asked Questions [wolfram.com]
Re: (Score:2, Funny)
like no other publisher is under litigation.
with this logic, you could never buy another book, ever!
Effective Security Policies (Score:1)
Well-defined and documented security policies are practically useless if they are not technically enforced.
IT departments need to stop faulting the user as "They forgot X when handling information of Y type" because not every user who will be handling sensitive data will be capable of remembering and understanding how sensitive data should be handled. For example, do you think every person you talk to at a call-center at an health insurance company or hospital is a technical person, or are they a low-i
Re: (Score:1)
Re: (Score:1)
Re: (Score:1)
Re: (Score:1)
actually, they are 100% useless is there aint no enforcement.