Slashdot Mirror
Why Upper Management Doesn't "Get" IT Security
Schneier is reporting that the Department of Homeland Security has decided to delve into why upper management doesn't "get" IT security threats. The results aren't terribly surprising to those in the trenches, stating that most executives view security as something akin to facilities management. "Thankfully", the $495 report (if you aren't a "Conference Board associate") helps tell you how to handle the situation.
Upper management would get it but they send the auditors to talk to middle management who doesn't get it. As such auditors decide that a company needs X because garbage in is garbage out.
Many of the upper management people I talk to know more about what we should be doing compared to what we are doing. The problem they have in overriding the auditors is the threat of the government and the shareholders. If they take the safe route the keep their jobs and stay out of jail. Actually the fear of the government is far worse that fearing the shareholders. (thanks to wonderful overreactions by Congress we get even more doing a whole lotta about nothing that ends up preventing us from doing what we should)
* Winners compare their achievements to their goals, losers compare theirs to that of others.
Most upper mangament in my view came into the field in the 70-80s and as long as it donst bother them, they dont care, so why should they care about IT in the first place! they think every thing will be fine as the IT sysadmin will take care of it
From the part-of-your-job-to-explain-it-in-their-terms dept.
Lets try this. When you forget to lock your Lexus and it's not there when you are ready to go golfing, that sucks. Almost as much as when you go to use the server and some hackers are using it to joy ride the net and sell all your customer records while you are liable. But unlike the car, where you can buy a new one, it's a pain in the ass to buy a new company image.
Of course CEO's don't want to spend a lot of money and time on security. Unless the company makes security software or hardware, it IS an expense. Computer security should be handled with the same priority as physical security (keeping facilities secure) and basic infrastructure (power, water, telephone, etc.). Any CEO that spends an inordinate amount of time on computer security will, and should be fired. Just because you, as an IT person, spends all day reading about security threats, does not mean that upper management should do the same. A good top level manager understands priorities, and handles them accordingly. IT security should be handled as an absolute requirement to run the business (like power and water), but should be handled with the minimum possible expense, since it does not generate any income.
As a manager, you have to understand that EVERYBODY is screaming at you about their particular area. The marketing people need a bigger budget. The maintenance people are wanting to upgrade this and that. The transportation people need new trucks. That's their job. It's a top manager's job to look at each of these recommendations, and prioritize them in a way that will do the best for the company.
Seems to me like this blog entry is just another example of IT people being too myopic to get any real handle on how a business is run. In case anybody is scratching their heads as to why IT people rarely climb up the executive ranks to manage large companies, this example illustrates that reason very well. (Usually, in large companies, the people running the show are from marketing or finance. Occasionally operations. Never from IT.)
1) Explain the effects of a DOS attack by shutting off power to the beancounters' servers.
2) Simulate the effects of spyware by displaying the contents of the PHB's um...photo collection along with his browsing history.
3) Demonstrate the impact of weak passwords by logging in as the PHB and sending off a few colorful resignation letters to the CEO on his behalf.
4) Emphasize the importance of reliable nightly backups by indiscriminately doing rm -rf everywhere. (you ARE root, aren't you?)
5) Using the custodian's account, log in and download the entire customer database into your ipod, load it onto an independent laptop, and use the data to e-mail oodles of spam.
Or you can just tell them the risk factors in which case they'll just stand in front of the swiss cheese and sing of how all the holes are theoretical.
I think management think if you spend the money and take the time to release a secure product .. you get behind, have a more expensive product, and lose in the market. Since it's enormously (and often infeasible) to certify a product as 100% secure .. where do you stop spending the money on security? If they waited for IE or Firefox to be 100% secure before ever releasing it .. we'd use other browsers (which may actually end up being either less secure or not as good).
.. i wont mention any products or websites that have had issues. But the point is, for all it's ranting about wanting security and reliability, it appears to me the market just doesnt forgive those who would spend the time and money on these things.
People have shown a willingness to put up with insecure half ass reliable products
And yes, this must change.
I guess I should have explained.
We, the taxpayers have paid for this paper, yet we also must pay for copies of the very document we paid for to begin with.
That's what I dont like. Akin to double-taxation.
(from the BuyMe screen liknked from schneider...)
survey by The Conference Board (sponsored by the U.S. Dept. of Homeland Security)
Why would they spend $500 on a report to help them get it?
"Thankfully", the $495 report (if you aren't a "Conference Board associate") helps tell you how to handle the situation.
Bruce isn't in the business for giving out his top notch observations for free.
Are any of us?
I'd say it's a pretty lame attack to point out the cost as a negative. Just admit that you're not interested in his opinion and move on.
IT security sucks for this very single reason: It takes effort.
The solution? Demand effort.
Tom
Someday, I'll have a real sig.
Slashdot Burying Stories About Slashdot Media Owned
The general problem with IT work is that if you do your job realy well, nothing happens. So you then have to deal with questions like "why did we spend all that money on y2k when nothing happened".
Its almost worth messing up from time to time just to show what would happen every day if you weren't there.
"Have you ever thought about just turning off the TV, sitting down with your kids, and hitting them?"
Upper management doesn't get IT because upper management doesn't get much of anything. They only see numbers, numbers they must play with until they add up to a plus mark.
Given all that we have been asked to give up in the name of security, the fact that this isn't free shows once again that Homeland Security is about money and power, not the well being of the citizens. Yes there is some private sector company involved, but if Homeland Security pays for it, then it should be a study done for the sake of, maybe National Security. And if that is the case then it should be distributed for free. More likely the case, that company is receiveing a return on a political favor (campain contributions)
We are all just people.
Don't try to talk ROI. You'll be talking to finance people who will see instantly that there's not enough data about quantitative risks to back up what you're saying.
Instead, calculate the cost of a breach. Then walk up the chain of command with the message "Like any risk, we can avoid it, mitigate it, transfer it to an insurance company, or accept it. If you do nothing you're accepting it. If you accept it then on the day a breach happens you will spend eleventy thousand dollars of company money. Do you have signing authority for eleventy thousand? If yes, here's the cost of a couple of mitigation options, and you're the boss. If no, you understand that I'm only going over your head because the decision has to be made at that level."
IT stuff is voodoo to most upper management, and I'm convinced IT shops get away with things they never would if the upper management understood IT as well as they understand, say, supply. I was upper management in two government organizations heavily dependent on IT. As a fairly competent computer user who likes to keep up with current events, I fought with our IT folks endlessly -- at least the management.
The first problem is IT quickly forgets that -- like everybody else except the people actually doing the core functions of the organization -- they are a support organization, not a control organization. They latch on to their ability to throw out security and voodoo computer terms to persuade the upper management to let them set policies. Upper management doesn't understand the policies at all, and often has no choice but to side with the IT pros no matter what the actual users want or need. As often as not, they then set policies that are purely for their convenience (for instance, wanting to standardize on Windows and a strict set of programs even though they support 25 or 30 different sections, some of which have been doing things like digital photography, desktop publishing and design on Macs for years). From the users' perspectives, IT makes using the actual IT resources as painful as possible to make their lives as simple as possible, and the fact that they're hampering actual mission accomplishment doesn't bother them.
Next, they have a sweet deal going where they set a bunch of standards that require certain certifications or skills, so they hire people who perpetuate those standards, and only buy things that are compatible with those standards. This then requires getting on an endless treadmill of more training, more personnel, more software, more hardware, etc. And all the while they make it clear that it's lunacy to buy anything that doesn't have vendor support because if it actually breaks they can't be expected to get it going again using only the training, hardware, software and people that they have brow beat management into paying for using money that *every other part of the organization* was crying for and could have put to good use, too.
Lastly, on a day-to-day basis, far too many of them think that, because they're IT, it's their right to be arrogant, socially or organizationally inept, or just plain weird -- and sometimes it's a combination, so you get a organizationally inept weird guy being arrogant. How many of those does it take to ruin a shop's reputation? (IT certainly has no corner on that market, I'll grant you).
I could go on here, but I'm sure I've pissed off enough people already. I came from the internal communications side of things -- journalism and later PR. In my field management always thinks they can do your job better than you can because, hey, it's just writing and talking. Eventually, I got promoted into management and in dealing with IT I saw that their best defense is that almost nobody in a position of leadership (being mostly older guys, half of whom had never launched a program that wasn't sold by Microsoft) understood what they hell IT did or what it took to get it done. So all it took was a good talker or somebody who learned to cite vague security mandates from higher headquarters to get much more of what they wanted than anybody else did.
Of course, it also left IT open to being weaker when their leadership was weaker (or less smooth). But I didn't run into that. I ran into IT shops that got more of their resource requests approved than anybody else, but didn't really realize it and kept whining for more even though their support curiously never got better no matter how much you spent on them. And for every new capability you read about on Slashdot, they came up with two new security policies that made using it impossible.
Now I'm back in the trenches and don't get to go to the meetings where the IT guys try to talk the boss into banning the USB drives everybody has taken to using because the e-mail
If only there where a set of colors we could code the threats by, then even the "upper" manager could understand.
You missed the point that the creation of the report (costs of writing it) might not have been completly covered by the grant. In fact it was probably put forward as a proposal this way: the govenment agency wanted a study done, and rather than paying a company the full price to do the work, they payed them half (or some other fraction), but at the end of the job the company gets to re-sell the report.
For the govenment department it costs less for the report they wanted. So they saved the taxpayers money.
How is that any different from getting a grant to write a book?
Sounds like a damn fine reason not to give people grants to write books then, unless they want to do so as U.S. Government employees, and allow the book to be a product of the United States Government (with their name on it, of course), and therefore in the Public Domain.
If public money is being used to fund the creation of something, the end product of that creation ought to be freely available to the public.
Do you think people would be quite so keen on funding the Smithsonian Institutions, if they charged admission fees? Probably not. I don't have any problem with the Smithsonian being publicly funded, in fact I think it's great; but making things halfway-publicly funded is just crappy, and generally gets the taxpayer less "bang for their buck" than if they just went all-in on half the number of projects, but funded them completely and 'owned' the results for the public, therefore making them free for anyone to enjoy.
"Ladies and gentlemen, my killbot features Lotus Notes and a machine gun. It is the finest available."
Then there are the good IT departments that get ignored by management.
For instance: The company I work for (and the reason I'm posting anonymously) is currently running our main website on a Windows server. From talking to our hosts, it seems that crypto is something the Windows world just doesn't do. By that I mean, we want to install new web software? (PHP stuff -- you know, new version of Drupal, Wordpress, whatever.) We can either pay them $75/hour or so, or do it ourselves, over FTP. Plain-fucking-text FTP.
It's a small company, so it didn't work when I was the only one who said anything about it. Then I got someone else to say something about it, and now we're actually talking about possible solutions. It's still not a priority, which is kind of understandable. But when someone sniffs our FTP password -- hell, when they simply hijack our connection, casually, really -- they will have credit card numbers for all of our customers.
Well said about the USB drives, but that is why I actually explain IT issues to management. I make it as simple as I just made it for you. It means I won't be able to arbitrate inane stuff simply to get more power, like you're describing -- in fact, I'm one of two people who work on a Mac; everyone else is on Windows, and I run some Linux servers to test on. But it also means that I will be able to get us the level of security we really need, and that's got nothing to do with a lust for power -- it's no skin off my back as it is, I can even mount it with curlftpfs for easy syncing of my test server to the live Windows server.
Just realize: it goes both ways. Management should "get" IT, and keep them on a tight reign. But management should actually listen to IT. Unless I'm going to be fully autonomous, I'd much rather help my boss to "get" it, rather than simply figuring out the right concoction of buzzwords to stuff in his ear to get my way.