Slashdot Mirror
Oracle Exec Strikes Out At 'Patch' Mentality
An anonymous reader writes "C|Net has an article up discussing comments by Oracle's Chief Security Officer railing against the culture of patching that exists in the software industry." From the article: "Things are so bad in the software business that it has become 'a national security issue,' with regulation of the industry currently on the agenda, she said. 'I did an informal poll recently of chief security officers on the CSO Council, and a lot of them said they really thought the industry should be regulated,' she said, referring to the security think tank."
Oracle are (rightly or wrongly) worried about competition from Open Source. Regulation of the software industry would be a major benefit to them in this. Anyone who didn't meet the regulators' criteria couldn't compete.
Of course the "patch, patch, patch" business plan is bad for consumers. But in truth, most software companies don't care about consumers. They care about making money. As it happens, most people really don't care enough about the subject to make the companies change.
One of the examples in the article asks, "What if civil engineers built bridges the way developers write code?" and answers, "What would happen is that you would get the blue bridge of death appearing on your highway in the morning." The difference here, however, is that civil engineers couldn't get away with making rickety bridges. You would find public outcry if it broke while people were on the bridge. In the software world, however, they scream and the companies just fix it with a patch and it shuts the consumers up. Saves a lot of money and time in testing at companies.
Wow, really nice slice on the Brittish.. FTFA
She claimed that the British are particularly good at hacking as they have "the perfect temperament to be hackers--technically skilled, slightly disrespectful of authority, and just a touch of criminal behavior."
It seems to me that the F/OSS industry has shown that fast, and effective patches can be applied, and that software we pay for has less then reasonable responses to such threats. I use F/OSS and I'm quite happy with the response they have to software problems. I don't expect it to be of NASA quality, just to be good, and it is. For the amount that you have to pay for Oracle et al, you expect fast resonses on problems. The problem is that they don't respond fast enough. There is NO bullet proof software, though I give a hat nod to the guys that wrote the code for the Mars rovers. Certainly, Oracle isn't showing that they deserve the price they demand, at least not in this respect.
I might be off topic, but all the F/OSS that I use, delivers what I pay for AND MORE. The software that I have to pay for is lacking. When you pay thousands of dollars, you expect patches in a timely manner, and before you get hacked. I think this is a big reason that F/OSS will continue to win hearts and minds across the world. Despite the financial differences, F/OSS actually cares, or seems to, and they do fix things as soon as they find out, or so it seems to me. They have a reputation to uphold. Without it, they will just wither and die. It amazes me that investors, stock holders, and customers are willing to wait for the next over-hyped release of MS Windows while they suffer the "stones and arrows" of the current version. It appears that no matter how bad commercial software is, people rely on it. Yes, of course there is more to the equation than this simple comparison, but I think this is important. If you weigh what you get against what you pay, F/OSS is a good value. The argument is old, and worn, but ROI is a big deal, and patches make a difference to ROI.
Is it really what the software industry needs? A set of rules to make things bullet proof.. which of course won't ever happen. That kind of mindset is totally wrong, even though the sentiment is in the right place, you can't regulate quality in this regard. Sure, you can make sure that all gasoline is of a given quality, but I don't trust the government to test and regulate software. The US government already has a dismal record of keeping their own house in order on this account, I don't want them telling me how to do anything or what I can and cannot sell, never mind what I can give away for free under GPL.
Support NYCountryLawyer RIAA vs People
"I did an informal poll recently of chief security officers on the CSO Council, and a lot of them said they really thought the industry should be regulated,' she said, referring to the security think tank."
Funnily enough, I'm just now reading Darrell Huff's book, "How To Lie With Statistics".
The problems with her poll are manifold.
Firstly, her group is composed of securiy officers who are on the CSO Council; might their views differ from security officers not on the Council? perhaps tending to be more of the belong-to-an-organised body sort? might perhaps therefore be predisposed towards regulation?
Secondly, of the officers on the Council, which ones did she ask? all of them? or did she have a bias to tend to ask those she already knows will agree? perhaps those who found it rather boring and aren't quite so pro-organised bodies just don't turn up at the meetings.
Thirdly, what's her position in the organisation? if *she* askes the question, are people more likely to say "yes" than they would to another person?
Fourthly, are people inclined in this matter to say one thing and do another, anyway? e.g. if you do a survey asking how many people read trash tabloids and how many people read a decent newspaper, you find your survey telling you the decent newspaper should sell in the millions while the trash should sell in the thousands - and as we all know, it's the other way around!
Fifthly, even if the views of members of the CSO Council truely represent all security officers, and even if they were all polled, who is to say the view of high level security officers is not inherently biased in the first place, for example, towards regulation?
So what, at best, can her poll tell you? well, at best, it can tell you that chief security officers who regularly turn up at meetings will say to a particular pollster, for whatever reason, and there could be widely differing reasons, that they think regulation is a good idea.
Well, I have to say, that doesn't tell us very much, and that's even assuming the best case for some of the issues, which is highly unrealistic.
Whose patches are infamously known to break stuff, released in 6 month batches (maybe just a mite too spaced out?), and so infamously poor at actually patching their bugs that they currently have an open, publically known 0day with no patch, because they screwed up patching it last time and it's still open?
And they think security patches are a poor model?
Maybe that's why they put so little effort into them. Maybe that's because they put so little effort into them. Maybe some people think of it as bridge maintainance, and they want to build the bridge perfect every time? When they can't even get patches right when they have six months between them? Fat chance.
Honestly, out of the people in the software industry, even Microsoft do a better job, security-response-wise, than Oracle. And when you're behind Microsoft in that department, you've really got a problem.
They need to make a serious effort at security response and treat it like a real priority, not show-ponying about regulation when, if they were regulated, they would still be completely unable to respond, but would point to poorly-drafted regulation as "tying them up in red tape".
She claimed that the British are particularly good at hacking as they have "the perfect temperament to be hackers--technically skilled, slightly disrespectful of authority, and just a touch of criminal behavior."
Sums me up perfectly old boy (well maybe not the technically skilled part)
This infuriates me to no end, when people use references they saw on the back of a cereal box beacuse they thought it was cute. FTA:
"What if civil engineers built bridges the way developers write code?" she asked. "What would happen is that you would get the blue bridge of death appearing on your highway in the morning."
Im sorry, but there are crazy people scanning my highway for open ports and i dont see script kiddies pinging my roads. Graffati aside, they are left alone. Code that is written works just fine if people dont try to over flow buffers and install rootkits. The bridge I see out of my window is fine because people dont hit it with sledge hammers.
Just my 2 cents . . . .
Since when does being a Socialist mean 'someone who has a different opinion than me'?
As soon as the management starts to then so will I. Or did you think unrealistic deadlines and bad overall designs come from the grunts?
They are the company who have the worst user interface tools on the planet.
The GUI's would have sucked in the 1980's.
Every SQL statement was designed by a dfferent person, with a different syntax.
If the guy expects us to assume he is an authority on the subject, he should clean up his own rubbish first.
Open Source Drum Kit, LPLC deve board - mjhdesigns.com
The difference is that software is expected to be cheap, released fast, and to run on all kinds of platforms. Sorry, that leads to errors. You can have software that never needs patching, you just have to take some concessions:
1) Development cost will be a lot more. You are going to have to spend time doing some serious regression testing, besically testing every possible compination of states that can occur. May seem pointless, but it's gotta be done to gaurentee real reliability.
2) Development time will be a lot more. Again, more time on the testing. None of this "Oh look there's a new graphics card out, let's get something to support it in a month." Be ready to have years spent some times.
3) Hardware will be restricted. You are not going to be running this on any random hardware where something might be different and unexpected. You will run it only on hardware it's been extensively tested and certified for. You want new hardware? You take the time and money to retest everything.
4) Other software will be limited. Only apps fully tested with your app can run on the same system. Otherwise, there could be unexpected interactions. The system as a whole has to be tested and certified to work.
5) Slower performance. To ensure reliability, things need to be checked every step of the way. Slows things down.
If you aren't willing to take that, then don't bitch and demand rock solid systems. I mean such things DO exist. Take the phone switches for example. These things don't crash, ever. They just work. Great, but they only do one thing, yoy use only certified hardware, they've had like one major upgrade (5ESS to 7R/E) in the last couple decades, and they cost millions. You can do the same basic type of stuff (on a small scale) with OSS PBX software and a desktop, but don't expect the same level of reliability.
The thing is, if your hypothetical bridge were software (and it's quite simple compared to software) people would expect to be able to put the same design anywhere and have it work, drive tanks over it and not have it collapse, have terrorists explode bombs under it and have it stay up and so on and have all that done on 1/10th of the normal budget.
Until we are willing to settle for some major compramises, we need to be prepared to accept patches as a fact of life. I mean hell, just settling on a defined hardware/software set would do a lot. Notice how infrequent it is to see major faults in console games. It happens but not as often. Why? Well because the hardware platform is known, and you are the only code running. Cuts down on problems immensly. However take the same console code and port it to PC, and you start having unforseen problems with the millions of configurations out there.
Me? I'll deal with some patches in return for having the software I want, on the hardware I want, in the way I want, for a price I can afford.
I write the OWASP Guide, which is used by basically everybody as the standard for web application security, and is the official standard of Visa, many governments, and so on.
She talks to CSO's who mostly are bean counters. They see money down the drain from patching. I agree with them - patching is inefficient and wasteful. But it's necessary as Oracle builds crap, buggy and insecure software. They are easily five+ years behind Microsoft in churning out safer software. Buffer overflows, high privilege accounts, public access to highly privileged library functions - all this stuff is easily 10-15 years old and should not be in Oracle 10g, but it is.
Oracle has time and time again outright refused to get on board with a secure coding program, often fixing just the little bug which gained root privileges, exposed all your data, or destroyed the database outright. Instead, they should be searching for all those types of bugs and fixing them in one hit. Davidson has more than enough time to address the root cause
She is holding software up to the standards of bridges. Bridges have tolerances and over-design built into them. Most software does not. Often to make artificial deadlines made by beancounters, software is shipped with bugs. Often the bugs are not found for some time and requires researchers to go find them. If it's not researchers, its the commercial 0day crowd. This is where Davidson shows she is an amateur and must be replaced. It's best for HER customers to be secure, and that means shipping secure software. Shipping insecure software does not prevent the 0day houses from creating exploits. Oracle's reputation as a solid data partner is worthless if we lose all our data to an attacker because Oracle suppressed the news from us, rather than fixes the problem.
It is simply unachievable to build bug free software for a reasonable cost. What is required is care, developer training in secure software techniques, and defense in depth. That is our tolerance and over-design. Oracle is sadly lacking. She has had five years to get their developers onto a program of building this into their platforms, and she's failed miserably. I will be interested to hear what standards they use, and if it's mine (OWASP Guide), or if they do their own based upon ours, or use Microsoft's.
I've called for her to step down more than once. When she attacked the good name of David Litchfield and NGS Software, I was outraged - this was like shooting the messenger that their "unbreakable" software was pure crap, which we already knew - but now know through his unstinting efforts that it is truly appalling and not fit for purpose.
If this latest "push" for too little too late does not work out, she should be sacked by the Oracle board for the good of all Oracle shareholders and customers. She's had more than enough time to make a positive change, and should make way for someone who really understands security.
Andrew van der Stock
Ah but most bridges don't fall apart that easily. Well no, most bridges are best on millenia old technology. The more advanced designs are designed to very fine tolerances.
Take that "new" superhigh bridge in france. It cannot support the weight of an ocean liner. Would collapse if you blew up one of the pillars and a nuclear strike within a mile would cause it to fall apart. Hell even a simple typhoon would do it.
Ah, but none of those things are likely to happen so the bridge wasn't designed for it.
That is the big difference between software and hardware. Even the simple thing of user supplied data is different. In software you need to check and check again every bit of data to make sure the user hasn't supplied the wrong kind of data. Hasn't the user put a 1 gigabyte of data in a bool field?
In the real world this is kinda easier to check. I think you would notice if a truck instead of being loaded with 10 tons was loaded with 10.000 tons. A clue might be the way its axels are buried in the asfalt.
So the bridge designer only has to design for the entire roaddeck being filled with trucks filled with lead and that is it. He can work with real world limits. The french bridge was really tested like this. It withstood the test and is in theory designed to withstand 2x the load. That ain't much of a tolerance but in the real world you can easily discount such a heavy load ever being put on the system. Someone driving up with an ocean liner on his trialer would draw attentention.
Not so with software. I can put anything I want in this input form and the software better be designed for it. I am not constrained by real world limits.
That is what makes software engineering so difficult, you need to account for every possibility. If you checked a piece of data and wrote it too storage then you need to check it again when you read it. This would be like a bridge engineer testing the steel, then having to check it every day to see if hasn't turned into porridge by an act of god.
Oh and one final note. A lot of software insecurity only happens under attack. Bridges don't exactly last long under attack. Blowing one up is amazing easily. Any army engineer can do it.
MMO Quests are like orgasms:
You may solo them, I prefer them in a group.
I prefer this scenario, if bridges were like software.
We all know you can build bridges out of spaghetti (surely you did it at engineering college?), so at a company like the one I work for, a college kid would be hired, lick the bosses arse and mention that he could build a much better bridge with the tools and new practices he learned at his university where he was taught the latest, cutting-edge technologies. Boss is impressed and asks for a prototype (not stupid this boss, prototype first, then ship it. He's learned loads on the 'how to manage technical people' course he went on)
So, new kid builds the prototype and, yes, it can carry a model car across. Even scales up to carry the stress test of a model lorry! Boss is impressed - "why can't you old guys do that?' he says, thinking of the praise he'll get at the next board meeting.
So they set about scaling it up to suit their customers, larger bridge, industrial spaghetti, held together with glue and installed in the customer's city, across the river. Customer is really happy with their upgrade, and after testing it with a compact saloon realise they can de-commission the old steel monstrosity. All's happy... until it rains. But, that's ok, its just new-bridge teething troubles, just requires a patch with some waterproof paint and rubber sealant.
Until a lorry decides to cross.. and it snaps, but again, just patch it up by reinforcing with some old-technology steel girders. Doesn't look so pretty and won't be as maintainable, but.. what the hell, the project manager declares it a success so the comapny is happy, and new customers are told that the company's flagship bridge uses only the latest cutting edge technologies.
Unfortunately, in the real world, software is not as visible as a bridge so new customers can only go with the marketing and sales waffle. Once they've bought it, its too late.
Given what Oracle's problem _is_, probably what they _really_ want isn't regulation of the "you must prove that your software passes this and that criteria to be allowed to sell it." (Which would also raise entry barriers for competitors.) I mean, really, if you were a company which takes five fucking _years_ to bother patching a security hole, and even then only when an exploit was widely publicized, you're not going to ask for a regulation that'll ask you to pull the product off the market until you fix it.
The kind of regulation they want is more like "you're an evil irresponsible hacker and going to jail if you disclose bugs in someone else's product." Yes, it's security by obscurity. But that way Oracle can happily spew bullshit about being secure and unbreakable, and never have to fix any bugs.
Basically Oracle doesn't give a shit if Corporation X's database is riddled with bugs and exploits. They just don't want the PHB's at Corporation X to know about it.
If it also results in some entry barrier, all the better, but that's not the main goal.
A polar bear is a cartesian bear after a coordinate transform.