So, people from within Iran have hacked the Dutch company DigiNotar, allowing them to issue fake certificates so they could listen in on Iranian dissidents and other organisation within Iran. This is a very simplified version of the story, since it’s all quite complicated and I honestly don’t even understand all of it. In any case, DigiNotar detected the intrusion July 19, but didn’t really do anything with it until it all blew up in their face this past week. Now, the Dutch government has taken over operational management of DigiNotar… But as a Dutch citizen, that doesn’t really fill me with confidence, because, well – whenever the Dutch government does anything even remotely related to IT technology, they mess it up. And mess it up bad.
It’s almost a commonly known fact in Dutch society that whenever our government – regardless of which coalition is in power – does anything even remotely related to IT technology, they will mess it up. It will be insecure, it won’t work as intended, it will break down, and it will go over budget at least 56 times.
At some point, our government decided, in all its wisdom, that out existing method of buying train and bus tickets was no longer adequate. The system worked just fine; for trains, you walked up to a machine or a service desk, handed over some money, and you got a (return) ticket of your choosing. For the bus, you bought a ticket which had several ‘zones’ on it; each zone corresponded to a certain area. So, between Alkmaar and my hometown, there were four zones. You always pay at least one zone, so in total, the bus driver stamped five zones off my ticket. These tickets came in various sizes (15 zones, 30 zones, etc.). If you traveled by bus/train more often, you bought a pass which you only had to show to the driver, and that was it.
This worked just fine, but the government decided that this should all be done electronically. So, they let a private company come up with a plastic card with a chip in it, which had a certain amount of money on it. Using NFC, you tap it against a turnstile, which would then register your starting location. Tap again at the exit location.
Sounds simple enough, but the system was hacked quickly, making it very easy to hack more money onto the cards, or to duplicate them. To make matters worse, travel information is stored by a private company, which has raised red flags regarding privacy issues. The system also doesn’t work properly, causing wrong amounts to be deducted from cards, and it has also increased costs of public transport. A supposedly safer version of the system has been planned, but the roll-out will take six years, and the costs of this roll-out will be transferred to travelers. The system is also not particularly friendly towards the elderly. The list of complaints is long.
In short, the system has been an unmitigated disaster.
Another example: crisis.nl. This website is supposed to provide citizens with information in the case of a disaster. However, the website cannot handle larger loads, and during a few recent disasters, the website was knocked offline very easily. The website cost a whopping ^a‘not500,000 to build, and now that it has proven not to work, the government is throwing even more money at the project.
In the meantime, a smaller company has built crisis2.nl, which delivers the exact same information, and can actually handle lots and lots of traffic which would’ve already killed crisis.nl – and at only 75% of the costs. The reason this company doesn’t get to handle this project is because the company is considered too small. The rules for government projects like this state that only large companies are allowed to accept them.
Then there’s the Dutch police. The problems with the Dutch police go far beyond just the computer systems backing them, but it is still a major problem. The plan was to invest ^a‘not46 million in modern computer systems for the police, but the new systems did not work, were hated by police officers, and have horrible usability. To make matters worse, local police stations did not adapt their workflow, and kept existing systems side-by-side. The end result is a major cluster*&*$. Costs have already risen to ^a‘not70 million – but this excludes costs for implementation at local police stations. Everything put together, police computer systems in 2009 cost ^a‘not770 million – and it still doesn’t work.
According to comprehensive studies among police officers and personnel, this has a direct and negative effect on the police’s ability to do its job. In other words, crimes aren’t solved, and criminals aren’t caught. Police offers tell stories of them still struggling with entering information of a crime into the computer system, while the perpetrator walks by – he’s already been set free. It’s… A disgrace.
And these are just a few examples. So, when the Dutch government sends out a press release stating they have taken over operation management of an IT-related company… Well, while this might inspire confidence abroad, it doesn’t inspire me with any. This story is far from over, and I wouldn’t be surprised if it goes even deeper than we know today.
Well, considering how incredibly bad the private company screwed up it’s not like it can get much worse.
Don’t say that… they’ll take it as a challenge…
They didn’t screw up – they implemented stuff that doesn’t work, and now will get paid even more to get it to actually work. Or maybe they’ll get it working in the next upgrade. The execs got some nice bonuses, lawmakers/police now have a way to track people…
Problems? What problems? Thom, you just worry too much!
With private companies you get to chose if you want their services. With the government it’s a monopoly, you don’t get to chose if you use it, nor if you want to pay for it.
Except I haven’t read anything about the Dutch government monopolizing the CA business. You’re still allowed to start your own CA business if you want to and you’re free to not use DigiNotar.
Of course you can choose it. And not only by, say, finding large enough like-minded group of people for a peaceful coup and/or shopping for a gov that is to your liking on the world marketplace (and if there isn’t any – tough luck, maybe humanity isn’t for you*).
It is fairly easy to use hardly any services, and pay for none – being simply below the taxation threshold …wait, what, you do want to live comfortably in an environment provided by modern society? Then don’t escape from what is just the “cost of doing business” – you don’t expect the landlord, who makes sure you have a comfortable place to live, to not get his rent money just because you don’t feel like it, right? Or likewise with comfortable utilities you’re so used to? (which BTW would be a disaster without regulation; not safe, immense waste of incompatibilities and duplication, etc.; with many people unable to choose water, electricity… most areas would be without these services if there were no intervention, as is still the case in many areas around the world)
*If humanity at large isn’t willing to fulfil your whims, “deliver you what you want” to use your words elsewhere from this thread, maybe it’s time so sign out …or at least not be a hypocrite, not live where you benefit abundantly from the comforts provided by an integrated society.
If the Dutch government would get only a few things right, they would be doing things better than DigiNotar and would prevent many other attacks.
I think the Dutch government could have one team in one organisation that handle offline signing.
That means it is not in any way connected to the online world like DigiNotar.
They check a number of things (simplified):
– they receive a request by PGP-signed email
– check if they are on the contact-list and PGP checks out.
– look at the name of the request and see if it oesn’t have *.google.com or other silly things like municipality X does not need to create a certificate for the website of municipality Y.
– call the people at the other end if they send the email
– check the numbers on the certificate request over the phone.
– create the certificate
– email it back, PGP signed.
Done, much more secure than what they had before.
Edited 2011-09-06 11:15 UTC
Cool, the Danish government was greatly inspired by the Dutch electronic traveling card… we get to have it too, and it also doesn’t work here.
Here in Latvia it does work (we only have to check in, not check out) but the maintenance costs are very high.
How does the bus know which amount to charge? Do you have to enter it upfront?
I’ve been in Riga (Latvian capital) about a year ago, and as far as I remember, you have a pre-loaded card that you swipe on the reader on entry only.
The trick is, that unlike similar systems elsewhere in Europe, Latvian transit does not differentiate tarriff – it’s exactly the same no matter the distance travelled. And you cannot get a “free” transfer – once you exit the bus/ trolleybus/ tram, you have to swipe it again on the next connection.
That is also how it works in Toronto, Canada when I was there a couple of years ago.
Independently of how you pay ofcourse.
Edited 2011-09-06 14:14 UTC
Yeah, that’s the way it works. A ride costs the same no matter which bus, trolley bus or tram in Riga you take and how far you go.
I also remembered another issue. One can buy single ride ticket only at the bus driver which is a bit more expensive (approx. 1 EUR) than buying 3 rides ticket at a kiosk (which costs approx 0.7 EUR per ride). The single ride ticket is printed on paper and the higher cost is to discourage people from buying tickets from the driver. The 3 ride ticket, however, is an electronic card which obviously has higher production costs and offering a single ride electronic card just isn’t cost effective.
Maybe it is to encourage your two places to use bikes even more than is already the case?
(well, from one other example – the real purpose of inner-city parking metering is to discourage people from using cars in the city centres, according to a buddy of mine who works at a public office responsible for it)
I see so many similarities on government acting that I could call The Netherlands as the european Brazil.
Japan has a similar way of paying for travel and one of my friends who stayed their thought it worked perfectly. But they might just be better at implementing things considering the trains are never late. In my country I’m sure we would get the same result as the Netherlands if we did the same.
Edited 2011-09-05 23:44 UTC
Even in Japan the trains run late…. Mostly to suicides or disasters…. The system works perfectly though, even when using a card on another train network or a combination of the above. It also gives you more functionality then just travelling. Buying stuff at kiosk stands etc. can also be paid for with the card…
Your country (or at least Uppsala region) does electronic tickets on an NFC card
I know, I was there !
Only for regional transport. For national line, it is paper or sms, bought online or at a shop. And for regional transport, the quality depends on the area, for instance in G~APteborg, you have to press a sequence of button in a certain order to get your ticket (with no documentation on board), and if you go out of town, and forget to validate at exit it will charge the full content of your card. And in ~A–stg~APtaland, the system is not too bad, but it is slow as hell, it can takes more than 10s before the machine validate your ticket.
There’s some mysterious ergonomy in Uppsala too. When you enter the bus, you must press a button stating which tarification you use (full tarification, children, etc…). The only thing which labels the buttons is a digit. Thankfully, the driver is here to explain you, but when she doesn’t speak English and you don’t speak enough Swedish, it remains an awkward moment.
Past this painful learning step, it works perfectly.
In national trains, what does this SJ card (called SJ prio IIRC) do ?
In France too, national trains don’t use NFC tickets. Guess it doesn’t matter so much, because the regions are so large around here that only few people use national trains frequently. The implementation of regional tickets also varies on a per-region basis, but I have to admit that I haven’t travelled enough outside of Ile-de-France to tell you if it changes a lot.
Edited 2011-09-06 07:32 UTC
The Prio card is not an NFC card, only a customer discount card. Lower fares, special offers, free newspapers/coffee in the bistro on sundays.
It doesn’t contain any ticket info, but you might need it on your person if you need to validate a discounted ticket bought via that card.
And it’s only for regional fares by train and only valid with one carrier (SJ).
Thanks for the explanation. I think I have confused with something else. I remember people just playing with a plastic card in the train, without a paper ticket or a mobile phone. SJ’s website mentions Pendlarbiljett and ~A…rskort, perhaps it was that.
Also…
Sure, but if I get it right SJ owns pretty nearly every national train in Sweden, and you have a good train network (except in the middle of winter), so that’s already something.
Edited 2011-09-06 11:49 UTC
It’s not only they are better at implementing things but that their society wouldn’t even think for a sec to try to steal from others. Most of Japan doesn’t even have locks at home because the idea of someone breaking into your house is… well, they just don’t think about that.
They have so much respect for each other that it gets ridiculous many times.
That was something I really loved about the japanese
The Dutch system is nothing compared to the Japanese system.
Dutch metro’s now feel like supermax prisons while Japanese are clean and friendly.
Doing customer service in the Netherland is horrible, me trying to merge three different cards together which I got from one transport agency for three different products was a nightmare. They deal with you like you are some kind of criminal… how dare you try to make your live easier by putting 3 digital products onto 1 digital card.
No give me the Tokyo or Helsinki system any day… and please let us just stop with the foolishness in the Netherlands.. it only will costs us large amounts of cash for a transport service which is not much better then cattle transport.
Like another poster said before… we Dutch will never be so civil as the Japanese are, so this type of system will never work for us…
It will just be a new version of the way we deal with fuel or telco prices… completely non-transparent and very volatile increasing prices… with less and less customer service….
(btw have you ever seen a Japanese gas station ? that is customer service!)
Sort of like South Korea with a conservative pro-American government right now. Pro-American conservative politicians in the National Assembly in Seoul don’t know anything about computers, softwares, and the tech industry.
Almost 1 paragraph of actual content about something that you admit you don’t really understand.
Followed by a rant about some failed projects that you have heared all the popular complaints about and wrote them down ad verbatim
Do you really think that standing in line for a trainticket, seeing that train you wanted to get on already leaving is better than the few seconds you have to spend “beeping” yourself through the gate? How much real abuse is taking place now and how much did the old systems get abused?
As far as I had heard of crisis.nl had problems with a person updating the page wrongly, not with “not being able to handle the load”.
The new police software not being integrated into normal workflow is a problem of the central government or the local government? Serious question, I don’t know .
The software for doing your taxes seems to work quite nicely by the way, even though it relies on certificates that were issued by Diginotar, you know, the company that this article is about.
I am not saying the Dutch government and IT are a good combination, but your examples aren’t very good. They are biased and they are not very related to the topic of taking control of a TRUSTED certificate authority
It sounds like the Dutch government’s problem doesn’t have anything to do with IT. It sounds like they have a problem with how they contract projects out.
Now I don’t know how things work in the Netherlands, but there are two problems with how it works in Canada. A huge problem is that bidders underestimate costs, because it is the only way to make a competitive bid. The mentality of outsourcing also means that the government has relatively little technical expertise to evaluate bids, meaning that they are almost dependent upon accepting the bids at face value.
Oh, and thank-you for the transit anecdote. My city is planning to move to a similar system. Ironically, many people can barely figure out how to use the existing system!
That is a problem not just in IT, but everywhere. Particularly in public projects those in charge favour cheap projects over slightly more expensive (but significantly better projects). Invariably this usually means that the cheaper project is either cutting corners or underestimating the costs which either they raise once granted the project, or, produce shoddy work.
Unfortunately this is required by law (at least in Belgium, but probably in most countries). Government projects must always choose the cheapest solution that fulfils the requirements. Which makes sense in a way: why waste taxpayer’s money on more expensive contracts?
However, in practice, companies abuse this the same way Ryanair cheats on you: hiding costs everywhere or just plain lying about it. It requires a lot of expertise to write a perfect contract and make the right decision in so far the law allows you. Most government workers do not have this expertise, especially not local ones). This results in very poor solutions, often never finished.
There have been a few big cases like this here. The government has now sued several of those companies but meanwhile they don’t have the money for an alternative solution (with the same risks) and lawyers aren’t exactly cheap either.
Edited 2011-09-06 12:22 UTC
That’s just par for the course. Bids are lowered to look good, and companies are hired based on how much money will be saved. Imagine the shock of the government (or corporate) users when they discover that things aren’t to spec and will now cost significantly more to support or change than expected. But hey, it looked cheaper.
(I had the pleasure of watching this on some outsourcing projects where vendors promised all sorts of wonderful savings – and the end users ended up shocked when making changes actually turned out to cost more and take longer than when they were in-house – and now they had a new stand-alone external system that didn’t even integrate as well with the other apps anymore. On the other hand, I had the foresight to get out while it was happening and didn’t have to be the IT chump picking up the pieces of a bad contract and broken system.
)
We have a similar public transport ticketing system here (Perth, Australia), and it rocks.
You don’t have to worry about how many zones you’re going to be travelling; the smartcard system works it out when you tag off. If you change your plans and decide to go further than you originally intended, you don’t have to get off the train and buy another 2-zone ticket and then wait for the next train; you still just tag off as normal at your new destination.
If you’re catching a bus, it’s even better; you don’t need to know how many zones it is to your particular bus stop. The smartcard reader works it out when you tag off and handles it all transparently.
It even works out the cheapest fare for you; if you travel three zones into town and then three zones out of town, the system knows that an All Day fare is cheaper than the two three-zone fares, and charges you for the All Day fare.
The Smartrider system can even be used to pay for parking at train stations and for unlocking the bike sheds at train stations. It is linked to the universities, so students automatically get a concession fare when studying full time, and as soon as they drop to part time or leave university they get charged full fare again.
If you load a larger value onto your card (for instance, $50) you get some discount off your fares too.
Oh: And the system works perfectly every single time.
Please don’t say “The paper tickets worked fine, why are they replacing it with a smartcard system”. When done properly, smartcard tickets are awesome because they make life a lot easier for travellers, and are a lot more flexible. They cut down on fare evasion and concession fare fraud (where students enroll full-time, get a concession card and then drop back to part time; that trick doesn’t work anymore with the Smartrider).
In short, smartcard ticketing is so good, you won’t realise how much you love it until you think about the olden days with paper tickets
Unfortunately, you hear a lot about countries where it hasn’t been done properly and there have been problems. But that’s a government problem for choosing the wrong company, not an inherent flaw in the idea.
Heh, gone back to paper tickets atm because my university is stupid, so I have to agree : contactless tickets with nicely picked plans are just awesome. You go wherever you want, without worrying, and you enter/exit the station much faster.
So what has Iran got to do with any of this? Is it just a convenient scapegoat? According to the BBC the vast majority of certs were issued to Dutch companies and individuals. Hell the govt. takeover is also because the Netherlands govt. was using this CA.
The old Dutch system for buses, with its strange folding-up tickets, probably worked well enough for locals. To tourists it was beyond baffling. Considering the size of your tourist industry, maybe a change was called for.
Well, the system is pretty much the same. It is just electronic now.
Thom: Well – whenever the Dutch government does anything even remotely related to IT technology, they mess it up. And mess it up bad.
Except when they regulate mobile phone companies!
What a success that was…
I vaguely remember Thom pushing that here, so do I hear some buyers remorse?
Knowing your love of free market theories, I guess what you advocate instead is a public transportation network, without any form of government backing, or even regulations to make sure that they cover the majority of the inhabited territory with decent service.
Or perhaps no public transportation at all. After all, stuff which can’t work well in a free market is not interesting, right ?
Neolander: After all, stuff which can’t work well in a free market is not interesting, right ?
I wouldn’t say not interesting, it might be. Certain things are necessary of course, i.e. the government has the monopoly on coercive force.
But for your particular example: if the free market doesn’t provide it, it means it can’t do it, either because it is forbidden by the government, or it cannot provide it at an acceptable cost.
So if the government provides that service, you incur a cost. At minimum the public should be aware that if the government steps it, the cost might potentially be draconian. If the government should do it, is obviously a political item.
A good reason to object is that the government uses coercive force to extract the money from its citizens. Coercion is generally bad IMO.
Thanks for this more precise and open-minded comment. I owe you one, will try to do my best.
In my opinion, every service which at the same time is expensive to provide and requires a universal reach to be useful, needs some form of government backing and/or regulation to work in a fashion that is socially and morally acceptable. Services in this category include :
-Public transportation
-Postal service
-Water
-Electricity
-Telecommunication networks
If one of these services is left to a free market, then the laws of financial rentability will do their job and the areas of high wealth/population density will get a high-quality service while the areas of low wealth/population density will get little to no service.
If we consider that every human being in a country has a right to a minimal level of service in one of these areas, then regulation must be introduced, as a force that makes sure market actors will provide this level to everyone, even when it economically means a loss.
This does not mean that the government has to own the company and dictate every single decision. Only a limited number of objectives must be ensured, and the company is left free to take care of other aspects of the situation. It’s not the color or brand of a bus that matters, it’s how much people it can carry, where it goes, and how often.
The drawback is that as you say, sometimes governments may abuse their power. Which is why justice must work in a fashion that’s roughly independent from the government (as is the case in many countries), and why governments must be regularly renewed through elections in order to ensure that they still match the demand of most.
Besides this “stick” side of things, governments also have a “carrot” at hand : they have a budget, fairly collected among the population, which may be used to provide backing when companies can’t realistically meet an economic objective alone. As an example, high-speed trains are often operated at a loss, and require financial backing from the whole population to become profitable while still remaining accessible to most travellers.
Fundamental research is another example of service which requires government backing to be operated optimally. It’s financially too risky for most companies to provide significant backing to it, even though in the end that work may benefit everyone.
Globally, government backing and regulation is necessary to meet demand for a higher quality of service than what is financially optimal.
Edited 2011-09-06 11:53 UTC
In a truly free market, what’s to stop somebody from coercion via economical domination?
IT ACTUALLY HAPPENED NUMEROUS TIMES, the economically advantaged group can use free market to enforce tyranny on others, for example: White Citizens Councils ( http://en.wikipedia.org/wiki/White_Citizens‘_Council ), you can’t rewrite history. Fact: White people used the power of the free market to destroy the lives of middle class black business people without having to resort to violence.
Thankfully, the government stepped in and stopped the oppression. Do you want the clock turned back on that one?
Again, it actually happened, the free market was used to oppress and destroy people, you can’t simply rewrite or overlook history to agree with your political/economic theories, to make your cherished ideologies seem more plausible. It’s astounding that people can discuss “market forces” and, nearly in the same breath, deny that the market has any kind of (“bad”) force.
Yes, yes, “harm” would be illegal, but what is harm? Does it harm me if you pollute? What if you use child labour and my moral code prohibits it, when you undercut me in the market, is that harm? What if you decide to form some White Citizens Councils and drive all blacks out of business? Certainly that would be considered harm, right? Or would it just be the free market at work? What would keep somebody from economically coercing and dominating others? Or is that considered OK, as long as that somebody uses market forces to oppress others, to destroy their lives?
Who regulates what is harm? Maybe, hm, we should form some central entities handling such stuff on our behalf?
Edited 2011-09-13 00:12 UTC
Unlike you, I don’t perceive the world as black and white. I can understand that what works in some cases, does not work in others. Government intervention can be both good and bad. In the case of mobile phones, it worked out okay. In the case of anything related to implementing actual technology (as opposed to just regulating said technology), things generally go wrong.
In the end, you still haven’t answered the question I posed you so many months ago: point me to a truly free market which works. Any luck with that yet?
Thom: point me to a truly free market which works.
Sorry, missed that. Your question is probably: point me to a free market that delivers me what I want.
But that’s not what markets are. Markets are purely the free and non-coercive exchange of private citizens. They might not produce you an outcome you’d like (mobile phone calls for 1 cent per hour for example).
I would offer food markets as a market that is pretty unregulated. The outcome might be that we got too many obese poeple, but probably better than people starving.
If you had lived only 400 years ago, you would have been a peasant bound to the land. Market economies have improved lived drastically.
This is what you come up with? The food market is probably one of the most strictly regulated markets in existence today! Everything from ingredients, to packaging, to information provided about the products, to the production process, to the kind of gloves and protective gear workers have to wear, and much, much more than that, is strictly regulated by the law. I don’t think there is any industry that is more strictly governed by laws and regulations than the food industry.
More and more I’m getting the idea that you read some fancy theoretical book cited in Economics 101 at university about the ideals of the free market, without really understanding what it means. If you cite the food market as an example of an unregulated free market… Holy cow.
Edited 2011-09-06 09:43 UTC
While $.01 calls might sound great, it’s the lack of regulation that can lead to $5 per minute phone calls if monopolies / price fixing aren’t controlled.
There’s a lot that free markets are great for (let’s produce X number of phones as dictated by a central planning committee isn’t such a great plan), but certainly protecting people from environmental health issues, setting safety standards, or preventing price gouging by unscrupulous mega-corps aren’t amongst those items.
For anyone who would say ‘well, if Company X is charging too much, Company Y will show up and offer a better priced service”, keep in mind that in a TRUE free economy, Company X may be large enough to implement unfair agreements with distribution / access channels, litigate, or buy out Company Y and never have to worry about competing on lower prices.
Oh come on, don’t you know this stuff sorts itself out in the long run and if some poor sods die in the meantime, well, that’s just the price of freedom.
Regulated market economies have improved lived[sic] drastically.
How easily you ignore many starving or malnourished (or with unsafe food and water) people around the world; mostly in places without functional administrations… It really doesn’t put in you in a good light.
In some truly unregulated markets, there is a hunger epidemic right about now. People DIE. Warlords capture aid, it’s more profitable for their free enterprises.
Glance at least once at a packaging of some of the gorged (to feed that obesity) food; contemplate how can you believe what they write about contents. And, somehow, we don’t really die regularly from food poisoning or food-borne pathogens any more… why is that? If a dozen or so would die in each city every week, it would lead to better efficiencies! It would self-regulate, promote the best providers of the month!
Look how nicely unregulated capitalism works: http://www.rootsweb.ancestry.com/~belghist/Flanders/Pages/phossy.htm …so efficient, profitable and rational to not bother about the lesser people around.
Oh, and our agriculture runs on fossil fuels ( http://en.wikipedia.org/wiki/File:Human_welfare_and_ecological_foot… ) …we certainly want to regulate that crap.
They put electronic tickets on NFC cards around London (Oyster Card), Paris (Navigo D~A(c)couverte), and Uppsala (Uppsalakort), and they work just fine, except for a early card death from time to time…
However, I don’t remember these systems ever being directly controlled by the government. In all three cases, it is a state-owned company, but this essentially means that it gets a big chunk of its funding from it and must work with it on some infrastructure projects in exchange, otherwise remaining basically independent. As good public transportations networks, for all their technical awesomeness, are unprofitable in essence, this sounds fair enough.
The only case where government has to actually strongly step in is when several transport companies can’t agree on a common transport pass standard, and it generally goes fine.
Edited 2011-09-06 06:49 UTC
It is not illegal to do this in Iran. So why now are you so upset that someone may be stealing? I think this is the funniest story in OSAlert in a long time. Now we see the effects of how some people view theft an others deny it. Good for Iran. They deserve to steal from the Dutch. The Dutch were too busy stealing stolen movies anyway.
Dude, seek professional help.
Once this holliday I was stuck in Rotterdam trying to get out. The ferry have had already left. It was 21, I had a bike and cash to pay for the ticket.
You know what, I couldn’t buy it!
The freacking ticket machines woudln’t accept neither paper money nor my non-pin bank card. The ticket for me + bike ended up being 20+ euro. Who in the world carries around that abount of change?
I had was literally at mercy of the train director, who as it appears couldn’t sell me the tickek anyway (cause the bike I guess).
Edited 2011-09-06 13:30 UTC
I and several others I know at large companies have removed them as a trusted ca root. You’ll find that the internet is much better at granting initial trust than restoring it to someone found unworthy of that trust.
That would have been kind of a problem if the CA is used by your government and you want to make use of their services. Most people also have no idea how the system works they just rely on others getting it right.
A ssl cert CA’s business is based on trust. Others must trust them to at a minimum keep their cert issuing authority out of the hands of bad guys. If they don’t as in the case of DigiNotar, then people like me stop trusting them. Then people who expect things to work, find they don’t and blame the company they are trying to connect with. Then that company switches CA roots to someone who is trusted. And they system works for everyone again.
So I guess you also removed Comodo, last time ?
To bad you can’t use a quarter of the whole HTTPS sites on the internet.
You see, it isn’t that simple.
I didn’t even know Comodo was hit.
http://www.infoworld.com/t/authentication/weaknesses-in-ssl-certifi…
You can’t reasonably block them without breaking most HTTPS sites. If I’m not mistaken, microsoft has also chosen them to do code signing certificates in win vista+.
I was referring to a different use of HTTPS other than the www. Which, I’d prefer not to delve into. But rest assured Comodo is blacklisted as is DigiNotar. In fact, we’re in the process of switching to our own white list, rather than the default ones you see in browsers.
But actually, I did personally remove it from all of my browsers. It hasn’t been a problem yet. Are there any big sites that actually use Comodo as a cert? Most I see are GeoTrust, Verisign,Thawte, Net Sol, TrustWave.
So, what do you do when you need to electronically communicate with some websites that use a certificate signed by some CA you don’t trust?
Bill Shooter of Bul,
“Then that company switches CA roots to someone who is trusted. And they system works for everyone again.”
I know you understand what is going on. However what you view as working system, I view as a broken model.
3rd party authentication, as with the CAs, is inherently problematic when the CA’s security is lower than that of the websites using SSL. As it stands, any CA has the technical ability to create a fraudulent certificate for any website. No matter what precautions SSL users/websites take, they are dependent upon *ALL* CA’s to not screw up.
The CIA probably was not a client of DigiNotar, and yet they were a victim of the leak. DigiNotar didn’t even bother to tell anyone about the leak for several weeks – if there are more leaked keys out there, we’d have no idea.
I don’t want to sensationalize this and blow the risks out of proportion, but 3rd party trust is a disturbing requirement of SSL.
I’d be a bigger proponent of a secure DNS based solution which guaranties that we are communicating with the registered owner of a domain name. Everyone with a domain name would be entitled to publish their own certificate in their DNS records and not have to use a CA for the privilege.
This would still require trust in one’s hosting provider to supply the legit certificate via secure DNS, however since trusting a hosting provider is implicit anyways, it doesn’t increase the scope of trust and it can be insourced to increase security.
You raise some good points. I would be in favor of a better system that wouldn’t allow any trusted CA to issue a cert for any site.
Given the current system that we have, the best bet is to restrict the number of CA’s that you trust.
Bill Shooter of Bul,
“Given the current system that we have, the best bet is to restrict the number of CA’s that you trust.”
Well yes, but that only applies to what you can control. There are problems with managing CA’s personally:
1. As a website owner, your choice of CAs doesn’t increase your security. The authentication of your website is validated by the list of CAs in your user’s web browsers.
2. As a user, it’s reasonable to want to trust only specific CAs where I can attest to their security. However in reality real websites will use CAs who’s security I cannot attest to. So, this may not be an option.
2b. Obviously you’re talking about blacklisting a select group rather than whitelisting a select group. But the problem remains that you are trusting CAs who’s security procedures haven’t really been attested to and could in fact be as bad as DigiNotar.
I’m not even sure how bad DigiNotar’s procedures actually were. All CAs are vulnerable to things like zero day exploits and disgruntled employees even when they do follow best practices.
1) As a website owner, you choose a CA that is used by large companies that your customers would want to use. If they are likely to trust those large websites, they’ll be likely to trust al certs signed by the same CA.
2) As a website user, the number of SSL enabled sites that I use are limited to a few, those few do use reputable large CA’s. Its actually quite easy, and with minimum side effects. If a site is signed with a ca root you do not allow, you examine the cert closely and determine if its really worth the effort to verify the identity of the website or to use a different website that provides the same features.
2b) Yes, this may be trusting CA’s that have just as poor security as DigiNotar’s, but reducing the number of ones that you do trust reduces your vulnerabilities, I think. In any case there are CA’s that are trusted by browser makers, that I do not trust who also do not sign any certs of any of the websites I use over SSL. Removing them is an obvious choice.
Bill Shooter of Bul,
1 – I think you missed my point. A bank/commerce site can choose whatever CA they want, but it doesn’t matter when 99% of their customers (purely made up) have default CAs in their browsers. It may not be the site’s fault, but users are never the less vulnerable through the weakest CAs in their browser. There is absolutely nothing you can do as a website owner to protect your users.
2 – That’s quite a hassle. Even for people who have the extra time and expertise to do it, it’s bad that they’d need to give up their online choices due to shortcomings of HTTPS.
2b – Even if we assume that it’s possible to audit the internal security of a CAs in a comparatively meaningful way, that knowledge is not really public. I certainly can’t tell if vendor X is more secure than vendor Y, so on what basis should I white/black list them? Popularity?
So, I don’t think it’s reasonable or helpful to ask normal users to manage their own CAs. If anything, CAs should be licensed and audited to ensure some kind of compliance with security protocols. Better yet, transition to technologies which take third party CAs out of the loop.
Edit: I guess another possibility would be to change HTTPS validation to require two valid certificates from two independent CAs. This would significantly reduce the attack windows when one CA is compromised.
This would be pretty good from a security robustness standpoint…I don’t think it’d be popular though.
Edited 2011-09-06 21:52 UTC
Oh, I thought you were asking me what I did, not what Joe computer user people should do.
But for argument’s sake,
1)If everyone were to limit their trust to the mythical”good CA’s” then yes, it would be a good idea for a website owner to use one of those. All of these points have to assume that is the case, or none of this makes sense.
2)It doesn’t have to be individuals, necessary. Browsers are already doing this, although they’re doing it quite poorly. It would be better if they had an interface like Adblocker that allowed you to choose from some preconfigured list of ca’s. Yes, this is trusting other people to trust Ca’s who trust websites. Which is a lot of trust going around, but I actually think it would be better.
2b)Yes, Popularity with big trustworthy sites. I trust large websites who have good security minds to select their CA roots. Other people should too.
Edit Comment:
That would be an improvement, but would require CA’s to work together.
Bill Shooter of Bul,
“Yes, Popularity with big trustworthy sites. I trust large websites who have good security minds to select their CA roots. Other people should too.”
This immediately rubs me the wrong way because I have a strong bias against promoting large corporate oligopolies. Anyway, assuming large popular CAs are generally more secure than small ones, then logically we should restrict our business and trust to large scale vendors. The problem I have is that this crushes smaller competitors, some of whom might actually be more secure and more deserving of our business.
Maybe a bank has vetted small vendor X as extremely secure and competitive, but the bank never the less decides to go with a larger vendor Y because Y is on the popular CA whitelist and X is not.
I have to admit that a few large & secure CAs seems more secure than hundreds of smaller ones. (I really hesitated to write this, such is my resentment for monopolies/oligopolies).
“Edit Comment:
That would be an improvement, but would require CA’s to work together.”
I was thinking that one would just acquire two certificates from two CA’s, and transmit them both to the client for validation. No CA cooperation needed. However I guess you were thinking of having two CA’s sign the same CSR to create one certificate, which would also work. I’d still prefer a DNS based solution though.
Edited 2011-09-07 03:15 UTC
Yeah, its not easy to be a a small company in tech. Maybe there needs to be something simular to PCI or HIPPA standards for CA’s. While those aren’t perfect, they’re better than nothing.
I’d be interested in SSL certs being in DNS records, although I’d need to think through the implementation. Registrars aren’t very secure these days either, looking at the recent Register Hack.
Bill Shooter of Bul,
“Registrars aren’t very secure these days either, looking at the recent Register Hack.”
Maybe true, however breaches at insecure registrars are naturally contained to their own customers. As a domain owner, I have control over who my registrar is.
I have no control over unaffiliated CAs issuing fraudulent certificates in my name (intentionally or accidentally).
“I’d be interested in SSL certs being in DNS records, although I’d need to think through the implementation.”
I would not rely on DNS to prove “identity”, only domain ownership (which turns out to be much easier to verify). Therefor the semantics between CA SSL and SSL over DNS would be slightly different:
The CA cert might tell me that I am communicating securely with “CitiGroup Inc”, whereas the DNS cert would merely tell me that I am communicating securely to the owner of “citibank.com” (which is sufficient, IMO).
However in theory, even the DNS certificates could be enhanced with “extended verification” by a CA. The benefit here is that a fraudulent CA certificate is harmless since the CA doesn’t have access to replace your DNS records – you have 100% control over your own domain. Wow, this is sounding even better than I initially thought.
That is kind of like DNSSEC with DANE works. No browser supports it though and the RFC is still being drafted I think (there have been similair efforts before).
DNSSEC would garantee that the IP-address and certificate information in DNS correct. DANE defined the format of how to store a certificate in DNS.
The biggest problem with DNSSEC is deployment because of compatibility.
There are a lot of networks that are not compatible. Many home users have a DSL-router that is not capable of handling DNSSEC. Operating systems like Windows XP do not support it.
So if you are a browser maker and need to check the DNSSEC-signatures and it does not work; you don’t know if there is a man-in-the-middle-attack or if it is the operating system, a DNS-server or a firewall that is not compatible.
If DNSSEC is deployed, it would be pretty easy to have the support for certificates in DNS in the browser and so on. Companies like registrars would be much more willing to sell DNSSEC to their customers too (which is an other issue right now, not all registrars are ready for DNSSEC. Not even every top level domain is ready).
Also some people think DNSSEC is to much like a one-CA-system. For example if something breaks everyone will have problems:
http://www.osnews.com/permalink?488737
“Even if we assume that it’s possible to audit the internal security of a CAs in a comparatively meaningful way, that knowledge is not really public.”
Every CA in the browser has to have a frequent audit to check if they still comply with the rules set out by the CAB-forum (CAB is Certification Authority/Browser Forum).
An external party comes in ones or twice a year and checks if they abide by the rules.
This is like a notary* they check if you have procedures in place to do all the tasks that are required.
But that is all they do, they do not check any technical stuff, just rules and procedures.
I think there is one organisation which does most of those, maybe 90%: WebTrust.
I believe, the rules did not say anything about transparancy and disclosure of breakins. Not even to the members of the CAB-forum.
* Ironically the organisations that started this discussion is called DigiNotar
(sorry, the whole thing became a bit large)
A long time ago there was one CA and people were not all that happy about that either.
DNSSEC (crypto keys for DNS) with DANE (which is a proposed RFC) would be the closest thing to what you talk about, is in a way a single CA-system.
DNS is a hierarchy, it starts at the ‘root’.
With ICANN at the top (root) and operations of the crypto handled by ICANN/IANA and Verisign.
The DNS root-servers however are handled by different organisations around the world. One is a large ISP (Cogent), one again is Verisign, one is the RIPE (European IP-addresses organisation), an other is the US department of defense. The list is here: http://www.root-servers.org/
The money to run ICANN comes from the US department of commerce (if I’m not mistaken). Although the department did sign a contract saying they don’t interfere with technical operations.
The money from IANA and RIPE comes mostly from the people that need the IP-addresses. IANA is like RIPE, they ‘lease IP-addresses’ to organisations like ISP’s that need them.
While they normally only tell DNS-servers where to find the DNS-servers for .com (which is Verisign) they could in theory point it somewhere else.
However DNSSEC adds crypto in the mix and access to the crypto keys is limited to a bunch of people from around the world.
As you can see it is complicated.
But there is a root and thus it is kind of similair to a single-CA-system. But a lot of different people and organisations have a say in different parts of it.
A lot of the organisations are US companies (because of historic reasons ofcourse) and thus the US has some power of those organisations.
Not everyone likes that, the Internet should be ‘owned’ by everyone.
DANE depends on DNSSEC being deployed and that deployment has been slow. Some currently deployed software and firewalls are not compatible. After all it is the largest change to DNS since it was created almost 30 years ago. Just an example, some operating systems and DSL-routers need to be fixed before everyone can use it.
Edited 2011-09-07 10:53 UTC
Lennie,
Wow thank you for the informative posts. Yes I am aware upgrades would be necessary and that DANE is one of the proposals.
I don’t actually think it’s that complicated, but then again I study this stuff closely.
“Many home users have a DSL-router that is not capable of handling DNSSEC. Operating systems like Windows XP do not support it.”
Really? That’d be a surprise to me since DNSSEC is just the existence of more records on top of DNS. If DNSSEC doesn’t work across a router, it implies that the router isn’t truly compliant with the DNS protocol. Not to say it’s untrue, but why would a manufacturer go out of their way to break their DNS stack like this?
“Also some people think DNSSEC is to much like a one-CA-system. For example if something breaks everyone will have problems:”
Well, the main difference would be that the root keys would not be vouching for people’s identity, only vouching for the accuracy of the DNS database, which we already implicitly rely on for the web to work anyways.
From my understanding of DNSSEC, verisign has zone-signing keys for the .com domain (with a relatively brief lifetime), but someone else can hold the key-signing keys – so it would require attacks to be successful on two fronts (in other words a completely broken DNSSEC would still be no worse than today’s DNS).
Personally I would have three independent DNSSEC key signing organizations with three master KSKs – and require that at least two of them agree in order for “verisign’s” ZSK to be valid. Cryptography redundancy schemes like this are very secure in practice.
Edit: In case it wasn’t clear, the intention of the 3 keys is that the corruption of one entity (say by the US government) is insufficient to corrupting the whole system.
We could make DNSSEC KSKs arbitrarily redundant: 7 KSKs world wide, and require that 4 of them agree on ZSKs in order to be valid.
Edited 2011-09-07 17:09 UTC
(as you liked informed posts, here some more ;-))
Well, DNSSEC isn’t just new types, certain types belong with each other. Which are the signatures and the data and flags. Which changes how the basic protocol works. The signatures also make the packets larger, a lot of the times larger than the old DNS limit.
The operating system change is an extra API-call (or change) to allow an application to request signed-answers.
So the operating system will request signed answers from the nameservers. Obviously the nameservers need to be upgraded to understand it to respond with signed answers if available as well.
On this page there is a presentation “DNSSEC Support by Home Routers”, which might give you an idea about what the problems are with DSL-routers:
http://ripe60.ripe.net/archives.php?day=thursday
This is the PDF:
http://ripe60.ripe.net/presentations/Dietrich-DNSSEC_Support_by_Hom…
Basicely they can’t handle the DNSSEC flags, they don’t have large DNS-UDP-packet-support and can’t handle the fallback method: TCP. It pretty much was never needed for regular DNS.
As you may know many of these DSL-routers have their own DNS-server and that is what is communicated over DHCP to the hosts behind the DSL-router. So they use the DNS-server and that is usually the one that can’t handle all this.
____
The root keys are both, in a locked machine called an ‘HSM’ which can be used for signing.
And for safety a copy of the key has been split up in 7 slightly overlapping parts and is kept by different people from around the world (Paul Kane (Great Britain) Dan Kaminsky (United States), Jiankang Yao (China), Moussa Guebre (Burkina Faso), Bevil Wooding (Trinidad and Tobago), Ondrej Sury (Czech Republic), Norm Ritchie (Canada)).
New keys are generated every few years, anyway have a look here:
http://www.root-dnssec.org/wp-content/uploads/2010/06/draft-icann-d…
http://www.root-dnssec.org/documentation/
It probably explains it better than I do. I just type what I think is right from memory.
And the video and documentation of the Key-singing are here:
https://data.iana.org/ksk-ceremony/
__________
Anyway a possible solution might be to use Convergence:
http://www.youtube.com/watch?v=Z7Wl2FW2TcA
http://convergence.io/
This basis system is where the browser asks others on the Internet if they see the same certificate.
With Convergence however the browser can ask for other information as well. So DNSSEC could be one of the things it asks about.
Even when you are in a network or on an operating system that does not support DNSSEC.
Edited 2011-09-07 22:56 UTC
Lennie,
Wow, how did you hear about convergence?
The convergence website is unfortunately void of details. However the youtube clip seems to tackle everything we’ve talked about here… great find! Definitely a very interesting approach, and I am very impressed overall – it’s a great look at the CS theory to see what’s possible.
He says you can configure notaries to verify the CA signatures cryptographically as normal, but I’m honestly not sure what this mode buys us. What difference does it make whether the CA cert is validated in my browser or on a trusted notary server?
The concept which I find most novel is the “perspective verification”, which verifies that my notaries are seeing the same (unverified) SSL certificate as myself. If I am the target of a middle man attack where the SSL certs in my traffic are forged, then the discrepancy would be detected with my notaries.
Hypothetically though, it could be pretty easy for a backbone provider or a country like china to do a man-in-the-middle such that all the notaries I have access to are compromised in the same way. This problem does not exist today with CA SSL.
I don’t remember, but I only looked into it recently. Thinking it was very similair to the Perspectives project: http://perspectives-project.org/
Which pretty much only handles the man-in-the-middle-attack.
I’m not sure what you meant about what he said.
But here are the basics about how it should work as I understand it:
– based on the idea of the Perspectives project the fight the MitM-attacks
– adds privacy by allowing connections to the notaries to be proxied
– adds trust agility. A way to realtime blacklist certificates, CA’s.
I think he means you can have a second or third CA also vouch for a certificate. Maybe that was the part you meant ?
Anyway, there are a lot of different things a notary could check for which makes it very flexible.
Lennie,
“I think he means you can have a second or third CA also vouch for a certificate. Maybe that was the part you meant ?”
That’s not what I understood. I understood having second and third *notary* vouch for a certificate. So the burden of verifying the authenticity of a given certificate is shifted from the browser to several notaries.
The question becomes, how does the notary validate the certificate? It can automatically validates the CA’s signature, but this is obviously no better than doing it in the browser. In other words, if diginotar’s master key is leaked, then the notary has no way to cryptographically distinguish between legitimate and fraudulent certificates. Either the notary trusts the CA, in which case it trusts the fraudulent certificates, or it does not trust the CA, in which case it blocks legitimate certificates. Neither is acceptable; this is the exact same problem we face in the browser, only we’ve delegated the problem to a notary.
The perspective certificate validation mode offers a novel approach to detecting altered certificates. The theory being that everyone not under the influence of a man-in-the-middle should see the same certificate. If anyone sees a different certificate, it was probably altered and should be considered fraudulent.
I do like that this mode is automated and completely does away with CA’s. This is pretty clever IMO, but not foolproof. Also, there are false positives (I speculate are caused by load balanced servers with different SSL certificates) which he referred to as the “citibank problem”.
We can combine these metrics heuristically as he has proposed, however when they disagree we truly don’t know which one is correct and we need an arbitrary tie breaker – we either trust the CAs over the notary consensus or visa versa. However once we do this, there is no reason to bother checking the less important factors at all (lookup short circuit evaluation).
Anyway, I am really glad you found this extremely relevant project. I think he’s pushing the boundaries of what’s possible with consensus based authentication and I’m really quite impressed that he built something that appears to work.
I think what he means is you go to a second CA and get your existing certificate added to their notary (they have a whitelist) and people have that notary in the browser ?
Anyway I noticed on the github site, someone already made a fork of the server code ( https://github.com/nevins-b/Convergence ) and added Google DNS:
http://googleonlinesecurity.blogspot.com/2011/04/improving-ssl-cert…
No code with DNSSEC yet though.
“I think what he means is you go to a second CA and get your existing certificate added to their notary (they have a whitelist) and people have that notary in the browser ?”
If that’s what he means, I’m not sure that standard HTTPS supports multiple certificates at the same time, many hosting providers may not allow this non-standard configuration. Even if I am wrong about this, I make the claim that since multi-CA websites are so unusual today, any notary which requires mutliple CAs for a website will fail for nearly every website in existence today.
(then again I may have misunderstood you).
Edit: If only he had information on the website, we wouldn’t need to speculate what he means!
“Anyway I noticed on the github site, someone already made a fork of the server code ( https://github.com/nevins-b/Convergence ) and added Google DNS:”
Interesting, in effect it turns google into a perspective notary. Although one with probably higher propagation lag (when a site owner legitimately changes a certificate).
I wonder what the practical implications are from a user perspective: “Oh this site is not working through convergence, but I still need to do my banking. I’ll give it a shot with standard SSL authentication…”
Edited 2011-09-09 10:05 UTC
I think it is like a spam filter, this will only work if you have enough rules. Only then it will do scoring properly.
That is why I wonder if this would scale.
It works really well for people with atleast some technical knowledge.
I wonder how well this project would scale.
Would ISPs setup a notary(proxy) for their customers ?
http://threatpost.com/en_us/blogs/microsoft-revokes-trust-five-digi…
FYI,I’m not the only one who no longer trusts them.
Edited 2011-09-07 20:13 UTC