Slashdot Mirror
What Would It Take To Have Open CA Authorities?
trainman writes "With the release of Firefox 3, those who have been using self-signed certificates for SSL now face a huge issue — the big, scary warning FF3 issues which is very unintuitive for non-technical users. It seems Firefox is pushing more websites in to the monopolistic arms of companies such as Verisign. For smaller, especially non-profit groups, which will never have issues with domain typo scammers, this adds an extra and difficult-to-swallow cost. Does a service such as this need the same level of scrutiny and cost since all that is being done is verifying domain and certificate match? This extra hand holding adds a tremendous cost and allows monopolistic companies such as Verisign to thrive. Can organizations such as Mozilla not move towards a model that helps break this monopoly, helping establish a CA root authority that's cheap (free?) and only links the certificate to the domain, not actual verification of who owns the domain?"
try it....
Can't the World Wide Web Consortium take over the job? Of course, Verisign will be all against it as it breaks their monopoly ...
Help a man when he is in trouble and he will remember you when he is in trouble again.
When Google Checkout came along, I figured I'd accept that too - so I started doing scripts on my web site to take Google Checkout payments.
This came to a screeching halt when I realized that Google Checkout payments (or at least automated CGI processing of them) would only be done through web sites with SSL certificates signed by one of the "Major Authorities".
I wasn't willing to shell out $100 (about half my yearly profit!) for the stupid certificate.
This FF3 problem is even worse - if you use SSL, your web browser would be screaming to your end-users that you're probably dealing with some hokey-untrusted individual!
Let's just say that in any respect, I won't be having any little buttons on my site recommending that people use Firefox...
First of all, what does this certification crap prevent?
I go to randommalwaresite.com, I get a certificate for randommalwaresite.com!
HURRAY!! Everybody is happy. WTF?
I run a small norwegian forum, and we use SSL. Since our income is around 100USD a year, which is donated by members, it would be very unfair to spend all of that on a SSL cert. However, how can one explain that there is no security risk involved in creating an exception when the browser so fiercly states that it is a huge security risk? It would be better if you just got a warning like "This site is probably not your bank"...
Assembling etherkillers for fun an profit
SSL certs are a great source of revenue. Why would someone want to make a free one.
To create a free one you would have to get Microsoft to agree. They would never do that for say Mozilla "which would a logical choice to do this."
I don't think Microsoft would do it for Google.
It is a way to print money. I wonder just how much revenue Microsoft and or Mozilla get from the different CA root Authorities?
See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
or create your own CA with a link on the http site to install that root cert on the browser.
$27 a year? (GoDaddy) $50 a year? (InstantSSL) etc.
Sorry, but if an organisation can't swallow around $50 a year then they have more serious problems that wanting SSL.
someone with a stuffed wallet. They essentially would have no more room in their pocket to earn money from people who simply want want credentials on their verified, secure web site. Unfortunately that isn't happening soon.
Godaddy has a very simple SSL cert option that only validates that the certificate issued matches the domain registration info, which is super cheap.
One entire point of SSL is to ensure that the user can trust the site they're connecting to. If I register citicardbank.com, my inability to get an SSL certificate for it without being traced by my phishing victims severely undermines my ability to rip people off.
The only way to get what you're asking for is to get a secondary protocol, somewhere between HTTP and HTTPS, that would provide privacy for the communication link but wouldn't promote the notion that the end domain is what it says it is. Whether such a thing is a good idea is open to question, even if it is desirable.
If push comes to shove, the only problem with the present regime is that it's expensive. There's increasing amounts of competition in that space, so you should expect prices to come down over time. Wait. .com domain names once cost more than what many SSL certs do today.
You are not alone. This is not normal. None of this is normal.
Why is this being brought up now as something new? IE7 has been doing practically the same thing since it was released. I agree that there should be something "open source", but this is far from new...
Are we talking about some sort of meta-CA or does the submitter have a stutter?
Someone could run a service where sites can list themselves to be verified... That way bank sites can still give the big scary warning if the certificate does not check out AND smaller sites can use self-signed certificates...
The real problem would be to get a neutral and secure way to host this site... (The current SSL method of verifying a site's identity might work in most cases...) In addition, administrators that add domains need to prove that they own the domain... Verification of this site is VERY important to protect against DNS based attacks...
The fact that there are "compan*ies* such as Verisign" means Verisign is not a monopoly. In Firefox, go to Tools, Options, Advanced, Encryption, View Certificates, Authorities. These are all valid CAs according to Firefox. As for being cheap, a quick check at GoDaddy's says you can get one from them for $30/year.
Anyone know the IE status on this? Did they buy themselves out of a warning, or some such? It's totally down Microsoft's alley to trick Firefox into screaming "LittleGuy.com suxxors t3rr0rIsts" while IE cruises along, users shrug and say "uhh... well, works for me when I use MS..."
My first Journal Entry ever, in 8 years! http://slashdot.org/journal/365947/aphelion-scifi-fantasy-horror-poetry-webzine
For all but the biggest transactions, most people couldn't care less about what the certificate says. Really, how many people check the certificate on, say, PayPal, to see that it's actually owned by them?
I'm all for breaking the monopoly of current root CAs, but for the most part, that's already being undertaken over at OpenCA, which is indeed trying to get included into major browsers. (Last I heard, they had problems with IE, but Mozilla and perhaps Apple were willing to let them try if they had several audits, among other things.)
Perhaps a better solution would be for Firefox 3 to detect self-signed certificates (separate from expired, or wrong-domain certificates) and warn the user that there's no good way to be sure that the people running the website are who they say they are, but that if all they want to do is connect and have an encrypted communication, have a simple (but slightly scary) button to proceed, once per session. That of course wouldn't protect against man-in-the-middle attacks, but that's the reason the root CA infrastructure is in place. Getting something like OpenCA in more browsers is probably the best (only?) fix for that.
While CA-validated certs are still somewhat stupid (my site is just as encrypted self-signed or not, though I see the points on the site of having CAs), namecheap.com does offer somewhat cheap SSL certs -- I've used them and it's been OK for simple stuff like adding a cert to my mail.* mailservers and such.
-Matthew Riley "TofuMatt" MacPherson
I have a website
FF3 appears to have these as an authority by default.
http://cert.startcom.org/
StartCom, the vendor and distributor of StartCom Linux Operating Systems, also operates MediaHostâ, a hosting company, which offered its clients, SSL secured web sites with certificates signed by StartCom for many years. That's where the idea originated: Free SSL certificates!
I think the optimum solution would be a cheap root CA who is also highly trusted.
I don't know who this would be - maybe someone like a traditional brick-and-morter "bank" which could vogue for an SSL certificate being validated in the same way that are able to link a bank account to a person, company, SSN, etc.
I was going to say also someone like Google.
The point is, if a CA-signed cert was $5, no one would be complaining, but if any 'ol shmucks signed certs were automatically accepted by your browser, the whole system wouldn't mean anything.
Can organizations such as Mozilla not move towards a model that helps break this monopoly, helping establish a CA root authority that's cheap (free?) and only links the certificate to the domain, not actual verification of who owns the domain?
How can anyone possibly establish that a given certificate is associated with a given domain without first proving that they do indeed have the (ownership) rights to establish said association?
What you are asking for can be accomplished via SecureDNS, you can enter the hash of the certificate in the DNS entry and Secure DNS ensures that only the authorized party can enter that association and verifies that it was not changed. SecureDNS facilitates a lot of these kinds of authentication issues by extending the rooted hierarchy of DNS names to securely dissiminate information, whether it be IP addresses of servers or public key commitments. See my paper "Layering Public Key Distribution Over Secure DNS using Authenticated Delegation" (ACSAC 2005).
It sounds like some people need to educate themselves on security and the reasons for SSL in the first place. Also take a look at the current situation on the internet - for example how do phishing sites currently operate?
One of the biggest reasons for using or trusting SSL is that you can trust that the website is who they say they are. If you give out certs without validation, you're not helping the community at all.
If you think just encryption is enough, you're wrong. People are rarely defrauded because their packets were intercepted. Using encryption on the internet is like using a armored car to deliver $5 from the man on a park bench to the hotdog stand on the corner. The endpoints are the biggest security problem these days.
All of the phishing attacks have to do with sending you to a malicious site that convinces you to enter your information.
There are cheaper SSL certs out there than verisign, do some shopping around.
Firefox is not trying to harm a small site. They are trying to protect the community from the phishing attacks.
You keep using that word. I do not think it means what you think it means.
There are more Certificate Authorities than just Verisign; e.g. Thawte, GeoTrust & GoDaddy.
GoDaddy charges $15/year for a single-domain SSL cert.
1. Step 1 - FACTOR algorithm in polynomial time
2. Step 2 - SSL is obsolete, and certificates are pointless
3. Step 3- PROFIT!
This is my sig.
The certification authorities really need to get together with the web browser vendors so the big scary warnings can be made trust-level-appropriate.
For example:
Domain confirmed: [green][yellow][red]
Responsible Party Identity Confirmed: [green with seal][green][yellow][red]
Where "yellow" meant unconfirmed or self-signed and not whitelisted SSL or an easy-to-fake or -steal ID such as a credit card, "red" meant revoked, expired, or invalid credential, and "green" meant a valid SSL or hard-to-fake or -steal personal ID such as a driver's license backed by a notary. "Green with seal" meant a financially-backed guarantee, something big banks would probably get.
Most small-time web sites would be either green/yellow or yellow/yellow, depending on if they had self-signed certificates.
The cost of a "no identity confirmed" green/red certificate shouldn't be much more than domain registration. A "yellow/red" self-signed certificate would remain free.
If people expect "green with seal" when dealing with major financial companies, "green" with most businesses, and "yellow" for personal web sites, they'll give the appropriate level of trust.
Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
"With the release of Firefox 3, those who have been using self-signed certificates for SSL now face a huge issue â" the big, scary warning FF3 issues which is very unintuitive for non-technical users. It seems Firefox is pushing more websites in to the monopolistic arms of companies such as Verisign. For smaller, especially non-profit groups, which will never have issues with domain typo scammers, this adds an extra and difficult-to-swallow cost. Does a service such as this need the same level of scrutiny and cost since all that is being done is verifying domain and certificate match? This extra hand holding adds a tremendous cost and allows monopolistic companies such as Verisign to thrive. Can organizations such as Mozilla not move towards a model that helps break this monopoly, helping establish a CA root authority that's cheap (free?) and only links the certificate to the domain, not actual verification of who owns the domain?"
There was this one word which means "exclusive ownership or control", but I can't remember what it means. Can anyone help me out?
The problem with SSL certificate is that what you're supposed to be buying is trust. Your 400$ is supposed to be for VeriSign to validate that (a) an entity of that name/address pair exist; and (b) there's supporting evidence that the applicant represents that entity.
The reiterate strongly: Certificates are about authentication not encryption!
This isn't cheap, it requires a fair bit of effort.
Also, the CA needs to be trusted in the first place. That's very gray, but even old VeriSign is a lot more trustworthy then "Joe Q. Random Computer Service Associates" with a PO Box in RU.
Most proponent of "free" CAs really want the little padlock without any concern about trust because they implicitly trust themselves. Suppose you did have a shall-issue free-for-all CA on the web. What value would you place on its certificates? Would you trust that entity to not have a compromised private key?
I manage a small ecom site for my father's company. He's being using a shared cert provided by his hosting company (free with the hosting account) for the checkout portion of his site. That was wonderful until IE7 came out and started shouting a frightening warning full of red Xs at the user.
But, I put in a little message to the site for IE7 users and we carried on.
Now, it seems FF3 will shoot the same bullet at us, along with MANY other small sites who can't afford the cost of a certificate.
The unfortunate thing is that this will likely make him give up on the site altogether. While it is a code-beast, it is still a nice source of extra freelance cash for me, and a part of his business.
This just kills me. I don't know ANYONE who actually checks who a certificate is signed to. As long as it's there, and you know what site you're on. That's all you need.
I'm really disappointed in this news.
They offer certs with domain validation for free. There are gentle attempts to upsell you to higher levels of validation, but their domain validated certificates work without errors. Look here.
If you want certs that are validated to your business' identity (instead of just your domain) and don't indicate in the DN that they were free, there is a small charge.
.sig: file not found
There's such a thing as a non-technical FireFox user? I've never met one; it almost seems to be reserved for people who "get it".
You don't get security if you switch of your brain. Something like I-refuse-to-think-but-want-to-have-it-secure ... forget it!
If I understand it right, the expensive authorities put some effort (do they?) into checking the identity of some person applying for a certificate. You pay for this work and on the otherhand you get a certificate which most browsers can verify immediately without shouting loud.
If you make yourself a certificate not using the those authorities, you need explicitely tell the browser (once) to accept the certificate. It is in my opinion good that the browser shouts quite a lot, because this makes people think a bit before they accept it.
Now would you think that a low buddget CA authority could/would provide the same trust as the more expensive ones? Would you trust it so much that you would automaticaly accept all certificates from this CA authority?
We need a Firefox extension that will add a toolbar under the location bar to always show who owns the certificate. Maybe also do a whois query and show who owns the domain.
IMHO Firefox should have a bar pull down from the top like the password saver or the pop-up blocker warning you its self signed. Enough to let you know, but not too much to disrupt you from actually using the site.
The current ominous warning is a bit much I think.
SecureTrust and XRamp were the most common I saw at a Web host.
A lot of people brought their own GoDaddy signed cert.
My last employer used Equifax signed certs.
I've seen a few RSA Security signed certs.
Verisign is the big name; Linux is the big name in not-Windows but you see a lot of Apple. Which company has plurality, and is it more than a percent difference from the runner up?
On a more th
Support my political activism on Patreon.
Not to mention there are a bunch of second level CA's that are very reasonably priced. I think trainman needs to do a bit more research. If you can't afford GoDaddy's prices, I don't think you really need to be concerned with your customers freaking out.
--
Luck is just skill you didn't know you had.
i don't want understand the people anymore... #$%#@#@!@#!
Keep an eye out on good bargains. Once in a while, CA's have really good deals to get some fresh customers. You can get certificates for as low as 10-30/year for up to 7 years. Still not cheap, but for a signed certificate that doesn't need to include fancy insurance/identification and such, 63 for a 7 year cert is a good deal.
It only takes one man to change the Wisdom of the Crowd to Tyranny of the Masses.
the foremost aim of an SSL cert is to encrypt the communication so 3rd parties cant eavesdrop.
it doesnt make a ZIT of difference if the site you are shopping from has a Verisign signed 256 bit certificate or a self signed certificate. almost all certs are encrypted with similar technologies encryption wise. if you are concerned with 'authenticity', you dont know a website or dont trust them or suspect them, you should NOT be shopping there in the first place.
yes, this move of firefox 3 is a VERY bad thing. it really pushes people to the arms of verisign, geotrust (which is verisign) and so on.
not only that, it will also force control panel companies like cpanel, which serve millions of website users through web hosts to have to force users of their services to pay for SSL certs for each server they use or let their users connect to their site control panels through unencrypted connections. that will eventually drive up prices in the high to mid end hosting market. which is BAD, since majority of people host their websites in such small business hosts with $3-4 bucks a month. the overall effect that will have is yet to be seen.
yes, this was a stupid move by mozilla team, unfortunately.
Read radical news here
This FF3 problem is even worse - if you use SSL, your web browser would be screaming to your end-users that you're probably dealing with some hokey-untrusted individual!
If you're not willing to lay out as little as $15 for an SSL-Cert that will work on FF3, you are a hokey, untrusted individual!
Conformity is the jailer of freedom and enemy of growth. -JFK
Instead of relying on centralized CAs, and implicitly trusting these privileged monopolies, we could shift to trust webs.
It's like a social network. You trust who your "friends" trust, and distrust who they don't. With weightings, so some friends' and enemies' associations (and dissociations) count more than others Because some people you trust in their content, but not their judgement of who to trust (and vice versa, but probably more rarely).
Trust webs can perfectly simulate the current centralized trust model. You can just set your trust web to always trust whoever, say, VeriSign trusts, and ignore everyone else, which is what we get by default today. But you could tweak your trust web to say "If my grad student distrusts a site, then ignore whether VeriSign trusts it".
Such a trust web could therefore just ship set up with the current CAs the only trusted authorities, and work exactly the same as now. But we'd each have the freedom (or our sysadmins, who could lock the trust web changes away from normal users) to emphasize whoever we actually trust to influence our automated trust.
Independent authorities could "watch the watchers". So investigators with a reliable track record could become important "second guessers" to the "offical" CAs. People could make their reputation by proving a trusted authority has less than 100% good judgement. And the whole system can become more robust, instead of fracturing as soon as different CAs have different trust levels for different sites.
The technique and some SW is already available, for apps like PGP and others that rely on a Public Key Infrastructure. What's necessary for the critical mass that makes such a system work is for a browser like Firefox to upgrade to a trust web, with an easy and reliable UI with sensible defaults. Then we're as strong as the trust network in which we embed ourselves.
--
make install -not war
IE has the same problem. In fact they were first to the table with the over-the-top warning.
It's especially hard on vendors who sell browser-based applications which run locally. The customer wants SSL, even on their local network, and even for non-sensitive data... But then they go to their local machine and get a big warning from Firefox or IE that their connection is insecure... But they don't want to pay for a certificate.
I assumed Microsoft did it to reduce competition for native and .NET apps from browser based apps, but I don't know what Mozilla's reasoning is... Just to copy IE, maybe?
The O.P. mentions "...monopolistic arms of companies such as Verisign."
Okay, look. The word "monopoly" has as its prefix the stem "mono-," from the Greek, meaning "one." That means there can only be ONE "monopoly."
A phrase such as "monopolistic company LIKE Versign..." is absurd on the face of it. If there are other companies LIKE Verisign, then there is no monopoly.
Is it REALLY that hard to understand?
This is an example of how the rising generation is so used to "buzz words" chosen for shock value, etc., and has gone completely away from clarity of speech and writing. What the O.P. means to say, really, is "I don't want to pay the going rate for this service, so I'll call Verisign 'a monopolistic company' because everyone knows 'monopolies' are bad, and that will communicate the 'badness' of 'companies like Verisign.'"
Oddly, the word "rhetoric," also from the Greek (rheteros, "a speech") used to be a positive appellation for the study of good, clear communication of thoughts and ideas. But it has also succumbed to the buzz-word dementia, and now usually means "empty words."
How sad.
Any technology distinguishable from magic is insufficiently advanced.
Looking at it again, it's just crap UI.
Let's assume there are still two or three people on the planet who don't use Firefox 3 and consequently have no idea what big scary warning you're talking about.
Also let's figure that those who are using self-signed certificates are at least somewhat likely to fall outside the ranks of "non-technical users".
IE 7.0.5730.13 Shows a drawbar on top with a Blue Shield and a pink page: Content was blocked because it was not signed by a valid security certificate. For more information, see "Certificate Errors" in Internet Explorer Help.
èåæç©
There are already plenty of providers selling crap "domain control only validated" certs. We (as SiteTruth) regard those as having no value, and we encourage others to do the same. If it doesn't have an "L" (location) field, it's worthless. The introduction of those crap "quick SSL" certs poisoned the whole cert industry.
It's a problem that certificates which verify business name and address cost too much. They ought to cost maybe $25 per year. Validation isn't that expensive. That's what registered mail is for.
There used to be some enthusiasm for "web of trust" schemes of certification, but since the bad guys organized into criminal networks, domain farms became popular, and it became easy to get phony GMail accounts in bulk, that approach is obsolete.
IE's warning is, if anything, even more scary. It does, however let you override it after clicking through a few warnings saying it isn't a good idea.
There's really two different types of certificates here:
You need one that verifies your real-world identity. You have to have some degree of technical knowledge to understand that the name in the Internet domain is absolutely worthless to establish trust. When I visit my bank's web site, regardless of their URL, they should be able to present some kind of certificate that clearly establishes that web site to be my bank's. It should be labeled and effectively impossible to spoof.
You need one that strongly connects the DNS hierarchy. We already have a single, trusted root in the DNS world. Why don't we layer automatically-generated certificates on top of that? The root signs TLD certs. The TLDs sign second-level domain certs. The second-level domain owners can sign certs for whatever they want beneath that. This requires very little extra work, especially compared to getting an SSL cert from a CA today. This can be completely automated, since you don't have to do any real authentication beyond what you've already done by giving out the domain assignment. This cert shouldn't have any information in it except for the domain name.
The latter class of certificate could even be trivially extended to include things like e-mail certificates (validating only the e-mail address, not the identity of the person owning it), or anything else that's based on a DNS name. All for far cheaper than what we have to pay today.
This needs to be transparent for it to work. You've already lost the vast majority at "root cert". They have absolutely no fucking idea what you're talking about. That isn't going to change.
If it's not in the default install, it doesn't exist.
The article suggests that CA's should provide certs for free. How do you propose these CA's stay and business??? There's no free lunch but somehow many web users (who likely have no experience running a business) forget that it costs *money* to provide a service, pay staff, pay rent, pay utilities, pay payroll taxes, etc. The typical response is something like, 'pay for it with advertising'. Well, for something such as a CA that will breed distrust and suggestions of conflict of interest (and likely such a site wouldn't get enough traffic to pay for even 2-3 people a full-time salary! Perhaps it should be a government provided service? well, try and get that through congress.
In the end, it costs money to verify and authenticate an organization, install and maintain the hardware and software to relay authentication requests to browsers, etc.
Firefox 2 popped up a warning dialog on self-signed certs, and that was just fine. People who didn't know what the fuck it was all about could just click through.
Those people aren't going to be able to navigate the process to add a cert in Firefox 3. It's effectively banned encryption for non-technical users.
Hard to say which is a more implausible conspiracy theory: this or the guy on the AMD story speculating that AMD's earnings report is a news plant from Intel.
What I'm listening to now on Pandora...
If anyone can get one, how do you verify whether they can be trusted or not? I thought the price put a premium on keeping out the riffraff?
The problem with this way of thinking is that SSL certs are used to address more than one threat model. For certain threat models, the degree of verification offered by organizations like OpenSSL and Verisign is worthwhile (assuming you actually get it). For others, it's not.
But consider this. Suppose Mozilla adds support for the free certificate authorities that only verify that you own your domain. So it will accept these certs and put a lock icon where it belongs.
The DNS is not secure. That is, the fact that you appear to own your domain is not securely authenticated. So this means that in theory, someone can spoof the person who is checking your cert request. That makes that person's machine the weak point in the chain.
How much would it cost to spoof this person? Well, you find the one who's least careful. You get control of his or her internet connection. Maybe it costs a couple of thousand bucks. And now you can generate a cert for any domain, and Mozilla will accept it without question.
So for your web site where you're just trying to keep passwords private, and not really worried about a serious cracking effort costing thousands of dollars, having that free CA is a great deal. The problem is that in order for you to get what you want, all the sites that actually need real security have to give it up.
So why not use the PGP web of trust? Validation by real people rather than the lowest bidder?
Vik :v)
And the profit is $1 million, as you would have solved P=NP.
Actually, it's not been proven yet that FACTOR is np-complete. If I understand correctly (and please a real Phd in Comp Sci step in), you need to show that an NP-Complete program could use your program you are trying to show to be np-complete as a subroutine to calculate itself, and in the case of FACTOR, no one has actually done it.
Still, you're right in the larger sense that any polynomial time solution to FACTOR would be a breakthrough that should shed large insight into NP-Completeness problems.
This is my sig.
... that even if open, a Certificate Authority Authority is really what you want.
What we need is a more standardized system for storing certs locally. The centralized system is probably preferable for banks and such, but for most, a simple "Install this certificate" dialog would be sufficient. That reduces the man in the middle problem to just the first access, which is just as secure as ssh.
It also arguably has advantages over the centralized system, since you know exactly why you're trusting the site. With the centralized system, you're just assuming that the certificate provider is legit. Compromise or spoof a certificate provider, you potentially compromise many sites. Spoof one site that has provided certs to its users, users all get the nasty error message. And that's about as good as it can get.
Of course, every time I've gotten a key-mismatch error message of ssh, my response has been to delete my key file. So it may not allow much more security, but it gives the user more room to make an intelligent choice without reducing usability.
Any distributable app with a web interface that wants encrypted communication needs to generate a self-signed certificate. This is a case where having *any* third party sign a certificate is simply not feasible. And most users can't be expected to know how to generate their own cert.
It was bad enough in the past, with the increasingly alarming browser warnings, but now FF3 (in some cases) *refuses* to connect without even giving the option of a bypass. The only solution is to dig through 6+ layers of preference menus/tabs to add an exception via a completely un-intuitive UI.
The *real* solution is to de-couple encryption from host verification. It's silly to jump through all these hoops when all you want is an encrypted connection.
There are many lower-cost SSL certs out there. GoDaddy has standard SSL certs for $29.99 /year. Is that cheap enough for you? Or is there some new functionality that FF3 has that no cert now has? (I'm not talking green bar/EV certs either)...
-m
http://www.invisik.com
> "monopolistic company LIKE Versign"
So you are saying only one monopoly can possibly exist at one time? So if the original Bell monopoly still existed and it was decided that Microsoft was a monopoly, suddenly Bell would no longer be one? Could you not say "Microsoft is a monopoly, like Bell."
http://en.wikipedia.org/wiki/GoDaddy#Controversies
This is to say nothing of a number of lower profile controversies and the fact that their entire site is a usability nightmare that seems largely designed to trick marginally informed customers into buying (and cause more savvy customers to explode in frustration).
Tweet, tweet.
Why would the Firefox people do anything about it when they charge good money to include root certificates in Firefox? In fact, the way Firefox has been pushing these stupid "extended" certificates and pretending self-signed means "OMG guaranteed scammer, run!" only shows how much they are in the CA's pockets.
PKI is, like, the man trying to keep me down. Skroo you verisign! I dont need to know that people i talk to over ssl actually are who they say they are. That's for, like, the man!
but the recipient is having problems getting phished or easily deceived into believing some other site is in place of what she has specifically asked for, there are not much you can do.
Read radical news here
You think Verisign et al reliably do that? How?
There was a /. story maybe a year ago about all sorts of obviously fake ones... what the major cert providers verify is that your payment cleared. Which is _something_ because there's SOME kind of traceability. But it's not much.
I don't really blame them, though, because the problem is fundamental. There's just no real way for them to verify someone is who they say they are, because we don't really have a definition of who that "we" is. It's not like the gov't issues you a social security private key at birth and each corporation too (not to mention going international)
So the thing keeping them secure IS the payment and the record of the payment, and the fact that so many people fall prey to phishing without a valid cert that no one cares.
*****
In my opinion, the best we can do is issue physically linked certs. Cryptographically identical to existing certs, this changes the people part - The certificate authority a) must require a payment, but there's no minimum they can charge b) mails a physical letter with a code c) makes an automated, repeating voice call with a code d) if both codes are entered and they own the domain, issues a cert for that contact info, which can optionally be used to generate certificates for multiple servers.
Now, the hard part is that you haven't verified IDENTITY at all, you've only verified contact information. So the browser would have to literally display this information, if it was one of these contact-certs (perhaps in a bar just below the URL bar) I say in 'these certs' because for these certs you're not even implying that you can trust anything except the
You COULD set this authority up with a relatively small expense. You might be able to write a FF extension to display the addresses. If you have reasonable internal security, you probably could get FF to add you as a trusted authority, at least FOR contact-certs.
That's not GREAT, but it's the best we can do for simply automation for general-purpose merchants/certs... beyond that it's trying to do credit and background checks the old fashioned way.
My only OTHER idea is that the FDIC/NCUA/etc ought to get together and create a CA for US banks. Then you could even make the bank-trusted bar a DIFFERENT color. And presumably the regulators have a secure way to talk to the banks. (I'm not suggesting that this be legally mandatory for the banks to sign up for or use, but I think there's no one who is more likely to be able to authentically verify the authenticity of a US financial institution than the US regulators...)
Looking for freelance Actionscript (Flash/Flex) or ColdFusion work and/or freelance developers. Email me, put Slashdot
So I'm sorry but I don't agree with that. First the warning from IE is much more accurate and non alarmist. It is different from the failed page message, while the FF3 ssl and fail to load page are very similar. This is rather annoying and the first time I saw it, I thought the URL was invalid. Also FF3 lets you visit the site as well.
Charles Wyble System Engineer
Isn't there some kind of free to the user service that can link a domain to an IP address so that when you "dial up" that domain you can be sure it always goes to the correct IP address and no other?
What?
I think FF3's cert thing is lamer and lamer
I've been thinking about this... and I'm happy to have FF3 mark the unsecure, secure, and EV-secure sites differently. But it's really, really lame to say that any self-SSL site is WORSE than a random non-SSL site. It's only the same. If they're going to go through the trouble of getting people used to trust markings, they should just mark the self-SSL sites like they mark the unsecure sites. Changing the URL bar to say:
(unverified) https:///
Would be enough, if they were changing the color/style of the secure sites. (Sure, don't give the self-SSL a lock icon. Fine.)
Looking for freelance Actionscript (Flash/Flex) or ColdFusion work and/or freelance developers. Email me, put Slashdot
Other people have said it in nicer ways, so mod me redundant, but this is one of the more uninformed posts I've seen here lately. Real certs do two things:
The warnings are there for self-signed certs because self-signed certs don't do #2. Who cares about the encryption if it's not necessarily a company you trust?
I can imagine a not-for-profit CA that does #2, and maybe can charge less money than Verisign for it, and maybe that can be accepted by the browsers by default after enough time. But it would still cost some money because step #2 is not zero work. But the most important thing is that the browsers should not be pushing this forward--the browsers should wait until that CA is up, running, and proven before they do add it to their default list.
Quit simply, if it is a small, close community and you want to use SSL, have them add your and only your certificate, with plenty of ways to make sure it's fine.
The problem with CACert, is that you are extending your statement to the greater internet, not just yourself. Adding the certificate for a website is one thing, adding a CA certificate of an organization whose validation process is not as thorough (for good reason) is another.
CACert might be good *if* the browser but a bar indicating the lower barrier of entry to get one. cacert would be more than adequate for most forums, mailing list archives, and non-financial account information on sites you don't have much to lose on. For financial transactions, I would want only to participate with a site that went through a more rigorous validation process than CACerts. Such processes require manual human attention to do right, which isn't free.
XML is like violence. If it doesn't solve the problem, use more.
I'm surpised no one mentioned this, but https://gmail.com/gmail.com pops up this alert in FF3, because the certificate is actually for mail.google.com. I'm surprised Google didn't fix this - especially considering how much money they give to Mozilla.
IE7 is worse, because its user interface does not ask the user if they want to add the site as an exception as Firefox 3 does. The end result is you get the big, scary warning in IE7 every time you visit the site, but you get it only once in Firefox 3 because you need to add the exception before it will let you proceed to the site.
Anyway, get a free cert from StartSSL and the problem is solved.
What a fool believes, he sees, no wise man has the power to reason away.
OK, OK already! We get it! It seems clear from the context that the word the author intended was more along the lines of "cartel" than "monopoly." The point is still well taken. These companies exploit a captive market with inflated prices.
you can justify anything regardless of its impacts elsewhere.
- if someone does not attentively type in a domain,
- if someone is still naive enough to click links on any email arriving and give out personal info whereas they never give their wallet or id to anyone in the street if asked to
then that someone has no business being on the internet.
excuse me, but thats not discrimination or anything else. its just the way it is. you dont let such people go around the town with a credit card and a wallet and an id in real life, and try to secure back pocket of their trousers. you shouldnt break more things than you try to fix in that fashion on the internet too.
Read radical news here
An SSL certificate is for encryption of a link. Absolutely not trust and absolutely not identity. If that were the case why in the world did the major certificate vendors come up with an "EV Cert" which claims to do what you say? As opposed to the certs they already offer which obviously don't offer that trust. Otherwise they wouldn't need the new EV certs.
>My only OTHER idea is that the FDIC/NCUA/etc ought to get together and create a CA for US banks. Then you could even make the
>bank-trusted bar a DIFFERENT color.
Corporate execs will scream about it, and do whatever it takes to make their company meet the definition of a "bank" just to have the color code.
-fb Everything not expressly forbidden is now mandatory.
I or anyone else can make a public key, and claim it belongs to GW Bush. The process of distinguishing the real public key for GW Bush from fraudulent pretenders is called "identification". True, many/most paid certs don't do a very good job of it, but a public key is worthless for preventing "man in the middle" attacks unless you have some form of confidence that key A belongs with person A.
This is a contender for dumbest statement in the history of security.
If Thawte and it's owner Verisign are so evil, why does Thawte offer FREE certs based on the web of trust model? Let's not get our certs in a bind here..
"For smaller, especially non-profit groups, which will never have issues with domain typo scammers, this adds an extra and difficult-to-swallow cost."
How terribly naive. non-profits are as target rich as any other company for domain typo and other scammers. Why wouldn't a scammer steal from the generous people that donate to non-profit causes? The small non-profit is far less able to track down and pursue justice against scammers and their donors are a generous, trusting bunch.
Your ISP could do it.
Or their ISP.
Or any one of the dozen hops between you and www.somebank.com.
Or anyone who happens to be in a non-switched-network (on a hub) between you and www.somebank.com -- in some cases, this means anyone else on your ISP.
Or anyone on the same wireless access point as you. (A guest over at your house, or someone in the same coffee shop.)
To anyone who thinks SSL isn't required, or that man-in-the-middle attacks on certificates aren't feasible, look at the above list and ask yourself if you trust all of these people.
Not only if you trust them not to pwn you, but if you trust them to keep their own machines secure enough so that no one pwns them first. (Maybe it'll be your friend's spyware-laden Vista laptop on your otherwise-secure wifi?)
What a properly-signed certificate means is that you have a much shorter list of people you're required to trust with your bank information -- that is, yourself, your own software, and VeriSign. Sucks that it's so centralized, but it is much better than being owned the next time you're on public wireless.
By the way, for those who don't know: GMail sessions can be hijacked if you don't use SSL. But to use SSL, it's trivial -- all you need to do is use https://mail.google.com/ to login.
Don't thank God, thank a doctor!
... is to drop the fundamentally broken X.509 PKI infrastructure, where any CA can sign certs for any subject, and switch to a DNSSEC-based PKI where signing authority is limited to subdomains of the authority. In the process, we end up with the ability to sign all the certs you want, for every host, if you like, and have SSL anywhere.
http://cert.startcom.org/
Company in Israel has the ideal to give out free certificates and last time I checked (about a year ago) they were on the brink of being accepted into the ranks of Verisign and similar companies (without the crookedness of course)
Everyone seems to be going on about authentication as though that were the most important function provided by SSL/TLS.
Personally, I've always felt the encryption was much more important and useful. I'm usually far less worried that the site I'm connecting to is an imposter than I am about people using the same public wireless access point "sniffing" my traffic.
I keep wondering whether it'd be possible to use PGP encryption instead. Obviously there'd need to be some browser-side support for this, but for situations where authentication is less worrisome than encryption, it would at least get us away from the "pay Verisign for permission to encrypt your website for a year" model that we seem to be headed towards. (I'm sure government agencies around the world love this idea - then there are only a few central "certificate" companies they need to flash a badge at to get a fake certificate for your domain name. Hey, if you're not into child pornography or terrorism you have no reason to object to this, right?...)
Failing that, at least allowing a much simpler "accept self-signed certificates" configuration option, with the "lock" icon indicating "your communication is safely encrypted, even though we cannot guarantee the identity of the website you're communicating with" rather than the Big Scary Pop-Up Window interface they're using now.
Hacker Public Radio is our Friend
This bullshit is just training users to accept any and all certs, making it easier to get a signed SSL cert and do MITM, then just rely on the user to click through the mismatch. Fuck you, Mozilla, you're making stuff worse at this point.
Okay, look. The word "monopoly" has as its prefix the stem "mono-," from the Greek, meaning "one." That means there can only be ONE "monopoly."
A phrase such as "monopolistic company LIKE Versign..." is absurd on the face of it. If there are other companies LIKE Verisign, then there is no monopoly.
OK then it's an oligopoly. Same difference (How's that phrase for clear communication?)
You have a small club of companies that can issue certificates. No other company can enter the club. The oligopolistic companies have almost all the advantages a monopolist would have. This point, at least is valid in the article.
Oddly, the word "rhetoric," also from the Greek (rheteros, "a speech") used to be a positive appellation for the study of good, clear communication of thoughts and ideas. But it has also succumbed to the buzz-word dementia, and now usually means "empty words."
Uh, please check your facts before you post. The word is Greek, but comes from rhetor meaning orator or teachecr of rhetoric. Rheteros does not mean anything in Greek or English as far as I could find.
How sad.
Indeed.
Two instances of verifying X.509 digital certificates used in SSL come to mind that led me to deny the online "vendor" their attention and/or business:
1) unknown vendor, first time visit, using SSL; checked certificate via (Mozilla) browser "check certificate" method; found a known CA; attempted to verify vendor cert through known CA - no can do; have to be vendor to verify (?!?)
2) another unknown vendor, first time visit, using SSL; checked certficate (as before) and found UNknown CA; checked certificate Policy URL, et cetera, to get some viable URL to CA; CA turned out to be an intermediary so attempted to track THEIR CA; eventually wound up with NO verification of vendor cert; sent them mail about lack of plausibility of their cert due to lack of verifibility of it and its CA(s); vendor had no clue about anything
Personally, I don't accept the "trust your friend, web of trust". Did your "friend" actually verify the authenticity of whomever? Did YOU grill your friend about verification? What means of authentication were used? Et cetera, et cetera.
Having a "provable chain of authenticity", which is partly what CA's AND certificates are supposed to provide, can be just a slippery, but there is at least one or more levels of authentication that can be verified (if everything is done correctly, of course). THEN it is the individual's responsbility to decide where to go from there.
Even if you have nothing to hide now, one day you may, and then you probably don't want to advertise the fact by sudden conspicuous switch to encryption.
You might want to encrypt everything possible simply to make life harder for those who listen everything in the hope of catching something valuable.
as their certs are free some people do user them to demo systems like www.test34.example.tld and when done revoke them {i know i use them for public beta's of ssl based systems before they get re-certed by the end-customer and used live} like during the look/feel testing and the development} so could easily see many revocations due to this
Why is encryption a good thing? I assume you want to prevent someone from intercepting or modifying your traffic, but if the end-point is unverified you might have a secure connection right to the eavesdropper.
Look up "Marketscore" from a few years back, they made a business of intercepting SSL traffic and reencrypting it out the other side.
No, that should be even easier, just set up a local root certificate rather than use self-signed certs. Each user installs that root (or has it preinstalled by IT) and everyone can connect securely to all your internal site with no warnings.
The purpose of only trusting root CAs like VeriSign is the trust that the CA did the verification work. Obviously, in practice, it doesn't work out that way. In theory, trusting an issuer like CAcert by default defeats the purpose of the system. The system is setup so that the CA verifies the identity of the person purchasing the cert(in theory) before issuing the cert. CAcert doesn't do that for the most part(unless you participate in the Web of Trust for which the nearest Assurer or Trusted Third Party may be quite distant and therefore unavailable). In theory, including CAcert and other CAs like it would be a huge mistake. In practice, it would be a good idea because from what I hear, VeriSign does as much verification as CAcert sometimes so why not include CAcert?
The main issue is, quite simply, that an SSL cert is used for tow things at the same time which don't always need to be in sync.
First, an SSL cert is a public key to start up an SSL session (public key begets public key begets shared symmetric key and hey, presto - we have an encrypted tunnel).
In addition to that, the SSL cert serves as a site identifier, and this is where the problem starts because that requires a chain of trust to work. I personally would prefer a system almost like the Web Of Trust (www.thawte.com/wot) so that there is a DEGREE of trust that can be injected into a cert, but not as much as is presently the case.
You see, there is nothing that tells me I can trust the CA itself. Why should I trust Verisign? At the point where Thawte was sold to Verisign, Thawte was IMHO both a LOT more secure in the way it issued certs as well as significantly more efficient, so if anything I would have trusted a Thawte cert more for server identity than I trust Verisign (not to mention the fact that Verisign being US is now another recipe for mistrust, but I digress).
It is in the interest of Verisign to pass off their certs as "secure" but few understand that "secure" just means "the result of a process we followed" so it depends on the process - and I can't find any public, independent audits of that mechanism. Ergo, I can't determine if I can trust the cert or not, and all the marketing in the world can't change that little fact.
In addition, research has shown that the average end user cannot distinguish a safe site from a fraudulent one, even just putting the padlock icon on the webpage itself is sometimes enough to mislead them into thinking a site is safe..
Commingling the two purposes is confusing, and FF3 doesn't help here. But they're IMHO just following this trend of wrongly trusting what a cert states to identify the identity of a site..
Insert
CACert has been supported by Opera for a long, long time. StartCom is not supported by Opera, nor is it supported by Konqueror. I personally prefer CACert over StartCom, so does the CCC, Indimedia and a whole lot of others.
I briefly looked at StartCom SSL again just now. I read the FAQ. "Validations of domain names and email addresses are valid for 30 days. After the 30 days they must be revalidated.". The unverified CACerts expire after 6 months, that is so short that all the work of keeping certificates up to date almost makes it not worth it. 30 days? I would not do anything but verify domain names. Yeah, I use CaCert SSL on 12 sites. All it costs is time and 1 IP pr SSL host, so why not.
9/11: Never forget it was a false-flag operation
Actually you can only get a certificate from CACert if you've been assured with enough points, and that's only supposed to happen after in-person ID verification by multiple members.
In other words, a key signing party. As I understand it, those are practical for people who routinely travel between large cities, but that has become more difficult with the quadrupling in fuel prices in the past four years. Worse, key signing party coordination web site Biglumber.com appears to be down (Network Timeout). So other than through key signing parties, how can I meet multiple members within reasonable bicycling distance of my home? Or should I just go the notary route?
What do you mean CAcert has no accountability? They have a web of trust in place that actually checks IDs person to person. Thats more than Verisign does. All they do is charge a credit card.
By using a credit or debit card, VeriSign (and Thawte and any other CA that VeriSign has acquired) outsources this identity verification to your bank or credit union.
A CAcert server certificate does exactly what it says it should, that the owner/controller of the domain is in control of the server.
Consider this second statement: The owner/controller of this domain is authorized under other applicable law, such as trademark law, to own/control this domain. If not SSL certificate chains, what framework is designed to support that statement?
Instead of relying on centralized CAs, and implicitly trusting these privileged monopolies, we could shift to trust webs.
How does one get inducted into a web of trust without traveling hundreds of miles to a major city where key signing parties happen? Or do you expect the majority of notaries public to sign keys?
But we'd each have the freedom (or our sysadmins, who could lock the trust web changes away from normal users) to emphasize whoever we actually trust to influence our automated trust.
The "or our sysadmins" part scares me. Home Internet access is moving slowly toward a model where the ISP is a home user's sysadmin.
To my knowledge certificate authorities that offer free personal certificates often rely on self-consistent web of trust mechanisms for identity verification. The web of trust model while having the advantage of allowing the issue of free or low cost certificates is badly implemented, inconvenient and full of security holes.
The web of trust is composed of existing certificate owners, this along with the requirement of having attained a minimal level of trust within the WoT is often the only qualification required to act as a guarantor (hence self-consistent). There are no guarantees that such guarantors are trustworthy and/or competent to verify an individuals identity and using so called 'trusted professionals' outside the WoT as guarantors often incurs a fee. Also efforts to attain a level of trust sufficient to qualify a customer for a verified certificate and/or to act as a guarantor often requires two or more fact-to-face visits thus discouraging uptake of verified certificates (in favour of unverified alternatives which could never be adopted by secure browsers) and more importantly restricting the number of available guarantors within the WoT.
In all the WoT model is too inconvenient to be seriously considered by individuals and small businesses thus how can free certificate authorities hope to compete with the big boys (Verisign and Comodo, etc...). Furthermore the level of security offered by such a nepotistic network of individuals is insufficient in my opinion to warrant inclusion of root certificates within a secure browser such Firefox, Opera or the big evil e.
Thus I would propose a free/low cost alternative (or perhaps complement) to the WoT mechanisms implemented by the free/low cost certificate authorities. Many commercial authorities use the preexisting infrastructure maintained by the credit and banking industry as a means of authorizing secure payments online. Many online retailers will only accept credit/debit card payments if provided with the correct billing address and some will not deliver to any address other than the verified billing address. Other organizations make use of this secure infrastructure to verify a customers age (online casinos, porn sites, etc...) by making a nominal charge to a credit card or in some cases crediting a customers bank account (usually a few very small credits) and requiring that they input the exact value of the credit(s). Thanks to my love of eBay, apathy towards online porn and hatred of online casinos I myself am now 79p the richer and the proud owner of a verified PayPal account.
For all those to thick to understand what I'm trying to get across here (infants, people over 40, Tory voters, business studies & management students, people in a persistent vegetative state. etc...) here is a blow by blow account of how one could obtain a verified personal certificate in the future.
Also as an additional precaution against abuse by credit card fraudster (who lets face it are the most likely abusers of such a system) CACert.org could also send a letter to George W's verified billing address requiring that George W respond via snail mail, perhaps even signing a simple 'identity contract' and/or sending photocopies of his passport, drivers license, household bills, etc. All of this has the added advantage that were George W (or the very cleaver fraudster that stole George W's credit card, house keys, passport, driving license and household bills) were to abuse any certificate issued to him CACert.org would be able to relay any verified information that the
"Nothing strengthens authority so much as silence." -- Leonardo Da Vinci
It's the validation which expires after 30 days, not the certificates. StartSSL implements a two stage process for validating attributes like domains, email, identity and organizations and for actually creating and issuing certificates. All certificates of StartCom are valid for one year including the free ones (Class 1).
Opera never shipped the Cacert root at any time nor does it now. Never ever!
If you're in the UK, one of the cheapest RapidSSL resellers I've found is Servertastic, who do individual certs at 7.12 pounds + VAT and if you buy them in batches of 10 like we've started doing, the price comes down to an impressively cheap 5.50 pounds + VAT each.
With Firefox 3 now scaring the living daylights out of the end user when it comes across self-signed secure certs, spending less than 10 pounds a year isn't a hardship for anyone, even for a non-profit org that the original poster mentioned.
One snag though - Verisign bought Geotrust who provide the RapidSSL certs that Servertastic use, so I'm afraid that you just can't avoid giving money at least indirectly to Verisign (albeit a lot less than buying one of their hugely overpriced secure certs).
The word appears in my copy of "Greek-English Lexicon" by Liddell and Scott, Ninth Edition (Abridged).
Didn't anyone tell you that only high schoolers who, with their small brain, didn't have time to learn Greek properly use the little Liddell? Anyone who has studied Greek with any proficiency eventually moves up to middle Liddell, since it has rarer vocabularies while not wasting space on the "irregular" forms that any worthy student ought to know.
Here's the page where you supposedly claim that this mythical word "rheteros" appears [1]. Do you see it? I don't [2]. If I were in an extremely generous mood, I might say that you meant the genitive form of "rheter", "rheteros" (both etas, not epsilons), but in that case, you got both the meaning and the grammatical function wrong: It means "of a speaker", not "speech".
And my point stands, whether you take it or not.
And your point is completely wrong. I was going to leave it off at the little nitpicking, but your hypocrisy (especially "Nowadays, it seems "one-upmanship" is far more important than scholarship. Your reply is Q.E.D." My criticism was a perfectly valid, fact-based, well-cited criticism of a factual remark you made. That comment of yours, to borrow your own word, is "one-upmanship", you hypocrite) is so disgusting that I have to dismantle your entire argument.
One, "rhetoric" never meant anything positive. If you mean "rhetoric" as the study of public speaking, it is at best value-neutral. The way nuclear technology is value neutral---it can be used for good, or great unspeakable evil. But, from the beginning, rhetoric always carried negative connotation. Do you know who the first "rhetors" (public speakers) were? If you had any sort of education befitting your posturing, you should know: they were the sophists, or their students, whose job, in the words of Socrates (to be granted, their enemy), was "to make weak argument stronger". The function of rhetoric was solely to convince others to what you are advocating, right from the very beginning, not to mention its English version. It didn't matter whether what you were advocating was true (it may well have been). It didn't matter whether it was for public interest (it may well have been). It didn't matter if you yourself actually believed it (you may well have).
The only way one could attach any sort of positive connotation to "rhetoric" was if you were a mercenary. Why? It gave you money and prestige being the demagogue, or a servant of one (that is, if one was any good at it, unlike you, Illbay).
Two, to get to the heart of your point, no, everyone (except you, apparently) understands English perfectly. I believe by "clarity", you were simply complaining about the changing meaning of words, such as "monopolistic". Sibling posters also brought up this point, so I'll be light on it, but if the word "monopolistic" has come to mean simply "evil" (or sharing other characteristics of a monopoly), then that is the correct English meaning, Greek root be damned (and I say it as a long-time student of Greek, well into college and graduate school). It's about time that you prescriptive grammarians realized that native speakers, as a group, can never be wrong. Native speakers DEFINE the language. If they want to use "monopoly" to mean what we mean by "giraffe" now, and if the majority do use it like that, then that is the correct meaning of "monopoly". Something like that happened many, many times over the years (for example, look up the word "apothecary", which is a perfectly fine English word, and see if its present meaning has any hint of what the originating Greek word meant).
So, to wrap up, the young generation speaks English just fine (or even better so, presumably because they write more blogging and such), and you, Illbay, are a freaking hypocrite.
If you aren't a lying hy
Who the hell do you think you are to tell people when they can or cannot encrypt their traffic?
You want to add a $50/year subscription to every cheap consumer router?
The problem here is that the commercial CAs are selling an artificial trust substitute.
Do you remember something called Aspartame? Sacharine?
Wonder Bread? (Well, my dad called that a bread substitute, even though it isn't a sugar substitute.)
Substitutes don't work like the real thing, and often can cause damage when used like the real thing.
Certificates are set up in hierarchies. In most real hierarchies, I do not trust the guy at the top. In the real world, I (should) only trust the guys I know, and that maybe as far as I can throw them. I only trust each one in certain ways determined by my experience with her or him.
And, as we have witnessed in the current US political situation, when we start letting trust transit "up" a chain, we buy ourselves a world of hurt. That was what we/they were supposed to be getting away from when we/they wrote royalty out of the Constitution.
True? True.
Does that help explain why certificate use by browsers is so messed up?
There is no way for you to control the effective semantics of a certificate. That means that you effectively make the browser vendor your king.
Until that's fixed, the commercial CAs should, indeed, check all the information they say they check, but the end result is an artificial substitute for trust that is just barely enough to let your browser function meaningfully in the best of situations. Throw in a little malware, and there's not a lot of difference left between verisign and cacert.
I'd say more, but I think I'll go write a bolg about it.
Computer memory is just fancy paper, CPUs just fancy pens with fancy erasers; the 'net is just a fancy backyard fence.
-1 Flamebait, huh.
(Chucks in a couple of Karmic Jellybeans)
Then I invite someone to post something from the Developer level on why these warnings are so fierce.
My first Journal Entry ever, in 8 years! http://slashdot.org/journal/365947/aphelion-scifi-fantasy-horror-poetry-webzine
Are the servers in question inward or outward facing? If inward then push the trusted cert, if outward, spend the bucks to get an internal CA set up wich has its root with one of the trusted CA's.
Or educate your users. Provide instructions for your users to add your CA as a trusted one.
TANSTAAFL GIGO Acronyms to live by!
I haven't used FF3 yet so I haven't seen this, but if they are displaying a bigger, scarier warning than FF2 did, then that is a UI design flaw (unless they are also displaying that same warning at the beginning of unencrypted sessions).
This is merely a bug that the FF team needs to fix. If they don't ever get around to fixing the bug, then perhaps they'll eventually get forked.
Onto your question:
Yes, they could, by understanding that what you want is what some people want and some people don't. What you're asking for is both reasonable and unreasonable, depending on what is at stake for the connection. When browser makers ever get around to recognizing these different needs -- that there are degrees to which an identity can be trusted, then the "monopoly" will vanish quickly.
What will happen is that the implicit WoT that comes with the browser (the user totally trusts the Mozilla team and the Mozilla team totally trusts these CA issuers) will become explicit, and the UI will show to what degree the other side is believed to be who they say they are as well as how that decision was made.
As copyright owner of this comment, I authorize everyone to defeat any technological measure which limits access to it.