First you were talking about open hotspots. Then you were talking about https. Now you are talking about ssl. But all the while you're still just talking about monitoring and restricting the activity of 3rd parties on 4th party systems. And it seems really important to you for some reason. Please, waste time and effort and money patenting your /spyware /chaperone system that monitors web activity with the intent of /creating consequences /for activity which you - or your intended customer - opines is "invalid". I doubt very many people will buy into it because there is no upside for them. Even when they alter it to fit their own agenda, they just anger their customers who can click OK for EULAs and enter logins, but cannot bypass your 3 Minute Hate. If it can detect an "invalid" certificate, then by changing a couple code lines (if even), it can detect anything else about an attempted site visit. Of course this ability is ancient now, but less evil implementations of it merely censor by blocking, which is bad enough. Yours is "educational" - and it's interesting that /you /put the quotes around that word yourself - for the purpose of taking up other people's time with propaganda. If it became common, it would become a mandatory advertising medium anytime anyone clicked on a competitor's site, or a site with bad reviews for your customer. If it became law, it would become a mandatory propaganda delivery system anytime anyone clicked on a site containing any kind of dissenting viewpoint. Are you hoping to create one of those conditions? If so, which? Because this sure looks like more than just wanting to manipulate lesser people into a system designed to reinforce your wishful feelings of superiority. There has to be a more compelling reason that you're this overly concerned about what 3rd parties do on 4th party systems. Which, btw, brings up the fact that your system is not equivalent to EULAs or logins or pay systems, because the connection provider has the right to set conditions for using their connection. Your spyware idea is to harass people who are using /other people's/ connections. I'm not an expert on web connection technology per se, but it seems that Tor would nicely wire around all SSL issues after the initial connection to the now-restricted hotspot. You certainly make a great case for using it, even if just on general principle. So what would you do about that? I don't think your grandmother wants you monitoring her activity. I don't think /anyone /wants you monitoring their activity. But you seem to want to do it anyway. And no one but me is saying boo to you. :-( As to the trivia: I personally have never had trouble from visiting a site with an "invalid certificate" of any kind, because that stuff simply isn't 100% maintained. Obviously I am careful where I go and what I click and download anyway. I do not so easily ignore "known malware site" warnings, and if in doubt about a site I reflexively check the web address. MyBank.Phishing.com and Phishing.com/MyBank do not get clicks from me. But that's all beside the point. On 3/20/2017 9:57 PM, Brien Dieterle wrote: > On Mar 20, 2017 3:36 PM, "Vara La Fey" > wrote: > > OMG!! > > First of all, you'd be mis-educating them if telling them that > certificate "validity" has any real meaning. (But now you're > talking about http.) > > I mean validity as in trusted roots that have been shipped with your > OS or browser. Surely you don't mean these are meaningless. AFAIK > they are very reliable as long as you never accept bogus certs. If > you accept bogus certs "all the time", I really hope you know what > you're doing. Pretty much any important site should have working SSL. > > There is a reason why all the browsers freak out when you get a bad > cert, but users still click "add exception". My captive education > portal would give real consequence to this with the 3 minute power > point slideshow and mandatory quiz. I wonder if this is already > patented. . . > > Second, why do you think you have any right to put speed bumps in > the way of people who are doing nothing to you? > > Plenty of businesses do this already for captive portals and forcing > users to log in, pay, or accept an EULA. They are already tampering > with your SSL connection in order to redirect you to the portal. I'm > just suggesting to use this technology for "educational" purposes. > > Third, if your grandmother needs internet "safety" education, just > educate her, or refuse to keep fixing the problems she encounters > in her ignorance - if she really is all that ignorant. I hope you > wouldn't install a browser re-direct without her consent, because > then you'd be just any other malware propagator with just any > other self-righteous rationalization. > > Well, I'm lazy. I'd much rather have an ongoing passive education > program for anyone that uses that router. Maybe only 1 in 1000 > requests trigger the "test", or once a month per mac address maybe. > If grandma fails the test I can get an email so I can call her up and > gently chastise her. "Grandmaaaa, did you accept a bogus SSL > certificate again? Hmmm?" > > As far as consent goes, I'm only talking about routers you own or have > permission to modify. That should go without saying. > > Fourth, if /you /need educational "speed bumps" on /your /router, > /you /are free to have them. One of the great things about freedom > - from government or from meddling busybodies - is that /you /get > to be free too. > > My post is in the context of businesses or individuals that provide > Internet to the public. Presumably businesses and individuals have > the freedom to do this kind of SSL interception, since they've already > been doing it for years without any repercussions. Personally I'm > disturbed that businesses will try to get me to accept their SSL cert > for their Wi-Fi portal, but I know the technology leaves little > choice. One trick is to ignore the cert and try again with a non SSL > address. > > It is pretty ironic that the first thing these captive portals ask > users to do is blindly accept a bogus SSL cert. It is really just a > sad state of affairs that we are literally training people to accept > bad SSL certificates. > > For years my Firefox has had an option to "always use HTTPS", and > I'm sure all other modern browsers do as well. Plus, Mozilla.org > has a free plugin - I think it's from EFF.org - called "HTTPS > Everywhere". It's all very easy to use, and will be almost > entirely transparent to Grandma. > > This won't do anything to protect you/grandma from bogus ssl certs. > Imagine connecting to a bad AP at Starbucks that is proxying all your > SSL connections. Your only defense is trusted roots and knowing not > to accept bogus SSL certs. If only we had a captive router-based SSL > education program... ;) > > > > On 3/20/2017 3:14 PM, Brien Dieterle wrote: >> A system like I described would just be an "educational tool" to >> encourage people to use HTTPS (properly). It wouldn't stop you >> from accepting bogus certificates-- just a speed bump. Now that >> I've thought about it I'd really like to install something like >> this on my grandparent's router. . . heck, my own router. . . >> >> On Mon, Mar 20, 2017 at 2:50 PM, Vara La Fey > > wrote: >> >> Oh HELL no!! What kind of hall-monitor nanny mentality do you >> want people to adopt?? >> >> I accept "bogus" certificates all the time because the whole >> idea of certificates is crap in the first place - they are >> NOT maintained - and years ago I got tired of that procedure >> warning me about "invalid" certificates for sites that were >> perfectly valid. >> >> I've never had a problem. Of course I'm also careful where I >> go, certificate or not. >> >> - Vara >> >> >> On 3/20/2017 2:12 PM, Brien Dieterle wrote: >>> Maybe every commercial router should do SSL interception by >>> default. If a user accepts a bogus certificate they are >>> taken to a page that thoroughly scolds them and informs them >>> about the huge mistake they made, forces them to read a few >>> slides and take a quiz on network safety before allowing >>> them on the Internet. Maybe do the same for non-ssl HTTP >>> traffic, etc.. . >>> >>> On Mon, Mar 20, 2017 at 1:55 PM, Matt Graham >>> > wrote: >>> >>> On Mon, Mar 20, 2017 at 12:29 PM, Victor Odhner >>> > wrote: >>> >>> I’m really annoyed that so many companies offer >>> open WIFI when it would be >>> so easy to secure those hot spots. Restaurants, >>> hotels, and the waiting >>> rooms of auto dealerships are almost 100% open. >>> >>> [snip] >>> On 2017-03-20 13:20, Stephen Partington wrote: >>> >>> This is usually done as a means to be easy for their >>> customers. >>> >>> >>> Pretty much this. Convenience is more valuable than >>> security in most people's minds. >>> >>> they’d be happy to do the right thing if we >>> could explain it to the right people. >>> >>> >>> I'm not sure this would happen. Setting up passwords >>> and then distributing those passwords has a non-zero >>> cost and offers zero visible benefits for most of the >>> people who are using the wireless networks.[0] And as >>> another poster said, what about football/baseball >>> stadiums? Distributing passwords to tens of thousands >>> of people is sort of difficult. "Just watching the >>> game" is not an option; people want to FaceTweet >>> pictures of themselves at the game. >>> >>> OTOH, the last time I looked at the access points >>> visible from my living room, almost all of them had some >>> sort of access control enabled. Maybe there's a social >>> convention forming that "my access point" ~= "my back >>> yard" and "open access point" ~= "a public park"? >>> >>> [0] Having a more educated user population would make >>> the benefits more visible, but it's very difficult to >>> make people care about these things. >>> >>> -- >>> Crow202 Blog: http://crow202.org/wordpress >>> There is no Darkness in Eternity >>> But only Light too dim for us to see. >>> >>> --------------------------------------------------- >>> PLUG-discuss mailing list - >>> PLUG-discuss@lists.phxlinux.org >>> >>> To subscribe, unsubscribe, or to change your mail settings: >>> http://lists.phxlinux.org/mailman/listinfo/plug-discuss >>> >>> >>> >>> >>> >>> --------------------------------------------------- >>> PLUG-discuss mailing list -PLUG-discuss@lists.phxlinux.org >>> >>> To subscribe, unsubscribe, or to change your mail settings: >>> http://lists.phxlinux.org/mailman/listinfo/plug-discuss >>> >> --------------------------------------------------- >> PLUG-discuss mailing list - PLUG-discuss@lists.phxlinux.org >> To subscribe, >> unsubscribe, or to change your mail settings: >> http://lists.phxlinux.org/mailman/listinfo/plug-discuss >> >> >> --------------------------------------------------- >> PLUG-discuss mailing list -PLUG-discuss@lists.phxlinux.org >> >> To subscribe, unsubscribe, or to change your mail settings: >> http://lists.phxlinux.org/mailman/listinfo/plug-discuss >> > --------------------------------------------------- PLUG-discuss > mailing list - PLUG-discuss@lists.phxlinux.org > To subscribe, > unsubscribe, or to change your mail settings: > http://lists.phxlinux.org/mailman/listinfo/plug-discuss > > > --------------------------------------------------- > PLUG-discuss mailing list - PLUG-discuss@lists.phxlinux.org > To subscribe, unsubscribe, or to change your mail settings: > http://lists.phxlinux.org/mailman/listinfo/plug-discuss