Doug Tygar, author of Why Johnny Can't Encrypt, has died(eecs.berkeley.edu)
eecs.berkeley.edu
Doug Tygar, author of Why Johnny Can't Encrypt, has died
https://eecs.berkeley.edu/news
8 comments
https://www.ischool.berkeley.edu/news/doug-tygar-tributes
Why Johnny Can't Encrypt: A Usability Evaluation of PGP 5.0 - https://people.eecs.berkeley.edu/~tygar/papers/Why_Johnny_Ca...
As an indication of the effect that the original 1999 paper has had on security research, and its legacy, let me also mention the paper "Why Johnny Still Can’t Encrypt" [1] from 2006, and the paper "Why Johnny Still, Still Can't Encrypt" [2] from 2015.
Perhaps in a few more years Johnny will finally be able to encrypt using email software that supports the emerging Autocrypt standard. [3]
[1] http://cups.cs.cmu.edu/soups/2006/posters/sheng-poster_abstr...
[2] https://arxiv.org/abs/1510.08555
[3] https://autocrypt.org/
Perhaps in a few more years Johnny will finally be able to encrypt using email software that supports the emerging Autocrypt standard. [3]
[1] http://cups.cs.cmu.edu/soups/2006/posters/sheng-poster_abstr...
[2] https://arxiv.org/abs/1510.08555
[3] https://autocrypt.org/
Mmm.
Repeatedly what makes a big difference has been ambient availability. Something is "just there" and you rely on it without even thinking about it - the way we mostly take clean water for granted in in the industrialised world.
I believe this was important for DSL and similar technologies often labelled "broadband" Internet service. Having 5Mbps rather than 50kbps seems important but what I'm confident really mattered was that the former was Always On, you didn't have to "go online" to check email or do a search or watch a video any more, and that changes how you think about everything.
For encrypted messaging that's delivered by systems like Whatsapp (and of course Signal) because everything is always encrypted all the time. It is not delivered when you need to explicitly "turn on" encryption, no real users will do that for routine conversations.
It makes a huge difference in transit too, transit engineers call it "metro service" - once you schedule at high enough frequency users change how they use the service. If there's one train a day people need to plan "OK, the train is at 11:23 so I need to be at the station a few minutes before that". When there's one bus a minute (some London peak bus routes) that never occurs to the passenger at all. They just catch a bus when they need one, because obviously there will be a bus, there is always a bus.
In many countries 4G wireless Internet has this sort of "ambient availability" outdoors in populated areas, but not so much in the wilderness or indoors.
EduRoam (and to a lesser extent GovRoam) do this for WiFi. The experience is that whether you're at your university or some foreign university maybe for a conference, WiFi just works the same everywhere magically.
Autocrypt seems to have some thoughts towards ambient availability but it's a long way from here. I wish you luck but if you can't reach that point it's probably not going to make a significant difference.
Repeatedly what makes a big difference has been ambient availability. Something is "just there" and you rely on it without even thinking about it - the way we mostly take clean water for granted in in the industrialised world.
I believe this was important for DSL and similar technologies often labelled "broadband" Internet service. Having 5Mbps rather than 50kbps seems important but what I'm confident really mattered was that the former was Always On, you didn't have to "go online" to check email or do a search or watch a video any more, and that changes how you think about everything.
For encrypted messaging that's delivered by systems like Whatsapp (and of course Signal) because everything is always encrypted all the time. It is not delivered when you need to explicitly "turn on" encryption, no real users will do that for routine conversations.
It makes a huge difference in transit too, transit engineers call it "metro service" - once you schedule at high enough frequency users change how they use the service. If there's one train a day people need to plan "OK, the train is at 11:23 so I need to be at the station a few minutes before that". When there's one bus a minute (some London peak bus routes) that never occurs to the passenger at all. They just catch a bus when they need one, because obviously there will be a bus, there is always a bus.
In many countries 4G wireless Internet has this sort of "ambient availability" outdoors in populated areas, but not so much in the wilderness or indoors.
EduRoam (and to a lesser extent GovRoam) do this for WiFi. The experience is that whether you're at your university or some foreign university maybe for a conference, WiFi just works the same everywhere magically.
Autocrypt seems to have some thoughts towards ambient availability but it's a long way from here. I wish you luck but if you can't reach that point it's probably not going to make a significant difference.
Which is named for "Why Johnny Can't Read" - a book from 1955 (by Rudolf Flesch, co-creator of the Flesch–Kincaid readability test)
That's very sad and AFAIK startling - he was still relatively young. I remember him being very friendly as a faculty member at CMU, as well as being whip-smart.
It is interesting that "Why Johnny Can't Encrypt" is the paper that appears in his obit link. This was coauthored with his student Alma Whitten (who later went on to the thankless task of being the main flak-catcher for Google on privacy issues, including what looked like some uncomfortable grillings on Capitol Hill).
It is interesting that "Why Johnny Can't Encrypt" is the paper that appears in his obit link. This was coauthored with his student Alma Whitten (who later went on to the thankless task of being the main flak-catcher for Google on privacy issues, including what looked like some uncomfortable grillings on Capitol Hill).