237 karma · joined November 10, 2010
I am not sure it is as simple as just better recommendations.
I am done with Facebook now. Bye.
"Besides a mathematical inclination, an exceptionally good mastery of one's native tongue is the most vital asset of a competent programmer."
Sorry, my own response wouldn't post inline - see http://news.ycombinator.com/item?id=2185676
Glad to hear you will be getting the code on Github. I look forward to it.
My first paragraph? You mean when I commended the project and it's motivation?
"But I'm not sure any of this is really relevant. The key question is: how sensitive is the data?"
If it identifies my participation in potentially anti-Government protest, then rather. All I have asked for is transparency and disclosure. If security is not so important, why have you gone to the trouble of the pitching for trust so heavily on your website?
"We can do nothing about the telcos using their geolocation features to track the whereabouts of phones."
Of course, you surrendor certain freedoms whenever you carry a mobile phone. But there is an important distinction between triangulating my cell position and this application.
Furthermore, I think my concerns have been concise enough. You have just refused to answer them. If you did that, there would be no more discussion. And that's why the original post has been consistently upvoted on this thread.
"I've extended an invitation to you privately to come along to a hackathon and to help us. And I extend it again."
When did you invite me the first time? No really, I have no idea.
Anyway, I might have taken you up on that offer if you didn't just call me a "self important pompous windbag".
That's a real mature and educated argument, thank you.
Congratulations on doing your project a momentous disservice.
"git push origin" doesn't seem like it will take too long.
What's the difficulty with license? (other than the App Store's incompatibility with the GPL).
"We have a shoe-string budget, no equipment beyond our own personal phones and laptops, no office space beyond what we can borrow for free. None of us are paid for our work on Sukey and many of us also have full-time careers."
I wont point out the benefits of actually being "open" in then. Limited resources can traditionally be overcome with collaborative development.
Again, I commend the effort. But I continue to urge nobody to trust a protest tool until these concerns are properly addressed.
These are not difficult questions and the continued absence of answers only confirms my fears for the project.
You continue to ignore my legitimate concerns, however.
* I urge nobody to trust a protest tool that is closed. Where is the source? Where is the documentation on its architecture? The official line is:
"It has always been our plan to make the code open source. Security is paramount and we feel it would benefit from being released to the public after each protest."
I'm sorry but retrospective release of source code is simply not good enough. Security software (which this is, essentially) needs to be out in the open from the start. Not after a protest has happened and all data collected. The wider community must be able to audit this software before it can be trusted. No exceptions.
* If the security depends on secrecy before a protest, is it dependant on a random seed? Can this seed be extracted from a binary (IPA, for example)?
* Sukey has one of the worst pages on "security" I have ever read (see http://sukey.org/security). To paraphrase:
"The team members involved on the security side are a mix of commercial information security experts and computer nerd under/post graduates who love nothing better than a complex algorithm."
That scares me. Complex algorithms are not normally conducive to secure applications. I want to hear about cryptographers - not computer nerds. I want to know what crypto implementations you are using.
Continuing:
"One of our key team members has technical commercial data security patents in his name and has provided information security consultancy to IBM, Lockheed Martin, and to the NHS."
Let's have a citation then please. Who is this expert and what (where?) are the granted patents?
"Your data is safe with Sukey."
Okay, prove it.
* How can we prove the application distributed on the App Store matches that which we (hopefully, in future) have the source for? Is there a mechanism for which we could compare checksums - if the provisioning certificates were also published? This needs to be considered. There is more to trust than merely publishing a source tree.
Let me finally restate my original point: I commend the effort (and I was once involved in a similar application). However, these are serious concerns that need to be addressed.