Your message sent my brain off exploring possible avenues, and it came back with this idea..
The problem you have is that an application that can both see the phone number you're receiving calls from could also connect to somewhere on the Internet. You probably don't care all that much that it can see said phone number, you care that there's a potential route from there to someone malicious or overly nosy finding out who is calling you. If there was a way at the programming level (and hence at the permissions level) of making that absolutely impossible, i.e., allowing a program to do those two things, but not get the phone number information into anything that goes out to the Internet, that would make a lot of worries and actual hacks disappear. Of course in your actual case the real solution should be for Android to split the permissions as you say.. Perl has something I never really looked into, tainted strings that can't be used in concatenation, I suppose that's a related idea, but where my poor befuddled brain was going was confidentiality via monads, and it turns out I'm not the first (or probably even in the first 10,000) to think of this. PDF paper, no Greek symbols but some Haskell code: http://www.cse.chalmers.se/~russo/publications_files/haskell22Ext-russo.pdf Anyway, I was kinda expecting you to complain more about the Java 7 problem you had uploading to Sonatype. :) On Wed, Sep 26, 2012 at 4:39 PM, Fabrizio Giudici <[email protected]> wrote: > The detail is the permission "Read phone state and identity". It allows to > do: > > 1. read the IMEI of the phone > 2. read my phone number > 3. know the status of the phone, e.g. the number of an inbound call > > This is at least what I understand. Now: > > * I understand that (1) might be useful for applications that want to count > the users (even though many app developers, like me, just do with a > self-generated UUID) > * I understand that the portion of (3) related to knowing that the phone is > ringing can be useful e.g. for pausing an interactive activity, pausing a > sound reproduction, etc... > > But I don't understand why these activities haven't been split in separate > permissions. While it might be legitimate to know that the phone is ringing > to pause a sound reproduction or such, it is not necessary to know which is > my phone number and the number I'm talking to. What I find unbearable is > that more and more apps are happily using this permission, often a newer > version of an app does while it didn't in the past. For instance, in the > latest round of updates I can see five app updates that are asking for the > first time this permission. Some without any reasonable motivation, and I'm > quite tempted to drop them. But this is very annoying! > > > -- > Fabrizio Giudici - Java Architect, Project Manager > Tidalwave s.a.s. - "We make Java work. Everywhere." > [email protected] > http://tidalwave.it - http://fabriziogiudici.it > > -- > You received this message because you are subscribed to the Google Groups > "Java Posse" group. > To post to this group, send email to [email protected]. > To unsubscribe from this group, send email to > [email protected]. > For more options, visit this group at > http://groups.google.com/group/javaposse?hl=en. > -- You received this message because you are subscribed to the Google Groups "Java Posse" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/javaposse?hl=en.
