The below link is a tough, but I argue a necessary, read. 
Here is the link that has sparked a firestorm, but I totally, 1,000 % agree with it. 

A dog isn't saved if it's still stressed to the point of attacking people unprovoked (like Rocco), and bites hard enough to kill any dog.

Shelters, rescue groups, and people need to start focusing on the dogs that should be in a home, who are behaviorally sound, instead of sending loose cannons into people's homes, even if it's an unpopular stance. Additionally, people need to be held accountable for their dogs actions. If their dog bites, it's likely because one of the following has happened: 

 1. The dog has been asking for space and never got it.  (Owner didn't know what to look for). The dog snapped. If it's before a level 3 bite, a training plan might help, or not, depending on the severity of the anxiety and the prognosis of that dog.
 2. The owner hasn't trained the dog. (Dog barks, lunges, growls, but never bit before, so "it's not an issue").  If it's before a level 3 bite, a training plan might help, or not, depending on the severity of the anxiety and the prognosis of that dog.
 3. The owner noticed that something was amiss but used punishment to suppress anxiety (which science proves doesn't work, and often backfires over time).  Alternatively, insufficient steps were taken to manage a dog who is having a hard time. If it's before a level 3 bite, a training plan might help, or not, depending on the severity of the anxiety and the prognosis of that dog.
 4. A rescue, shelter, breeder, or Craigslister knowingly passed an unstable, feral, or behaviorally unsound dog to someone else (passed the buck so someone else could deal with the problem).  If it's before a level 3 bite, a training plan might help, or not, depending on the severity of the anxiety and the prognosis of that dog.
 5. The dog has a medical condition (blindness, deafness, lyme, arthritis, pre-seizure activity, or any number of conditions) that can't possibly be trained out. Some medical conditions can cause aggression, and some are contributing factors.
 6. The dog was defending itself from abuse, teasing, or a perceived threat.
 7. The dog is "wired wrong", is feral, unstable, overstressed, unable to cope, aggressive, undersocializaed, or homed in an environment/family where it will not thrive. We don't have to wait for a bite to determine if these dogs will likely harm a human or dog: that's what behavioral testing is for. This is the one that we are seeing much more of as trainers, and it's disheartening. 

 No matter what the reason, we have an epidemic of dogs going into homes in which they are not suited. There are thousands of dogs the need rescue and do very well with families, who become man's best friend, and love life to the fullest. There are dogs who do really well with a behavior plan and do well with management, and advocacy. Then, there are the dogs who need to be put humanely to sleep because the stress of living every day is hell, a hell that we owe them to get out of with dignity and prevent them from causing irreparable harm to a child, another dog, or any adult victim. 

 It's awesome that people want to save dogs, but what isn't helping is the idea that it's our duty to save every one of these dogs in shitty situations. We can't. We shouldn't. There are dogs who deserve a home, who genuinely love people, or can acclimate comfortably with a very strict routine in an environment that is suited to them. Some dogs, like Sadie, need to be medicated and trained every day to be successful, but she is generally a happy dog, even if certain things stress her out.  But it's not our duty to save dogs who are killing kids, dogs, and live their entire life as a ball of stress and anxiety. If Sadie were no longer happy, was stressed out, behaviorally unsound, or I was unable to keep her, my family, my kid, other dogs, and my community safe from her, than it's on me as her owner (damned public opinion) to humanely end her life so she doesn't harm another. She doesn't even need to physically harm to the point where the long arm of justice would crack down - if I, as her owner, see that she is too stressed to function, even without a bite, then it's on me to be accountable. 

To the people who are crying out for the dog who was run off by a cat to "please let me adopt him, I can help him!", go to a shelter and help a dog that is behaviorally tested. Put your effort into ending breed bans so dogs like Lennox don't get euthanized for being a Pitbull. Put your effort in a healthy place, not to rescue a dog who prey-stalked a kid, shook it to create substantial harm, and has had a reputation for biting people in the past. I know you want to help, so help in a meaningful way. Help the shelters in your area get behavior testing so dogs like the one above can get the help it needs BEFORE it attacks a kid completely unprovoked. Go into schools and talk to kids about the DoggoneSafe program. Volunteer at your local shelter and teach these dogs to "sit", "down" and "stay". Work on preventing dogs from getting adopted on roadsides out of state lines, or work with your shelter to make sure when dogs are homed, they have support if things aren't going well. If you want to help, help in these ways. I encourage you to do these things. We need this. 

Education is the part that is missing in many of these dog-bite cases. I recently worked on a case where the owners really just didn't know about barrier aggression, and inadvertently made things worse. The dog is now doing well with a training plan. Some do, some don't, but we need to behavior test and be honest with potential adopters about the dogs they are getting, and to humanely euthanize those that clearly won't be successful. To truly save dogs, we need to know where to focus efforts. Education, behavior evaluations, and accountability from owners/rescue groups/shelters/breeders is really where it's at. I have low regard for those shelters who are consistently passing the buck to owners so they can keep their "no kill" status, but to what end? That's really the terrible part in all of this.