It should have been done in a very different way. It should have been more cooperative (e.g. "clarity officers" who have a masters in English or French, and who pour over statements and responses to ensure that it cannot be misinterpreted or misrepresented, without changing the core meaning).
The purpose of federally funded scientific research is not and cannot be to support the policy objectives of the currently sitting government.
There hasn't been a single example that had anything to do with the sitting government's policies or agenda. The government had no particular agenda regarding factory farming salmon or rock snot. Though from an overall government perspective for the health of an industry such as salmon (a cross-party industry), they want the message coherent.
This would have been easily achieved by transparency in the process.
They were also, rightly or wrongly, seen as actively de-funding work that could be seen to challenge policy objectives. They had ample opportunities to resolve such impressions but chose instead to present an arrogant face. I can't have a lot of sympathy for that.
Even if you aren't actually meddling, if you have the perception that you are, you have a problem.
The purpose of federally funded scientific research is not and cannot be to support the policy objectives of the currently sitting government.
So a real answer to the putative problem might be to have an arms length communications office review communication for accuracy, but one that does not directly answer to the PMs office or federal PR. What Harper did was far more draconian than that. At the same time he initiated other damaging actions to federal science, so you can hardly fault people for not trusting the motives. He could have solved this easily by making things arms length, and chose not to, as well as choosing to be very opaque about the whole process. If for no other reason than that, his government deserves all the criticism it has had on this account.