curious - The only thing I have to add to Pierre Lindenbaum 's precise answer is that, while you are considering and reading about Ensembl canonical transcripts, I would go ahead and read about MANE select transcripts (Morales et al., Nature, 2022; PMID: 35388217) and the methodology used to devise it as well. There is no substitute for the manuscript, but it could be helpful to consider a primer as well: https://www.ncbi.nlm.nih.gov/refseq/MANE/.
I think this will help give you key insight and background into related bioinformatic issues.
That ... doesn't make much sense. Did you even Google the term before creating a post here? You could have even asked ChatGPT!