I am trying to synchronize the blockchain on an old laptop using Debian + Raspiblitz and its desperately slow. I remember this website where you could download the entire blockchain via torrent, what happened to it?
If you refer to torrent which contain file bootstrap.dat, there's no interest for it due to security/trust concern and no longer offer faster sync/validation time.
"It is advisable to download from the network but this is a way for those who struggle with downloading or have poor connection."
and how is it any more irresponsible to be downloading and running pull request code that is "such a risky move" and "perhaps lower risk"..
If you validate the downloads there is no issue.
TBH I don't know why they don't offer a more up-to-date way to download this and then validate there are a lot of people who still struggle and bootstrapping the data helps.
Because it isn't any faster to use the bootstrap.dat file. It has not been faster to use it since 0.10.0 when headers first sync and out of order download were first introduced. The bootstrap.dat file was useful prior to that, but it no longer is.and how is it any more irresponsible to be downloading and running pull request code that is "such a risky move" and "perhaps lower risk"..
If you validate the downloads there is no issue.
TBH I don't know why they don't offer a more up-to-date way to download this and then validate there are a lot of people who still struggle and bootstrapping the data helps.
The reason it is slower to use the bootstrap.dat file is because it just contains the blocks. Your node still has to read them off disk and validate them. It also happens to be one big file and the node does not know where each block is in that file, so it can only go through it one at a time. This is much slower than how syncing works now where multiple blocks are downloaded in parallel, and because the node is the one writing them to disk, it also knows exactly where in each file the blocks are stored so they can be pulled up later for the final step of validation. This parallelization also allows validating things that are independent of each other to be done in parallel which speeds up the sync.
It is important to note that the primary slow down in syncing is due to the processing, storage (disk I/O), and validation of blocks, not the network speed.
Is there a similar project somewhere?
Check https://prunednode.today/, but it only offer pruned data.
It's the best upgrade you can get. In case you already have an SSD and just use the HDD for block storage: move your chainstate directory to the SSD. It makes all the difference! 
