op_return outputs can be pruned because they are not spendable.
putting a hash on in the witness script data won't make things better
(it would actually make them worse) and it definitely doesn't help
"block size bloat".
I think I'm missing some context, but if you're using op_return purely
for timestamping I would recommend using pay 2 contract  instead.

On Tue, Aug 14, 2018 at 8:34 PM, Christopher Allen via bitcoin-dev
<bitcoin-dev@lists.linuxfoundation.org> wrote:
> On August 5, 2018 9:11:26 PM UTC, Lautaro Dragan via bitcoin-dev
> <bitcoin-dev at lists.linuxfoundation.org> wrote:
>>Should we actually be using the BIP process to claim a prefix?
>
> I recommend against using an op_return prefix, as they allow for transaction
> censorship.
>
> In fact, in our case, where we use an IPFS hash in an op_return, we remove
> the IPFS multihash prefix information to post a “bare” SHA256 hash to look
> like many other hashes being posted in op_returns, to minimize any ability
> for a miner to identify our transaction. The more projects that do this the
> better — a form of herd immunity.
>
> Longer term I’m looking for more responsible ways to publish this hash, for
> instance have the hash be in the witness script data, so that it can be
> easily purged from nodes that do not wish to preserve it and prevent block
> size bloat. However, to do so everyone has to do it the same way, ideally
> have it look like any other transaction. I’ve not quite seen a solid
> proposal for best practices here.
>
> — Christopher Allen
>
> _______________________________________________
> bitcoin-dev mailing list
> bitcoin-dev@lists.linuxfoundation.org
> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev
>
_______________________________________________
bitcoin-dev mailing list
bitcoin-dev@lists.linuxfoundation.org
https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev