Question regarding the overrall state of Tribler #8024
Unanswered
TheOverpassArsonist
asked this question in
Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
I was doing some searching and found tribler mentioned with respect to someone asking about using Bittorrent through Tor and while I'm definitely impressed with it in concept, there have been issues pointed out that I can't see have been solved or addressed. Notably, this post currently occupies the 19th and 20th citations on wikipedia (the 20th being a comment thread for the post) and identifies several issues. While I don't necessarily agree with the seemingly quite critical take away of the post, many of the issues raised do seem to be legitimate and I can't find information on if they've been addressed. (according to their comment in late 2023 the author does not believe they have been)
For convenience, below are all of the points mentioned in that article, ordered roughly by order of mention, but grouped if they are co-relevant. I've attempted to cut out as much as possible for the sake of brevity, but the post is about 10 pages long and a lot of different points are raised so I was only able to cut it about in half.
(for the purposes of not having this post be 5 pages longer than necessary, I moved this section to rentry)
Additionally I'm curious how safe Tribler would be against the potential threat of "store now, decrypt later" future Quantum attacks. Obviously BitTorrent itself offers no protection here, and frankly very few services even try to offer Q-safety, but if Tribler could provide some confidence for Qsafety that'd be a massive bonus for people who legitimately do need to worry about governments and other large-scale threat actors trying to get at them.
Again, I don't agree with (what reads to me as) the fairly critical tone of the post, but it does still raise quite a few issues that seem to be valid from an outsiders perspective and I can't easily tell how many of them have been addressed over time. I imagine some of them definitely have been, others likely not, but I can't find any mentions of a lot of them aside from this one post and the author themselves seems to believe many have not been solved. (if their comment on 2/7/2023 is anything to go by at least) Tribler unfortunately doesn't have great user-facing documentation (a problem which I do completely understand, spending time documenting, especially for a project still being developed, can be a hard ask) so it's hard to know which of these problems are still issues and which have been solved, mitigated, or may never have even been a legitimate issue in the first place. I'm massively for what Tribler is trying to do/be (I even floated the idea of a fully private/encrypted torrenting network in my head for quite a while as a concept-problem since it's a really complex and interesting and challenging problem to try to solve even in theory, let alone practice) but it's really hard to figure out just parts of it are/aren't encrypted, secure, private, etc; what functionality is/isn't missing with respect to other torrent clients; etc, especially as an end user who isn't already super involved with these subjects. (though I suppose even comparing it to a torrent client is a bit of a mistake in the first place since any given node is doing way more than a simple torrent client needs to.)
On a completely off topic note; is there anyway to update files in Tribler and use a symmetric key for transparent password protection? It would be really interesting to be able to basically be able to have a virtualized mesh-storage system between devices that could automatically stay in sync (Sort of like a Mesh-NAS or something) without needing to rely on some external service. With channels and the like it seems like it could already be implemented over Tribler feasibly, though not necessarily performantly. I might even end up taking a look at making a proof of concept myself if it's not already possible. I imagine to get good stability/performance I'd need to integrate more tightly with Tribbler's systems to sync file states/update times, etc. without sending too much unnecessary data, At that point it's more like creating a file sharing program that happens to be Tribler-compatible, but I just discovered Tribler a bit ago and I'm not sure how transparent the data transfer is so I'm not sure if that level of integration would break compatibility or not. If it's able to just send completely arbitrary data without trying to operate on it then it wouldn't be too hard and could probably be implemented directly in Tribler, but if there is a weaker seperation between the data and logic then it would probably be a lot more involved and be best served in a seperate program - or just shelved entirely if doing so ends up being way more impractical then I think. I mean you'd also want to have some fallback for local-communication to not stress the network unnecessarily, (and again, kill your token balance) then you'd probably also want to be able to use different node-depths when only communicating with trusted devices, etc. so I can see how, if there is a greater coupling between the data and logic, it'd quickly become wildly impractical to retool it this heavily while trying to keep compatibility with the rest of the network. If there is a lot more seperation though the possibilities feel borderline endless. It wouldn't be the best option for highly demanding applications but for everything from sending some files to friends/family one time to establishing a simple and secure connection to long-term embedded devices you could use basically the same underlying network just by configuring a few settings. If the latency was low enough it'd be interesting to even try to implement a VPN or something over it.
Beta Was this translation helpful? Give feedback.
All reactions