Is hydrus botnet? It's github page redirects to a MITM'd cianigger/hypno.is version of github and the dev may or may not be a furfag. The software itself uses python and opencv which may or may not be botnet even if brainlet/pajeet tier. From my initial inspection it doesn't seem like botnet outside of the whole source code being MITM'd thing.
Hydrus users and questions
For context see archive.fo
Probably not.
Huh? " github.com
Clearly you haven't been long enough on Zig Forums.
Not an argument.
So? I am not sure where did you find this hypothes.is link, on >>>/hydrus/ every link seems to be clean. Elaborate please.
I agree that those questions are really shitty--so much so that I wonder if it's reverse shilling, but Hydrus, for as much of a good idea as it is, especially if it eventually supports ipfs, is marred by very incompetent programmers.
For one thing, it's Python, which is very robust for what it is, not to mention portable and very accessible (both of which meaning a large userbase with a large gradient of competency), but just because any imbecile can pick up and become proficient in Python doesn't mean Hydrus doesn't need to be well-written. In fact, Hydrus is an excellent example of what it means to repeat bugs and design issues that should have been learned from decades ago. It totes much of the problems that plagued second-gen p2p programs like eMule, whose peak in popularity was over a decade ago, yet Hydrus still doesn't learn from that. The main developer uses Windows for God's sake. I think the only reason why he uses Python is because it's the only language he knows--all the benefits of Python were simply a mere coincidence rather than choice.
It's licensed under wtfpl. Adolescent contrarianism aside, there are reasons why we don't use joke licenses for projects whose outcomes we care about. We live in a time where copyright is an insurmountable obstacle; we can't all be a retarded edgelord like Julian Assange--especially when you're anonymous. In fact, if public domain is what you want, there are licenses that relinquish your copyright powers far better than the wtfpl.
And the bastard won't even ship it on pip for "reasons". Those reasons probably being that he's too scared he would be ousted as an idiot.
It's not a secret that he is incompetent. The main problem is the he is unwilling to admit his shortcomings and work with others. The program improves at an incredibly slow pace and every update introduces more bugs and bloat. I like the idea of Hydrus and it has some nice features, but the it's going nowhere and will remain always a "hobby project" with that attitude. Also...
I know. And it sucks because Hydrus, for all its shortcomings, has the kernel of an idea that is so damn prescient I would argue that it's the end-all solution for free, distributed media consumption and stratified metadata in lieu of websites whose respective development models are kind of skewed in a way just by nature of http to be proprietary and disconnected from its source code even if it is ostensibly libre (e.g. danbooru, reddit, etc.), yet resources are squandered on a bloated, poorly-written program with a server that dies every time its client does and a UX philosophy that would make even the most veteran designers weep like a little girl.
lol. kill yourself redditfag
...
redditors use mit and bsd because wtfpl triggers them and because they have heard that public domain doesn't exist in europe.
It's ironic that you would resort to such a seemingly unrelated strawman when you yourself write like a Redditor. Plus, public domain doesn't exist anywhere, nor is wtfpl public domain: it's copyright. In fact, the wtfpl is relatively strong compared to some free licenses. One could argue that its susceptibility to exploit makes it an even stronger license than some moderate copyleft licenses without nearly as much control.
spotted the kike
Are you trying to convince me or yourself?
convince you of what?
maybe you should try convincing me that public domain doesn't exist with facts
Is there anything like Hydrus, but not Hydrus?
Calibre is basically the Hydrus of books. You have a bloated, front end that takes a dozens seconds just to load when all you wanted and ever really needed was the simple command line utility that you can't build separately. Plus, the developer is a stubborn asshole who hates conventional things like support for arbitrary directory structure just for the sake of hating it.
There are a lot of p2p tools that are completely different implementations that use the same network. A good example is iMule or Nicotine+, so it would be theoretically possible to make another Hydrus client, but you'd have to write it yourself.
kek, it might not be botnet but it's shit for a whole different reason.
It still hurts.
I just wanted to assign tags to my memes.
The advantage of Hydrus is that it lets you crowdsource the tagging of your memes. It's especially useful if you're the special kind of loser who saves hundreds of slowjack memes for the sole purpose of perpetuating meaningless flame wars subsisting on nothing but namecalling. It's just too bad that tagging is contained within Hydrus, filenames are renamed to their checksums which sucks for interchangeability with actually good image viewers and managers, and there's no special tag for an images licensing and copyright holder.
If you were to integrate Hydrus with ipfs, that could mean a lot for a kind of Universal, distributed booru that doesn't need a website or special client to facilitate access.
It's bloated shit for something as simple as file management.
That's all you have to fucking do for file management.
It's a useless gui frontend made in a slow as shit language with server capabilities for things that don't need it.
Also, like fuck is hashes going to help when it comes to tags, half the shit i got is captures/edits that i made myself.
Any small difference will make the hash system completely irrelevant and you'll have to tag it yourself anyway.
Also.
It copies over all your files into another directory doubling the space used on your harddrive
This is as degenerate as tag based music storage. It's basically iTunes for jpegs.
That is incredibly stupid. Especially for art. There are a million criteria by which you can sort art. What are you going to do? Rename your directories according to that information? Okay, sure, let's just pretend that things like lyrics don't exist. How do you procure that information? A tool like beets is incredibly helpful for people who care about file hierarchy. You know how beets gets that info? Tags. Tags let you change your directory structure and names based on arbitrary criteria.
Also, what kind of autist do you have to be to think that tags and directories are mutually exclusive?
Believe it or not, not everybody wants their music sorted by the artist tag. Some want it sorted by albumartist, something most autistic music players can't accommodate. What if the compilation tag is toggled? Some people want to browse by album artist but see the artist tag when they query a single track. That's not an insane thing to want. Sometimes you sort by genre, composer, year, all sorts of random things. How are you supposed to accommodate that? How do you embody all that information in a file structure without keeping a separate info cache? And how would you standardize that? How do you think that cache affects parsing info?
Also,
You're just baiting, aren't you?
You write like I'm being apologetic to Hydrus. Just because I like an idea that Hydrus implements doesn't mean I like Hydrus itself. I'm telling you not to conflate the two.
It's archive.is that is redirecting through hypothes.is. It's a work-around for sites blocking archiving, there's nothing sinister about it.
The developer of meguca.org
I think you got that backwards. Most normie players can't sort by album artist, most autistic ones can.
t. Autist who sorts by album artist.
As a fellow albumartist autist, I can say the exact audience. For many music players in the Free World, they can't even sort by metadata. That is especially true the more extensible and well-authored they are. Look at EMMS: artist is hardcoded as the search criteria and there's no way to fix that unless you go in there and change things yourself. The only one I can think of that can sort media indiscriminately and is still really thorough is gMusicBrowser.
Tagsistant is a tagging filesystem implemented with sqlite and fuse. You can move files to the tags you want with paths like store/animated/cat/@/. The "@" means no more tags, and each directory between "store" and "@" is a tag. The order of tags doesn't matter, since it's not hierarchical. The directories are dynamically created from your tags. Since it's a filesystem, it can be mounted and hosted with ftp and such.
Most image viewers support some kind of primitive tagging. The one thing Hydrus has going for it is that the tagging is automatic and shared between users.
To this day, only Quod Libet can really fulfill my music tagging autism.
gmusic has a quod libet preset.
found the (((jew)))
Sometimes I wish I wasn't a LARProgrammer.
Not efficient at all when working with thousands of pictures. It's actually easier to use than it seems, but as it's still extremely cumbersome without gazillions of handmade scripts.
I'd like to give it a try, but won't Tagsistant kill my hard drive though with billions of folders and reduntant copies?
bump