TagSpaces – Open-source personal data manager | Hacker News Reader

TagSpaces – Open-source personal data manager

tagspaces.org

Very nice, but a stopper for me is that if you have a pdf with the name "foobar.pdf" on your disk and tag it with TagSpaces with the tags Sciences, Thermodynamics you end up with a file "foobar[Sciences Thermodynamics].pdf". The tags are directly encoded into the file names.

Yeah, me too. I don't quite get the appeal of the tags living in the file name. Yes, it means that there isn't any need to maintain a database, but it also means that a) things become very unpleasant if I want to use anything but TagSpaces to browse my files, b) if I want to use anything to rename my files (especially automatically, i.e. a script) I have to take extra care not to muck up the tags, and c)if I want to share files with other people or organize files that I don't strictly own, then everyone else has to deal with me trashing their file names for my own convenience.

So, yeah, show stopper. Also, what's so bad about a database? Yes, the file system is very good at keeping track of files, but there are databases that largely don't get in the way, like sqlite, cdb, etc. They will sync across servers reasonably well with things like dropbox, as long as you don't have lots of people trying to concurrently do stuff from multiple places. If that's what you're doing, this kind of tool isn't right for the job anyhow.

This is a stopper. I don't want to pollute the filenames with this information. Using a sqlite db or something even simpler would have been more preferable imho

Hm, sqlite3 would be slow for large collections of small individual files.

However their approach is awful. Imagine what will happen if you tag video/audio files.

Saving the tags in the file name has many advantages. First of all your tags are not vendor locked. You can use the search capabilities of your operating system to search for the tagged files. And last but not least, just by simple synching your files with for example dropbox oder owncloud you are getting your tagged data on all your devices... You don't care about complex db sync mechanisms...

Better approach:

* Metadata: read/write tags as file metadata (depending on the file format) most common formats like document, image, audio, video have metadata support. http://en.wikipedia.org/wiki/Metadata

* Sidecar file: http://en.wikipedia.org/wiki/Sidecar_file and http://en.wikipedia.org/wiki/Extensible_Metadata_Platform and the Apple Spoghtlight files http://apple.stackexchange.com/questions/6707/how-to-stop-os...

* Alternative file streams: http://en.wikipedia.org/wiki/Fork_(file_system) and http://en.wikipedia.org/wiki/Extended_file_attributes

(first option is preferred)

File/data management is one of those things that are very fundamental to what computers do and one of the first applications for computers, but where advancements have been quite limited. We are still mostly stuck on simple hierarchical structures with short plain text string as the sole identifier for files. Files themselves are mostly opaque blobs, file managers usually having limited info about their content.

Admittedly file managers these days at least can show thumbnails and in some cases even previews of the file which is great. They also are capable of extracting some metadata of files, but that still feels bit tacked-on in most cases. Versioning is also available in various forms, but that too could certainly be better integrated to the system.

Imho it would be interesting to see a system which would forgo our current conventions of files and directories, and instead would be more strongly be based on the concept of tags, metadata and objects.

But I got sidetracked... I meant it is nice that people are trying to rethink even a little how we manage data, even if this particular project doesn't really hit the mark for me personally.

Remember the "Information at your fingertips" speeches of Bill Gates (Comtex '90 and '94) and the Cairo operating system, its ObjectFS filesystem and later WinFS?

https://news.ycombinator.com/item?id=7720028

I'm too young for Cairo, but I do remember Longhorn and WinFS, and indeed the marketing and ideas of those have shaped my view of what the systems could/should be today. Didn't BeOS also have some sort of object-database-filesystem-thingy? I remember some chatter about that, but didn't actually catch any first-hand experience at the time.

edit: Longhorn concept video: https://www.youtube.com/watch?v=b9ifQvQCO7Y

Nevertheless, consider watching Gates' Comdex keynote videos.

Have you seen the WinFS trailer: http://www.youtube.com/watch?v=fZ_4p5iUc88

You can try out BeOS open source successor that comes with a BeFS compatibility called HaikuOS. I suggest you to read the BeFS book written by the main developer who later developed the Spotlight search engine for MacOSX 10.4+ and later iOS. http://en.wikipedia.org/wiki/HaikuOS , http://en.wikipedia.org/wiki/BeFS , http://www.nobius.org/~dbg/practical-file-system-design.pdf , http://en.wikipedia.org/wiki/Spotlight_(software)

Both NTFS and the newer ReFS (http://en.wikipedia.org/wiki/ReFS ) offer almost equal capabilities to BeFS. And it's a real shame that Microsoft hasn't been able to come up with a decent GUI. Even Vista had a better advanced search dialog integration than the successor's Windows 7 and 8.x

Can this be used together with Camlistore (http://camlistore.org/) ?

First of all I should say that I've been following camlistore and if it reaches its goals, I think it's going to be amazing.

However, it's nowhere being user friendly yet and the command-line based API changes all the time.

I suppose you could use them together because you can mount your files in camlistore to the file system with cammount but it would not be easy to make them sync tags and metadata.

This looks very interesting indeed. There is a definite need for good cross-platform, open-source information management software. It's all too easy to get stuck in one manufacturer's information silo.

This seems like a perfect companion (or replacement) for Notational Velocity.

This is a program named after a feature. Doesn't really do a good job of conveying what it's for IMO.

Also since it uses the AGPL it's a no-go for me.

Why is software licensed under the AGPL a no-go for you?

Just curious, why is that? Why AGPL would prevent you from using this?

The ugliness in the WordPress community over themes and plugins. http://www.red-sweater.com/blog/825/getting-pretty-lonely

I don't see how themes/plugins have anything to do with an (A)GPL licensed end-user application that doesn't support any sort of plugins (at least not to my knowledge).

Even if you did write plugins for apps and license/sell them, that would stop you from using a useful app just because you couldn't do it for that app?

I'm having a hard time following how AGPL affects you as a user of the app, not someone who's selling addons for it and not as someone who provides the app over a network.

It seems you could modify the source and run your own version without releasing your modifications and without violating the (A)GPL.

1. This type of app just cries out for plugins.

2. Yes, because I don't want to spend time learning a system that I can't extend if extending systems is my livelihood.

3. I don't want to support a community that shuns people who disagree about licensing.

4. This feels against the spirit of open source to me, to avoid having people use code, and provide feedback.

I don't believe that anyone understands the GPL including its creators. Too many things in it are vague.

GPL: Do whatever but the work is to never have restrictions added.

AGPL: Anyone using/visiting the service need to have access to the source code.

They are very simple. The only reason people can be confused by those two is FUD, and lawyers who find such statements too simplistic.

Yes, you can modify and run versions privately without having to release source code.

The AGPL want that anyone who uses a service also have access to the source code. If one runs a modified AGPL service that only one person have access to, then the set of people who need to have access to the source code is that one.

Don't over fixate in a single data point. I give my family a pretty good life thanks to that WordPress community, and all my colleagues do too. Both in the services and products space.

That "ugliness" you mention has worked really well for my career and I know more people that think the same than people that think the opposite.

i love bootstrap to death, but am I the only one getting Bootstrap fatigue?

Reply on news.ycombinator.com