Filesystems

[CWood]

The wiki contains lots of information about existing file systems, and also mentions that there are many FAT based file systems being created by osdevvers. The reason for this issue? People don't know how to design an FS. The solution is quite simple. Practical File System Design. A book on rubber ducks (haha duh!) that uses BFS (Be File System) as an example. Perhaps someone could link to the PDF in the FS section of the Wiki, and/or condense the book down to a shortened "taster" into FS design, to put into an article, with a link to the PDF (I can do this, but it would probably take a while, as I have a number of other commitments as well...)

Just my 2c on the subject, feel free to object/criticize (constructively, ofc)

[OSwhatever]

It's not only here FAT is popular but also in commercial systems. The reason: FAT is simple to understand, implement and it works almost anywhere. Just take USB mass storage for example where VFAT is commonly used.

I see no fault that you implement a FAT filesystem as the first filesystem support for your OS. It makes sense actually.

FAT on the other hand is not a contemporary filesystem. For you who want to design a new filesystem, Practical File System Design is a good read. I personally recommend that you have a look at BTRFS since it is a good example of a modern filesystem. XFS is also another modern (and extensive) filesystem.

Filesystems is such a large subject it could have a forum like this of its own. Designing operating systems AND creating a new modern filesystem is a huge task. Most of us stays in either subject since each of them are big enough.

AND: if you can link Practical File System Design from here, do it.

[Brynet-Inc]

The author published his book as PDF on his website, but it might a good idea to link to his website as he also has a tarball there with some of the source he references in the book.

He hasn't given up copyright though, nor does he mention any redistribution rights. I imagine extracting anything from it and posting it on the Wiki wouldn't be a good move on our part.

[CWood]

Perhaps, then, a "summarizing" of it. Rephrase it, to skirt around the copyrights, while getting the useful content onto the wiki, for all to see.

[Brynet-Inc]

Perhaps, then, a "summarizing" of it. Rephrase it, to skirt around the copyrights, while getting the useful content onto the wiki, for all to see.

Or you could write something original and not try to pass someones work off as your own.

[CWood]

I wasn't intending on passing it off as my own. I would cite it correctly, attribute it to its author, and link to the original content. I don't want any credit for the content, I just thought that the readers of the wiki might benefit from having the content of the read present in the wiki, rather than have to hunt for it (as I did)

[Brynet-Inc]

I wasn't intending on passing it off as my own. I would cite it correctly, attribute it to its author, and link to the original content. I don't want any credit for the content, I just thought that the readers of the wiki might benefit from having the content of the read present in the wiki, rather than have to hunt for it (as I did)

So contact the author and ask for his permission, seems like the obvious thing to do.

[CWood]

I wasn't intending on passing it off as my own. I would cite it correctly, attribute it to its author, and link to the original content. I don't want any credit for the content, I just thought that the readers of the wiki might benefit from having the content of the read present in the wiki, rather than have to hunt for it (as I did)

So contact the author and ask for his permission, seems like the obvious thing to do.

(sheepish) It does indeed seem like the obvious thing to do... I will get in touch with him tomorrow, no more time today, but with any luck, I will be able to upload the book to the wiki, hopefully.

Although I still feel if we do place it on the wiki, it should be summarized and condensed. As, while it is an interesting, albeit important, read, it can be very dry in places, and a lot of details are reitterations of what has been said previously. While repetition drives the idea home, it is the reason why there are >250 pages in the book, and it seems to labor a lot of points... not too sure on that one. What do you guys think?

[Brynet-Inc]

I'm pretty sure he agrees to free distribution of this book (it says so on the page) and you can link just to his home page - it's fairly trivial to get the book from there for anyone who can read plain english sentences.

Yeah, but we can't assume that he's allowing redistribution, he still retains copyright. All that we can assume is he's allowing people to download it from him. He hasn't explicitly permitted anything else.