resurrecting dead filesystem
resurrecting dead filesystem
Hi all
I am considering resurrecting a filsystem type that is no longer properly supported by a lot of os'es and has serious limitations. this fs is adfs which was used on the bbc master 128 and acorn archimedes. linux supports it (badly) and it has various issues such as needing compaction every so often as files must be contiguous, and limitations on number of objects in a directory (47) so... is it even worth it, even to keep historical filesystems alive? What do you think?
I am considering resurrecting a filsystem type that is no longer properly supported by a lot of os'es and has serious limitations. this fs is adfs which was used on the bbc master 128 and acorn archimedes. linux supports it (badly) and it has various issues such as needing compaction every so often as files must be contiguous, and limitations on number of objects in a directory (47) so... is it even worth it, even to keep historical filesystems alive? What do you think?
Last edited by brain on Sat Feb 25, 2012 8:45 am, edited 1 time in total.
- Combuster
- Member
- Posts: 9301
- Joined: Wed Oct 18, 2006 3:45 am
- Libera.chat IRC: [com]buster
- Location: On the balcony, where I can actually keep 1½m distance
- Contact:
Re: resurrecting dead filesystem
If you want your old acorn up and running again and you need that driver to get there, please go ahead. That situation also means it's not actually dead for you.
Actual necromancing is frowned upon. (Especially on fora. )
Actual necromancing is frowned upon. (Especially on fora. )
- gravaera
- Member
- Posts: 737
- Joined: Tue Jun 02, 2009 4:35 pm
- Location: Supporting the cause: Use \tabs to indent code. NOT \x20 spaces.
Re: resurrecting dead filesystem
Well, limiting the number of entries in a directory is already a no-no, but on small embedded devices maybe, this may be a viable filesystem? Not too sure, but good luck
17:56 < sortie> Paging is called paging because you need to draw it on pages in your notebook to succeed at it.
Re: resurrecting dead filesystem
I'm certainly willing to hand out some intertube geekpoints to you if you succeed. Also it's sad to let it die. If you could provide some real documentation on it for future generations while enjoying yourself, why not?
Re: resurrecting dead filesystem
Yeah it may be fun only problem is that adfs assumes 256 byte sectors, modern hard disks and floppies have 512 byte sectors so if I was to be historically accurate I would have to translate sectors splitting them into two virtual adfs sectors... there's also a filsystem size limit of about 40 megs on adfs L format (bbc micro)... Limitations abound! As my os is a modern reinterpretation of the bbc micro os though it would make some sense.
Re: resurrecting dead filesystem
I don't know about this, but since wikipedia says acorn used 3½ inch, couldn't you just get your hands on some Acorn 640kb floppy with 256byte sectors and make the FDC read it? Or is the support for odd disk and sector sizes like these not supported by the "modern" floppy controllers?
Re: resurrecting dead filesystem
Well I was more going for hard disks than floppies, I do have some old 5.25 inch 640k L format floppies laying around but no hardware to read them, I was hoping to phase out floppies from my os and maybe just work on images as native loopback devices, this may make more sense and get around the sector size issues. modern floppy controllers do support 256 byte sectors but to th best of my knowledge modern hard disks don't, nor do they support other semi-required archaic things like mfm encoding... (required only to read hard disks formatted by an Archie or beeb)
Re: resurrecting dead filesystem
Wouldn't you need that hardware to investigate the structures and document the unknown parts of the FS? If there is documentation, and you mean dead as in no good implementation, then I don't really see any good reason anymore. With documentation it's never to late, and it will never truly "die".
Re: resurrecting dead filesystem
The FS is very well documented, even the official os/hardware instruction manual documented the fs format, its a shame they dont do that any more. Its only dead in that nobody really implements it any more...bubach wrote:Wouldn't you need that hardware to investigate the structures and document the unknown parts of the FS? If there is documentation, and you mean dead as in no good implementation, then I don't really see any good reason anymore. With documentation it's never to late, and it will never truly "die".
Re: resurrecting dead filesystem
is it even worth it, even to keep historical filesystems alive? What do you think?
Data is more important than anything else. There still might be a use case to recover disks containing data with this file system. Please pardon my ignorance, since I really do not work with file systems/storage. Digital Presevation is still an active area of research.
--Thomas