View Single Post
Old 11 March 2009, 18:13   #17
mr.vince
Cheesy crust
 
mr.vince's Avatar
 
Join Date: Nov 2008
Location: Hawk's Creek
Age: 48
Posts: 1,383
Quote:
Originally Posted by alexh View Post
Wow, I cannot believe that no-one other than myself cares that SPS titles are not truly preserved.

I can understand why the dumper, and the analyser are not released open source, it helps maintain the quality of available IPF images. But the emulator plugins?
Aren't these two completely different things?

IPFs are truly preserved. They have all the information present on the original disk. We keep copies in different places, and we also give the IPFs back to the contributors, which can also store them. Or distribute them - it's up to them, not us.

Now for the lib. It's available for all major platforms. I see that some people are scared because it is not open source, still there are many projects using it. HxC even uses it for making the original content available as flux data to a real computer and C20 once again uses the HxC sources to convert the IPF data to its own intermediate format called .C20. I don't see anything stopping anyone from doing the same and archiving the games in another format he likes (if he does not like IPF that is). That would of course mean writing and designing your own format.

Being open source does not mean everything is better. In regard to PDFs, you now have the light and the dark. By opening the file format, Adobe has lost control over it. This of course also freed users to chose from whatever application they want to use for viewing. I really like that. But it also leads to documents being misinterpreted, i.e. looking somehow bad, with elements missing or simply shuffled around. I also tend to get some PDFs that do not display properly with Preview, included with MacOS. This is not Adobe's fault (you just can not blame them for this, which I regret, because I really do not like their domination in regard to monopoly and pricing). It's because different codebases did not include particular functions or, even worse, did implement things the wrong way.

Do you really want to give preservation into the hand of novices, that can code but have no understanding of protections or similar? Do you want them to mess with the codebase, just because it is open source? Do you want them to reverse the library to also write IPFs? To see dozens of IPFs of the very same title, called "alt1", "alt2", "+8", "fixed"...

While "slightly off", "shuffled" or whatever still enables you to read most of the text in a PDF document, "slightly off" or "shuffled" in regard to flux data being interpreted means reading wrong data. Copy protections won't work anymore, programs will read corrupted data files etc.

In fact, UAE is open source. I have not seen many useful distributions with any functionality I am currently missing. I am very sure that if Toni should ever abandon the project, there will not be many people available to fill the gap. Writing C code is easy. But emulating a complex machine with several custom chips (down to the gate level) is as well rocket science and requires a large amount of knowledge. That's not inside the source, but Toni knows what to change where without breaking other things (ok, even happens to him from time to time).

The SPS library takes a different approach. It is available to anyone. On many platforms. You can do with it what you like (well, mostly). You can read the data like it would be coming from a real floppy. But we can make sure, the format itself is consistent, and the data is consistent as well.

I just do not get the point why this would be a showstopper for preservation.

And to be honest: Of course all modern operating systems might vanish. Somewhen. I still bet there'd be emulators for them. Somewhere, somewhen. So you could still use. This would however mean there is nobody of the SPS team around anymore... no heier to the legacy. I just do not think we'd not be giving away the source one day before the end is near. We have been thinking about this, but it's simply not necessary right now and we have to make sure we can differentiate between IPFs made by us and IPFs made by others (which could be done with the source).

We're currently looking into adding a signed digest added to the IPFs so users can be sure they are getting the real deal and not anything forged or changed. Please note that this is not DRM, so please have a read on what signed digests / hashes are used for.
mr.vince is offline  
 
Page generated in 0.04324 seconds with 11 queries