Hi everyone,

after Wandrell pointed me towards the disk image project I got very interested but had a few questions I asked Mr Creosote. Well, to make things short, we got into a lengthy discussion. Basically I was thinking it would be a lot easier for people to support the project and submit images if it felt more open. I was worried about the lack of any possibility to also download the images and even more about the lack of any statement indicating that people will be able to access this archive some day in the future. It felt a lot like "submit your stuff and if you're lucky you'll get access to it too", despite the knowledge that The Good Old Days always cares about the community.

Of course one reply I got was that it would just be too expensive (traffic) if people could easily "leech" full sets from the site and that it would not be a good thing anyways (which I disagree on because in terms of preservation imho sharing and spreading is THE most important part). So the real question might be: How could we achieve sharing the images without increased costs, and is that even wanted by the people who already contributed?

@Mr Creosote

If you feel our conversation is not presented correctly or you want to add something I'd be glad if you speak up (Yeah, I know it's your forum and you can do what you want anyways, just wanted to encourage a conversation)

No need to correct anything from my side. This is obviously a very complex issue, and it should be obvious to everyone that what you posted can only be seen as a shortened version of the views we both have. I'm curious what others have to say, too

Only one central point to add so far: As with everything on this site, this project is a case of 'the people who do the work decide'. If there are people willing to put time and effort into it, many things are possible and imaginable.

I personaly added a lot of disk images to that list and I didn't expect to get anything. I made the images as my private backup and I'm glad that I found someone who was interested in them too so I shared them with TGoD. It's like a online backup of my collection. If I ever loose one of my games I can count on Mr. Creosotes conservation skills and get my copy back.
If everybody who contributes only once to the list could download all images I would expect traffic costs exploding so that's no real choice.
You are missing one of these games? Write an test and the Disk Image will be put online free for everyone who's interested.

I was thinking on asking Creosote to hand me all the images we have (I don't have all of them, and much less sorted) and share them a bit to try to get more people interested. As, like you say, it's hard to help a project where you feel you won't get anything.

That's for example the reason of trying to start the project on UG, even thought I fear that will just double the work. But that will attract some more people.

I think T-Pow's idea is good. You help to the project, you can download the project. In Replacement Docs there is a similar idea (you upload ten documents, you can download from the big list of still not edited manuals), and people would feel rewarded for the effort.

That's not what he said though, is it? His suggestion was to write a review to be able to download that specific title. He specifically mentioned 1-title uploaders should not be able to get everything (which I fully understand seeing the traffic costs). A free to download solution would only be an option if it doesn't create highers costs (I'm wondering though what kind of host this site is on? I've seen unlimited space/traffic offers for less than $150 a year)

I think what you're talking about is something like a ratio. Like you may download x times as many as you upload. I think such a system would probably work quite well. Maybe structure it like this

level A (less than X uploads, 20 maybe?): no downloads other than your own uploads / on request
level B (more than X uploads): 5*X downloads
level C (more than Y uploads, 100 maybe?): unlimited downloads (aka FTP access?)

That way any substantial contributor would be able to get the whole set if he wants but people with only a couple of games could use it as a backup like you said.

@T-Pow
"You are missing one of these games?"

That's not point. I haven't really checked the list but I should have all those as installed Abandonware. Not that I even play those, I've already got more original games than I can play. For me a project like this is more of an archival thing. Think of it as an eternal backup of all games in their original form. To get an idea what I mean:
http://www.igda.org/preservation/

I was thinking on a more simple "You upload once, you can download everything always".

Ah I see. I don't think Mr Creosote will like that

On a different note: Does anyone know a tool that's able to merge two images? Shouldn't this be possible technically? It would be very helpful in cases where we have two people who dumped the game but both with errors on different tracks. The track-based structure should theoretically allow merging both images using the good tracks only (unless both have the same bad tracks). For rare games this might be the only option to recover them?!

I don't want to kill the discussion before it even began by writing something long-winded covering life, the universe and everything, so just a couple of important points for now.


In fact, I think his point is very good. You keep pointing to that link, but what do we find there? Not a single word about the necessity to make archives public. I quote:

This is already met with the current setup. While you may argue this isn't perfect (agreed), there is no contradition between what TGOD does, what T-Pow says and what the IGDA group proposes.


Yes, and I've seen Xoom offer unlimited space / traffic for free in 1998. Have you ever tried using any of these offers in an 'unlimited' fashion? You're bound to have mail from your host very quickly.


Will you set it up? Keep in mind that this means everyone who uploads 20KB will be able to freely download gigabytes of stuff. And experience tells people don't just download everything once. A week later, they'll be too lazy to search for some files on their own hd, because they can't quite remember where they put them, and just download them all again - because it's so much more convenient.

If we're talking about a quota system, why not just quid pro quo, i.e. 1:1? Very simple and somewhat fair. At least when ignoring for a moment that obviously, people will use either no compression or the worst compression they can get away with for upload (to get more out of their uploads) while files in the download area will obviously be compressed the best way possible. Nobody would be able to grab the whole collection then? Who says that needs to be done from a central location? Four contributors could easily get everything with their combined quota and then share the stuff they got with the others. What I'm trying to get at: Please try to think about decentralized systems as well.

A totally different suggestion which I already mentioned in the discussion with hydrox: It would be simple (and I can't think of any reason against it) to offer access to the disk image catalogue based on the site's contribution rating. Just like access to the faster download server for CD images is currently handled that way.

Sorry, we were obviously replying at the same time Regular dd will do that. See the man page for the options 'count', 'seek' and 'skip'.

huh? only if they together contributed more than 50% of the total amount of images. Otherwise that would be impossible. Imagine a situation where 50 people have uploaded ~50 images each. That's 2500 images in total. One single person of those has his own 50 and may download 50 with a 1:1 ratio (I don't get why you want to do this based on KB instead of number), so he can have 100 of those 2500. Even with perfect collaboration (no one uses his quote to download stuff one of the others uploaded) you'd need 25 of the total of 50 people to download everything and share it so all 25 have a full set. This will never work. Bottom line: 1:1 doesn't allow acquiring full sets.

About the hosting, no, I never tried to max out one of those unlimited offers. I'm aware that a host might contact you if you produce way too much traffic. But then again, have you actually experienced this? Do you know what the limit might be? I have no idea how much traffic you produce with this site but then again you got to remember that 200 floppy images is still less data than let's say your Dracula Unleashed download. I can hardly imagine this project's traffic could even compare to your normal traffic. But I agree a decentralized solution would be best.

About the IGDA link, I'm just using it to show what preservation is and why we need it. Not because of their methods. They are part of the industry and have to respect all copyright issues. That's why they cannot even think about public access to such an archive.



no idea how that current system works so I can't comment on that

@dd

well I'm not a linux/unix user but to do that one would have to know where a certain track in the IMA starts and ends, right? Does anyone know how an IMA file is structured? Does it contain info on where a track was or is the track structure completely removed in the image?

Which is why 'normal' users download that game from an alternative server...


You get points for contributions to the site (see T-Pow's profile, for example). At the moment: More than 100 points equals access to the better server when downloading bigger files, so that one's 'flat rate' - once you reach that number, you're good. Another simple way would be to give access to one game of the disk image collection per X points (ordered by the ID listed on the overview page) - but that's just an example.


dd is available for MS-Windows and MacOS as well (link in the images FAQ). Are you saying you're using neither of those three?

As for the rest of your doubts concerning 'combining' images, I'm not sure what the problem is. While you're reading an image from a floppy disk, dd will tell you which sectors or tracks are faulty, so that information is available.

I don't know how many games you have hydr0x but I have send in roundabout 80 Images in two or three weeks. So it isn't that hard.

Where do you want to preserve the games for mankind? In your home or do you have your own site which could get some sort of partnerside which shares it's content with TGod?

I think you would be better off with allready installed versions which can be found all around the web because in my opinion you don't won't to conserve the games for future players. You just want to own your personal copy.

please, don't even start that discussion, I'm a real game collector. My PC game count is somewhere >400 and my total game count is somewhere above 1500. That's original, boxed games. If you don't trust me, as an example here's a pic of my snes games:



Also, I actually do have a huge abandonware set, I don't need any disk image to play a game.

"but I have send in roundabout 80 Images in two or three weeks. So it isn't that hard."

What are you even trying to say here? I never said uploading is hard.

@Creosote

I'm using WinImage to create my images and it also does tell me which track/head had errors. What I don't know is how tracks and blocks (that's what those dd commands use) are related.

Okay, so that system basically would be similar to my ratio proposal in so far that as soon as you hit a certain number of uploads (reflected in those points) it allows you to download freely from the set from a FTP, right?

So if uploading isn't that hard, why do you diskuss instead of putting on some Images and see what you get for it?

I don't really see the need for yet another user level system (FTP) when on HTTP level, the infrastructure is already there and allows for complete automation.

There isn't. I just said FTP because that's what we were talking about before

@T-Pow

because

a) I don't want to put energy into a project if I'm unsure if it's the right concept

b) I'm not mainly discussing these points because of me but because I want to make sure more people will be encouraged to join

What concept is there on making an image of a floppy? You have just talked about the need of preserving games, yet you seem to distrust the idea of preserving games.

The first idea that should come to mind to anybody who owns old floppies, old disks which are decaying fast, is preserving them so they aren't lost.

If after doing that you want to upload them to us, great!

If after doing that you want to upload them in any other place, great!

If after doing that you want to upload them you want them to be well preserved and then upload them here, there and anywhere else you see fit, ever greater!

That's exactly the way I found TGoD. I allready made Images and some of my Disks where broken. So I searched for sides which offer Images and guess what. I only found TGoD. So the public doesn't seem to be interessted in Disk Images.
Except Mr. Creosote nobody whas willing to help finding images to exchange my defektive disks that also seems to be an indicator how much most people concern about preserving games.

On the other hand TGoD isn't an image site. Every game which is added must have an test and screenshots a least. That's the whole concept of the site if im not terrible wrong.
So Images are just an little bonus.

Let's quickly recap which options we've got so far. I'm trying to stay neutral in this post, so please correct me if I get something across incorrectly.

- Keep everything as it is, i.e. make disk images publically available when the respective game gets added to the site (which everybody can do). Simple, because it requires no work. Critisized, because it's supposedly not open enough.

- Set up an FTP server hosting all the files. Every contributor gets his own account. Upload / download quota based. Exact quota circumstances to be discussed. So far, nobody has volunteered in setting this up and maintaining it.

- Set up an FTP server hosting all the files. Every contributor gets an account (doesn't really need to be individual) and gets the right to download as much as he wants after uploading something. So far, nobody has volunteered in setting this up and maintaining it.

- Hand out file access based on site's point system. Easy, because infrastructure is already there. Details (what is worth what and so on) would have to be discussed. Potential problem / advantage (depending on point of view): Disk image project is completely embedded into the site, i.e. disk image contributions alone probably won't give you access to a lot of stuff. On the other hand, other content contributions to the site will.

- Use torrents to seed the collection. Could be used in addition to any of the other points. Problem: No direct incentive for contributions in this case (just, depending on the tracker, to seed the collection as it is).

For all options but the first one, we'd have to clear things up with the previous contributors (T-Pow is one of them). They sent in their disk images under the assumption that the games would be handled the way described in the first point - that was the initial announcement when we started this. Not impossible to contact them all again, but something to keep in mind.

Don't worry. I do have backups of most of my games, including an off-site backup. I was talking about the concepts of collecting and sharing them which Mr Creosote just listed



But that's exactly my problem. If you want to only make an image available if there's a review than imho that's just wrong and the site probably not the right place to undertake such a project. This is also exactly not the kind of concept I'd like to contribute my images to.

It's also untrue that there aren't other images out there, it's just a little hard to find them. There's the TOSEC image set and there are other sites where people share images on request.

@Mr Creosote

Yeah, I guess that sums it up quite well. Are you sure the contributors submited their images under the premise that they only become available for download when the game goes online as a review? Cause the upload function on the site doesn't mention this at all. In fact it doesn't mention what happens to the images (which was another problem I mentioned in our discussion).