Re: D9090 back to life !

From: Rob Eaglestone <robert.eaglestone_at_gmail.com>
Date: Tue, 28 Jan 2014 14:57:50 -0600
Message-ID: <CABNTyr-HufCU54Hi0hYk9C5QDN4A4Fpw3UCrutbU=zrjTL-sdw@mail.gmail.com>
The 'degenerate' case here of course is a file scan for a Magic String,
potentially parametric, which returns the appropriate image factory...



On Tue, Jan 28, 2014 at 2:22 PM, Rob Eaglestone <robert.eaglestone@gmail.com
> wrote:

> > the major problem with that approach is that it doesnt work on disks
>> that are
>> > completely filled with data (which isnt really unlikely, lots of cracks
>> and
>> > demos use the dirtrack for files, for example) - which rules it out as a
>> > generic solution.
>>
>> I think another big advantage is that it modifies the data contents
>> themselves. This is something I personally do not want to have. For me,
>> one use of a disk image is to preserve data that was once on a real
>> disk. Here, changing the disk data that should be preserved is no
>> option.
>>
>
> Granted. There are disks which ought never be modified.
>
> However, it seems like the best solution is going to follow The Burrito
> Principle (80% of the meat is in 20% of the burrito).
>
> In other words, a solution which will work on 80% of all diskette images,
> and which yields 100% backwards compatibility for those diskettes, is most
> likely the one to pick.  So, fiddling with Directory Track N-1 is a valid
> solution.  Tacking on a "trailer" block is more robust but slightly less
> valid (programs might choke on the file size).
>
>
>
>
> On Tue, Jan 28, 2014 at 1:47 PM, Spiro Trikaliotis <
> ml-cbmhackers@trikaliotis.net> wrote:
>
>> Hello,
>>
>> * On Tue, Jan 28, 2014 at 07:59:10AM +0100 Groepaz wrote:
>>
>> > the major problem with that approach is that it doesnt work on disks
>> that are
>> > completely filled with data (which isnt really unlikely, lots of cracks
>> and
>> > demos use the dirtrack for files, for example) - which rules it out as a
>> > generic solution.
>>
>> I think another big advantage is that it modifies the data contents
>> themselves. This is something I personally do not want to have. For me,
>> one use of a disk image is to preserve data that was once on a real
>> disk. Here, changing the disk data that should be preserved is no
>> option.
>>
>> Also, someone suggested changing the BAM, because it can be rebuilt by
>> "V"alidate. No, I do not like this, because the BAM contains info, too.
>> There can be an error (accidentially or intentionally) that might serve
>> a purpose.
>>
>> Thus, please, DO NOT MODIFY THE DATA CONTENTS SOMEONE MIGHT WANT TO
>> PRESERVE.
>>
>> I do not like to have it optional, because people will not know about
>> these options, and they might accidentially use it.
>>
>> If people do not like a container format, another option might be to add
>> another file beside the original one. For example, have a file
>>
>>    MYSUPERDUPERDISK.d9060
>>
>> and a meta-file
>>
>>    MYSUPERDUPERDISK.d9060.meta
>>
>> (or, if you want to play with ADS on Windows, use
>> MYSUPERDUPERDISK.d9060:meta ;)
>>
>> Regards,
>> Spiro.
>>
>> --
>> Spiro R. Trikaliotis
>> http://www.trikaliotis.net/
>>
>>        Message was sent through the cbm-hackers mailing list
>>
>
>


       Message was sent through the cbm-hackers mailing list
Received on 2014-01-28 21:01:18

Archive generated by hypermail 2.2.0.