Re: Archiver able to handle corrupted data and to show

On 4/18/05 4:54 AM, in article 3cheg6F6lbtg4U1@individual.net, "Claudio
Grondi" <claudio.grondi@freenet.de> wrote:

> According to what I have read from other postings to
> this newsgroup and on the Internet, StuffIt is a
> very controversial

Only in that it has recently broke "new ground" - lossless compression of
already compressed data - jpeg specifically in the first release.  I'm not
sure how that's a bad thing.  It's been proven to work by 3rd parties, and
is shipping commercially now - if there was any initial controversy, it's
long since been put to bed.

> archiver and except postings which
> look to me like advertizing (in my eyes it is just not
> possible to recover large area of damaged data at
> the cost of 1-3% larger archive) I have not heard
> much good about it.

We'll I guess I won't point you to things like the PC Magazine's Editor's
Choice awards, etc, because you would think that is "advertising". :-)

> By the way: I have already got WinRAR and
> in its headlines it promises same features, but
> going into details of its helpfile shows, that:
> "1% of the total archive size, usually allows the
> recovery of up to 0.6% of the total archive size of
> continuously damaged data" and
> "it is impossible to recover files, which fail the
> CRC validation" (in case the process of
> repairing using the recovery record failed)
> even if "it is still possible to recover
> undamaged files, which were inaccessible due
> to the corrupt archive structure." (but not in
> case of solid archives)

RAR has a similar feature - but I don't know enough about their
implementation to comment on it.

Stuffit allows for different options (how much overhead to include - and
thus the amount of damage that can be recovered from:

1/128 - 1 overhead
1/64 - 3 overhead
1/32 - 6 overhead
1/16 - 14 overhead
1/8 - 33% overhead

> Beside this, Stuffit is like WinRAR not free and
> there is no source code available allowing to
> investigate if it actually does what it promises to
> do.

Since when to you need source code to test if something works or not?  You
may need source code to see HOW it works, but not IF it works.  And yes, RAR
and Stuffit are generally not free  - it cost money to pay engineers to
create this stuff after all.... (there are trial/shareware versions of the
stuffit compressor - and StuffIt decompression is free)

> In addition I can't evaluate Stuffit without
> exposing any credit card information, so:

Not true - You can download the trial right here - no credit card required:

http://nct.digitalriver.com/0135/standard.html

- Darryl

> Any other hints?

Sure - pay less than $40 for StuffIt (or rar) to get existing software that
solves the original problem you posted, or you can always spend a few
ws/months/etc writing/testing/etc your own Error Correcting code in
combination with a public source tar+bzip, and end up with something that is
close to this one feature of the commercial applications - depends on what
your time is worth to you.

- Darryl

> Claudio
>
> "Darryl Lovato" <Darryl@Lovato.com> schrieb im Newsbeitrag
> news:BE88BBB8.333648B5%Darryl@Lovato.com... 
> not 
> even 
> enabled, 
> will 
>
>