"Xz format inadequate for long-term archiving"

Maitreya · Guru Joined: 11 Jan 2006 Posts: 441

So apparently XZ is now BAD™ : http://lzip.nongnu.org/xz_inadequate.html

What are the thoughts on this, is over exaggeration or does it have some merit? (Like is the developer disliked or something)

Maitreya · Guru Joined: 11 Jan 2006 Posts: 441

Ah, I think I can answer my own question:

erm67 · l33t Joined: 01 Nov 2005 Posts: 653 Location: EU

Since that is what the developer of lzip data compressor thinks about xz it would be also interesting to hear what the developers of xz think about the lzip compression format :-)

_________________
Ok boomer
True ignorance is not the absence of knowledge, but the refusal to acquire it.
Ab esse ad posse valet, a posse ad esse non valet consequentia

My fediverse account: @erm67@erm67.dynu.net

erm67 · l33t Joined: 01 Nov 2005 Posts: 653 Location: EU

BTW the standard for long-term document archiving is https://en.wikipedia.org/wiki/PDF/A and it only allow the (de)flate compression algorithm (used also by gzip) since it is considered to be patent free.
_________________
Ok boomer
True ignorance is not the absence of knowledge, but the refusal to acquire it.
Ab esse ad posse valet, a posse ad esse non valet consequentia

My fediverse account: @erm67@erm67.dynu.net

mike155

Maitreya · Guru Joined: 11 Jan 2006 Posts: 441

Maitreya · Guru Joined: 11 Jan 2006 Posts: 441

Ant P. · Watchman Joined: 18 Apr 2009 Posts: 6920

Ah, this crank again. If you get into compression performance, ck's lrzip beats the crap out of both xz and lzip in speed and ratio anyway (even without -z), and has done for about 15 years. zstd outperforms them on all fronts simultaneously too.

lzip is still taking angry little pot-shots at a format that was obsoleted long ago, which shows you how relevant lzip is in today's world: it never was.

Dr.Willy · Guru Joined: 15 Jul 2007 Posts: 547 Location: NRW, Germany

Ant P. · Watchman Joined: 18 Apr 2009 Posts: 6920

I'll concede that, but I'd also prefer the 10 seconds of extra download time over 20 seconds of CPU time with possible OOM.

Dr.Willy · Guru Joined: 15 Jul 2007 Posts: 547 Location: NRW, Germany

NeddySeagoon · Posted: Fri Jun 07, 2019 5:12 pm Post subject:

Once you move away from parchment or clay tablets, nothing has lasted even 25 years.

We don't have a long-term archiving format.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.

Amity88 · Posted: Sat Jul 27, 2019 4:15 pm Post subject:

@Neddy,
Don't forget the cave paintings

That was one good time tested long lasting data archive
_________________

Goverp · Veteran Joined: 07 Mar 2007 Posts: 1993

For archiving, surely the answer is "no compression". That way (a) you don't get problems if you lose the program, and (b) the archive contains redundant information that can help you detect and correct corruption, and (c) even if parts are corrupt, other parts will be salvageable.

An equally important part of archiving is the physical medium. Some of us will remember the batch of Ampex (?) tapes that went "read once" when the binder aged. Anyone still got important information on Zip disks? Or backups on floppy disks? (I've a backup on an 8" floppy, some punch cards, and some paper tape.) Good luck reading them.

I remember some researchers at Southampton university coming up with a very dense archive on glass that sounds ideal, providing enough people used it to ensure a perpetual supply of reading hardware. But like most such ideas, it hasn't caught on.
_________________
Greybeard

NeddySeagoon · Posted: Sun Jul 28, 2019 10:10 am Post subject:

Goverp,

I still have an APPLE ][ that works with the original floppies. Its nearly 40 years old.
Paper tape is not an issue. I can read that by eye :)
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.