Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Don't Use RAR

group.miletic.net

Post reply on HN
Print view View original

131–140 of 218 posts

Re: Don't Use RAR

#131

Post by grrrrlpower »

throwaway81523 wrote: post #73
nikanj wrote: post #46

Earlier quoted context omitted.

If you've got the resources to mount a credible legal defence against Microsoft

Microsoft ironically made its fortune by selling software that ran on IBM PC compatible x86 boxes, built around PC BIOS's that started from a clean room reimplementation of the reverse engineered IBM BIOS. IBM itself was not very significant in that market except at the very beginning. Clones for me, but not for thee.

I look back now at how weird it was going to the store looking for the IBM compatible computers amongst the Amiga, Apple's, Commodores and such.

Re: Don't Use RAR

#132

Post by nix23 »

wongarsu wrote: post #130
nix23 wrote: post #104

Earlier quoted context omitted.

Why would you try to compress pictures....in your case zip is as good as everything else (and less problematic to open)

Ok, bad example, imagine I would be sending you ... text files or something. The point I was trying to make was how viable it is, not how useful it is.

Yes please zip there too since i don't think it would be gigabytes of text.

If sending gigabytes of text please use lzip ;)

Re: Don't Use RAR

#133

Post by pid-1 »

grishka wrote: post #82
throwaway81523 wrote: post #10

The rar license prohibits writing a compatible compressor? What kind of nonsense is that? I clicked the license link and there is a scary looking anti-reverse-engineering clause, but 1) that sounds legally dubious given how the program is distributed; and 2) there is a FOSS decompressor, and studying that may be sufficient to write a compressor without examining the non-free official compressor. The article is right…

In some countries there's an exception in the copyright law if you're reverse engineering something to provide compatibility.

Also in many countries software / algorithm parents do no exist, so you could write your own rar utility.

Re: Don't Use RAR

#135

Post by account42 »

ksec wrote: post #100

If you have half decent internet connection, Compressing and Decompressing the files is going to take more time than simply just sending it or downloading it. I still remember we use to compare compression tools and format on compression ratio. But I cant record the last time I care about any of these any more. If it needs to compress just use zip. The same happened to Audio Codec as well. Video Codec is where we sti…

> If you have half decent internet connection, Compressing and Decompressing the files is going to take more time than simply just sending it or downloading it.

This might be true for transmitting files 1-to-1 (and even then, for every connection speed slower than memory there is likely a compression algorithm that makes sense) but most files are distributed 1-to-many which is why algorithms like deflate that might be slow to compress but are fast to decompress make sense.

For example, Linux distributions would not keep switching between gz, bzip2, xz and now zstd for their packages if the choice was pointless.

Re: Don't Use RAR

#136

Post by sumtechguy »

lifthrasiir wrote: post #56
wyattpeak wrote: post #43

Does anyone know why RAR is better compression-wise than the two more modern standards? It doesn't seem like it's been updated in a long time, and I would have expected modern software from two of the best houses in the world to handily beat it. Is it that good, or is it a matter of priorities, or something else?

> Does anyone know why RAR is better compression-wise than the two more modern standards? It simply uses a much slower algorithm (prediction by partial matching, PPM) in the highest setting. Those modern standards are designed to be fast enough and improve the status quo in that performance target. If they were only concerned about the compression ratio there are tons of other algorithms that would handily beat RAR a…

7zip also has a ton of settings where you can get it in the same ballpark of compression ratio. But LZMA2 usually consumes more memory to do better. It also takes a decent amount of time the more you fiddle those settings. The defaults on 7zip are fairly tame but 'good enough'. I found in my cases if you turn on solid archiving and duplicate file matching you are usually just as good as RAR or usually better. But costs speed.

Re: Don't Use RAR

#137

Post by kazinator »

"Furthermore, the small differences in the resulting file size matter less over time as bandwidth is increasing and increasing fast."

Firstly, this argument, if true, would add up to: if you're on Windows, don't use anything other than ZIP.

The argument is not true, because although increasing storage and communication resources make compression less relevant for small files, there is content which is getting larger because of available bandwidth, including aggregations of smaller content.

Re: Don't Use RAR

#138

Post by 2ion »

mikl wrote: post #72
lifthrasiir wrote: post #60

Earlier quoted context omitted.

Zstandard and Brotli are also not an archiving format, which the layperson normally expects to be a "compression" format, so the comparison is a bit off. But that doesn't really matter because there is 7-zip which is an archiving format AND compression format AND competitive to RAR AND open-source.

Yeah, you’d need to use something like *nix `tar` to do the archiving part. I think newer versions of tar already have support for Zstd.

a tar.something is just a tar archive passed through a compression filter like zstd or gzip. Anyway, tar needs metadata extensions in order to allow random access; most tarballs you'll find don't. So you need to read 5TB of data to find the file at point 5TB in an archive of size 10TB. Which is why as a general purpose archiving format, tar needs to be applied to use cases where it works. It gets old really quick if you don't have random access and random access requirements, plus, remediated by SSDs these days for a bit, it's really time consuming.

Re: Don't Use RAR

#139

Post by anthk »

throwaway81523 wrote: post #10

The rar license prohibits writing a compatible compressor? What kind of nonsense is that? I clicked the license link and there is a scary looking anti-reverse-engineering clause, but 1) that sounds legally dubious given how the program is distributed; and 2) there is a FOSS decompressor, and studying that may be sufficient to write a compressor without examining the non-free official compressor. The article is right…

Unar is a libre alternative.

Re: Don't Use RAR

#140

Post by kazinator »

mimsee wrote: post #125
kstenerud wrote: post #36

Earlier quoted context omitted.

From the winrar download page you can download the binaries directly without agreeing to any license, which means its clauses wouldn't apply. It's already been established in the courts that merely using a piece of software does not imply acceptance of licensing terms; the parties must actively agree to it. That's why so many companies push in-your-face agreements that you must actively click "I agree" to.

Could I then download a zip of torvalds/linux from github and use it without abiding by the license as it’s never thrown in my face?

Yes. The GPL is not a license which dictates how you can use software, such as that you must not read it, or use ideas from it.

EULA restrictions like "no reverse engineering" are not rooted in copyright law, and borrowing ideas is in the realm of patents.

EULA's try to leverage copyright law in order to impose non-copyright restrictions, using the concept that if you violate any of the arbitrary non-copyright restrictions, the thereby violated license lapses, and that license is the document which allows you to have a copy of the software (not the fact that you paid).

Post reply on HN