An error was unable to be described usefully



  • @powerlord said:

    The patents that cover the format Zip uses only recently expired (they were the same ones Unisys was actively suing companies that used the GIF image format over).

    Having said that, I use .tar.gz and .tar.bz2 all the time... on *NIX machines.

    pkzip 1.0 used a compression method called "Shrinking" that was based on the patented LZW.  In 1993  PKWARE released pkzip 2.0 which got rid of all the miscellaneous compression methods of zip 1.x and replaced them with a single new compression method called "deflating". 

    Which was my whole point.  Since 1993  there has been a compression format that has been freely available to anyone who wants to use it.  For iexample, the ability to uncompress zip files is built into Firefox (those "jar" files are actually zip files).  There simply is no excuse to be a retard and continue messing with tar.gz nonsense.  It isn't 1992 anymore.

     



  • .tar.gz always gets on my nerves.



  • @El_Heffe said:

    Which was my whole point.  Since 1993  there has been a compression format that has been freely available to anyone who wants to use it.  For iexample, the ability to uncompress zip files is built into Firefox (those "jar" files are actually zip files).  There simply is no excuse to be a retard and continue messing with tar.gz nonsense.  It isn't 1992 anymore.

    Well... tar has existed since at least 1988, and gzip since 1993. They have also been freely available to anyone who wants to use them. And while I'm not sure what Firefox has to do with this, but it can uncompress gzip streams (because they're used in HTTP). I don't see your point.



  • @Spectre said:

    @El_Heffe said:
    Which was my whole point.  Since 1993  there has been a compression format that has been freely available to anyone who wants to use it.  For iexample, the ability to uncompress zip files is built into Firefox (those "jar" files are actually zip files).  There simply is no excuse to be a retard and continue messing with tar.gz nonsense.  It isn't 1992 anymore.

    Well... tar has existed since at least 1988, and gzip since 1993. They have also been freely available to anyone who wants to use them. And while I'm not sure what Firefox has to do with this, but it can uncompress gzip streams (because they're used in HTTP). I don't see your point.

    .zip ships by default on every OS. Every user you're likely to encounter, unless they're running something really ancient, can handle .zip files without any additional software downloads. Every user.

    That is not true of .tar, .gz, .rar, .7zip, or whatever crazy alternate format you prefer. Downloading additional software for an operation that's likely to only occur *once*, *ever* is a huge pain in the ass. If you use anything other than .zip, you're being a huge pain in the ass. You're forcing the recipient to spent significantly more time Googling/installing/dealing with shitty UIs/turning off spyware attached to the installer/etc than they spend actually opening the archive.

    The disk savings of using a "more efficient" compression scheme *do not matter*, because it's overweighed 50,000 times by the annoyance of having to find a piece of software to deal with it. Only a complete inconsiderate asshole would think otherwise, which probably explains why so many people here are sending around .gz files.



  • @Spectre said:

    Well... tar has existed since at least 1988, and gzip since 1993. They have also been freely available to anyone who wants to use them
    And only a tiny portion of the populations uses them.  Mostly people who are apparenty stuck in 1988.
    And while I'm not sure what Firefox has to do with this
    It's an example of a program which can natively handle .zip files and is available for many platforms, which means there's no excuse for *nix tards to continue using .tar.gz nonsense.
    I don't see your point.
    If you use anything other than .zip, you're being a huge pain in the ass (See Blakeyrats post above this one --- OMG I'm agreeing with him!).



  • Zip can preserve file names, paths, and... VAX/VMS file version IDs. It cannot preserve special attributes (symlink, fifo, so on) or file ownership.

    That's why unixoid use tar (which preserves all these, but doesn't compress at all), then compress the tar file with gzip, bzip2, or xz. The ones stuck in 1988 still use COMPRESS (*.Z).

    Now, I totally agree that "TAR inside an archiver" sucks. It's not a seekable archive, because to the "outside compressor" it's all a single file stream. So, to reach a file in the archive, you need to uncompress everything in front of it. TGZ fans tend to claim it's a good thing because it allows for a few % better compression (the data table covers whole archive, not per-file). RAR and 7Z offer similar behavior as an option, called "solid archive". Off by default.
    Worst of all, TAR was meant for mag-tapes, a linear medium. So, as opposed to all modern archivers, it doesn't create a file index at beginning (or, end, in case of ZIP) of the file - you need to read (uncompress!) the whole archive to even get a complete LISTING OF CONTENTS!

    What baffles me is that there's still no archive format that CAN preserve all the metadata and doesn't have the crappiness of tar. Or is it, but no one heard of it?



  • @bannedfromcoding said:

    What baffles me is that there's still no archive format that CAN preserve all the metadata and doesn't have the crappiness of tar. Or is it, but no one heard of it?

    SQL Server .bak files do that. EDIT: Of course they're only designed around backing-up databases, so I doubt they'd work in the general case.



  • @blakeyrat said:

    .zip ships by default on every OS
     

    As I said earlier, those OS-based Zip programs don't (always?) create/open big zip files (over 4GB uncompressed) so you'll have to use a better program anyway.



  • @Zemm said:

    @blakeyrat said:

    .zip ships by default on every OS
     

    As I said earlier, those OS-based Zip programs don't (always?) create/open big zip files (over 4GB uncompressed) so you'll have to use a better program anyway.

    Ok, but in 99.99999999999% of the time it's fucking fine. If you're the 0.00000000000001% then go ahead and use whatever, ok? And then you're not being a jackass.

    And no I didn't count those zeroes.


  • Garbage Person

    All the major third-party archivers support self-extracting archives now. Use the fucking feature, you assholes.



  • All the major third-party archivers support self-extracting archives now. Use the fucking feature, you assholes.

    No, please don't. A) You're asking me to run an .exe B) This isn't cross-platform


  • Garbage Person

     @nexekho said:

    All the major third-party archivers support self-extracting archives now. Use the fucking feature, you assholes.

    No, please don't. A) You're asking me to run an .exe B) This isn't cross-platform

    Oh gods, not an exe! It's not like we've spent the past 15 years figuring out all this antivirus crap for any reason at all, right?

    And cross-platform is irrelevant - if you have the archiver, it'll open it anyway. Which, incidentally, means that if you don't want to run the exe, you can open it anyway with the fucking archiver.

    Furthermore, if you're distributing software binaries, this argument should be irrelevant.



  • I dislike ZIP files because they always take an eternity to open when there's a few hundred files in them (there's a process that creates about a thousand of them daily, and these files are archived weekly since they're very compressible, but are needed a few times per month - when the script was packing these into ZIP files, it always took several minutes to open the archive, regardless of the tool used; after I changed to .7z files, the archives take around 2 seconds to open over the network - all users that need these already had 7-zip installed, because Windows' built-in ZIP support took even longer to open them).



    As for 7-zip's GUI, from what I've read, it's more of an accident that it's present at all - the author doesn't care about it at all. I still find it much better than WinZIP, though my preferred GUI archiver is WinRAR (even though I only ever use it to decompress archives).



  • @El_Heffe said:

    @Spectre said:

    And while I'm not sure what Firefox has to do with this
    It's an example of a program which can natively handle .zip files and is available for many platforms, which means there's no excuse for *nix tards to continue using .tar.gz nonsense.
     

    Last time I checked, jar files were [url=http://en.wikipedia.org/wiki/JAR_file]Java Archives[/url].  While they are indeed really zip files, Firefox doesn't open them but launches the Java Plugin to open them.

    Unless Firefox has coopted the extension for something else, that is.

    As for tar.gz, it follows the UNIX philosophy of splitting large tasks into smaller components.  In this case, combining a bunch of files into a single file and compressing files are done as two separate steps.  In fact, I'm glad you mentioned Firefox, as gzip is supported cross-platform as well, namely by any [url=http://www.w3.org/Protocols/rfc2616/rfc2616-sec3.html#sec3.5]HTTP 1.1-compliant web browser[/url], which includes Firefox.



  • Somehow I stumbled into this thread.

    @powerlord said:

    While they are indeed really zip files, Firefox doesn't open them but launches the Java Plugin to open them.

    It depends on the URI. If you use jar: prefix where the protocol part goes and a bang to separate the path of the jar from the path of the archived file, you can open jars. In fact, you can open any zip files. Right now, I've got a tab open somewhere to [url]jar:file:///C:/search/jdk-6u24-docs.zip!/docs/api/java/lang/Object.html[/url].

    I'll hold my Unixy comments regarding the tar.gz vs zip discussion. Apparently arguments of culture, utility, and technical superiority doesn't count unless you are one of those mouse-addicted Windows luser types in the 99.9-repeating percentile majority.



  • @Xyro said:

    Filed under: secret SSRR reference

    I almost started wondering if you actually have a C:/search folder.



  • @El_Heffe said:

    @Spectre said:

    Well... tar has existed since at least 1988, and gzip since 1993. They have also been freely available to anyone who wants to use them
    And only a tiny portion of the populations uses them.  Mostly people who are apparenty stuck in 1988.

    And people actually working with Unix systems. Most UNIX-related stuff is transported/released in that format, and I've used tar.bz2 for archival purposes. RAR is good for archiving as well, and has a compression algorithm which is on par with Bzip2... and most people I know use that one as well.

    Moot point then, because anyone having WinRAR can already open tar.gz or tar.bz2 files. MacOSX supports tar.gz out of the box, it looks like it is only Windows that doesn't support that one... even then, you could add support with Services For Unix if you really want to have free OS support for that.

    I have to use RAR at work, RAR at home, having a PC without winrar is an oddity.



  •  @danixdefcon5 said:

    @El_Heffe said:

    @Spectre said:

    Well... tar has existed since at least 1988, and gzip since 1993. They have also been freely available to anyone who wants to use them
    And only a tiny portion of the populations uses them.  Mostly people who are apparenty stuck in 1988.

    And people actually working with Unix systems. Most UNIX-related stuff is transported/released in that format, and I've used tar.bz2 for archival purposes. RAR is good for archiving as well, and has a compression algorithm which is on par with Bzip2... and most people I know use that one as well.

    Moot point then, because anyone having WinRAR can already open tar.gz or tar.bz2 files. MacOSX supports tar.gz out of the box, it looks like it is only Windows that doesn't support that one... even then, you could add support with Services For Unix if you really want to have free OS support for that.

    I have to use RAR at work, RAR at home, having a PC without winrar is an oddity.

    WinRAR isn't the only tool that can open tar.gz files (and .tar.bz2 files)... you can also use WinZip, PKZip (it still exists!), and 7Zip.  Having said that, 7Zip's Extract here shortcut doesn't properly deal with programs that are tarred and gzipped/bzipped and just leaves the tar file there, which you must then Extract Here again.

    Then again, I think it was already mentioned how bad 7zip's GUI is.


  • ♿ (Parody)

    @powerlord said:

    Then again, I think it was already mentioned how bad 7zip's GUI is.


    I've never claimed to be an expert on usability, but I've never had a problem using 7zip (and the previous mention was about as informative as this one), either via GUI, command line or shell extension. What's the big deal?



  • @danixdefcon5 said:

    I have to use RAR at work, RAR at home, having a PC without winrar is an oddity.

    Last time I used WinRAR has to be at least 10 years back. I'm angry whenever someone packs stuff as RAR, not because I can't unpack it, but because there's no proper official kioslave for it yet (far as I know).


    I'm 100% fine with zip. I'm 100% fine with .tar.* too. If I release things for Windows, I make sure to include a .zip archive. I use .tar personally to pack MP3s, for the lulz. People who require others to use RAR, 7z or Unix formats are just assholes. You can always make 2 or more different packages if you prefer one over the other, you know?


    Other than that, I find nothing wrong with zip. Yes, maybe there are things that pack better. And for specific purposes, other packers are definitely superior. [i]But you shouldn't use some rare archiving format for public distribution of shit or distribution to an audience not definitely known to be able to unpack it just because you prefer it for some religious reason![/i]


    Can we kill this thread now and move on?



  • @ender said:

    I dislike ZIP files because they always take an eternity to open when there's a few hundred files in them (there's a process that creates about a thousand of them daily, and these files are archived weekly since they're very compressible, but are needed a few times per month - when the script was packing these into ZIP files, it always took several minutes to open the archive, regardless of the tool used; after I changed to .7z files, the archives take around 2 seconds to open over the network - all users that need these already had 7-zip installed, because Windows' built-in ZIP support took even longer to open them).

    That's probably because Zip in its sheer genius puts the "master directory", i.e. list of contents, at the end of file. Only worse thing is .tar.<whatever> which doesn't have such a list anywhere, you just need to full-scan the file (uncompressing it to RAM on the way) to know contents.

    @ender said:
    As for 7-zip's GUI, from what I've read, it's more of an accident that it's present at all - the author doesn't care about it at all. I still find it much better than WinZIP, though my preferred GUI archiver is WinRAR (even though I only ever use it to decompress archives).

    Well, RAR is a closed format, and only official RAR/WinRAR can (sometimes) repair the archives made with the redundant recovery data option, as far as I know. Also, how good does 7zip handle multipart RAR archives? I don't think I ever tried...



  • @bannedfromcoding said:

    Also, how good does 7zip handle multipart RAR archives? I don't think I ever tried...

    It works fine in the present version.

    Also my machine does not have winrar installed, the horror!



  • @bannedfromcoding said:

    Well, RAR is a closed format, and only official RAR/WinRAR can (sometimes) repair the archives made with the redundant recovery data option, as far as I know.
    I really don't care, as my only use for it is to open archives directly from browser. When creating archives, I always use command-line tools, regardless of which format I'm using.@bannedfromcoding said:
    Also, how good does 7zip handle multipart RAR archives? I don't think I ever tried...
    No problems at all in my experience.



  • @bannedfromcoding said:

    That's probably because Zip in its sheer genius puts the "master directory", i.e. list of contents, at the end of file.
    To be fair, .zip was created in 1989 when it was fairly rare to deal with files larger than what would fit on a floppy.  It's one of those design decisions that sort of made sense at the time



  • I'm planning on subverting the system and installing zipeg on my work laptop.



  • For desktop use I right click and choose extract here. Or select some files and do right click compress. That's about all the UI I want for a compressing and decompressing archives.

    For other 99% I use archives, I use tar.bz2 or tar.gz. Because it's on servers. Servers that run linux. and I can't be arsed to install pkzip, or whatever the tool is called that does zip on linux.

    Also, gzip has the handy feature of being able to compress streams, so I can copy a file pipe it trough gzip and send it to another computer to be saved on disk. Which has proved handy once or twice when I had to backup a whole lot of data from a server that didn't have enough space left to make the backup locally first. 

     


Log in to reply