Jump to content

zero0w

Members
  • Posts

    864
  • Joined

  • Last visited

Posts posted by zero0w

  1. I think it was rushed. The 9.1 installer can hang at certain locations, and then yesterday I found that inputting in Chinese does not work right. Well, I'll see if I can fix it. But Mandrake 9.1 definitely is different from 9.0 in many regards. If Mandrake can make it thru another release, I hope they can concentratre on bug fixing rather than adding new features next time.

  2. Interesting. Actually Ardour is among the most promising multi-track audio recording and editing applications. However, it's still not finished and very beta in quality. Some other apps to keep a look out over there: I think a new version of Rosegarden is coming out soon. Thac's rpm page Sound section provides most of what you will need.

  3. aRTee: Well at least on the development side of it Linux will be heading to over 50% soon. :D

     

    http://www.linuxjournal.com/article.php?sid=6755

     

    However, I still maintain developers need a good cross-platform strategy in the beginning. wxWindows is a good start.

     

    Check out these Open Office Impress presentations of wxWindows and see how good it can be used to develop for cross-platform apps:

     

    http://www.wxwindows.org/docs.htm#presentations

     

    Of course developers have reservation of the on-going maintainence of an open source GUI widget without a company support. They may prefer Qt. Either way wxWindows and Qt (also GTK) will provide common GUI frameworks for cross-platform apps development.

  4. << Highend 3d package learning edition on Linux >>

     

    1. Softimage EXP for Linux Released! (Registration required)

     

    http://www.linuxartist.org/article.php?sid=197

     

    Prepare to have broadband access. It's a 170+ MB download. :P

     

    < Support forum for Softimage EXP >

    http://forum.softimage.com/forum.asp?FORUM_ID=5&MID=1

     

    2. On a side note, you can also download Houdini Apprentice (learning edition) here:

    (New version 6.x is out!)

     

    http://www.vislab.usyd.edu.au/sidefx/

     

    Enjoy!

    Now we'll see if Aliaswavefront will make Maya PLE available on Linux...

  5. Upgrade of all packages and proper config changes should take care of most problems. But thats very tricky. About, KDE-3.1 on LM9.0, texstar compiled them on LM9.0 so they work ok on LM9.0. If you try to install LM9.1 KDE rpms on LM9.0 (or vice-versa), it won't work.

     

    Of course not, LM9.1 has its own KDE-3.1 compiled against glibc-2.3.1. Probably the supermount in the kernel-2.4.21 has some tweaks to work with KDE-3.1 in 9.1.

     

    As for solving dependency hell, well it will probably take more time than a fresh-install so I will not go into that. Upgrading all package * should * work but... it might involve some quirks somewhere. It seems some package such as Broadcast 2000 is gone in Mdk 9.1. So the dependency resolution might not all solved with missing packages.

     

    In final analysis, fresh install is still recommended. But /home can be kept I think.

  6. 1) Anyway UPGRADE seems to be a bad option for 9.1. The leaps from KDE 3 to 3.1, glibc 2.25 to glibc 2.3.1 both introducing some big changes. But again I think it's perhaps Mandrake's way of managing things. Other users have been reported to be successful upgrading to KDE 3.1 from Mdk 9.0 using Texstar rpm and such. So maybe the UPGRADE path is not thoroughly tested from the Mandrake team?

     

    2) The installer in 9.1 final has problems, if you tweak too much away from the default option you will have funny things come up. I did have repeated freezes at the same location after finising change of selection of system services. So, basically something's wrong with the installer. I did not however, try a text-mode install to verify if it happens there too but why or how would I know to do that in advance?

     

    3) Screen font Anti-aliasing slows performance and may not work with some older display card (that's what I heard). So that might explain something. Of course, to verify this, you may want to try Red Hat 9 and see if the problem is due to anti-aliasing but not due to vendor's packaging flaws.

     

    4) Fresh re-install is recommended anyway. I have mysterious (eg. seg-fault) problem in MDK 8.2 which was solved in 9.0. Also, you may want to try using XFS instead of ext3 as filesystem. My harddrive seemed to have data corruption with ext3 in my first installation of Mdk 9.1, not sure it's a general phenomenon or it's simply my system config's 'uniqueness'.

     

    5) btw, it's usual to install more than once to get a new distro release to work the way as you like. You may want to give it more trials. Also others reported problems with ALSA sound driver not installed properly - that's true: Mdk 9.1 will not install ALSA as default, you need to change it manually.

  7. Read the article about debunking the myth; not bad, but it fails to notice that there is really a very small market, and the money is still on the MSwin side...

     

    Well, guess it is "install base" rather than "market share" then. Since Linux is basically free, the money will mostly go to support service and contract development projects rather than the OS product itself. It probably will never reach that level of money MS is able to charge their users. I do not have high hope it will unseat M$ in the short term (1-3 years). If the install base can reach 20% by then I think it is already achieving something. In the long term, well things can change anyway.

     

    What I do know, is that linux is taking over stuff, both from windows and unix, in companies worldwide.  

    I am an electronics designer, and I know that all design software from Cadence for instance is being ported as we speak, and should be available (all their software) on linux before this summer....!

     

    Great, thac's mdk rpm page hosted some electronic software package, haven't looked into that in detail myself :o .

     

    Basically, developing for MS is supporting their monopoly. Using MS software is supporting their monopoly (because you count in the statistics).

     

    I would not go that far. But MS incorporating DMCA in Palladium architecture is taking things too far. And Volume Licensing 6 in the middle of economic downturn wouldn't get any cheers but the recognition to the reality of lock-in to pay more. MS Windows 2000 is okay but other OS product lines each have their flaws and problems in terms of instability or intrusive features (XP spying users data).

     

    Supporting the MS Monopoly is shooting yourself in the foot in the long term. As a software vendor, you cannot compete with MS, they will extinguish you. As a user, they will milk you.

     

    I think MS monopolistic strategy is bad to users, and at sometimes bullying OEM is unfair to competitors. But developing on MS platform, for now, is a better way to have people purchasing your products because people CAN use it. Once again, the only realistic strategy is multi-platform if the developers want to have more revenue from the widest user base possible.

     

    Personally, 2002 was the linux on the desktop year, anyone putting it any later than that is just behind.... ;)

     

    I would say kernel 2.6 will fill in the puzzle of multimedia capabilities in terms of pre-emptive kernel patch and a much more efficient scheduler. It reduces latency in video/audio editing and network audio stream. To multimedia authoring kernel 2.4 is still a bit lack in that prospect. Well if you're a Linux kernel hacker guru and patch the kernel yourself you certainly can claim 2002 has all the source you have. But a lot more user space apps is coming in 2003 and later. And I believe mplayer 0.9 final is very important in viewing QuickTime video, which is to be released soon.

     

    Next milestone is the Windows-off-the-desktop-year. 2007 anyone? ;)

    (I know, won't happen and as far as I'm concerned, doesn't need to, as long as they have less than 50% marketshare by then..)

     

    Hmm... I can't see that far, but maybe we will have kernel 3.x by then :) .

     

    BTW can't get to the site in your sig, what's up with wthat?

     

    RatedPC.com may be down, try again later. :)

  8. zero0w, check googles zeitgeist: http://www.google.com/press/zeitgeist.html, and you will see that linux marketshare really is way behind. Note that google doesn't just check for the browser id but for the OS id which is generally not faked (but surely can be), so much more reliable.

     

    Ok, I guess google speaks for most people accessing the internet.

    Perhaps you are right, Linux desktop still has some time to catch up. I just read somewhere it will surpass Mac soon but your google statistics show it may be a bit too optimistic. Perhaps Mac OS X is really a good product...

     

    I noticed that statistics is for Feb 2003, I am curious if it will stay the same by the end of 2003 though, kernel 2.6, mozilla 1.5+, Open Office 1.1 and possible improvement in other multimedia capability such as Xine 1.0 and Mplayer 0.9 final will push Linux desktop even more to the front stage, I believe (btw I am working on an article of these interesting developments).

     

    There are two other posibilities those statistics don't show though:

    (1) people who don't use google - eg. a lot of people in China probably will use something other than that for their relatively localized web-surfing in only Chinese sites. They may have other useful portals for searching Chinese materials. (2) people who don't have internet access - well that's probably very few nowadays.

     

    However Chinese piracy rate is very high so I don't think the missing representation in non-google-access would increase the market share of desktop Linux much.

     

    However, it should be clear that the market share based on sale is not appropriate to Linux. Since you can install as many copy to the computers as you want to. Here's another article I read at LinuxWorld which help debunking some market share myth on this issue:

     

    Debunking the Linux-Windows market-share myth

    http://www.linuxworld.com/2003/0314.petreley.html

     

    Anyway, I hope 2003 will be the year where we will see a turnaround of things on desktop Linux.

     

    BTW I like the GIMP, only used PS very few times, and I'm nowhere near a professional, but the point is, people who want to do photo editing on linux will use the GIMP, and not consider buying PS anymore once it's out. Exceptions will prove this... ;)

     

    Color accuracy is extremely important in printing and publishing. If your printout does not match what you've seen on the monitor, it will be big trouble to print publishers. So far GIMP is not designed to have high color precision or calibration accuracy. GIMP-Print Project might change that, but I am not too familiar with it. Besides, color adjustment and calibration is simply plain hard work between the printer manufacturers and the software (color library, in particular) developers. It is test and test and until the color on monitor matches the color values stored in the printing hardware, driver, and the graphic software. It's extremely boring work I am afraid. It's probably not fit for open source developers where they are developing for challenge and interest. But perhaps I'll check out the GIMP Print project to find out more. Either way, printer driver support will need to be improved if Linux is to be considered important in print industry.

  9. About the tiny marketshare: well, apple has a tiny marketshare, compared to MS, but Adobe still ports, because Mac seems to be the platform of choice (or used to be). Linux does not have a small marketshare compared to apple, although still smaller.

    And by not porting, they are just keeping the marketshare down, until there is an OSS alternative, and then who are they going to sell to??

     

    GIMP can work great at web graphic, but for truly desktop publishing where color accuracy matters, Photoshop still rules. I don't know if GIMP-Print will have anything to do to change that yet. Besides, the current user-base will still prefer Photoshop as more powerful and productive than GIMP, with the large developer base of plugins and filters (well I guess that's not entirely true, but again GIMP does not have as much publicity anyway).

     

    As for Linux marketshare, it's a myth. Maybe Linux market share survey should have counted the number of online downloads and P2P transfer head count of it, in addition to retail sales. I think by 2004 Linux should have surpassed Mac in user market share, when many user space desktop apps are becoming more and more mature and the much needed publicity being more openly addressed.

     

    So bad management on one hand, low risk taking on the other.

    Not sure what I would do as a company...

     

    Planning a strategy for multi-platform.

     

    The truth is, even Microsoft is surprised at the number of companies which are still using NT4 and Win98. Many of them simply won't do anything to upgrade their IT infrastructure when they works good enough. So it makes no sense to leave MS market altogether.

     

    I expect in the near future we will see more articles and books on developing cross-platform apps, as there is really a demand of it. A good cross-platform GUI widget is instrumental in this regard.

  10. Wow... this was a good game!

     

    Back in 1996 where I only have a Pentium 100 it was a very enjoyable game. Wish it can be ported to Linux soon and bring up some nice - ... oops -... violent shooting... memory :P .

  11. I see. Audacity is still not as sophisticated as the big guys nor as Ardour.

     

    But the question remains, they need to choose a GUI widget and support this port in the future. Um, right now it seems Qt has a lot of momentum. Perhaps Qt and wxWindows will both contribute a lot to Linux port of software in the future. Being supported by a real company, using Qt is safer in the eyes and mindset of business world. Although wxWindows proved its worthiness after 10 years of continuous development.

  12. Hey SoulSe, you have any experience of Final Scratch?

    I heard about it for sometime and didn't manage to play with it. How it works on Linux? Any online radio-station using that?

     

    I completely agree on getting a start, but open source world is so segmented that developers don't know which GUI to support. As I have said before, a common wrapper platform such as wxWindows (http://www.wxwindows.org) will save them more trouble in the long run. wxWindows can wrap against Win32, GTK , X11 Motif, and Cocoa on Mac OS X. In case GTK developers lose interest, wxWindows can wrap itself against another (new) GUI widget on Linux later. The shift of risk to a middleware GUI wrapper will be a benefit to many commercial developers out there. The downside I've seen so far is the small developer base of wxWindows, and also the lack of publicity. As you may know Audacity is based on wxWindows and has ports on both Windows and Linux.

     

    Maybe you can recommend the CoolEdit developers to have a look at wxWindows too. :)

     

    Also check out the Open Office Impress presentation of wxWindows at here:

    http://www.wxwindows.org/docs.htm#presentations

  13. I hope they will though. But I don't think it's as easy as that.

     

    I hope companies realize they should start new project on a cross-platform target from now on, maybe using wxWindows as the GUI wrapper against each platform.

     

    In fact, Adobe and now Macromedia both have cross-platform ports of their software (eg. Photoshop and Dreamweaver both on Windows and Mac OS X, the latter is coming to Mac soon). That being said, porting to Linux should not be too difficult to them. The problem is support, because they don't have control on what GUI/DE users have installed on their systems, or not. And perhaps also the pressure from Microsoft. In fact I will say three 3rd-party software vendors (and their products) will decide the acceptance and possibly the fate of (massive) Linux desktop, they are Adobe, Macromedia, and Intuit (Quicken and TurboTax). Too many people are using software products of their offerings. If they start to port to Linux, I am sure the rest of the industry will follow. Still cross-platform using wxWindows will be a more realistic goal, and co-existence has its merits.

×
×
  • Create New...