Difference between revisions of "Cross Compiling FileZilla 3 for Windows under Ubuntu or Debian GNU/Linux"

From FileZilla Wiki
Jump to navigationJump to search
(exyBzxVqlyMaaM)
(awDwDhjVPDc)
Line 1: Line 1:
−
Back in 2001-2005 when I used Libranet (or even from 1999-2005) when Libranet was commercially avlaaible (and also freely avlaaible for testing), it was one of the finest distributions avlaaible.  It had the good, solid Debian software behind it, but it had some excellent administration tools that made it really easy to manage.  For a while, when it faded, yes, there were easy distributions, but not too many with good administration tools in the Debian space.  Kanotix had a few, but not as good as Libranet.  Sidux was the first distribution to bring back some good tools, but fighting over project goals and sources of income and support tore the group and team members apart multiple times.  A distribution closer to the original sidux, with some similarites to the original Libranet called siduction is one of the finest distributions based on Debian Sid.  Libranet, however, was generally built upon Debian Testing. I personally think that a distribution that could easily allow you to build your system from Stable, Testing, or Sid would be great; I know of one called antiX core, where you can grow your system from scratch and select your Debian repo as you go; it's a great idea, but perhaps a bit raw for some tastes. The antiX base aims just a bit higher and is simpler. Grabbing good ideas from Libranet, siduction, and antiX, the best of each, and eventually build a couple of base platforms, perhaps one that people could just install and use (a modest, but full featured system), then a more minimal, but complete system, however one that has few applications, instead providing a simple basis for building your own, makes a second good choice, then, like antiX core, a system containing the core system essentials and a set of easy tools to build your own customized system would be ideal.The antiX systems target older systems.  I think a great model for LibraNext would be to initially target newer systems.  If it grows, having one branch for newer systems and another for older systems would be nice.Whether one or two of these ideas could actually be accomplished would greatly depend on who is avlaaible to develop, test, promote, and provide ideas.  One suggestion would be NOT to re-invent all of this, but grab the best ideas from the best distros; for me and my personal preferences, I'd lift the best from the likes of antiX, siduction, and MEPIS with proper permission.  I'm sure other distros have great pieces; integrating them together in a light, cohesive set would be great.  Perhaps even stuff from the original Libranet could still be useful if they are freed up and made avlaaible.
+
I dunno.  The leap from 16-bit to 32-bit was Something That Needed To Be Done, as there were too many restrictions in a 16 or 20 bit adrdses space.  Plus Microsoft used Win95 as the springboard into 32 bit, so that really smoothed things out I think  it gave the hardware vendors a bit of incentive to put out 32-bit drivers.  AND I think MS did a better job of making the backwards compatibility work (although it wasn't until Win98SE that it was worth bragging about) than they've done with 64-bit. Although some of that fault may be Intel's (I'm not real clear on the hardware restrictions for 64 bit) or maybe not.I don't see (yet) the same driving force behind the upgrade to 64 bit. It still seems a bit of a luxury, IMHO, and not giving the same magnitude of benefit as the move to 32 bit did.  Thing is, 16 bit ran out of steam, but 32 bit hasn't yet.  Sure there are some edge cases, and the ever present server stuff, where 64 bit is a big bonus, but for general office use, I'm still not convinced it really offers that much MORE than can be had with 32-bit.  At least not until MS pushes the Windows 9 minimum RAM requirements to over 4 gigs

Revision as of 14:49, 5 March 2013

I dunno. The leap from 16-bit to 32-bit was Something That Needed To Be Done, as there were too many restrictions in a 16 or 20 bit adrdses space. Plus Microsoft used Win95 as the springboard into 32 bit, so that really smoothed things out I think it gave the hardware vendors a bit of incentive to put out 32-bit drivers. AND I think MS did a better job of making the backwards compatibility work (although it wasn't until Win98SE that it was worth bragging about) than they've done with 64-bit. Although some of that fault may be Intel's (I'm not real clear on the hardware restrictions for 64 bit) or maybe not.I don't see (yet) the same driving force behind the upgrade to 64 bit. It still seems a bit of a luxury, IMHO, and not giving the same magnitude of benefit as the move to 32 bit did. Thing is, 16 bit ran out of steam, but 32 bit hasn't yet. Sure there are some edge cases, and the ever present server stuff, where 64 bit is a big bonus, but for general office use, I'm still not convinced it really offers that much MORE than can be had with 32-bit. At least not until MS pushes the Windows 9 minimum RAM requirements to over 4 gigs

Retrieved from "https://wiki.filezilla-project.org/wiki/index.php?title=Cross_Compiling_FileZilla_3_for_Windows_under_Ubuntu_or_Debian_GNU/Linux&oldid=23868"