| 1 |
On Thursday 26 May 2011 05:50:14 Walter Dnes wrote: |
| 2 |
> On Wed, May 25, 2011 at 07:13:41PM +0200, Volker Armin Hemmann wrote |
| 3 |
> |
| 4 |
> > On Wednesday 25 May 2011 08:46:48 Indi wrote: |
| 5 |
> > |
| 6 |
> > and have you ever heard of 'code reuse' or 'modularity'? |
| 7 |
> > |
| 8 |
> > It seems - no. |
| 9 |
> > |
| 10 |
> > Because KDE itself might be huge. But once loaded the apps are pretty |
| 11 |
> > small - because they reuse code. kmail does not have its own html |
| 12 |
> > engine. It does not matter where you type your text etc pp. |
| 13 |
> |
| 14 |
> Sorta like Internet Explorer in Windows. It "loads" a lot faster and |
| 15 |
> lighter than Firefox or Opera. That's because ie.exe is merely a "front |
| 16 |
> end" to a bunch of libraries that are loaded at boot time, which |
| 17 |
> contributes to the boot process taking do long. Starting ie.exe takes |
| 18 |
> hardly any time, because 90% of the app is already loaded. |
| 19 |
> |
| 20 |
> > Overall KDE uses LESS ram then most 'lightweight' solutions. Because |
| 21 |
> > xterm&abiword&some odd pager&thunderbird don't look so good anymore. |
| 22 |
> > |
| 23 |
> > This gem is a couple of years old, but still a worthy read: |
| 24 |
> > |
| 25 |
> > http://ktown.kde.org/~seli/memory/desktop_benchmark.html |
| 26 |
> > |
| 27 |
> > |
| 28 |
> > Read it. Seriously. |
| 29 |
> |
| 30 |
> I don't know how good "exmap" is, but my personal experience is quite |
| 31 |
> different. Between Fall 1999 and Summer 2007 I had a Dell Dimension |
| 32 |
> with a 450 mhz PIII and 128 megs of *SYSTEM RAM* (no not the video card). |
| 33 |
> It was actually quite usable to the very end, with Blackbox WM, and |
| 34 |
> running a few apps. Meanwhile, KDE (and GNOME for that matter) would |
| 35 |
> take forever to load and make the system crawl after that, even with 1 |
| 36 |
> or 2 apps loaded. |
| 37 |
|
| 38 |
I remember running Slackware on a Pentium 1 100MHz laptop with 128M RAM. The |
| 39 |
speed was of course glacial unless I was running only a console with no X. |
| 40 |
KDE would load and run, as long as I didn't push it too much. Fluxbox was |
| 41 |
more respectable. |
| 42 |
|
| 43 |
In contrast, MSWindows NT4 would load and run better as it was a more light- |
| 44 |
footed OS. MSWindows 3.1 was blisteringly fast and MSDOS, well ... |
| 45 |
|
| 46 |
However, life moves on and with the cost of hardware coming down software has |
| 47 |
moved towards larger, all bells and whistles, DEs. The change in design |
| 48 |
philosophy from KDE3 to KDE4 made things worse for those of us who do not want |
| 49 |
everything and the kitchen sink thrown in, but still want to use some KDE |
| 50 |
apps. |
| 51 |
|
| 52 |
Thankfully, the move to the KDE meta ebuilds has provided some compensation |
| 53 |
against a full blown monolithic KDE. |
| 54 |
|
| 55 |
Personally, I'm grateful that Linux devs continue to develop exceptional |
| 56 |
software and so I don't have to use MSWindows. On the other hand I have |
| 57 |
always preferred more lightweight WMs to the full enchilada of KDE and Gnome |
| 58 |
and wish that KDE devs retained the KDE3 design philosophy, or afforded us a |
| 59 |
light(er) option. |
| 60 |
|
| 61 |
PS. I'm not sure that Linus is using Gnome. I recall him bitching that the |
| 62 |
Gnome design approach (which unfortunately KDE imitated) was not the right |
| 63 |
direction to evolve linux in. |
| 64 |
-- |
| 65 |
Regards, |
| 66 |
Mick |
Archives