RE: A killer app?

[ian . mondragon]

> oh?  well then what's the point of a the "pure GNUstep environment" that
> people are constantly pushing as needed?  isn't the point having a
> 'superior' API to utilize it as much as possible?  isn't that exactly what
> GNOME has done with GTK?  KDE with Qt?  to some degree, windowmaker apps
> with WINGs?
> 
> i didn't think that this ideal GNUstep environment only consisted of
> major apps such as word processors, web browsers, email clients,
> spreadsheets, and drawing programs.  i think i could easily turn the
> tables and argue that even MORE time has been invested in creating "newer,
> better" <insert app here> with other toolkits that went nowhere (in the
> grand scheme of things)...and here we are in a position to start from
> scratch, and utilize all of everyone else's mistakes & glory moments.
> look at a fraction of the choices: lyx, abiword, kword, etc?  netscape,
> mozilla, galeon, konqueror, etc?  gnumeric, kspread, abacus, xess, siag,
> oleotk, lemur, abacus, etc?  elm, pine, mutt, kmail, balsa, aileron,
> tkrat, exmh, postilion, tkmail, arrow, mahogany,  etc?  
> 
> <soapbox>
> i think that even more than the apps themselves, we need to have
> _organization_ and a very well charted plan of action, so that we don't
> fall into the yet-another-___ category.  we need to form some sort of
> group structure, which takes into account the realistic time that
> developers will be able to devote to a project(s), and divies out
> responsibilities according to that and, of course, skill level.  yet
> another old concept that's perfectly applicable to our situation.  i've
> mentioned it 2 or 3 times already here in the discuss-gnustep group, but
> nobody's said anything, so i'm assuming the GNUstep world is even smaller
> than i originally thought.
> </soapbox>
> 
> i completely agree that we need these things more than any other little
> app...well, aside from project & interface builders...but the whole
> "environment" thing seems to be a predominant theme in the GNUstep
> underground, and i don't personally see it as such a moot point.  i
> initially started creating the things i mentioned above NOT because i
> brainstormed to think of something usefull for the world, but because i
> wanted a GNUstep version of something that i *personally* thought would be
> usefull.  i could have done these things in python/tk and finished in 1/10
> of the time, but i didn't.  the terminal preferences?  i want a gui
> configuration & central repository for my various terminals, as opposed to
> saving the long, ungainly commands in my wm menu or a small script for
> every single one.  the user manager?  freebsd's adduser script is in perl,
> and, while i don't have a gargantuan network at home with hundreds of user
> accounts to maintain, i would like to see an interface to this aspect of
> the system that mirrored NeXT's usermgr.app...because i, myself, like it.
> the shell thing?  i'm sick of the .rc and .conf files that constantly grow
> in numbers, and i want to see if i can actually do something about it,
> which i don't see as a terrible thing, and it's also proving to be one
> hell of a learning experience about ALOT of aspects of the system...it
> might even prove usefull someday...who knows?
> 
> @end
> 
> - ian mondragon
> 
> -----Original Message-----
> From: Helge Hess [SMTP:helge.hess@skyrix.com]
> Sent: Tuesday, January 23, 2001 3:53 PM
> To:   discuss-gnustep@gnu.org
> Subject:      Re: A killer app?
> 
> Hi,
> 
> I mostly disagree. I don't need YASW (yet-another-shell-wrapper), I'm
> quite happy with bash, yast and/or available LDAP tools.
> Having a UserManager doesn't give me any additional value (indeed the
> additional value of the NeXTstep UserManager is not the UI, but the
> netinfo backend, which I already have with PAM and LDAP).
> 
> You write that "we just need to do the same" like KDE and GNOME. This is
> wrong, since those are applications where GNUstep UI can play out little
> advantage - and, more important: they are already done in KDE and GNOME.
> You will never get "$15 Million" if you try to do sth which is already
> done twice (actually 4 to 10 times).
> 
> What I miss on Linux are productivity applications done in a consistent,
> interoperable manner. Diagram, OpenWrite, Improv, Mail, OmniWeb,
> WetPaint and for development, IB, PB.
> These are not new concepts, but stuff which is really missing on Linux.
> OpenStep in theory provides some advantages for implementing the above.
> 
> Helge
> 
> ian.mondragon@bankofamerica.com wrote:
> > 
> > >Let's look at the Unix tools and invent graphical equivalents ( <- is
> > >that english?)
> > >Build a graphical awk, groff, less, split, join and more
> > >This would mean no single killer app, but a killer environment. There
> > >are ofcourse some things that need to be solved like:
> > >How to pipe stuff from one to another command ?
> > >How to select all options and how to interface things right ?
> > 
> > i agree for the most part.  while i feel that a "killer app" would be a
> > great plug to get everyone to install GNUstep on thier systems, these
> things
> > tend to take a horrid amount of time, effort and coordination.  and look
> at
> > what that's done to mozilla <grin>.  i've been doing pretty much just
> this -
> > slowly piecing together smaller, more usefull tools that just about
> > everybody could/would want to use.  and they're not exactly new concepts
> by
> > any stretch of the imagination...a user manager app, a terminal
> preferences
> > pseudo-app (*NOT* a new terminal), a host manager app, an app that
> utilizes
> > the GNUstepDefaults in an attempt at replacing the overwhelming amount
> of
> > .rc and .conf files floating aroung the *nix variants...
> > 
> > nothing exactly mind boggling.  what we need is utility.  one of the
> reasons
> > that the GNOME and KDE *environments* are enjoying such success is that
> they
> > group together the usefull things that users need in a common place like
> > this.  we just need to do the same.  then maybe we can all come together
> and
> > work on something that would deliver a GNUstep environment (in it's
> truest
> > sense) to the end-user...and possibly get, say, $15 MILLION investments
> > every once in a while *cough*ximian*cough*.
> > 
> > >That means not creating the next NeXT or MacOS but to create the
> > >environment that is new.
> > 
> > i don't think anybody *really* wants clone NeXT in an absolute sense.
> the
> > thing is that NeXT did a whole lot of things right...and that's what we
> want
> > to do too (i'm pretty sure i can speak for most of us on that one).
> > 
> > @end
> > 
> > - ian mondragon
> > 
> > _______________________________________________
> > Discuss-gnustep mailing list
> > Discuss-gnustep@gnu.org
> > http://mail.gnu.org/mailman/listinfo/discuss-gnustep
> 
> -- 
> SKYRIX Software AG  http://www.skyrix.com/
> Join the team:      http://www.skyrix.com/de/jobs/index.html
> 
> _______________________________________________
> Discuss-gnustep mailing list
> Discuss-gnustep@gnu.org
> http://mail.gnu.org/mailman/listinfo/discuss-gnustep