Re: [NTLK] Shout out to developers

From: Jon Glass <jonglass_at_usa.net>
Date: Sun Sep 03 2006 - 09:06:12 EDT

On 9/3/06, Ivan Kowalenko <ivan.kowalenko@gmail.com> wrote:

> False. I point you to VLC, MPlayer, MythTV, OpenOffice, SNES9x, MAME,
> and several others.

My take on Open Source.... for compatibility--Open Source, for
capability--commercial. Not that OSS isn't capable, but typically it
doesn't work quite so seamlessly or as well as commercial software.
For instance, I tried the Open Office variations of
PowerPoint--Impress, both in Neo Office and Open Office.Org on my Mac,
and neither could produce. In fact, my PP presentations I imported had
to be drastically chaged, as some simpler transitions and builds I had
done were not possible in Impress, or didn't work the same, plus fonts
didn't work correctly, so I had to change fonts as well, and then, in
presentation mode, neither had the ability on my aging G4 processor,
to create smooth presentations. Granted, as far as cross-platform, and
multiple formats, I think that OOo blows away Microsoft, and its
cross-language/multi-language abilities makes Office look positively
"American" (you have heard the joke about people who speak three
languages--tri-lingual, two languages--bi-lingual and one
language--American) but when it comes to user interface, all too
frequently, the commercial software beats OSS. And, I would add, the
exception merely proves the rule in this case, so pointing to
"projects" like Mplayer doesn't prove anything except that it is a
notable exception to the OSS "rule." ;-)

BTW, I read somewhere that one problem with OSS is the "project"
mentality rather than shipping "product". In some things, it isn't a
big deal, but in productivity software, it is. People want software
they can trust to be stable and do what they want it to do, and by
stable I don't mean just "not crashing" but not in a constant state of
flux, never finishing.

I guess I say all that to say that even if this is Open Source, it
needs to be viewed as product not project.

> > By way of example, the best possible outcome for the Mac side would
> > probably be an iSync plugin. This is, by definition, not cross-
> > platform. Similarly, on Windows, you would want Outlook
> > integration. Outlook doesn't exist on the Mac, and even if it did,
> > their plug-ins would not be interchangeable.
>
> This is the biggest problem. In this case, I'd say that there should
> be separate projects. Perhaps, say, a libnewt, which handles
> communication with the Newton. An iSync Plugin could be built around
> that, as well as a modern Newton syncing app for Windows, and even
> Linux. libnewt would have to be an OSS project, everything else
> doesn't really need to be.

I think that one example of how "cross platform" works is NCU itself.
It had the plug-in architecture with platform-specific plugins for the
appropriate apps.

However, I was of the impression (misimpression?) that in Windows, NCU
still works, and most likely will.... Is there any software still that
works with it? Granted, people probably want Outlook sync, but not
everybody uses Outlook. Is there any other built-in address book for
Windows that is system-wide like on OS X? I think this needs to be
taken into consideration for any cross-platformness.

> For an example of this kind of system, I point to libgaim, which
> handles multi-protocol IM agents, on top of which gAIM, XFire, and
> AdiumX have been built.

Yes, but IM software isn't an address book or calendar. It's gotta be
a bit simpler, especially since IMing is all about communication, and
address books/calednars aren't...

> > So, I'm not sure if I'm advocating picking one platform, or having
> > two separate projects, but beware the potential cross-platform
> > albatross...
>
> You're right, about cross-platform being a problem (the Matroska
> project is a good example), but if the system is broken up into
> several projects, it should be feasible. Just make one reliable
> communication library, document it well, and keep it Open Source.
> Then we might have some luck.

You know, isn't NewtSync open source already? Are its problems indemic
to the software itself, or to the plugins as they stand currently? If
it's the plugins, would it not be simpler to fix those than re-write
something new? Or is it the fact that you need a new comm package on
the Newton? I'm afraid that whatever gets written will need a new
Newton package or extension, regardless. Thus returning to the old
bootstrapping problem...

However, other thoughts I have. I was looking at Daniel Padilla's site
this morning, and noticed that he had been playing with XML and the
Newton. If the core package that was done (application or
what-have-you) digested everything through XML, could it not be
extensible to whatever the other side would be, whether iCal, Address
Book or Outlook or whatever?

I don't know, but it just seems to me that whatever the problem is,
it's much deeper than we are thinking, because great minds like
Daniel and Paul have nor Dan Rowley or NowhereMan, nor others have
figured it out. DCL is dead, Newtsync seems to go nowhere, what's
next?

I don't want to seem like a wet blanket. That's not my point or
purpose. We need clarity more than anything, and that's my goal.

-- 
 -Jon Glass
Krakow, Poland
<jonglass@usa.net>
There is no such thing as public opinion. There is only published
opinion.   --Winston Churchill
-- 
This is the NewtonTalk list - http://www.newtontalk.net/ for all inquiries
Official Newton FAQ: http://www.chuma.org/newton/faq/
WikiWikiNewt for all kinds of articles: http://tools.unna.org/wikiwikinewt/
Received on Sun Sep 3 09:06:17 2006

This archive was generated by hypermail 2.1.8 : Sun Sep 03 2006 - 18:30:00 EDT