lilypond-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: battle-plan for 2.5 development


From: Johannes Schindelin
Subject: Re: battle-plan for 2.5 development
Date: Wed, 17 Nov 2004 13:26:24 +0100 (CET)

Hi,

On Wed, 17 Nov 2004, Ralph Little wrote:

> Certainly if, as Erik says it could be boxed up for one-step
> installation so that the Cygwin layer were packaged up with it, novice
> users would find it more acceptable.

There is a way to package cygwin on a CD and preselecting packages by
modifying setup.ini. However, this is huge. Fits on a CD, though...

> I did try to install Cygwin once in the past just to see what it was
> like and it was clunky to say the least. Ordinary users just don't want
> that hassle. A lot of issues raised on the users groups seem to be
> related to Cygwin and getting it up and running.

Agree.

> Seeing Lilypond in the shops in a box would be so cool and would be nice
> to see some of the revenue returned to assist our friends Han-Wen, Jan
> et al.

Agree.

A few remarks regarding Windoze:

- if the TeX backend goes away, then it gets a lot easier: the whole TeX
system is huge. And maybe at one stage we could even revert to MinGW,
which is much lighter on resources. However, AFAIK MinGW lacks some
important functions like fork().

- when the TeX backend goes away, it gets harder: I think that the best
Lilypond editor on Windoze to date is jedit (shameless plug (TM)). For now
it depends on point-and-click! And it would be difficult at best to
integrate GNOME support into jedit...

- if you want to go for the Windoze market, you absolutely have to make
the program fully clickable (again, I would tend to just use lily4jedit,
which has a flurry of new features lately). Because jedit is written in
java, this is relatively easy (you could even write a note editor in it),
and it also runs on other platforms without changes!

- the most important feature for professional Windoze users, the killer
feature so to say, would be a competent payed-for support. This is what
makes the difference between a good open source program and a successful
good open source program.

Ciao,
Dscho





reply via email to

[Prev in Thread] Current Thread [Next in Thread]