Message-ID: <20010608131855.10142.qmail@softhome.net>
References: <3B1F8DC8.22195.1FA053B3@localhost>
In-Reply-To: <3B1F8DC8.22195.1FA053B3@localhost>
From: StefanRieken@softhome.net
To: marco@marco94.f2s.com, Bob@softhome.net, Wilson@softhome.net,
	"Jr." <Postmaster@Inmail.com>
Cc: gopher@complete.org
Subject: [gopher] Re: Status Gnopher
Date: Fri, 08 Jun 2001 13:18:55 GMT
Mime-Version: 1.0
Content-type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 8bit
Sender: gopher-bounce@complete.org
Errors-to: gopher-bounce@complete.org
Precedence: bulk
Reply-to: gopher@complete.org


Yay, fanmail! ;-)

This week I got two emails, one from Marco Rooney and one from Bob Wilson,
Jr. asking me basically how things go with Gnopher. (Thanks Marco and Bob!
Your authentic pluche stuffed Gnopher mascottes will be sent to your home
address ASAP ;-)

It seemed pratical to me to answer both these mails at once. On top of
that, I decided to CC: gnopher@complete.org because this email was going to
be a LISR (Long Informative Stefan Rant) anyway. (So if you're planning to
make a reply, be careful to select the adresses you want to reply to...)

First off, Marco Rooney writes:

> Hi Stefan,
> 
> I happened to stumble across your Gnopher website. However, the 
> screenshots are not available. Also I would like to know if you are 
> still developing Gnopher?

The screenshot is put back in its place again. I got no idea how it
disappeared in the first place... weird.

The second question is one which I'll try to explain a little bit further
here.

The initial planning was as follows:

0.1   "Works For Me"                    initial release; works for me :-)
0.x   "Works For X"                     beautification, completion and
                                        portability releases
1.0   "Works For Gopher+"               considered complete, Gopher+
support
1.0.x "Works Even Better For Foo"       updates


1.1   "Works For My Nautilus"           initial Nautilus integration
1.1.x "Works For X's Nautilus"          beautification, completion and
                                        portability releases
2.0   "Works For Nautilus"              considered complete (with Gopher+?)

After 0.1, I got a few emails from people who said "Works For Me". That
meant that I didn't have to fix any stuff, yay! So my only reason for
making more 0.x releases is beautification. However, I am still stuck with
the same old problems that I was stuck with before. That is: I *do* know
how to set a default size to a GnomeMDI window, and I *do* know how to
complete the menu stuff, but the ways to do it are either hacks, or plain
ugly, and I don't like the idea of uglifying my code too much just because
it's written in GNOME ;-)

Let's face it, there are two major problems with the current GNOME API: the
menu subsystem and the GnomeMDI subsystem. The menu system, well, there are
actually two of them: plain GTK+ (which isn't really complex but very
unelegant to use because you have to create and modify each menu item
seperately) and the GnomeUIInfo stuff (which isn't that bad). The problem
is that it's not easy to convert GnomeUIInfo structs to GtkMenu objects and
vice versa. This might not be such a problem if it weren't so that the
GnomeMDI subsystem mixes both of these subsystems. Grr. This, together with
a few other things (little inconsistencies, the need to "hack" to get some
things done) make me thing that also the GnomeMDI system isn't really
practical. (Indeed a lot of apps simply avoid using it.)

Beside the GUI problems, I have time problems. I have a tendency to tire
myself with more projects than I can actually manage. I have come to
realise this, and now I am trying to learn myself to focus on just a few
things (being school and just one project) at the time, and to enjoy the
spare time for myself. That should take off most of the stress and
pressure.

Another factor is that I got a little bit disappointed by the abilities of
Gopher+ when it comes to CGI-like things. Gopher+, with all its metadata
stuff, is actually a kind of powerful protocol, but CGI-like behavior is
still its blind spot. We've been trying to find a solution on the mailing
list a while ago, but it wasn't really successful. I began wondering what
the use of the Gopher protocol was in a modern working environment.

And well, a last thing is that I didn't think it would be useful to
implement Gopher+ if Gnopher still only handles one file type anyway
(text). Yet the best way to implement more file type recognition stuff is
making Gnopher a Nautilus component -- this way Nautilus should be able to
handle all the file types for you.

So what are my plans for Gnopher?

Well, I was kind of planning to drop my initial release plan and to start
making Gnopher a Nautilus component ASAP, that is, as soon as the GnomeVFS,
etc. API docs started to appear. The stupid thing is that there hasn't been
any sign of these API docs ...UNTIL TODAY. I just checked out
developer.gnome.org to have a few facts confirmed for this email, and
there, coming out of nowhere, they all were! The fresh API docs!

So my new plan is like this:

Time: this afternoon
Project: play around with the new GNOME API docs

Time: in my good time :-)
Project: Nautilusify Gnopher

As this is a cool thing to do (Gnopher is well-designed on the inside so it
shouldn't cause much problems), it will probably be done very soon (to
avoid stress, the "fun factor" is my new priority system for my hobby
projects). But I really can't make promises.

(And concerning the lack of good CGI-like stuff in Gnopher+: once I get my
hands on a lot of free time (that'll be around 2078) and Free Gopher+
server code, I might hack in a good solution afer all ;-)

OK, I think that covers most of what I wanted to say, although maybe not
necessarily in a clear order ;-)

Greets,

Stefan

P.S.: In November I will have an internship period of half a year, which I
would like to spend abroad. If anyone of you would know a placement for me,
I would be very happy to hear! I study Information Technology, and would of
course be happy to provide further details when necessary. Thanks in
advance.