Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 22 Jan 2000 14:55:38 +0800
From:      Greg Lehey <grog@lemis.com>
To:        Albert Yang <albert@achtung.com>
Cc:        small@FreeBSD.ORG
Subject:   Re: New approach to picobsd
Message-ID:  <20000122145538.A390@mojave.worldwide.lemis.com>
In-Reply-To: <3888D5CF.329989@achtung.com>; from albert@achtung.com on Fri, Jan 21, 2000 at 01:55:27PM -0800
References:  <3888D5CF.329989@achtung.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Friday, 21 January 2000 at 13:55:27 -0800, Albert Yang wrote:
> Hey there fellow Picobsd'ers!
>
> I like the fact that I'm seeing development on picobsd again.  I know,
> we all have that things called "a job" that takes up most of our time,
> but I think we can keep picobsd alive.
>
> My suggestion though was that there is currently no clear outline
> (as far as I can tell) of the structure and development of picobsd.
> What I mean is, actually making this into a well defined project
> instead of a loosely held patch kit.

There's a good reason for this.  I'll discuss further down.

> The thing I was thinking about most, was defining what "flavors" of
> picobsd should have what functionality.  Documentation is sparce,
> and I would like that to change as well.

I put in a man page recently; at the moment it's only in -CURRENT.

> You people are obviously more technical than I am, and as much as I
> would love to help on the technical side, I'm not that great of a
> programmer, I do webpages and graphics design, but I want to see
> this thing work.

You can certainly help if you can give us good directions.  My last
update to PicoBSD based on a particular need from a customer, but at
the same time I tried to make it as general as possible.  In
particular, I put all the functionality of the fixit disk in the
configuration, as well as some things which didn't fit on fixit.

The real issue is: do we want a one-disk PicoBSD or a two-disk PicoBSD
(in fact, it's one-disk or multi-disk)?  One disk is becoming
increasingly difficult to manage; as Andrzej pointed out, he has not
been able to get things to work on -CURRENT.  I think we could create
a *really* minimal -CURRENT version if we omitted all hard disk
support, but the direction is clear: the kernel is getting bigger, and
the feasibility of getting both a kernel and a file system on one
floppy is diminishing.

I think the answer here is: we want both, which is what we currently
have.  If we have only one disk, space is *very* tight, and it's
likely that everybody will need his own configuration.

If we have two disks, things look a whole lot better.  The version I
have still has space.  Jordan Hubbard has asked for nethack, but
possibly there would still be space after that for things that might
interest you; but you need to define them.

The way the current 2 disk configuration works, you can have as many
floppies as you like.  The first one boots and asks you to insert
additional floppies.  You can stop whenever you want, but obviously
the more you read in, the more functionality you have.

> I'm more business oriented and so I would not mind at all keeping a
> list of what feature requests are suggested, what feature designs
> are suggested, improvements, patches etc... I have my own webpage if
> that is what it takes to keep this thing cranking.

Andrzej has a web page on www.FreeBSD.org.  It's not well linked; we
need to attend to that (Wolfram, are you listening?).  Thanks for the
offer, but we've found in the past that URLs starting with
www.FreeBSD.org are easier for most people to find.

> For example, we say that the networking version needs natd, ssh, etc..
> and we list out what consitutes a network version, and the same for the
> dialup etc..

I can't even recall how much of this is in the current /custom
subdirectory.  Certainly I agree that ssh should be in there; natd is
more the sort of thing you would want to run on a server, which might
then be allowed to have its own disk.  On the other hand, I frequently
travel (I'm currently in the air between India and Singapore) and I
can recognize the advantage of being able to hijack some machine
somewhere and use it to connect to the net without overwriting its
disk.

> Trinux has taken a modular approach to all this.  I don't think that
> is neccessary, but at the same time, a more well grouped development
> effort might make it better with quicker turnaround time.

We have a modular approach, too.  It's so modular that it looks
completely unstructured :-)  You go into the crunch and crunch2
directories and choose your programs.

> It's just a thought, and again, if you guys need some pretty looking
> website, let me know, I can cough that up.
>
> BSD currently doesn't recognize my modem, and so if I wanted to use the
> ppp picobsd, then I have to use one of my old 28.8 ones.  Ouch.  Does
> the network version support DSL?

I think that depends on which version you use.  My understanding is
that DSL is an Ethernet interface, so you don't even need ppp.  If
you're using the custom version and having trouble, I'd be interested
in knowing what problems you're seeing.  But don't confuse interest
with a commitment to fix it :-)

In summary: what else do we need in PicoBSD?  Do we need to do it in
one disk?  Otherwise, what is a good base 2-disk solution?  And if
that isn't enough, what could be on the third disk?

Greg
--
Finger grog@lemis.com for PGP public key
See complete headers for address and phone numbers


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-small" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000122145538.A390>