Re: [Gnu-arch-users] Utterly painless arch?

[john]

Hi All,

I've been lurking on this list for several months, watching as arch has been
accellerating.
I am a very simple user of CVS.  I have a score machines in 4 locations with
about 50
critical source code directories.  Each application controls hardware (think
motors and data
acquisition systems), fixes are done on location and synced to other
locations with similar
hardware using CVS and a central repository.  As a CVS user I have only ever
used import,
checkout, update and commit commands (well init as well but that was years
ago).

I still have only a basic understanding of how branching works.  I
understand is conceptually
mind you, but don't have my hand on how to use it at work.

Scaling down is a very important concept.

John

----- Original Message -----
From: "Zack Brown" <address@hidden>
To: "Tom Lord" <address@hidden>
Cc: <address@hidden>
Sent: Tuesday, September 09, 2003 2:18 PM
Subject: Re: [Gnu-arch-users] Utterly painless arch?


> On Tue, Sep 09, 2003 at 11:16:43AM -0700, Tom Lord wrote:
> >     > From: Zack Brown <address@hidden>
> >
> >     >> a) screwing around with =tagging-method is, indeed, the number
one
> >     >>    problem newbies encounter
> >
> >     > That's why the default should be something that requires no user
> >     > action, but just works reasonably; and then if they want better
> >     > behavior, they should be able to change it later.
> >
> > What's your opinion about the user-action of `tla add'?  That's about
> > all that will be required using what I've proposed and, in that
> > regard, it's the same as CVS.
>
> Sounds perfect.
>
> >     [re the revision namespace]
> >
> >     > It's just an additional inconvenience. Before anyone can start to
use
> >     > tla, they are forced to develop a fairly deep understanding of its
> >     > organizational infrastructure. While this organization may be
good, it
> >     > becomes part of a tla boot-up barrier that can't be avoided.
> >
> > The thing is, though, that while you may have talked to people who
> > pick that issue to gripe about -- a couple of years of experience
> > suggests that, really, it's not a serious barrier to adoption and does
> > have plenty of benefits in practice.  So, the utility/cost of trying
> > to soothe the people you've talked to (about this specific issue)
> > doesn't look (at least as things stand) very large from over here.
>
> I think fundamentally, we agree. People who need tla will (a) be able to
> grok it relatively quickly, and (b) be able to adapt to its conventions.
>
> All these "painlessness" issues could really be classified under the
rubrik
> "downward scalability". How *little* can a person know, and still make
> effective use of tla? And what is the fewest number of commands needed to
> get started on any project? And how easy is it to mop up a project, once
> the user has decided not to work on it anymore?
>
> Not all ideas for downward scalability are so good. It could be that tla
> naming is exactly as it should be. But I think the *concept* of downward
> scalability is one that should be considered in tla's design, especially
> since it seems to have got upward scalability pretty much nailed.
>
> Be well,
> Zack
>
> >
> > -t
>
> --
> Zack Brown
>
>
> _______________________________________________
> Gnu-arch-users mailing list
> address@hidden
> http://mail.gnu.org/mailman/listinfo/gnu-arch-users
>
> GNU arch home page:
> http://savannah.gnu.org/projects/gnu-arch/
>