All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Mike \(mwester\)" <mwester@dls.net>
To: <openembedded-devel@lists.openembedded.org>
Subject: Re: Getting Started -Makefile
Date: Tue, 20 Nov 2007 09:32:18 -0600	[thread overview]
Message-ID: <008f01c82b8a$89c8ab50$6e01a8c0@twilight> (raw)
In-Reply-To: 1195551213.6337.9.camel@localhost.localdomain


> From: "Richard Purdie" <rpurdie@rpsys.net>
>
> I think Holger's point and the general feeling of other OE devs, me
> included is that the current wave of makefiles hide too much. I had to
> solve this problem with Poky and I did it with a script you source which
> sets up a magic environment. I don't expect users to know/care about
> what the script does. I do want users to end up typing "bitbake xyz"
> though. Why?
>
> Users end up realising that everything is either an SCM command ("svn
> up" for Poky) or a bitbake one. Neither are complicated programs to
> learn. Yes monotone is more complex than svn but all you really need to
> know is "mtn pull;mtn up" and even then, understanding why there are two
> commands actually teaches you something important about the SCM. Hiding
> them from the user just creates a  phobia of using the real tools.
>
> Cheers,
>
> Richard

I know bitbake.  I use bitbake.  I debug bitbake (more often than I want to,
actually).

Yet I also type "make all" to build things.

Why?

Not because I have a "phobia of using the real tools."  Not because I'm
stupid.  Not because I'm philisophically opposed to bitbake.  It might
perhaps be because I'm so old that I have moss growing on me, and that I've
been typing "make <target>" since rocks were young and dirt was just
invented.  It's true that my brain thinks "bitbake <thing>" and my fingers
type "make <thing>".

I ask the OE team why are they are philosophically opposed to driving
bitbake with make?  Make has driven software development since near the dawn
of time.  Make has been used to set up environment variables, adjust for
unusual or tricky build environments, and generally to ease the life of a
developer for eons.  It works very well for its purpose.  So why would
anyone encourage a development approach that throws out such a useful tools
and forces a user to manually set environment variables, and remember an
obscure command to invoke a build?  Of course a serious developer should
know generally what goes on underneath, but that doesn't mean that they
should do all the steps manually!

Unless it is the intent to create a barrier to entry?

Mike (mwester)




  parent reply	other threads:[~2007-11-20 15:34 UTC|newest]

Thread overview: 56+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-11-19 12:08 Getting Started -Makefile David Farning
2007-11-19 14:33 ` Koen Kooi
2007-11-19 23:51   ` Rod Whitby
2007-11-20  9:12     ` pHilipp Zabel
2007-11-20 17:57       ` Tim Bird
2007-11-20 20:00         ` Cliff Brake
2007-11-20 20:38         ` Philip Balister
2007-11-20 23:25         ` Florian Boor
2007-11-21  2:24           ` Paul Sokolovsky
2007-11-21  3:38           ` Rod Whitby
2007-11-21  2:24         ` Paul Sokolovsky
2007-11-21  5:44           ` Rod Whitby
2007-11-21 17:45             ` Paul Sokolovsky
2007-11-22  1:35             ` Michael 'Mickey' Lauer
2007-11-22  4:20               ` Lorn Potter
2007-11-22  8:14                 ` Esben Haabendal
2007-11-23  1:05                   ` Lorn Potter
2007-11-21 17:58           ` Tim Bird
2007-11-21 18:29             ` Paul Sokolovsky
2007-11-21 23:17               ` Rod Whitby
2007-11-22  0:44                 ` Tim Bird
2007-11-22 20:29                 ` Paul Sokolovsky
2007-11-21 19:32             ` Philip Balister
2007-11-21 19:32             ` Philip Balister
2007-11-20 21:16       ` Rod Whitby
2007-11-21  6:20       ` Esben Haabendal
2007-11-21 17:57         ` Paul Sokolovsky
2007-11-21 23:04           ` Rod Whitby
2007-11-22  8:40           ` Esben Haabendal
2007-11-20  9:39     ` Richard Purdie
2007-11-20 11:06       ` Rod Whitby
2007-11-20 12:30         ` Koen Kooi
2007-11-20 21:20           ` Rod Whitby
2007-11-20  9:58     ` Koen Kooi
2007-11-20 11:11       ` Rod Whitby
2007-11-20 12:27         ` Koen Kooi
2007-11-20 21:19           ` Rod Whitby
2007-11-20 14:39         ` Paul Sokolovsky
2007-11-20 23:20       ` Florian Boor
2007-11-20 11:44     ` Rod Whitby
2007-11-20 13:10       ` Koen Kooi
2007-11-20 21:11         ` Rod Whitby
2007-11-20 23:06           ` Lorn Potter
2007-11-19 14:41 ` Holger Freyther
2007-11-19 18:32   ` Tobias Pflug
2007-11-19 21:08   ` Lorn Potter
2007-11-20  9:33     ` Richard Purdie
2007-11-20 10:46       ` Marcin Juszkiewicz
2007-11-20 15:32       ` Mike (mwester) [this message]
2007-11-20 17:04         ` Marcin Juszkiewicz
2007-11-20 17:29         ` Holger Freyther
2007-11-20 17:29         ` Koen Kooi
2007-11-20 23:12         ` Lorn Potter
2007-11-22  1:32           ` Michael 'Mickey' Lauer
2007-11-22  1:28         ` The truth about OE team being afraid of 'make' (was: Getting Started -Makefile) Michael 'Mickey' Lauer
2007-11-22  3:15           ` Chris Larson

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='008f01c82b8a$89c8ab50$6e01a8c0@twilight' \
    --to=mwester@dls.net \
    --cc=openembedded-devel@lists.openembedded.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.