Dear Graeme Russ,
Hi Marek,
On Fri, Oct 5, 2012 at 10:28 AM, Marek Vasut marex@denx.de wrote:
Dear Albert ARIBAUD,
Hi Marek,
Comments based on the assumption that we want to sync with the Linux tools.
General comment/hypothetical question: would it not be simpler to patch the existing Linux tools in-place so that we can use them on the U-Boot tree?
Yes, it is a good idea. I'll do that. The problem is, replies to my patches do documentation mailing lists are slow, that's for one thing.
The other, much more grave and unpleasant is that we're way too far behind the DM schedule. I don't know what to do, but since pushing stuff upstream goes much slower than I expected, I will soon be left with no option other than forking u- boot, finishing the university project and -- at the end, without the team -- merge the stuff slowly back upstream.
If I were in your position, I would not hesitate for a minute to do a fork.
I do, bloody damnit! Just remember how it ended last time -- the result was a stupid useless piece of crap which was never merged back. Lot of wasted time etc.
The beauty of the FLOSS model of development is that forking is trivial and if the fork has worthy features, merging later will meet with general enthusiasm from the community. The merge will probably happen faster than you think (and may actually happen in parallel so that when your project finishes, more than half of your work has already been integrated)
See above.
And once you fork, all the bottlenecks of global community approval disappear.
And the fork will be sheets of crap, one on another ... because there'll be no proper review.
My position is that I DO NOT WANT TO DO IT ... unless I'm pushed to do so by the circumstances. And I'm reaching that point :'-C
Regards,
Graeme
Best regards, Marek Vasut