http://www.20thingsilearned.com/open-source/1

Cheers,
--Ken Ritchie (Atlanta)
;-)

On Fri, Dec 17, 2010 at 8:13 AM, Oleksandr Nikitin <[email protected]>wrote:

> ?Can't agree more with Thomas.
>
> For most of the programmers out there, the best way to 'get in touch' with
> the project is to see the actual guts;
> leave the shiny presentations for sponsors and marketing people ;-)
>
> For example, the OMeta release was hugely successful - I find more and more
> applications of it, in completely
> different projects from different areas of computing, while other parts of
> FoNC remain only barely known, if at all,
> to people outside of this mailing list.
>
> -- Oleksandr Nikitin
>
> -----Original Message----- From: Reuben Thomas
> Sent: Friday, December 17, 2010 2:45 PM
> To: Fundamentals of New Computing
> Subject: [fonc] Show Us The Code!
>
>
> I started reading with interest the October 2010 STEPS Progress
> Report, then as soon as I got to the first screenshot, was overcome by
> a familiar feeling of depression: I strongly suspected, and quickly
> confirmed, that there was no code I could try.
>
> We've been over this ground in the past, and have seen two related
> arguments for not following the dictum "release early, release often":
>
> 1. "It's not ready".
>
> 2. "We don't have time to support people trying to build non-production
> code."
>
> To which the answers are, respectively:
>
> i. You don't understand "release early, release often".
>
> ii. You don't understand "release early, release often".
>
> Subscribers to this list do not, on the whole, care if code coming out
> of VPRI is buggy, non-portable, pre-alpha, or will only run correctly
> at a certain phase of the moon. We will quite happily help each other
> build and play with it. We understand that you have better things to
> do than support it, and also that in a few months time you may have
> thrown it all away and started again.
>
> I, personally, often don't even try building such code; I just want to
> know that other people can!
>
> You still seem to be stuck in the mentality of wanting to produce a
> finished product. But in computing we all know there is no such thing.
> Releasing unfinished, half-baked code will NOT turn people off. In
> fact, the effect is exactly the opposite: a huge marketing WIN!
>
> If you still doubt me, just look at the number of questions about
> building Ian Piumarta's idst repo, which has only seen one, trivial,
> commit this year.
>
> Releasing a technical report on software without code is like
> publishing a review of a new piece of music but not the score: it's
> tantalizing, but essentially useless. By waiting until you think you
> have something worth releasing, you are losing years—YEARS!—of free
> marketing and buzz. And you need it! Competitors doing merely
> evolutionarily cool stuff will consign your entire project to the
> dustbin of history, if you don't get people excited about what you're
> doing, and if we're lucky, someone else will pick up the ideas in a
> few decades when they have dwindled to a mere stepwise advance, just
> before they become actually redundant.
>
> The history of computing is littered with just such lost
> opportunities; please don't become yet another.
>
> --
> http://rrt.sc3d.org
>
> _______________________________________________
> fonc mailing list
> [email protected]
> http://vpri.org/mailman/listinfo/fonc
>
> _______________________________________________
> fonc mailing list
> [email protected]
> http://vpri.org/mailman/listinfo/fonc
>
_______________________________________________
fonc mailing list
[email protected]
http://vpri.org/mailman/listinfo/fonc

Reply via email to