[Linux-aus] RELEASE: DON'T LET YOUR VENDOR'S ROADMAP LEAD YOU UP THE GARDEN PATH
Con Zymaris
conz at cyber.com.au
Wed May 12 17:24:01 UTC 2004
On Wed, May 12, 2004 at 06:39:25PM +1000, Jeff Waugh wrote:
> <quote who="Con Zymaris">
>
> > Yes, but think about the differences between FOSS and closed.
> >
> > 1) who sets the roadmap?
>
> FOSS: The developers (be they company or not, in many circumstances mixed).
>
> Closed: The company (developers, marketing, etc).
>
> What's the difference? Are you implying that FOSS roadmaps are better at
> satisfying *users*? Seriously? The funniest example of this can be found
> wholly within the FOSS world - GNOME and KDE. Maybe users do count... But
> which users? :-)
Let me provide you with some context.
In recent months, senior Microsoft execs have stated, on several
occasions, that a core difference between FOSS and closed source, is that
the latter has a real company, projecting a real roadmap ahead, as a great
asset for users of that vendor's platforms.
I disagree. I'm saying that architectural roadmaps, as have been used by
firms like Microsoft, are a great disadvantage to users as they have, more
often than not, left users stranded when that vendor does an about face
and flips to a new, different roadmap. Product roadmaps are less of an
issue.
So, the mere fact that Microsoft is touting their roadmaps as a
'positive', and as an advantage over FOSS, is a message that needs to be
'corrected' in the public perception. Thus this the press release.
If Microsoft hadn't opened their mouths to tout a phantom advantage,
neither would I to tear that phantom down.
>
> GNOME "changed the roadmap" when 2.0 came out. Although it has been quite
> successful outside the technical user community, at least within it, the
> change raised some ire.
Yes, there are instances were FOSS projects 'jump', with annoying
discontinuities. But there are, pound-for-pound, more of these, with more
heart-breaking effects on users, in the closed space.
You guys remember Sun pulling the plug on SunOS? Then Solaris on x86? I
can point to a million lines of VB code that I know of, which will not
compile from one version of the compiler to the next. You show me where
this has happened with GCC, or Python or PHP.
>
> > 2) what recourse do users have to deviate from this roadmap.
>
> FOSS: Massive engineering effort to fork, or use another tool entirely.
>
> Closed: Use another tool entirely: go buy something else.
Assuming you can get your documents and data out...
Microsoft touted Blackbird to tens of thousands of content publishers in
the mid-90s. Anyone who spent time and effort on that platform ended up
with nothing usable 18 months later on. Once again, show me where this has
happened, to such a large extent, in the FOSS space.
>
> > 3) How many of these are orphaned when there is a large installed base of
> > generally happy users?
>
> FOSS: "I am just a user, I can't maintain it myself! I am orphaned!"
>
> Closed: "I am orphaned!"
>
> > Also, for FOSS, these are per-project roadmaps, not all-encompassing
> > system frameworks which require a big retooling job to migrate to or
> > adopt.
>
> But realistically, there are still single-point-of-failure projects in FOSS
> land... and when they change, so does everyone else. :-)
Indeed, no one claimed there wasn't. FOSS projects, however, do not claim
that the existence of a roadmap as a marketing advantage over competing
platforms, because, as we all know, roadmaps can change. The illusion of
platform stability that a roadmap offers users, is therefore a dangerous
thing.
>
>
> [ Anyway, I'm not arguing the validity of your point, just pointing out that
> the finger of accusation might come straight back at us. :-) ]
>
You know my motto on that score Jeff ;-) Bring 'em on...
--
_____________________________________________________________________________
Con Zymaris <conz at cyber.com.au> Level 4, 10 Queen St, Melbourne, Australia
Cybersource: Australia's Leading Linux and Open Source Solutions Company
Web: http://www.cyber.com.au/ Phone: 03 9621 2377 Fax: 03 9621 2477
More information about the linux-aus
mailing list