Please set your mailer/editor linewrap to 68-75 characters. I strongly
recommend 72 as a good default.
While many mail clients will accomodate unwrapped text:
- Some don't. Be considerate.
- Many more fail to wrap and attribute quotes properly.
- Many web-based list archives render unwrapped text as very long
lines, e.g.:
http://lists.debian.org/debian-devel/2003/debian-devel-200309/msg00568.html
Thank you.
on Fri, Jan 14, 2005 at 12:40:15AM -0800, ken keanon (kenkeanon@yahoo.com) wrote:
> Whether we call it 'free' or 'open', 'commercial' or 'propriety' is immaterial as long as we know what we are talking about.
>
> As I said before, I'll continue to compare and contrast.
>
> Let's consider one aspect of software, i.e. technical support.
Sure.
> In the 'free' software world, this of course is done freely. You have
> a problem, post it online and you may or may not get an answer. No one
> is obliged to answer it within a certain time, if at all. If you do
> get an answer, if it turns out to be wrong, you can't hold anyone
> accountable. The person may not be qualified to answer the question.
> The email addr may be available, but you don't know the person behind
> it. He/she could be just a kid trying to impress the crowd.
You get what you pay for.
Incidentally: on this list, an *aweful* lot of the content is GPG/PGP
signed, as my messages are. Which gives you at least a measure of
authentication as to who's responding. My record is easy enough to
search for online.
If you want to _buy_ tech support, you can. Most core devs will quote
you their contract rate if you need support.
And there's no guarantee of support in the "paid" world either:
- Software support terms can change, without notice. See Quicken and
Microsoft for EOL statements. Hell, Free Software does this too,
just ask my $PRIOR_GIG who'd "standardized" on RH 7.3 "for several
years", just before that product EOLd.
- There's no guarantee of a fix, or a timely fix, in proprietary
software. There's a lot of WONTFIX, "feature, not bug", and bug
denial (look through any six months of third-party Microsoft-related
product security alerts).
- Much proprietary support is worthless, incomprehensible, inaccurate,
incomplete, or inappropriate. And the stuff that isn't costs
$REAL_MONEY.
- Much _worthwhile_ proprietary support is just the same as what
you'll find in Free Software: free, mailing list, Usenet, or web
board based discussions. With all the concmittant issues of
reliability and trust. Except that most proprietary support boards
have little understanding of PKI and signed mail. OT: My own
introduction to community support models was for an enterprise-level
product selling for $50k-$100k/year, or better, on a subscription
basis. All-you-can-eat support, but often the best advice was what
you'd get from other users, online. After that experience, the
GNU/Linux model was *very* easy to accept.
> In the 'commercial' world, I know of an online technical support, that
> provided by HP Computers but is free for all as long as it is related
> to HP computer products. I have seen how it works. When a question is
> posed, the response is quick and the person answering the query has a
> human face.
I'm assured that the monkey-faced population on debian-user is a *very*
small percentage....
> On the questioner part, he/she has to give a feedback on the quality
> of the answer by giving points to the person who provides the answer.
> HP ranks the technical support staff with the total points each have
> accumulated and this in turn affects their performance review. Which
> do you think is a better system?
Hrm. HP support....
HP World users disdain offshore support move ...
http://www.computerworld.com/managementtopics/outsourcing/story/0,10801,95386,00.html
> Let's swing to the higher end of the spectrum, that of innovation. One
> thing that can be safely said about the 'free' software world is that
> it has not led in innovation.
Chimera: Innovation is *hard*. Neither proprietary nor free software
development supports it well.
What Free Software *does* do is provide an environment in which
alternatives can thrive, compete, and the best be selected from among
them. In the same way that evolution isn't a process which describes
_how_ variance in individuals comes to be, but how such variants are
promoted, relative to one another, the open source development model
provides a competitive environment favoring better options.
For a surprisingly relevant explanation of this process, read Jarred
Diamond's _Guns, Germs, and Steel_.
> OS? There was UNIX before Linux.
Guess what: there are Free Software implementations of:
- Unix (GNU/Linux, *BSD, Hurd)
- legacy MS Windows (ReactOS)
- BeOS (Um, somewhere)
- DOS (FreeDOS)
What you mostly hear about, though, is GNU/Linux. Why? Well, it works,
it's usable, it's stable, it's extensible. There's a large base of
existing software that can run on it (or be made to run readily).
And: Unix was originally developed under conditions strongly similar to
what is today called Free Software or Open Source. For a number of
reasons, among them that the idea of copyrighting software hadn't been
established yet (that happened in 1976), and AT&T was prohibited by
antitrust settlement conditions from selling computer operating systems.
The result: a Unix-style OS is strongly suited to the open source
development process.
> Firefox? Apache? Openoffice? All these have commecially innovated
> counteparts that existed before them.
First: how you fund and initiate a project has little bearing on its
ultimate relationship vis-a-vis free vs. proprietary. Back to the
evolution analog: project initiation is the incubation stage. For
novel projects (or species) what you want is a relatively contained,
nurturing environment. Typical examples are academic projects,
organizationally-funded projects (e.g.: NSF, NCSA, CERN), or internal
corporate development (Perl). And sometimes, opened proprietary
projects (OpenOffice).
Incidentally, of the projects you list:
- Firefox was an independent effort based on Mozilla, which was a
from-scratch rewrite of Netscape's browser, which was based on the
original Mosaic browser, itself an academic research project at the
University of Illinois National Center for Supercomputing
Applications.
- Apache "a patchy webserver" was based on NCSA, but evolved
massively. Netcraft still tracks NCSA's browser share. It's
minuscule. The NCSA project itself was somewhat collaborative,
dating from the dawn of the modern Free Software era.
- Openoffice is the one project which emerged from a proprietary
product. Incidentally, it's taken a lot of heat on the basis of
this origin. The codebase is huge and byzantine. The application,
previously monolithic, has been largely (though not wholly)
modularized. It's a very strong example to my mind of why freeing
proprietary software is hard (the Netscape example is another).
Free software and proprietary development models tend to be very
different, with strong impacts on architecture.
> Why? Its in the nature of things and the way they are done. Software
> is a complex entity. Software innovation requires individual
> discipline and teamwork. It can't be left to a group of unpaid
> volunteers who want to do things in their own free way and their own
> free time.
Tell you what: Name the ten biggest software innovations of the past 30
years, and where they came from.
> For the 'free' software movement to leap-frog the 'commercial' world,
> it has to come up with an innovation which has a competitive advantage
> over its 'commercial' competitor.
Well (starred items predate free software but follow similar
methodologies):
- The free software development model itself
- The World Wide Web *
- The Internet *
- Unix *
- Rsync
- Bittorrent
- A Package Tool (apt), Debian's package management system
Truth is, though, that most software is pretty fundamental, and changes
are evolutionary, not revolutionary. There's a lot of stuff which in
aggregate is pretty cool. Knoppix, say. But it's not, itself, a wild
innovation. It's an assemblage of things which themselves may or may
not be revolutionary:
- CDROM
- Bootable CDROM
- Loopback filesystem
- Compressed filesystem
- Package management
- Hardware autodetection
- An available software base
The result, though, is pretty damned impressive: 1200+ software
packages, totalling > 1 GiB, on 750 MiB of storage, updated every few
weeks, booting in a couple of minutes on a wide range of hardware to a
usable desktop / technical system, suitable for both the total n00b and
domain experts.
While there are no *technical* reasons why such a project isn't possible
(if not particularly easy) under legacy MS Windows, I note a number of
the reasons why social and licensing imperatives get in the way, and
strongly limit the utility of such tools as do exist:
http://twiki.iwethey.org/twiki/bin/view/Main/WindowsRescueDisk#Why_So_Few_Windows_Rescue_Disks_
http://twiki.iwethey.org/twiki/bin/view/Main/WindowsRescueDisk#_http_www_nu2_nu_pebuilder_BartP
> How long do we have to wait for that?
By my count, negative 36 years.
Peace.
--
Karsten M. Self <kmself@ix.netcom.com> http://kmself.home.netcom.com/
What Part of "Gestalt" don't you understand?
The black hat community is drooling over the possibility of a secure
execution environment that would allow applications to run in a
secure area which cannot be attached to via debuggers.
- Jason Spence, on Palladium aka NGCSB aka "Trusted Computing"
Attachment:
signature.asc
Description: Digital signature