From: Dragontamer on

rhyde(a)cs.ucr.edu wrote:
> Dragontamer wrote:
> >
> > I'd have to disagree with you on this one. OSS do have innovative
> > ideas.
> >
> > The Abiword project for example, is beginning to work on AbiCollab,
> > which would be a plugin that allows users across the web to edit the
> > same document in realtime.
>
> Can you honestly claim that this idea hasn't been floating around for a
> long time and that commercial apps haven't tried this? What I'm
> calling "innovation" here are brain-storm ideas resulting in concepts
> that have never before been done.

Abiword is run by volonteers, they don't got a paycheck. I find it
hard to believe that they would make something before Microsoft
unless they had a head start on the "innovation".

If Microsoft thought it up, they would have put it into MS Word
before Abiword probably existed. But to my knowledge, there
are no other word processors that exist that have even attempted
this idea.

> > http://www.abisource.com/~uwog/abiword/abicollab.htm
> >
> > As for something like VB or Delphi... I dunno much on Delphi, but
> > GNUstep seems to be making strides in the direction of VB, somewhat
> > copying Cocoa and NextSTEP... but still goes along in that direction.
>
> But that's exactly the point I'm making -- the OSS crowd are *copying*
> stuff rather than invented brand-new, never-before-thought-of, kind of
> stuff.
>
> Now don't get me wrong, I'm not saying there aren't *any* new ideas in
> OSS. There are lots of small things that are innovative and come out of
> OSS, but the really big stuff, that requires a concerted effort to
> develop, generally comes from commercial development.

I answer this later... I answer posts backwards, starting from the
bottom up. Sorry. :-p

> > Debian's apt system is great, as is FreeBSD's port system. Truely
> > innovative to just do "aptitude upgrade", and all the programs
> > are upgraded together, to resolve dependancies and so forth.
> > (supposedly FreeBSD's port system is better... but I haven't used
> > it yet)
>
> Having spent considerable time installing FreeBSD over and over again,
> I'm not real impressed by FreeBSD's "ports" that forces me to swap CDs
> a couple of dozen times because it installs things in the order *it*
> wants to, rather than their order on the CDROMs. Forgive me for not
> being impressed.

Try it when you have a net installation. Auto Downloads == great
stuff.

> > Fasm's "Virtual" keyword is another innovative little feature developed
> > on an OSS product.
>
> First of all, I'm not entirely sure it's as innovative as you think. I
> dimly recall a feature like this in IBM's mainframe assemblers. But the
> point is irrelevant. I'm not talking about new features in a
> traditional product. I'm talking about completely new ways of solving
> problems. Delphi, for example, isn't just Pascal with a few new
> features. It truly represents a new programming paradigm over previous
> languages (well, except for VB, of course).

You mean, like Perl (OSS Artistic License)? A totally new programming
paradigm that focuses on text and utility using "natural language
constructs" ?

Or TeX which lead the way into digital typesetting? Yes, Knuth
is a ****ing genious, but that doesn't change the fact that TeX is Open
Source, and that his innovation is thus, technically, OSS. Knuth wasn't
paid at all to make TeX.

And the way TeX was written, using WEB programming language, was
truely innovative. Literate programming, integrating typesetting and
compiling, while... erm, IMO, stupid, is an innovative idea, and OSS.
(The TeX source code is wonderful to read though :-p Haven't
read a lot of it, but its still cool to read :-) Literate programming
to the MAX)

> > There is certainly innovation in OSS products. Do you not think your
> > HLA is somewhat innovative? :-p
>
> Not like Delphi. HLA is more of a collection of existing good ideas all
> put into one package. Most of the "innovations" in the product are like
> "virtual" in FASM. Small ideas that are cool in conjunction with the
> other features, but aren't much of a standalone innovation. About the
> only thing in HLA I know of that I've never before seen are the
> context-free macro facilities. The parts of HLA that make it good for
> education were well-established ideas that predated HLA -- using
> HLL-like control structures (MASM & TASM), having a decent and
> substatial library (The C stdlib, the UCR stdlib for 80x86 assembly
> programmers), supporting reasonable data types (MASM & TASM), and
> having a good compile-time language (IBM's HLAsm and MASM/TASM). HLA
> is not a radical shift in the way you would do assembly language
> programming. It's just a culmination of good ideas all packaged
> together.

As was Sliced Bread.

The inventor of sliced bread did not invent bread, nor slicing. But the
packaged result was truely innovative </ibm commercial>

Point is: the majority of good ideas are just packages of old and other

ideas, possibly outside the field. (Genetic Algorithms == AI + Biology
for example)

Now, I haven't used Delphi myself, but if it is half as good as its
reputation (and pricetag :-/ ) then I guess its gotta be great.

> > But anyway, there are ways to keep OSS and still make a business model.
> > Cedega, which is based on WINE, sells builds their product for a
> > monthly
> > subscription, while the code remains free on their CVS server. So
> > its not necessarily mutually exclusive.
>
> Well, we were actually talking about how OSS is killing commercial
> companies that have invested tons of money in software R&D. The problem
> with the OSS model is that OSS developers work on what they *feel* like
> working on. There is no incentive, like a paycheck, to get them to work
> on stuff that wouldn't be their first choice and to keep them working
> on a (large) project when it ceases to be interesting. Even companies
> like IBM, which has invested a *ton* of money in OSS, probably wrote it
> all off as a marketing expense rather than a development expense.

Hmm. I just don't think ego is the main insentive to OSS development.

Okay, HLA: its incentive is to teach students assembly language. (or
at least, I assume thats your insentive) Else, to help support
AoA32, which really goes back to goal # 1.

TeX: The incentive was to make TAOCP look better, better than any
other digital
From: Betov on
"Dragontamer" <prtiglao(a)gmail.com> crivait
news:1154319268.261620.158970(a)i3g2000cwc.googlegroups.com:

>
> rhyde(a)cs.ucr.edu wrote:
>> Dragontamer wrote: [...]

Trying to convince a Right Extreeme and a religious dement,
of the vertues of the Open Source Mouvement?

Good luck.

:]]]]]

Betov.

< http://rosasm.org >


From: randyhyde@earthlink.net on
Dragontamer wrote:
>
> Abiword is run by volonteers, they don't got a paycheck. I find it
> hard to believe that they would make something before Microsoft
> unless they had a head start on the "innovation".

I'm sorry, I don't remember claiming that all innovation comes out of
Microsoft. (innovation==Microsoft, now that's kind of funny).

>
> If Microsoft thought it up, they would have put it into MS Word
> before Abiword probably existed. But to my knowledge, there
> are no other word processors that exist that have even attempted
> this idea.

The only reason Microsoft hasn't done this is because Apple didn't do
it first :-)
(okay, running away rapidly.)

Seriously, though, Microsoft has done their share of innovations, but
most people *don't* hold up Microsoft as the paragon of innovation.


> >
> > Having spent considerable time installing FreeBSD over and over again,
> > I'm not real impressed by FreeBSD's "ports" that forces me to swap CDs
> > a couple of dozen times because it installs things in the order *it*
> > wants to, rather than their order on the CDROMs. Forgive me for not
> > being impressed.
>
> Try it when you have a net installation. Auto Downloads == great
> stuff.

Yes, I've done that. How is this "innovative"? I've been doing this
with Apple, Microsoft, and Linux OSes for quite some time. Perhaps
FreeBSD is a little slicker in some respects (I don't know), but this
is hardly an "innovative" concept. At the very best, it's an extension
of an existing idea.



> You mean, like Perl (OSS Artistic License)? A totally new programming
> paradigm that focuses on text and utility using "natural language
> constructs" ?

Take a look at SNOBOL4 sometime (a programming language from the 1960s.
Then take a look at ICON. (1970s).

And being forced to use PERL on ocassion at work, I hardly feel that it
uses "natural language constructs". PERL is like the BASIC language of
the 2000s -- every time they need something they just clutter up the
language with yet another keyword using inconsistent syntax.

>
> Or TeX which lead the way into digital typesetting? Yes, Knuth
> is a ****ing genious, but that doesn't change the fact that TeX is Open
> Source, and that his innovation is thus, technically, OSS. Knuth wasn't
> paid at all to make TeX.

Actually, Knuth *was* paid to produce TeX. You might recall that he is
a University professor (at Stanford) and professors are paid to do
research. Granted, he wasn't *forced* to do this job by work, but it's
not like he wasn't getting paid to do the work. And TeX wasn't Knuth's
real innovation -- MetaFont was. After all, Tex is just an expansion of
the nroff idea that predated it. And finally, the creation of TeX
predated the OSS movement by many years. It was a single-person
project, by a person who was being paid for his time spent on the
project. That's not the OSS model.

>
> And the way TeX was written, using WEB programming language, was
> truely innovative.

Uh...
TeX was written in Pascal.
TeX predated the WEB programming language by many years.

> Literate programming,

Again, research coming out of Universities by people who were *paid* to
do this research. Not the OSS model at all.

> integrating typesetting and

Sorry, dude, I was around then. Knuth wasn't the first one to do this.
The graphic arts field was doing this already. And, of course, there
was the troff program that did typesetting from a computer that predate
all of this (developed at Bell Labs, I might point out). Knuth did
some cool things, like MetaFont. But again, this is not an example of
an OSS triumph. It is a result of university grants (typically from
commercial concerns) and student fees paying professors and grad
students to do the research and work.



> compiling, while... erm, IMO, stupid, is an innovative idea, and OSS.

The OSS model didn't even exist back then.



> As was Sliced Bread.

But was it really?
Was their really a single person who came up with idea? Or was it just
an obvious idea that lots of bakers developed independently? I.e., do
you have the *name* of the guy who invented sliced bread. Can you show
it was an unobvious idea that did not exist in any form prior to that
point?


>
> The inventor of sliced bread did not invent bread, nor slicing. But the
> packaged result was truely innovative </ibm commercial>

Was it "The inventor" or was it a obvious idea that several people
invented independently?


>
> Point is: the majority of good ideas are just packages of old and other
>
> ideas, possibly outside the field. (Genetic Algorithms == AI + Biology
> for example)

With one kicker -- a *non-obvious* combination.
I.e., *can you patent it*? (assuming the patent system worked the way
it's supposed to).


>
> Hmm. I just don't think ego is the main insentive to OSS development.

Anytime someone suggests that they don't want to release something as
Public Domain because they want their name attached to it, or that they
don't want other people taking credit for their work, that's your first
clue that ego is involved. Ego *drives* OSS development. Make no
mistake about that.

>
> Perl: A tool designed to process large amounts of I/O data quickly.

Yeah, that's why it's interpreted :-)

>
> BitTorrent: Innovation to the whole structure of the P2P
> network.
BitTorrent is certainly an example of an innovative product, for sure.
Cheers,
Randy Hyde

From: f0dder on
randyhyde(a)earthlink.net wrote:

>> And the way TeX was written, using WEB programming language, was
>> truely innovative.
>
> Uh...
> TeX was written in Pascal.
> TeX predated the WEB programming language by many years.

Hm, how is WEB related to TeX then? When building TeX from source on a linux
box a bunch of years ago, I seem to remember that part of the process was
running some "web2c" program?

TeX is a pretty nifty idea, but IMHO it's too arcane to be really useful.
Becomes more human with things like LaTeX, but still - unless you're using
some pre-packaged template, there's *some* amount of manual setup you need
to do.

It's also a bunch of bother to change fonts and such (no, not changing fonts
in the traditional Office style, but setting up a TeX equivalent of a CSS
style). It would be nice if there was something as intuitive as XML/CSS that
could give as pretty output as TeX...

>> Perl: A tool designed to process large amounts of I/O data quickly.
>
> Yeah, that's why it's interpreted :-)

To be fair, the perl runtime is compiled C, not interpreted perl :) - and
the regexes are "compiled" as well... speed does seem fair enough for what
it does. And while I'm not too fond of it's hackyness, it certainly is a
useful tool for getting jobs done fast.

>> BitTorrent: Innovation to the whole structure of the P2P
>> network.
> BitTorrent is certainly an example of an innovative product, for sure.

Innovative, ho humm. Perhaps my sense of time is schewed, but weren't there
a bunch of p2p apps available before bittorrent showed up?


From: rhyde on

f0dder wrote:
> randyhyde(a)earthlink.net wrote:
>
> >> And the way TeX was written, using WEB programming language, was
> >> truely innovative.
> >
> > Uh...
> > TeX was written in Pascal.
> > TeX predated the WEB programming language by many years.
>
> Hm, how is WEB related to TeX then? When building TeX from source on a linux
> box a bunch of years ago, I seem to remember that part of the process was
> running some "web2c" program?

That came much later.

>
> TeX is a pretty nifty idea, but IMHO it's too arcane to be really useful.
> Becomes more human with things like LaTeX, but still - unless you're using
> some pre-packaged template, there's *some* amount of manual setup you need
> to do.

The sad part is that WYSIWYG editing happened with the Lisa and the
Macintosh just a short time period after TeX appeared so other than on
systems that didn't support bit-mapped graphics and fonts (e.g., *nix
systems), the need for TeX evaporated overnight. That's also why TeX
became popular in academia and on *nix systems -- because that's all
they had for quite some time.


>
> It's also a bunch of bother to change fonts and such (no, not changing fonts
> in the traditional Office style, but setting up a TeX equivalent of a CSS
> style). It would be nice if there was something as intuitive as XML/CSS that
> could give as pretty output as TeX...

SGML?
I'm not sure it's intuitive, but it's certainly got the power (e.g.,
Framemaker was able to store files in this format).

> > BitTorrent is certainly an example of an innovative product, for sure.
>
> Innovative, ho humm. Perhaps my sense of time is schewed, but weren't there
> a bunch of p2p apps available before bittorrent showed up?

Yeah, but the "massive # of peers to me while I upload to others" was a
pretty cool idea.
Cheers,
Randy Hyde