From: rhyde on

Dragontamer wrote:
..
>
> Meh. C# and .NET seemed pretty innovated to me. The pieces
> of it of course; were quite

Of course, most people will argue that C# was created to counter the
Java threat. While there were probably some tiny innovations in the C#
language (I can't say, I don't know that much about C#), I think it's
fair to say that the Java/VC++->C# transition was nowhere near as
radical as the Pascal->Delphi transition.

Of course, Microsoft *has* innovated in programming languages. VB is a
prime example.

>
> Meh. Given its popularity, it seems like people like something about
> Perl. There must have been something that "clicked" with perl
> that wasn't there with SNOBOL4. But I'll look at it later.

Yes. It was the modern BASIC language.
BASIC was somewhat popular, too. That didn't particularly make it a
good language from a language design point of view. And it wasn't
exactly innovative, either.


>
> [snip]
> > 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.
>
> Woah woah woah.
>
> It was under my impression that people get paid in the OSS model.
> Not just donations mind you; but actual wages. The XEmacs project
> for example; had programmers working on updating Emacs full time.

That's the exception, not the rule. The basic OSS model is people
contributing their time.


>
> Now... if you are talking about Freeware, and not OSS, then yeah.
> The only thing freeware can do is copy other ideas.

???
That certainly isn't true. Anyone with an idea can implement that idea
and give it away. There could even be *some* innovation taking place
here. And as freeware tends to me more of the "single individual
pursuing an idea of their own" there's a greater chance of "radical
innovation" then when you have a distributed team all trying to agree
on what should be done.

>
> But in OSS, you have people getting paid to say, expand
> and improve the WINE project so that MS Word works on Linux.

Expand and improve isn't exactly "innovation".

There is no question that once a piece of free (OSS other otherwise)
software becomes popular enough, some organizations (e.g., IBM) are
willing to pony up money to help support the product. But by then, the
chance for innovation is long since past.

And I wouldn't hold up WINE as an example of "radical innovation". :-)

>
> But then when the tool is used and people start liking it, you
> get additions to it later.

This is all true and well and good. But I don't see how that's
supporting your claim that OSS produces all these great innovations.

>
> Linux may be developed like a Bazaar, but so far, it seems
> like you need a pretty big Cathedral before anything gets
> done anyway. Bazaar OSS projects tend to be the very popular
> very sucessful projects. (Apache, Linux, so on, so forth).

The whole "Cathedral vs. Bazaar" thing has gone down the toilet with
respect to Linux. The bottom line is that *small projects* (as Linux
was, in the early days) can be developed in a decentralized,
"programming in the small", manner. But once the project gets large, as
Linux is today, those techniques start to fail. The promise of the
bazaar was that you would get quick updates to bugs and new features
would be added rapidly. Gee, look how long it's taking to get *minor*
revisions out of Linux these days. Definitely a cathedral, not a
bazaar. And Linus (for good reasons) rules over the Linux development
and totally controls the "mainstream" release. Hardly the
decentralized model that Eric Raymond was preaching about.

>
> But a good amount of OSS projects are still done by paid
> programmers; or at least, a small dedicated group of
> people.

Most OSS projects are done by individuals. A product has to prove
itself before many people are interested in jumping on board and
helping out. But that is completely orthogonal to innovation.


> I think we have different definitions of OSS model. My thoughts
> of OSS is simply any business model that ends up creating
> software that is released in a OSS license

Yes, we have different ideas.

>
> > > 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.
>
> I think there is a double standard going on here.

Why?

>
> Delphi and VB were not the first programming languages either.
> What exactly do you see in them thats innovative? Maybe that
> will settle the argument a little.

The innovation isn't that they were programming languages, but the way
they integrated the visual element into the language. One could argue
that some of these ideas came from smalltalk and the smalltalk project
at Xerox PARC, but the bottom line is that Delphi and VB represented a
radical change in the way GUI apps were written.

TeX was an evolutionary change over systems like troff that came before
it.

>
> But TeX has glue, automatic kerning, hyphenation, etc. etc.

And many of these features were in troff.

>
> > > compiling, while... erm, IMO, stupid, is an innovative idea, and OSS.
> >
> > The OSS model didn't even exist back then.
>
> Some would argue that everything at that age was OSS.

Not what we think of as OSS development today (e.g., Eric Raymond's
C&B)

>
> In that case: it is Ego that drives game developers to put their names
> in the credits and/or title of the game.

Certainly.

>
> Just because ego is one incentive, does not mean it is the only
> incentive, or even the most important incentive. Example: XEmacs
> was developed to update modernize Emacs, and to be repackaged
> into a compiler/IDE package. In this OSS model, they sold XEmacs
> as part of an integrated package and made money off of it.

And, as you claim elsewhere, they were paid to develop it. Once someone
is getting paid, the dynamics change completely.


Cheers,
Randy Hyde

From: Dragontamer on

rhyde(a)cs.ucr.edu wrote:
> Dragontamer wrote:
> .
> >
> > Meh. C# and .NET seemed pretty innovated to me. The pieces
> > of it of course; were quite
>
> Of course, most people will argue that C# was created to counter the
> Java threat. While there were probably some tiny innovations in the C#
> language (I can't say, I don't know that much about C#), I think it's
> fair to say that the Java/VC++->C# transition was nowhere near as
> radical as the Pascal->Delphi transition.
>
> Of course, Microsoft *has* innovated in programming languages. VB is a
> prime example.

Actually, Microsoft bought VB from some other guy who innovated it.

But thats no different from hiring a guy to innovate and then
waiting for the deliver, really (except the much lower risk and other
unimportant business things :-p )

[snip]

> > > 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.
> >
> > Woah woah woah.
> >
> > It was under my impression that people get paid in the OSS model.
> > Not just donations mind you; but actual wages. The XEmacs project
> > for example; had programmers working on updating Emacs full time.
>
> That's the exception, not the rule. The basic OSS model is people
> contributing their time.

Most OSS models that _work_ and not just make vaporware involve
money.

> > Now... if you are talking about Freeware, and not OSS, then yeah.
> > The only thing freeware can do is copy other ideas.
>
> ???
> That certainly isn't true. Anyone with an idea can implement that idea
> and give it away. There could even be *some* innovation taking place
> here. And as freeware tends to me more of the "single individual
> pursuing an idea of their own" there's a greater chance of "radical
> innovation" then when you have a distributed team all trying to agree
> on what should be done.

Hmm... It seems like you've got a very very different view of what
OSS is than I do. I'm beginning to think this all comes down to what
you think OSS development is.

> > But in OSS, you have people getting paid to say, expand
> > and improve the WINE project so that MS Word works on Linux.
>
> Expand and improve isn't exactly "innovation".

But it requires innovation to expand and improve a product.

If you didn't invent anything or do something new, then you
really didn't improve it at all. Let alone expand the product.

> There is no question that once a piece of free (OSS other otherwise)
> software becomes popular enough, some organizations (e.g., IBM) are
> willing to pony up money to help support the product. But by then, the
> chance for innovation is long since past.

Wait, so you think Freeware projects have a chance for innovation...

While paid OSS projects may not?

Hmm....

> And I wouldn't hold up WINE as an example of "radical innovation". :-)

Yeah. But I'm just pointing out that people can make OSS and still
get paid.

> > But then when the tool is used and people start liking it, you
> > get additions to it later.
>
> This is all true and well and good. But I don't see how that's
> supporting your claim that OSS produces all these great innovations.

There aren't "all these great" innovations, because innovations
by default are rare :-p

I'm just saying that there is nothing to stop an OSS project from
innovation.

> > Linux may be developed like a Bazaar, but so far, it seems
> > like you need a pretty big Cathedral before anything gets
> > done anyway. Bazaar OSS projects tend to be the very popular
> > very sucessful projects. (Apache, Linux, so on, so forth).
>
> The whole "Cathedral vs. Bazaar" thing has gone down the toilet with
> respect to Linux. The bottom line is that *small projects* (as Linux
> was, in the early days) can be developed in a decentralized,
> "programming in the small", manner. But once the project gets large, as
> Linux is today, those techniques start to fail. The promise of the
> bazaar was that you would get quick updates to bugs and new features
> would be added rapidly. Gee, look how long it's taking to get *minor*
> revisions out of Linux these days. Definitely a cathedral, not a
> bazaar. And Linus (for good reasons) rules over the Linux development
> and totally controls the "mainstream" release. Hardly the
> decentralized model that Eric Raymond was preaching about.

I thought the point of the Bazaar was that you can find many many
unofficial patches for your product; and the main guy who is
making it just ties it all together.

Cathedral is that everything is planned from the beginning. The Bazaar
simply points out that Linus (and his team) is not the only guy that
comes up or implements the ideas.

Anyone can add their idea to Linux, and many have.

> > But a good amount of OSS projects are still done by paid
> > programmers; or at least, a small dedicated group of
> > people.
>
> Most OSS projects are done by individuals. A product has to prove
> itself before many people are interested in jumping on board and
> helping out. But that is completely orthogonal to innovation.

Course most. Most projects period are done by individuals,
and are propriatary.

That doesn't mean that the _good_ ones or the ones
that succeed are done by individuals. You may get the one
or two that do, but not really.

> > I think we have different definitions of OSS model. My thoughts
> > of OSS is simply any business model that ends up creating
> > software that is released in a OSS license
>
> Yes, we have different ideas.

Hmm. Maybe you can start by telling me your definition of OSS ??

> > > > 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.
> >
> > I think there is a double standard going on here.
>
> Why?
>
> >
> > Delphi and VB were not the first programming languages either.
> > What exactly do you see in them thats innovative? Maybe that
> > will settle the argument a little.
>
> The innovation isn't that they were programming languages, but the way
> they integrated the visual element into the language. One could argue
> that some of these ideas came from smalltalk and the smalltalk project
> at Xerox PAR
From: rhyde on

Dragontamer wrote:
>
> Actually, Microsoft bought VB from some other guy who innovated it.

First I'd heard that, but okay. Given Microsoft's history with BASIC,
this is not something I would have expected.


> >
> > That's the exception, not the rule. The basic OSS model is people
> > contributing their time.
>
> Most OSS models that _work_ and not just make vaporware involve
> money.

At some point or another, perhaps. But remember, Linux, GCC, Emacs,
stuff like that, was written for the love of the thing, not because of
money.

>
> Hmm... It seems like you've got a very very different view of what
> OSS is than I do. I'm beginning to think this all comes down to what
> you think OSS development is.

The bazaar side of the C&B.


> > Expand and improve isn't exactly "innovation".
>
> But it requires innovation to expand and improve a product.

I'm talking about the type of quantum leap represented by products like
Delphi and VB. That's what started this sub-thread, remember? Borland
leaving the tools business and some of us bemoaning another victim of
OSS. There have been very few tools that were as innovative as Delphi
and VB. These were not small or incremental improvements over Turbo
Pascal and QuickBASIC.

In some respects, one could say that Turbo Pascal was a radical
innovation over the command-line tools before it, but there were some
limited IDE systems at the time (e.g., FORTH) and, of course, the
smalltalk research.


>
> If you didn't invent anything or do something new, then you
> really didn't improve it at all. Let alone expand the product.

Oh, I'd argue that maintenance (e.g., fixing defects) improves the
product. But incremental improvements are rarely "innovative" except to
marketing departments.

>
> Wait, so you think Freeware projects have a chance for innovation...

I think that indivduals, working in isolation, have a much greater
chance of creating innovative than the Bazaar model the OSS embraces.

>
> While paid OSS projects may not?

"Paid" changes the dynamics. At that point people aren't doing it for
love or ego. That's a *big* deal.


> I'm just saying that there is nothing to stop an OSS project from
> innovation.

The nature of the "Bazaar" model is what gets in the way of innovation.
The more people you put onto a team, the more conservative they'll
become. Innovation usually requires risk taking that just isn't
possible when you're trying to please a lot of different people.

> > The whole "Cathedral vs. Bazaar" thing has gone down the toilet with
> > respect to Linux. The bottom line is that *small projects* (as Linux
> > was, in the early days) can be developed in a decentralized,
> > "programming in the small", manner. But once the project gets large, as
> > Linux is today, those techniques start to fail. The promise of the
> > bazaar was that you would get quick updates to bugs and new features
> > would be added rapidly. Gee, look how long it's taking to get *minor*
> > revisions out of Linux these days. Definitely a cathedral, not a
> > bazaar. And Linus (for good reasons) rules over the Linux development
> > and totally controls the "mainstream" release. Hardly the
> > decentralized model that Eric Raymond was preaching about.
>
> I thought the point of the Bazaar was that you can find many many
> unofficial patches for your product; and the main guy who is
> making it just ties it all together.

Perhaps you should read Raymond's book.
And read "A Second Look at the Cathedral and Bazaar", too.
http://www.firstmonday.org/issues/issue4_12/bezroukov/



>
> Cathedral is that everything is planned from the beginning. The Bazaar
> simply points out that Linus (and his team) is not the only guy that
> comes up or implements the ideas.

Actually, Bazaar means quite a bit more than that.

> > TeX was an evolutionary change over systems like troff that came before
> > it.
>
> ObjC + OpenStep/NextStep lead the way for that. According to
> http://delphi.about.com/cs/azindex/a/dhistory.htm
> Delphi was first released in 1995 (correct me if I'm wrong).
> NextStep had GUI programming tools
> in 1989. Not an open source innovation; but when viewed like
> this, Delphi was simply an evolutionary step from NextStep.

I've programmed NextStep. It didn't have the live data tools present in
Delphi. That was the true innovation.

>
> And I'm sure we can see the evolutionary step that lead to
> NextStep if we look into it. But I'm not too keen on that history.

Sure, smalltalk.

Cheers,
Randy Hyde

From: ArarghMail608NOSPAM on
On 1 Aug 2006 18:21:05 -0700, rhyde(a)cs.ucr.edu wrote:

>
>Dragontamer wrote:
>>
>> Actually, Microsoft bought VB from some other guy who innovated it.
>
>First I'd heard that, but okay. Given Microsoft's history with BASIC,
>this is not something I would have expected.

I found this in VB1 -- the gooey version:

Portions of this product were developed
for Microsoft by Cooper Software, Inc.

--
ArarghMail608 at [drop the 'http://www.' from ->] http://www.arargh.com
BCET Basic Compiler Page: http://www.arargh.com/basic/index.html

To reply by email, remove the garbage from the reply address.
From: ArarghMail608NOSPAM on
On 1 Aug 2006 18:21:05 -0700, rhyde(a)cs.ucr.edu wrote:

>
>Dragontamer wrote:
>>
>> Actually, Microsoft bought VB from some other guy who innovated it.
>
>First I'd heard that, but okay. Given Microsoft's history with BASIC,
>this is not something I would have expected.

First that I had heard of it also, However:

http://www.cooper.com/alan/father_of_vb.html
--
ArarghMail608 at [drop the 'http://www.' from ->] http://www.arargh.com
BCET Basic Compiler Page: http://www.arargh.com/basic/index.html

To reply by email, remove the garbage from the reply address.