Article: Ousterhout on open source, Tcl
Neil Madden
development:
http://arstechnica.com/articles/culture/free-software-summit-10th.ars/6
The interviews are quite brief, but there are a few interesting
thoughts. Not much about Tcl, except this snippet from John:
"... I started a company around Tcl in 1998, but it became clear within
a year or two that we couldn't create a very large business. In order to
create a business around an open-source package, there needs to be a
truly huge user community (in the millions), such as Linux or MySQL. Tcl
had a community of hundreds of thousands, but that wasn't enough for a
business...."
-- Neil
davidn...@gmail.com
> John Ousterhout and others reflect on 10 years of "Open Source" software
> development:
>
> http://arstechnica.com/articles/culture/free-software-summit-10th.ars/6
>
> The interviews are quite brief, but there are a few interesting
> thoughts. Not much about Tcl, except this snippet from John:
I think he's wrong about OS not creating much end user stuff. My wife
now uses Ubuntu almost exclusively, and while it has some rough spots,
that simply wouldn't have been possible 5, let alone 10 years ago.
EL
> I think he's wrong about OS not creating much end user stuff. My wife
> now uses Ubuntu almost exclusively, and while it has some rough spots,
> that simply wouldn't have been possible 5, let alone 10 years ago.
But Ubuntu is a system. I think he means more programs like photoshop,
design software or other specialities that are not really mainstream.
And then he is right: as soon as complicated and uncommon technical
application know how needs to flow into the program, it won't be open
source. If I knew that I can sell the special niche program that I
develop for several 10,000 € per license (because the niche market pays
that)... then I would be stupid if I made it available to the public for
free.
Eckhard
skuh...@web.de
> davidn...@gmail.com schrieb:
> But Ubuntu is a system. I think he means more programs like photoshop,
> design software or other specialities that are not really mainstream.
Photoshop is not mainstream?
> And then he is right: as soon as complicated and uncommon technical
> application know how needs to flow into the program, it won't be open
> source.
Not true, there are some. From my former job I remember things like
itk/vtk
and some DICOM-Libraries and Viewer. Those are used for medical
imaging
software, which is IMHO a very uncommon type of software and needs
much
special and technical know how. I'm sure, there are many other
examples.
> If I knew that I can sell the special niche program that I
> develop for several 10,000 � per license (because the niche market pays
> that)... then I would be stupid if I made it available to the public for
> free.
Granting gifts is not necessarily stupid. One may have another
business-model in mind, for
example selling service to the users of the free software. If the
software is really that special
as you assume, then there would be not much competition in the market
niche, so that
giving the source away for free and implement further features on
request is not much risk.
Besides, giving something for free to the world may also simply be a
sign of philanthropy... (okay, just kidding)
Regards
Stephan
Donal K. Fellows
> Photoshop is not mainstream?
Not like word processing is mainstream. There's lots of people about who
never ever need to edit a picture.
Donal.
skuh...@web.de
That's true, but where starts mainstream? I would say, in the field of
image
editing, Photoshop is mainstream as MSOffice and OpenOffice are for
word processing. If word processing is the lower water mark for
mainstream, then
there exists very few mainstream software... Also I think, that
comparing
a category of software ("word processing") with one application
("Photoshop") from
another category ("image editing") is not a valid analogy.
Stephan
Robert Heller
>
> EL schrieb:
> > davidn...@gmail.com schrieb:
>
> > But Ubuntu is a system. I think he means more programs like photoshop,
> > design software or other specialities that are not really mainstream.
>
> Photoshop is not mainstream?
In a sense it is not. At least not the 'professional' version. This
is true of most of Adobe's offerings, esp. given the price tag (on the
order of $500 retail for pro versions). Most home / casual users use
various cheaper lower-end programs. OTOH, GIMP (the open source
alternitive to Photoshop), seems to be doing alright.
--
Robert Heller -- Get the Deepwoods Software FireFox Toolbar!
Deepwoods Software -- Linux Installation and Administration
http://www.deepsoft.com/ -- Web Hosting, with CGI and Database
hel...@deepsoft.com -- Contract Programming: C/C++, Tcl/Tk
EL
> and some DICOM-Libraries and Viewer. Those are used for medical
> imaging
> software, which is IMHO a very uncommon type of software and needs
> much
> special and technical know how. I'm sure, there are many other
> examples.
You are talking about libraries, not about complete turn-key
applications. Which is what I mean.
> Granting gifts is not necessarily stupid.
No, absolutely not. But you must carefully think about what and also how
you make these gifts.
> One may have another
> business-model in mind, for
> example selling service to the users of the free software.
Yeah, but honestly: how many can live from that? To have good services
is not enough if the product doesn't get attention. And at the point
when the product gets attention, it is usually so good that services are
rarely needed.
If you are the maker of the product, there are some other questions: do
you want other (unknown people) to modify your code? Who takes the risk
if the code base breaks? What about documentation, bug and feature
request management... about creating a user community and user base for
the open source part? You won't get any bussiness, if you don't have a
user base for the OSS part that doesn't require your services.
It's not at all easy to be successful with Open Source Software,
especially not in commercial markets. The few OSS examples that take off
are peanuts, compared to the broad mass of applications and projects
that exist... just take a look at sourceforge and the masses of crap you
get there between a few pearls.
> If the
> software is really that special
> as you assume, then there would be not much competition in the market
> niche, so that
> giving the source away for free and implement further features on
> request is not much risk.
Apart from the risk that you are likely to do your work for competitors.
They can also implement features on request, based on your Open Source
product ;-).
Eckhard
Eric
> skuh...@web.de schrieb:
>
> ...
> Yeah, but honestly: how many can live from that? To have good services
> is not enough if the product doesn't get attention. And at the point
> when the product gets attention, it is usually so good that services are
> rarely needed.
DRH may have got this right with SQLite
skuh...@web.de
> skuh...@web.de schrieb:
>
> > and some DICOM-Libraries and Viewer. Those are used for medical
> > imaging
> > software, which is IMHO a very uncommon type of software and needs
> > much
> > special and technical know how. I'm sure, there are many other
> > examples.
>
> You are talking about libraries, not about complete turn-key
> applications. Which is what I mean.
Well, then OsiriX in the same field (medical imaging) is what you
mean.
Though it's not a mainstream software, of course.
> > Granting gifts is not necessarily stupid.
>
> No, absolutely not. But you must carefully think about what and also how
> you make these gifts.
This is true. What I doubt is just, that it isn't possible to make a
living from OpenSource
software or successfully run a company with OpenSource. I think, the
problem is that
company founders in that field are often good hackers but bad business
men. There is
a very interesting Article on "Joel on Software" about that ("The
Development Abstraction Layer" -
http://www.joelonsoftware.com/articles/DevelopmentAbstraction.html) I
sure don't know,
if this fits Ousterhout, but the point is, that if he failed with his
OpenSource-Company,
it doesn't mean, there is no way. I think, the problem is often a
business-model, that simply
does not fit into the idea of OpenSource.
> > One may have another
> > business-model in mind, for
> > example selling service to the users of the free software.
>
> Yeah, but honestly: how many can live from that? To have good services
> is not enough if the product doesn't get attention. And at the point
> when the product gets attention, it is usually so good that services are
> rarely needed.
I would say, this is only true for real mainstream software. Of course
very few people,
if any, make a living from offering service to OpenOffice. So for
mainstream software
you may be right. I don't say, that all or the majority of programmers
can live from
OpenSource software, but some do.
> It's not at all easy to be successful with Open Source Software,
> especially not in commercial markets. The few OSS examples that take off
> are peanuts, compared to the broad mass of applications and projects
> that exist... just take a look at sourceforge and the masses of crap you
> get there between a few pearls.
True, but not all of them tried to make money from it.
> Apart from the risk that you are likely to do your work for competitors.
> They can also implement features on request, based on your Open Source
> product ;-).
This is, what competition means: you have to better to win.
Regards
Stephan
Arnt
> And then he is right: as soon as complicated and uncommon technical
> application know how needs to flow into the program, it won't be open
> source. If I knew that I can sell the special niche program that I
> develop for several 10,000 € per license (because the niche market pays
> that)... then I would be stupid if I made it available to the public for
> free.
I think the line lies elsewhere: you (almost*) cannot make money of
being the only writer of an open source program/library/... You can
make money of supporting software written by a large community, and
even help writing it, if a large enough user base exists. But as soon
as it is technical enough and uncommon/niche enough that you will not
find anyone contributing, going open source will bring no benefit, and
open up possibilities for competitors to start from your work,
outpricing you.
* Unless you go the route of mozilla/firefox etc, and install (e.g.) a
default search engine where the searches return some money to you.
Only works if you are a sufficiently large project however, and even
then, this needs to make you enough money to fund developing a large
project. mozilla/firefox certainly need plenty of volunteers anyway.