opinion on Willem programmer?

[Xiaofan Chen]

On Sun, May 11, 2008 at 6:13 AM, Mark Rages <markrages@...> wrote:

> On Sat, May 10, 2008 at 3:58 PM, William Chops Westfield <westfw@...> wrote:
>>
>> On May 10, 2008, at 11:40 AM, Mark Rages wrote:
>>> Can you explain how the GPL does not allow you to change
>>> the software to suit your needs?
>>
>> "My needs include being able to keep my modifications to the open
>> source material proprietary when I distribute the software."  "I need
>> to keep distribution difficulties in check, and cannot afford to
>> distribute my embedded product in a form where it can be re-built
>> with modified versions of the open-source libraries that it uses."
>>
>> GPL advocates may claim that these are not VALID needs, or argue
>> about the difficulty of meeting them within GPL, but it tends to get
>> my hackles up whenever someone else is ready to tell me what *I* need.
>>
>
> The GPL is concerned with the software users' rights, not progammers.
>

Ok, let's talk about the end user. On the PC software side, I think
GPL is quite good. The end users are in general happy. The
programmers are in general happy. GPL V3 changed a bit so that
some end user are happier and some are less happy (eg: TiVo).

My main problem with GPL/LGPL comes when I consider the use of
embedded software in a MCU. This has been discussed before and
even BAJ think this GPL/LGPL do not suit that well. In that end, you
can not incorporate GPLed software easily into your project without
exposing your own codes unless you only use your device in-house.
So you, as an end user, can not change the GPLed software to suit
your needs in a way. So in the end you do not use it even though
technically it may be quite suitable.

So in the end, you do not really see GPLed firmware pieces in
small MCUs like PIC. You do see that embedded Linux being
used by bigger MCUs where it is easier to seperate the kernel
space and user space. Even there, the end user need to be
very careful about seperating the kernel and the user codes.

By the way, to me GPL is a very good ballancing protecting
the rights of the developer and the end user. I think it is not
true that it only concerns the end user.

For example, Linux kernel developers have a weapon to
wipe out the possibility of in-kernel proprietory driver by export
EXPORT_SYMBOL_GPL symbol.


Xiaofan
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Xiaofan Chen]

On Sun, May 11, 2008 at 9:35 AM, Xiaofan Chen <xiaofanc@...> wrote:
> So in the end, you do not really see GPLed firmware pieces in
> small MCUs like PIC. You do see that embedded Linux being
> used by bigger MCUs where it is easier to seperate the kernel
> space and user space. Even there, the end user need to be
> very careful about seperating the kernel and the user codes.
>

But GPL with an exception clause can be used.

For example, GPL+LE - (GPL plus Library/Runtime Exception)
is considered good by many sdcc developers.
http://sdcc.sourceforge.net/wiki/index.php?page=Library+License+Selection


Xiaofan
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Paul Hutchinson]

> -----Original Message-----
> From: piclist-bounces@... On Behalf Of Olin Lathrop
> Sent: Saturday, May 10, 2008 8:36 AM
>
> Byron Jeff wrote:
> > It's the none at all that's the real problem. Without a license, no
> > one has any right to do anything with the source for your software.
>
> Are you really sure?  I am not a lawyer, but if there are no restrictions
> listed then it seems to me there are no restrictions.  Many things on the
> internet have no explicit copyright and are routinely copied.  I don't
> remember anyone claiming that something without a explicit copyright was
> illegally copied.

IANAL either but, my employer's IP legal firm went over it in detail with
me. This was back in 1991, the first time we wanted copyright protection. We
had patents and trademarks but hadn't needed copyrights until then. Here's
what I learned from the IP lawyers.

Without an explicit license all works automatically receive complete
copyright protection. This even applies to a finger painting by a three year
old child. The moment the child creates the finger painting it acquires full
copyright protection that lasts until 70 years after the death of the child.

This is not how the laws worked in the USA prior to 1978 when I entered the
work force. Back then the default status was public domain, you had to
explicitly request copyright protection if you wanted it. Starting in 1978
the default changed to full copyright protection which harmonized the US
laws with most of the rest of the world. In 1989 the requirement of placing
identifying marks on a work was removed so, anything distributed after 1989
receives full protection without even identifying marks. AFAIK, no-one has
yet won a copyright suit without having identifying marks and you must
explicitly file with the copyright office if you want to get statutory
damages from an infringement lawsuit.

So, if you copy something from the internet that does not have an explicit
public domain release statement or a license granting you rights, it is
likely a violation of copyright law to distribute it in any way. If the work
you copied and distributed has no identifying mark you won't be sued for
damages but you can be made to stop distributing it. There is a concept of
fair use but it is an inadequately defined legal concept generally subject
to interpretation by the courts on a case by case basis.

A few reference pages on US copyright:
http://www.templetons.com/brad/copyright.html
http://www.templetons.com/brad/copymyths.html
This is Brad Templeton's personal site, he's Chairman of the Board of the
EFF.

http://www.copyright.gov/
http://library.findlaw.com/1999/Jan/1/241476.html
http://www.eff.org/issues/intellectual-property
http://creativecommons.org/

Paul Hutch
Proud EFF Member, join us at http://www.eff.org/

<snip>

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[William "Chops" Westfield]

On May 10, 2008, at 1:32 PM, Byron Jeff wrote:
>> "My needs include being able to keep my modifications to the open
>> source material proprietary when I distribute the software."
>
> I don't think there's any viable resolution to this one. The right  
> answer
> is that if you need to keep your software proprietary, then you  
> need to
> write is all yourself.

Nonsense.  I just need software with a different license.  There are  
open source licenses that allow this, and there are LOTS of people  
selling proprietary software libraries that allow me to re-distribute  
their code in MY proprietary product.  Usually it's pretty expensive,  
and sometimes the license terms are unattractive in other ways  
(royalties can be a real pain to calculate.)   But these alternatives  
DO exist.


> The GPL and to a lesser extent the LGPL are licenses targeted to  
> protect
> the end user, not the developer. They are both really unabashed  
> about that.


I have trouble thinking of a developer as anything other than a  
"user" of a library-like piece of software.  I guess LGPL tries to  
make that distinction, and doesn't do TOO badly except for embedded  
systems.


> The problem is that if you change the license significantly, and  
> end users
> get screwed, with very limited access to source, and little or no  
> ability
> to use that source in any productive way.


This assumes "end users" and "software" such that source is useful to  
the end user.  In particular, it assumes end users than are  
programmmers.  See the comment earlier about Olin's "open source" not  
being too useful cause it was too hard to do anything with anyway  
(being written in odd-looking assembly language, after all.)  And in  
many cases, the end user is "less screwed" by needing to pay for a  
supported proprietary piece of software than they would be by needing  
to support free source software themselves.  It is nearly insulting  
to equate "needs to pay for something" with "screwed." Perhaps it  
made more sense when so many "computer users" were starving college  
students.

BillW

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Olin Lathrop]

Byron Jeff wrote:
> The problem is that if you change the license significantly, and end
> users get screwed, with very limited access to source, and little or
> no ability to use that source in any productive way.

Now you're arguing what a good license should be.  "Good" depends on your
goals and measures.  The FSF wants to see executable software with no
restrictions.  That can be useful, but they ignore or usually not even
acknowledge that enforcing that decreases the choices of the end users.

If you really wanted to do something "good" for the end users then I think
using the GPL is not the right approach.  The GPL does force source of any
software that is derived from it to be open.  However to many that
restriction is too costly, so they don't use GPL code.  In the end the goal
of better and lower cost choices for the end users has not been served as
well as it could have been.

I was going to stop replying on this thread, but what made me respond was
your implicit statement that not having source to a program means you're
screwed.  This is of course totally rediculous.  A very tiny minority of end
users might derive some additional advantage from having access to the
source code for a app they are using, but the vast majority wouldn't know
what to do with it, and most of the ones that do have things to get on with
and don't want to bother messing with it.  Source code accessibility is way
overrated.

If you really want to make things as good as possible for end users, you
need to allow people to make a buck by doing so.  Most developers just don't
have the luxury like Richard Stallman has of forgoing compensation for
creating software.  If you want to harness their power, then you need to let
them make a buck in the process.  I'd much rather see five commercial apps
for a given task than one free one created by university students and
midnight hackers.  The competition between the five commercial apps will
keep them on their toes, and because they can afford to pay software
developers, there are far more people working on creating the software you
want.  Of course none of this prevent those who want to do it on the side
for free from doing it.  If the commercial apps are reasonably priced and
serve the need well, then there is no incentive to create a free app and no
need anyway.

Developing software costs resources.  Ultimately the end users of the
software need to pay for those resources somehow.  In Richard Stallman's
model, it is difficult to force the end users to pay, so there is less
development with GPL code because the people doing the developing can't
afford it.

If someone really wanted to take the moral high ground and try to provide
the best possible situation for end users of software, he'd let people use
his code any way they want.  This is basically what I try to do with my
source except in cases where I fear it will cost me compared to not making
it free.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
[phone].  Gold level PIC consultants since 2000.
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Olin Lathrop]

Byron Jeff wrote:
> The type of freedom that you keep referring to takes away rights from
> the
> end user. If you have the right to make modifications, distribute the
> product but not the source, then every user that you distribute to
> lose the ability to make further changes and share those changes.
>
> Zero sum game.

Not at all.  The two don't add up to zero by a long shot as I see it.  See
my previous message.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
[phone].  Gold level PIC consultants since 2000.
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Olin Lathrop]

Paul Hutchinson wrote:
> So, if you copy something from the internet that does not have an
> explicit public domain release statement or a license granting you
> rights, it is likely a violation of copyright law to distribute it in
> any way. If the work you copied and distributed has no identifying
> mark you won't be sued for damages but you can be made to stop
> distributing it. There is a concept of fair use but it is an
> inadequately defined legal concept generally subject to
> interpretation by the courts on a case by case basis.

So does that mean the various sites that carry copies of the PIClist traffic
are in violation of the copyrights of the individual posters, of the
"PIClist" whatever that is, of MIT?  These sites to real harm because they
make it easy to mine for email addresses, whereas James' site tries to be
careful about that.  I don't remember agreeing to give up my copyrights to
anything I write here, so who owns the rights to messages I post?


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
[phone].  Gold level PIC consultants since 2000.
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Gerhard Fiedler]

Mark Rages wrote:

> You are talking about distribution.  You are still free to change the
> GPL software to suit your needs.  You are required to provide source
> code only if you distribute derived works.

Isn't distribution a frequent need for those who try to live from writing
software? :)

Gerhard

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Gerhard Fiedler]

William "Chops" Westfield wrote:

>> The problem is that if you change the license significantly, and end
>> users get screwed, with very limited access to source, and little or no
>> ability to use that source in any productive way.
>
> This assumes "end users" and "software" such that source is useful to
> the end user.  In particular, it assumes end users than are
> programmmers.  

In a way, they always are or can hire some. The problem is when they can't
do that to modify the sources, they are at the mercy of the one who wrote
the software. I am still using CodeWright, a quite nice universal code
editor originally written by Premia. Over the years it went through some
hands, to end up with Borland and being scrapped by them a while ago. Now
basically all users are "screwed"; if it were an open source product with
the appropriate license, they wouldn't be. (This is a programmer's editor,
but it could be anything else, not related to programming.) It still works
for me, but the effect of my experience with it is that when I switch, I'm
very likely switching to an open source editor -- even if I should find a
commercial one that's more attractive. It may be more attractive now, but I
may get "screwed" again without recourse.

> And in many cases, the end user is "less screwed" by needing to pay for a
> supported proprietary piece of software than they would be by needing to
> support free source software themselves.  

Of course. But then that user wouldn't have to worry about open source
licenses, so the "screwed" WRT that doesn't really apply to them. They can
get "screwed" in other forms (like the users of CodeWright).

> It is nearly insulting to equate "needs to pay for something" with
> "screwed."

I think the "screwed" means that the chain of being able to make use of
derivated works of the original software is broken. Say there is an open
source web server. Someone makes some changes to it and doesn't distribute
the source code. Now say that changed server becomes some kind of standard
-- the users of that server can't change that server to fit their needs,
not even the authors of the original software; the chain has been broken,
which in this sense may be called "they're screwed" by some :)  To me it
seems it is the maintenance of this chain that's the objective for much of
the open source movement.

Gerhard

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Xiaofan Chen]

On Sun, May 11, 2008 at 10:57 PM, Gerhard Fiedler
<lists@...> wrote:

> > And in many cases, the end user is "less screwed" by needing to pay for a
> > supported proprietary piece of software than they would be by needing to
> > support free source software themselves.
>
> Of course. But then that user wouldn't have to worry about open source
> licenses, so the "screwed" WRT that doesn't really apply to them. They can
> get "screwed" in other forms (like the users of CodeWright).
>
> > It is nearly insulting to equate "needs to pay for something" with
> > "screwed."

Take note often you still need to pay for open source software
companies for support or packaging. Redhat and Novell charge
quite high cost for their client Linux workstation software.

So paying for something is not the discussion topic here.

> I think the "screwed" means that the chain of being able to make use of
> derivated works of the original software is broken. Say there is an open
> source web server. Someone makes some changes to it and doesn't distribute
> the source code. Now say that changed server becomes some kind of standard
> -- the users of that server can't change that server to fit their needs,
> not even the authors of the original software; the chain has been broken,
> which in this sense may be called "they're screwed" by some :)

> To me it seems it is the maintenance of this chain that's the objective
> for much of the open source movement.

That is a good objective. Availability of source code (no need to be using
an official endorsed open source license)  does have an edge here.

Example (post 34 and 36)
http://forum.microchip.com/tm.aspx?m=336190&mpage=2

Even though AVIX seems to be a good RTOS, I will think twice
using it since it is from a small company and no source codes
are available.

Regards,
Xiaofan
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Byron Jeff]

On Sun, May 11, 2008 at 09:22:23AM -0400, Olin Lathrop wrote:
> Byron Jeff wrote:
> > The problem is that if you change the license significantly, and end
> > users get screwed, with very limited access to source, and little or
> > no ability to use that source in any productive way.
>
> Now you're arguing what a good license should be.  "Good" depends on your
> goals and measures.  The FSF wants to see executable software with no
> restrictions.  That can be useful, but they ignore or usually not even
> acknowledge that enforcing that decreases the choices of the end users.

No I'm arguing that licensing is a zero sum game when it comes to rights.
Every right that you grant to one group of folks in a transaction takes
away a right from another group.

> If you really wanted to do something "good" for the end users then I think
> using the GPL is not the right approach.  The GPL does force source of any
> software that is derived from it to be open.  However to many that
> restriction is too costly, so they don't use GPL code.  In the end the goal
> of better and lower cost choices for the end users has not been served as
> well as it could have been.

I'm usually not a fan of the GPL precisely for this reason. The GPL works
OK when you're talking about an application, but it fails miserably on
infrastructure code (libraries, OSes, etc) for precisely the reasons you
outline.

In general the LGPL is closer to the right mark for infrastructure. It
mandates that changes to the infrastructure have to be propagated, but that
use of the infrastructure is unrestricted.

It's only failure is in embedded systems where application and
infrastructure are comingled as it reverts back to the GPL in that
instance.

> I was going to stop replying on this thread, but what made me respond was
> your implicit statement that not having source to a program means you're
> screwed.  This is of course totally rediculous.  A very tiny minority of end
> users might derive some additional advantage from having access to the
> source code for a app they are using, but the vast majority wouldn't know
> what to do with it, and most of the ones that do have things to get on with
> and don't want to bother messing with it.  Source code accessibility is way
> overrated.

It's way overrated until the vendor disappears and something is broken. Or
when something is broken and the vendor won't fix it. Or when you need a
feature that the vendor isn't interested in implementing. While there may
only be a small segment of the user community that falls into those
situations, without the source, they have no recourse except to start over
with another vendor, or to pony up dollars to try to entice the vendor to
help them solve their problem, if they are still around to do it.

Everyone isn't you Olin. EmbedInc is clearly a competent, well run,
customer oriented company. But unfortunately for every one of you, there
are a ton of folks who simply want to get the end users' money and walk
away.

And my original comment was directed to software that started its life
marked as open source. I'm a big believer in the fact if you wrote the
software, you can do anything you like with it. The issue always arises
when the two cultures clash, because always in the middle of that clash is
a developer that wants to take software that has an other or community that
wants it freely available and wants to somehow encumber it.

That's the change in license I'm talking about. And that's when the end
user get screwed.

>
> If you really want to make things as good as possible for end users, you
> need to allow people to make a buck by doing so.  Most developers just don't
> have the luxury like Richard Stallman has of forgoing compensation for
> creating software.  If you want to harness their power, then you need to let
> them make a buck in the process.  I'd much rather see five commercial apps
> for a given task than one free one created by university students and
> midnight hackers.  The competition between the five commercial apps will
> keep them on their toes, and because they can afford to pay software
> developers, there are far more people working on creating the software you
> want.  Of course none of this prevent those who want to do it on the side
> for free from doing it.  If the commercial apps are reasonably priced and
> serve the need well, then there is no incentive to create a free app and no
> need anyway.

You may be right. But that means I have to trust those developers to meet
my needs. And while the vast majority of the population has no choice in
the matter, there are enough of us out there to fix and augment what we
need if necessary, without having to continually pay for the priviege.

And the commercial apps are rarely reasonably priced. The happens because
of the reverse flow of screwing. End users will simply steal the software
that they need and not pay for it. It's wrong. It's awful. It's unfair. And
it happens every day. So to compensate commercial developers jack up the
price of the legit sales to cover for the stolen copies.

> Developing software costs resources.  Ultimately the end users of the
> software need to pay for those resources somehow.  In Richard Stallman's
> model, it is difficult to force the end users to pay, so there is less
> development with GPL code because the people doing the developing can't
> afford it.

Actually what happens is that the development is spread over a larger pool
of developers each of whom contribute less than full effort to the project.

What you miss is that these developers will develop the software regardless
of being paid or not, because the software meets their own needs. By
sharing the source, they can share the work required to get the development
off the ground. It works because the genesis of the software was never
based in money, it's based in need.

> If someone really wanted to take the moral high ground and try to provide
> the best possible situation for end users of software, he'd let people use
> his code any way they want.  This is basically what I try to do with my
> source except in cases where I fear it will cost me compared to not making
> it free.

The problem with your high moral ground is that virtually instantaneously
someone unscrupulous will take your code and embrace, extend, and suppress
the modification for their own benefit. What better way to make a profit
than to take something you get for free, value add, then sell it through
artificial scarcity? Then no one but them gets the benefit of the
modifications.

That's why the LGPL is nearly the correct middle ground. Any developer can
use LGPL infrastructure to build new stuff. They code they add to the
infrastructure is shared with everyone downstream, but code that just uses
the infrastructure does not have to be. That's equity for both the
developer who contributes the LGPL code, and the developer that uses the
LGPL code to build their own projects.

There's no right answer in a philosphy discussion, simply points of view.
If users trusts the developers, then simply buying the product is the right
way to go. However, there are several levels of costs associated with that
decision. Same with using open source infrastructure, including the quality
of code issue you raised above.

BAJ
>
>
> ********************************************************************
> Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
> [phone].  Gold level PIC consultants since 2000.
> --
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Byron Jeff]

On Sun, May 11, 2008 at 09:33:07AM -0400, Olin Lathrop wrote:

> Paul Hutchinson wrote:
> > So, if you copy something from the internet that does not have an
> > explicit public domain release statement or a license granting you
> > rights, it is likely a violation of copyright law to distribute it in
> > any way. If the work you copied and distributed has no identifying
> > mark you won't be sued for damages but you can be made to stop
> > distributing it. There is a concept of fair use but it is an
> > inadequately defined legal concept generally subject to
> > interpretation by the courts on a case by case basis.
>
> So does that mean the various sites that carry copies of the PIClist traffic
> are in violation of the copyrights of the individual posters, of the
> "PIClist" whatever that is, of MIT?

Yup. Every post you write is copyrighted. I probably could get away with
fair use because I'm commenting on your post. But technically unauthorized
copies are a copyright violation.

>  These sites to real harm because they
> make it easy to mine for email addresses, whereas James' site tries to be
> careful about that.  I don't remember agreeing to give up my copyrights to
> anything I write here, so who owns the rights to messages I post?

You do. Say for the sake of argument that James wanted to publish a PicList
Book with selected writings from all of us. He'd have to go negotiate
copyright assignments for that material from each of us. Otherwise we could
sue for copyright infringement and damages.

BAJ
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Paul Hutchinson]

> -----Original Message-----
> From: piclist-bounces@... On Behalf Of Olin Lathrop
> Sent: Sunday, May 11, 2008 9:33 AM
>
> Paul Hutchinson wrote:
> > So, if you copy something from the internet that does not have an
<snip>
>
> So does that mean the various sites that carry copies of the
> PIClist traffic are in violation of the copyrights of the
> individual posters, of the "PIClist" whatever that is, of MIT?

PICLIST and MIT can not claim copyright on an individual's posts so, no for
them.

For individual posters, it is possibly a copyright violation. Courts have
not had an opportunity to rule on this issue. Currently there are two
popular schools of thought on this. One is that by posting you are
implicitly granting a license with broad copying and redistribution rights.
The other is that you are commanding, not consenting, to create a number of
copies without an implied license for further copying. Until a court has an
opportunity to rule on this issue it is uncertain whether or not this would
qualify as a violation of copyright. In addition to this, there is the whole
issue of fair use rights that also have not been ruled on in this context in
the courts.

> These sites to real harm because they make it easy to mine for
> email addresses, whereas James' site tries to be careful about
> that. I don't remember agreeing to give up my copyrights to
> anything I write here, so who owns the rights to messages I
> post?

You absolutely retain the copyright on all messages you post that do not
explicitly place the information into the public domain.

AFAIK, trying to use copyright as a way to stop some of the bad archives has
not been brought to a court. It is possible that this strategy could work
but there are a number of large hurdles to overcome. An email address, or
any other non-creative work, is not covered under copyright so even if you
pursued this there is no way to stop them from publishing the email address,
you could only get the message content removed under copyright law.

A major hurdle that has prevented this issue from ending up in court for a
ruling is the cost to bring this to court. If the author did not receive
copyright registration on their post from the copyright office before
posting they can not receive statutory damages or reimbursement of legal
costs. So, even if they win they are not guaranteed to receive money to
cover the expenses. The poster could receive some money if they can prove to
the court that the copyright infringement caused a loss of revenue from the
copyrighted work. In the context of email list and usenet archives this
would be nearly impossible to prove in a court. This leaves the only remedy
from the court of having the post but not the email address and other
non-creative portions of the post removed from the archive.

Paul Hutch


--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Herbert Graf-2]

On Sat, 2008-05-10 at 15:01 -0500, Mark Rages wrote:

> On Sat, May 10, 2008 at 1:54 PM, Herbert Graf <mailinglist4@...> wrote:
> > On Sat, 2008-05-10 at 13:40 -0500, Mark Rages wrote:
> >> Can you explain how the GPL does not allow you to change the software
> >> to suit your needs?
> >
> > I suppose it depends what your needs are.
> >
> > GPL has often been considered very "virus" like, it's presence in your
> > project effects much more then just the chunk of GPL code you're using.
> >
> > If your needs are keeping your projects' code closed (i.e. a very common
> > situation is NDA agreements restricting how open you can be with your
> > code), there are very few ways you can make use of any GPL code.
> >
> > TTYL
>
> You are talking about distribution.  You are still free to change the
> GPL software to suit your needs.  You are required to provide source
> code only if you distribute derived works.

Umm, ya. Basically if you want to distribute/sell your product, and it
has ANY GPL code in it, you have to release the source. For many this is
a perfect example of not being able to change the source to suit your
needs.

FWIW, I'm a BIG fan of open source, but I've NEVER liked the GPL, that's
just my personal opinion though.

TTYL

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Herbert Graf-2]

On Sat, 2008-05-10 at 20:32 +0000, Byron Jeff wrote:

> On Sat, May 10, 2008 at 04:58:33PM -0400, William Chops Westfield wrote:
> >
> > On May 10, 2008, at 11:40 AM, Mark Rages wrote:
> > > Can you explain how the GPL does not allow you to change
> > > the software to suit your needs?
> >
> > "My needs include being able to keep my modifications to the open
> > source material proprietary when I distribute the software."
>
> Yup. That's a tough one. It's real difficult to balance that against the
> fact that a large base of incoming software was freely given you, yet you
> want to restrict the outflow of changes to others.

More often then not the person DOESN'T want to restrict the outflow, but
they are forced to by NDAs.

Perfect example of this is many chips out there require an NDA to access
the programming spec, and in many specific areas there are NO chips that
don't require an NDA. As a result, ALL code that uses information from
that spec MUST remain closed. So basically, if you want to create a
product in a certain area, you CAN'T use GPL at all, and I find this to
be a shame.

FWIW I've had to pass on GPL code because of these sorts of issues.

TTYL
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist