Hi,
I'm just reading about the latest big thing from apple for the iPhone
and found this quite interesting.
http://iphone.akamai.com/solutions.html
"
The segmenter also creates and maintains an index file containing the
list of short media files that were created. These files are placed on a
web server.
A media player built into the iPhone OS is provided a link to the index
file, it then requests the media files in order and plays them without
any pauses or gaps between segments "
Just out of interest do we have anything like that for open source?
Doesn't seem like it would be particularly difficult to build.
Cheers.
--
Patrick Shirkey
Boost Hardware Ltd
Arnold Krille:
> On Sunday 09 August 2009 14:12:56 Kjetil S. Matheussen wrote:
>> No, as I said, the solution is very simple: Don't install a 64 bit OS.
>> That's what's causing your problems, apparently.
>
> Helloo! Please leave the 20th century where it belongs. Nowadays 64bit are
> normal with new computers and its a terrible waste not to use them (reminds me
> of "640kB should be enough for everybody").
>
> If the combination of 64bit and 32bit-binary-only apps is broken, its the
> developers or the distributions task to fix it. And the users task to point out
> the problems and demand the fix.
>
> Looking away by only using 32bits is _not_ the solution. And please stop
> telling that "solution" to other people. We are in the 21th century after
> all...
>
Seems like you have some issues about something.? The case is that
a 32 bit OS works, while a 64 bit OS don't. So until the 64 bit
issues are fixed, using a 32 bit OS is a perfectly fine solution
to the problem. I installed a 32 bit OS because I didn't want
the kind of trouble Patrick has, and others who don't want
that kind trouble should do the same.
Dear linux-audio developers,
I have created New Project https://sourceforge.net/projects/impro-
visor/ for Impro-Visor, which is its correct name. I will populate the
source later today, as I need time to get acquainted with their
system, but I have to be off right now to another important meeting
for the afternoon and can't do it instantly. The source I will post
will be our version 4.0.
Regarding the assertion made earlier by another that I did not contact
sourceforge about the fork 'improvisor', see the forwarded message.
The fact that SF did not remove the other project immediately does not
mean they won't. Not everything happens at lightspeed. It would be a
true indication of courtesy and cooperation if the creator of that
project were simply to remove 'improvisor' as a possible source of
confusion. If not, I will consider the options regarding this action.
Forking at 3.39 would not really be a problem for me, but it seems
that there would be less stress and duplication of effort overall if
we were to proceed as I am suggesting. A year of development has been
put in between 3.39 and 4.0. (Why so long, you ask? For one thing
because of changes started by students, but not integrated, sometimes
take a long time to integrate.)
And yes, I do wish to be in control of my own project, which has been
in existence for four years. It seems reasonable to me. I DO intend to
accept contributions, however.
Thanks for your patience and understanding.
Regards,
Bob Keller
Begin forwarded message:
> From: "SourceForge.net Support" <sfnet_ops(a)corp.sourceforge.com>
> Date: July 27, 2009 9:20:36 AM PDT
> To: Bob.Keller(a)hmc.edu
> Subject: Re: Project: Improvisor has been reported as inappropriate
>
> Hello,
>
> Based on your complaint, it appears that you may wish to report a case
> of copyright infringement. SourceForge.net deals expediently with
> reports of copyright infringement.
>
> To report copyright infringement, please use our DMCA Notification
> Procedure as per
> http://sourceforge.net/apps/trac/sitelegal/wiki/DMCA%20Notification%20Proce…
>
> Regards,
> Chris Tsai
> SourceForge Support
> sfnet_ops(a)corp.sourceforge.com
>
> PS. When you submitted this report, you did not leave us any contact
> info to reply to. As such, I've taken the liberty of looking up your
> e-mail the Harvey Mudd College directory search. I trust this was not
> a problem.
>
Robert Keller
Csilla & Walt Foley Professor
Computer Science
Harvey Mudd College
Chris Cannam
> >> ... but it's probably
> >> illegal and certainly unethical to redistribute someone else's work
> >> without attribution (a basic necessity of copyright which the GPL
> >> doesn't disclaim).
The BSD license originally required attribution. The GPL people objected,
calling it the "obnoxious BSD advertising clause ... and forced this clause
removed.
So it's fair to say the GPL does NOT require attribution.
Jeff McClintock
On Friday 07 August 2009 20:10:14 you wrote:
> On Fri, Aug 7, 2009 at 5:34 PM, Raymond Martin <laseray(a)gmail.com> wrote:
> > Not at all. There is even evidence in the FSF documentation somewhere
> > exactly
> > about this point and they vehemently disagree with any attitude like
> > that. We
> > all know very well the situation of Emacs, Xemacs, and various other
> > forks.
>
> The FSF is not the law. I suggest you look up Trademark Law to realize why
> you are wrong, and why you are subject to a lawsuit for knowingly creating
> a product that is infringing on an already existing trademark(Regeristered
> or Unregeristered would make a small, but only small, difference in this
> case), and can easily be confused as such. In fact the case against
> forcing you to change it is rather strong because not only is your product
> nearly identical in name, it is nearly identical in function and can be
> easily confused with the original.
Another fool. Trademarks apply to commercial interests, the program is
non-commercial in nature. Thus it would be very difficult for anything
to be done about this for creating a free program from a free program.
Maybe you should check the fact that there is already another program called
improvisor that exists that is not Impro-Visor, it is a commercial company
that could claim trademark infringement against Impro-Visor.
So if anybody is in a problem it will be Impro-Visor first.
Raymond
I'm not a developer, just interested in Linux audio development
because I use Linux audio software almost daily, and as such I've been
lurking on this list for awhile. So this is just a practical
suggestion / brainstorming idea, not meant to incur flames (I wish I
could heat my studio by the flames this list generates). One
complaint I've seen raised a number of times is that in the world of
Linux, especially the audio realm, there are too many choices and not
enough decisions made. In the wider scope, The Linux Foundation is
trying to address this by maintaining a standards base to help promote
open standards across mainstream distros... So here's the idea: why
not make one of the responsibilities of the Linux Audio Consortium be
establishing and maintaining a standards base? As far as I can tell
from http://linuxaudio.org/about , it's not currently one of the
Consortium's roles. I.e., there could be a formal document that says
in a nutshell, "If you want your Linux audio application to integrate
seamlessly on most audio-centric distros, here's what it should
support. And, if you want your pro-audio-centric distro to be
seamlessly compatible with all these great apps, here's what it should
include and how it should work." Of course it would be a voluntary
standards base, and every developer / distro team can still do
whatever they like, so that innovation can continue... but as
protocols/interfaces/frameworks/whatever are developed and show their
merit, they can be included in the standards base, and
old/deprecated/redundant things removed, so that people aren't wasting
their time supporting old code. So to sum it up, *someone* needs to
make some tough decisions, or Linux audio will continue to stagnate...
why not the Consortium? It's the only real Linux audio "authority" or
central point of contact that I know of.
I'm sure there are many reasons this is a silly idea -- lack of time,
disagreements on what should be standardized, etc. -- but as I said
it's just an idea. I'd love some constructive criticism, even if it
just leads to a completely different idea getting implemented. By the
way, I apologize that this is basically just another "here's my idea
about what everyone else should do" kind of e-mails. I could maintain
a website and/or provide free hosting for the standards document(s),
if need be.
-Sean Corbett
blacktownsound.com (<- pardon the shameless spam)
On Sat, Aug 8, 2009 at 5:22 AM, James Cameron<quozl(a)us.netrek.org> wrote:
> On Fri, Aug 07, 2009 at 07:44:46PM -0600, Thomas Vecchione wrote:
>> You misunderstood what he was saying. Â The GCC compiler IS included on
>> the installation disks as part of the XCode program. Â You can also
>> download this program off of the internet(The download is over a gig
>> by the way last I looked). Â The complaint is that Apple only bundles
>> gcc with the XCode program and a lot of useless stuff that aren't
>> necessarily used by people like us. Â But they are used by most Mac
>> developers, and Apple I am sure wants to encourage their use and
>> 'their' way of doing things.
>
> Ah, I see, thanks. Â Yes, an encouragement. Â I get the point. Â An open
> ecosystem doesn't need to offer the same encouragement.
Yeah, that was my point. I checked my Debian testing gcc first and
verified that the download is far less than 1 MB.
I believe the latest XCode is over a gig, but my experience wasn't
quite so bad; it was an older ibook, discs long gone, but using an
outdated version that was only 300 some MB, still too much, and
significant given the speed of the comp.
Maybe here's a task for some beneficent soul like me, to compile some
GCC tools on OSX and offer small downloads.
-Chuckk
--
http://www.badmuthahubbard.com
On Fri, Aug 7, 2009 at 7:21 PM, James Cameron <quozl(a)us.netrek.org> wrote:
>
> The installation discs for Mac OS X already contain software licensed
> under the GNU GPL (e.g. bash), so the additional obligations under the
> GPL would not have been the reason to exclude a compiler.
>
> It seems much more likely that it was the size. Several hundred
> megabytes that aren't required by most users should always be omitted
> from the installation discs, so that more room is made for what most
> users require.
>
> The same is done with Ubuntu and Debian installation images. Most of
> the tools needed to build Ubuntu are not included in the installation
> image, and require downloading "apt-get build-dep $packagename".
>
You misunderstood what he was saying. The GCC compiler IS included on the
installation disks as part of the XCode program. You can also download this
program off of the internet(The download is over a gig by the way last I
looked). The complaint is that Apple only bundles gcc with the XCode
program and a lot of useless stuff that aren't necessarily used by people
like us. But they are used by most Mac developers, and Apple I am sure
wants to encourage their use and 'their' way of doing things.
Seablade
David Robillard:
>> Let's just fix the interaction between pulse and jack and be done with
>> it.
>
> That is one solution.
>
>> It's harmful to suggest that it things are less than they are
>
> Read the user posts in this thread, or ths post that started it. Things
> are as they are. Pretending everything is fantastic when it's not
> doesn't help either.
>
Well, in my opinion, everything really is fantastic right now.
Pulseaudio interacts perfectly with jack, pulseaudio doesn't screw
up jack in any way (as far as I can remember observing), and
all programs using jack, alsa or pulseaudio make sound, simultaneously,
without any notable latencies issues with alsa programs. Maybe
I was lucky, and I'm probably not a newbie either, but at least
I'm not _pretending_ everything is fantastic: it really is.
A person coming directly from win/mac probably wouldn't have
succeeded though, since jack needed to be configured and
I had to manually install the pulseaudio jack sink, edit
various files in /etc/, plus perhaps doing other stuff
I don't remember anymore.
standard ianal disclaimer applies.
at the risk of starting another zillion-mail thread, here's how i
understand the gpl to work under german laws (which should be almost the
same in most other countries):
a) there is a difference between the license and the copyright. what
many people fail to grasp is that the original copyright holder (if it's
actually a single person or closed group) is not bound by the license -
s/he can choose to distribute the product under any terms s/he sees fit.
the license applies only to people re-distributing the product.
so it does not make sense to accuse an original author of not providing
makefiles, scripts etc, for original code. specifically so in the light
of the famous "AS-IS", "WITHOUT FITNESS FOR ANY PARTICULAR PURPOSE"
paragraph.
ripping out makefiles from re-distributed code is another issue.
b) there is a difference between license and evangelist lingo and the
actual *contract* that is agreed upon by licensor and licensee.
the GPL cannot be viral in the sense that any code that incorporates GPL
automatically becomes GPL as well.
what happens instead is this:
* you use GPL code in your product so as to make your product a "derived
work".
* by including the GPL code, you are accepting the GPL (as stated in the
license). the original author has issued a standing offer for a contract
(the GPL), and your use of the code indicates your acceptance. thus, a
contract is formed between the original author of that code and yourself
(through "konkludentes handeln" or "implied agreement by action" in
german legalese, no signing of papers or verbal exchange needed here).
* you fail to comply with the GPL by not releasing your product under
the same terms.
* this means you are violating the license of the included GPL code and
are re-distributing the included GPL code without permission.
so this is a *breach of contract*, and anything that follows is about
the original author's copyright, not your own, or how you license your code.
you are now liable to a copyright infringement suit, a cease and desist
and possible damages.
from this point on, there are two ways to remedy the situation:
* you can stick to the original contract and comply with the GPL in the
future. since this happens to mean your code has to become GPL as well,
it is what causes the myth of the "viral" GPL.
* you can rip out the GPL code and continue to use your old licensing
terms (after you have settled the infringement suit).
all this means that if somebody uses GPL'd code without releasing under
GPL him/herself, s/he is in breach of contract with the original
author(s). it *does* *not* *mean* that you can now assume the entire
package is up for grabs under the terms of the GPL.
best,
jörn
