[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Discussion - non-free software removal



Steve Langasek <vorlon@netexpress.net> writes:

> On Thu, Nov 14, 2002 at 11:35:28AM +0100, Mario Lang wrote:
>
>> Well, I partly agree with you, but I dont think kicking
>> non-free as a whole is a good idea.  I'd really love
>> if every piece of software I need would be free, and in main,
>> but there are exceptions.  See #167189 for an example.
>> Mbrola is the ONLY speech synth available right now
>> which supports substantially more languages then just
>> english.  This is a very important thing for
>> allowing visually impaired computer users to work with
>> Linux.  I'd be happy if they'd release sources of course,
>> but it appears they are limited by a contract they
>> have with another company, so it seems we dont have
>> any chance to free mbrola.  So as long as no one
>> develops a diphone synthesizer which comes
>> with 25 different langauges under the GPL, I'd really like
>> to see non-free stay.
>
> I'm no expert on speech synth technologies, but I know 'festival' is a
> speech synthesizer that does support more than one language (though not
> yet 25, AFAIK).  Is festival feature-poor or poor-quality compared to
> mbrola, or is it just a matter of implementing the other languages?

It is partly a matter of both unfortunately.  Festival itself is very
slow, and therefore in reality absolutely unusable for a screen reader.
`flite' is an attempt to speed things up, but flite only supports one voice
right now, and that is very poor quality.

On the other hand, it's not "just a matter of implementing the other languages",
that is actually alot of work, and from what happend till now I get the
impression that either the free available tools to create such "voices" and "languages"
are not really available, or it is very hard to use them or time
intensiv, because I didnt see a single mail like "hey, I've just
created this new german voice+lexical rules for festival..." till now.

Not to say I prefer a non-free solution here, I'd much prefer to see
more DFSG-compliant development in the area of speech synthesis, but
unfortunately, that is currently just not the case.

>> > Do we serve our users by making it easier for them to become dependent
>> > on new forms of non-free software?
>
>> I agreed to the social contract, as you did, and I remember
>> clearly when reading no.5 I thought to myself:
>> "yes, that is a good thing to have."
>
> I also agreed to the Social Contract in its entirety.  However,
> acknowledging that our users sometimes need non-free software, does not
> automatically lead to the conclusion that continuing to add more non-free
> software to our archive is in our users' best interest.

I really don't get your last sentence.  I've read it several times, and it
makes no sense to me.  How can discontinuing to add certain packages
to non-free be in our users' best interest?  (besides, the actual
request was to remove all non-free packages!)  I mean, come on,
non-free already exists.  I'd understand the level of resistance toward non-free
if we were discussing the new creation of the non-free area.

Let's actively try to replace non-free packages by programs suitable
for main, and file RC bugs against non-free packages which can be
more-or-less completely replaced by main packages.

-- 
CYa,
  Mario | Debian Developer <URL:http://debian.org/>
        | Get my public key via finger mlang@db.debian.org
        | 1024D/7FC1A0854909BCCDBE6C102DDFFC022A6B113E44



Reply to: