Re: [linux-dvb] Multiproto API/Driver Update

From: Manu Abraham
Date: Sun Sep 14 2008 - 18:36:46 EST

Andy Walls wrote:
> On Mon, 2008-09-15 at 01:01 +0400, Manu Abraham wrote:
>> Andy,
>> Andy Walls wrote:
>>> Manu,
>>> Though I can't read much German, after looking at the website I
>>> can't help but think that you as well have financial interests driving
>>> your actions. If so, then your statements display quite a bit of
>>> hypocrisy.
>> To your utter disappointment as i should say, i am not working for any
>> vendor, but just get device support out to the community.
> Not to my disappointment. I'm glad to hear it. Someone who appears to
> have an EE background without corporate bias can be an asset to the
> community.
>> The domain is owned by Julian Scheel who runs Jusst
>> Technologies, just happened to offer me hosting for me repositories for
>> my work, using full ssh access, so that my workflow is easier.
>> Not that i have anything to do with otherwise. OTOH, i do have
>> the patches at
>> Maybe Julian can comment on this to make things more clearer on the
>> financial interests.
> Then what I perceived was wrong. My apologies.
>>> Manipulating (i.e. stalling) the timing of Multiproto being merged into
>>> the v4l-dvb tree or kernel, for you or your employer's gain, would be
>>> little different from the motivations you allege Steve of having.
>> I am not manipulating any timing of multiproto being merged. In fact i
>> had been away, for a few months due to certain reasons, that you are
>> perfectly aware by now as far as i can understand.
> I was aware you were away. For what dates I do not know (I have emails
> from you in May 2008). For what reasons, I do not know for sure (nor do
> I feel is it my business).
>> So the points that
>> you raise are quite baseless.
> Not entirely, there is a basis for the timing point. The pull requests
> seemed to have come in short order when confronted with a competing
> proposal. Yet the project had been ongoing for at least over a year (as
> far as I can ascertain). Here's a gripe about delays from Jan 2008:
> There seemed to have been no other visible motivation for the pull
> requests except competition.

I got back on the beginning of September.

>>> Since the major gripe I'm reading on the list "is that multiproto has
>>> taken too long" and since it seems to me the only thing that shook it
>>> loose was a competing proposal, please save the venom for when you
>>> actually have some clear moral high-ground to stand on. I don't see it
>>> from here.
>> Crap, just read above.
> OK, then you do have some high ground. But you also had essentially a
> monopoly position and now you have competition. That is not crap.

Monopoly, competition .. sounds nonsense to me.

>>> As for the technical superiority of either API proposal; that probably
>>> just doesn't matter. I've seen policy/political decisions force
>>> suboptimal technical solutions at work time and time again. If you
>>> really believe you have a superior product technically; then perhaps you
>>> should work to make it superior politically as well. Mud-slinging can't
>>> be a good long term strategy toward that end.
>> I don't have to do any mud-slinging, just wrote the exact facts out here.
> No, you are mud slinging. Let's count the derogatory terms you use in
> addressing your competition in the following quote:
> "No need for you to break the compliant devices in favour of your
> mediocre cards. As i wrote just above, the STB0899 is not the only one
> device using the said features. Also i can guarantee that the CX24116
> (HVR4000) is the most handicapped DVB-S2 device that you are basing the

Conexant themselves mentions what their demodulators can do. (In fact,
they stopped their satellite demodulator businesses and sold it to NXP,
AFAIK) I don't know what you want to add more into it, what Conexant
hasn't. Only basic 8PSK NBC mode of operation. The DVB-S2 specification
and supported devices do a lot more than that.

> DVB-S2 API on: and i can guarantee that what you do will be just be
> broken as you have done for other devices in the past."

> "Also i do not understand, why you have to make a lot of noise to port
> the STB0899 drivers to your crap, when all your cards work as expected
> by you with the multiproto tree. I don't see any reason why the STB0899
> has to be ported to the handicapped API of yours, handicapping the
> STB0899 based devices."

True it is.


To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at