[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-video
Subject:    Re: [video4linux] New API spec
From:       "Bill Dirks" <dirks () rendition ! com>
Date:       1998-07-26 1:25:45
[Download RAW message or body]

-----Original Message-----
From: Alan Cox <alan@cymru.net>

>> Alan, is there a deadline for when we should have the new spec
hammered
>> down?
>
>A week or so would be nice. Im hoping we dont change anything
drastically
>and cause breakages.


I think we can make some very worthwhile improvements, but it will cause
breakages. But they will be easy to fix.

We can keep the old bttv spec in place separate from Video for Linux if
you need the legacy bttv stuff to continue working.

I posted an early draft of some proposed changes to the spec under the
subject "New API spec proposals". It includes, for example, an improved
video image format structure. Obviously adopting that will break
existing code, but, jeez, it's not hard to adapt code to the new
structure!

Most of my proposed new or changed data structures have 'reserved'
fields in them, so in the future changes can be made without breaking
anything. So part of the proposal is about avoiding this very delimma.
In the long run it's worth it to make these kinds of changes before tons
of applications have been written/ported.

There will never be better time to fix the API. Now is our chance. I
know it's painful to have to modify working code. I have to modify my
driver too. It's not fair to the people who will be writing (and
porting) v4l applications and drivers in the future for us to ship a
spec that we know has problems we can easily fix but don't because we
don't want to break a couple early programs. I think everybody
understands that the current spec is experimental and subject to change.

I would really like to see Video for Linux be a solid, well-thought-out,
high quality foundation for all kinds of video applications. Video
applications on personal computers have been novelty or niche items in
the past, but I think they will become more and more commonplace and
important in the near future. There are many applications that could be
applied to Linux if the foundation were there. Linux is highly respected
for many kinds of applications; it should be true for digital video too.

Have a look at the proposals. If everyone agrees the proposals are a bad
idea, I'll drop it. Or else we can work out something we all agree on.
Keep in mind that *if* the initial spec is well designed and
sufficiently extensible we can add more features and refinements later
without breaking code anymore. And that will mean we don't have to solve
every problem this week.

Bill.

------------
To unsubscribe from this list send mail to majordomo@phunk.org with the
line "unsubscribe video4linux" without the quotes in the body of the
message.

[prev in list] [next in list] [prev in thread] [next in thread]