Wacom based devices and the mt kernel protocol.

Wacom based devices and the mt kernel protocol.

[Mohamed Ikbel Boulabiar]

Hi,

Now we can find many multitouch devices based on wacom Wacom panel/digitizer
Bamboo Pen&Touch, HP tm2, Fujitsu LifeBook T4410 & T4310, etc...

it seems that these devices are supported in some way in Linux Kernel,
but I want to know if they are compliant with the MT Linux protocol.
N-Trig based devices and Stantum ones are the most compliant, so Wacom
situation needs to be clarified specially considering they are HID
devices (meaning similar to the others devices which respect the
protocol). This impact very much the decision of a Linux user wanting
to buy wacom mt device.

Ping can you clarifie the situation of wacom based devices ?

Thanks,

ik.

On Tue, Jun 15, 2010 at 7:01 PM, Henrik Rydberg <rydberg@euromail.se> wrote:
>
> g number of intelligent multi-contact and
> multi-user devices, the need to send digested, filtered information
> from a set of different sources within the same device is imminent.
> This patch adds the concept of slots to the MT protocol. The slots
> enumerate a set of identified sources, such that all MT events

Re: Wacom based devices and the mt kernel protocol.

[Ping Cheng]

Hi Mohamed,

I am glad that you've initiated this email thread. Otherwise, I would
have to find a proper way to explain the current status of Wacom MT
support. Please see my comments inline.

Thank you,

Ping

On Fri, Jun 18, 2010 at 4:20 AM, Mohamed Ikbel Boulabiar
<boulabiar@gmail.com> wrote:
> Hi,
>
> Now we can find many multitouch devices based on wacom Wacom panel/digitizer
> Bamboo Pen&Touch, HP tm2, Fujitsu LifeBook T4410 & T4310, etc...
>
> it seems that these devices are supported in some way in Linux Kernel,
> but I want to know if they are compliant with the MT Linux protocol.

As most of you know, the current Wacom two finger touch devices have
been supported on Linux since last Oct.  However, they are not
supported with the current in-kernel MT protocol, aka, MT type A
protocol.

I know you would ask me why. The reason is simple: it doesn't support
the Wacom case - finger tracking.  If you wonder why I don't work with
the other developers to support Wacom MT case, I can assure you that I
am working with them. We will support Wacom MT devices once the
appropriate protocol is available.

> N-Trig based devices and Stantum ones are the most compliant,

That's fair since they don't need to track the fingers.

> so Wacom situation needs to be clarified specially considering they are HID
> devices (meaning similar to the others devices which respect the
> protocol).

Well, it is not a pure HID issue.  It is more of how we want to
support MT on Linux issue.

> This impact very much the decision of a Linux user wanting to buy wacom mt device.

I think end users have already got the Linux support from Wacom.  Can
you share the specific issues that are preventing Linux end users from
buying a Wacom MT devices? If I didn't misunderstand your point here,
you are talking about Linux users, not Linux developers/hackers,
right?

> Ping can you clarifie the situation of wacom based devices ?

What else do you need me to clarify? I am all ears and I am ready to
clarify any thing that puzzles you.

Ping

> Thanks,
>
> ik.
>
> On Tue, Jun 15, 2010 at 7:01 PM, Henrik Rydberg <rydberg@euromail.se> wrote:
>>
>> g number of intelligent multi-contact and
>> multi-user devices, the need to send digested, filtered information
>> from a set of different sources within the same device is imminent.
>> This patch adds the concept of slots to the MT protocol. The slots
>> enumerate a set of identified sources, such that all MT events
> --
> To unsubscribe from this list: send the line "unsubscribe linux-input" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
--
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Wacom based devices and the mt kernel protocol.

[Rafi Rubin]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 06/20/10 01:35, Ping Cheng wrote:
> Hi Mohamed,
> 
> I am glad that you've initiated this email thread. Otherwise, I would
> have to find a proper way to explain the current status of Wacom MT
> support. Please see my comments inline.
> 
> Thank you,
> 
> Ping
> 
> On Fri, Jun 18, 2010 at 4:20 AM, Mohamed Ikbel Boulabiar
> <boulabiar@gmail.com> wrote:
>> Hi,
>>
>> Now we can find many multitouch devices based on wacom Wacom panel/digitizer
>> Bamboo Pen&Touch, HP tm2, Fujitsu LifeBook T4410 & T4310, etc...
>>
>> it seems that these devices are supported in some way in Linux Kernel,
>> but I want to know if they are compliant with the MT Linux protocol.
> 
> As most of you know, the current Wacom two finger touch devices have
> been supported on Linux since last Oct.  However, they are not
> supported with the current in-kernel MT protocol, aka, MT type A
> protocol.
> 
> I know you would ask me why. The reason is simple: it doesn't support
> the Wacom case - finger tracking.  If you wonder why I don't work with
> the other developers to support Wacom MT case, I can assure you that I
> am working with them. We will support Wacom MT devices once the
> appropriate protocol is available.

Er, the A protocol supports both tracked and untracked fingers (just wouldn't
mix the two for a single device).  Still, you probably have a point about
waiting for the B protocol to make its way upstream.


>> N-Trig based devices and Stantum ones are the most compliant,
> 
> That's fair since they don't need to track the fingers.

N-Trig doesn't have hardware tracking, Stantum does.

>> so Wacom situation needs to be clarified specially considering they are HID
>> devices (meaning similar to the others devices which respect the
>> protocol).
> 
> Well, it is not a pure HID issue.  It is more of how we want to
> support MT on Linux issue.

Side note for Ikbel, its not clear the wacom devices actually stay HID compliant
when the full protocol is active, and the basic modes are really basic (and not
very useful).  As my understanding of the way HID and usb stuff communicate in
general, I've been growing less convinced its a protocol issue, more like an
initialization problem.  Still I'm not quite sure.

>> This impact very much the decision of a Linux user wanting to buy wacom mt device.
> 
> I think end users have already got the Linux support from Wacom.  Can
> you share the specific issues that are preventing Linux end users from
> buying a Wacom MT devices? If I didn't misunderstand your point here,
> you are talking about Linux users, not Linux developers/hackers,
> right?

I think normal users wanting to use linux for MT are generally frustrated and
confused at the moment.  Certainly the protocol between the wacom kernel driver
and the wacom user space driver is way below the level that most are following.

As developers, I think we have a strong interest in seeing wacom migrate to a
shared protocol.  But our interests are in pushing the functionality beyond
what's currently available, and we just want to see everyone get equal access to
features as we create them.  I think to convince Ping (if he weren't already),
and the others working on the Wacom code, we'd have to show off some of what
he's going to gain by switching, and at the moment, as a community, I don't
think we are quite ready, maybe soon.

So, Ping, consider that I and others will have cool things to show you in the
near future, and if you start migrating you will get some cool new functionality
when the two waves of progress collide.  Also, I think Henrik's B protocol has
some cool advantages which you've demonstrated considerable interest in.  So I
gather you do intend to switch when its convenient?

>> Ping can you clarifie the situation of wacom based devices ?
> 
> What else do you need me to clarify? I am all ears and I am ready to
> clarify any thing that puzzles you.

Will Wacom be moving to more than 2 finger support in the near future?  Do you
already have some products on the market?

Rafi

> Ping
> 
>> Thanks,
>>
>> ik.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkweaooACgkQwuRiAT9o609XfACeN7OgUTXusOztPHwoHbTEqF7F
VwwAoKIedssO79JIzeOnkegjsBu2fLkl
=QyOK
-----END PGP SIGNATURE-----

Re: Wacom based devices and the mt kernel protocol.

[Ping Cheng]

On Sun, Jun 20, 2010 at 12:22 PM, Rafi Rubin <rafi@seas.upenn.edu> wrote:
>>> N-Trig based devices and Stantum ones are the most compliant,
>>
>> That's fair since they don't need to track the fingers.
>
> N-Trig doesn't have hardware tracking, Stantum does.

Sorry for my ignorance. I don't have much time to keep track of what
other devices are doing. Keeping all of my own balls rolling is more
than enough work for me :).

>>> so Wacom situation needs to be clarified specially considering they are HID
>>> devices (meaning similar to the others devices which respect the
>>> protocol).
>>
>> Well, it is not a pure HID issue.  It is more of how we want to
>> support MT on Linux issue.
>
> Side note for Ikbel, its not clear the wacom devices actually stay HID compliant
> when the full protocol is active, and the basic modes are really basic (and not
> very useful). As my understanding of the way HID and usb stuff communicate in
> general, I've been growing less convinced its a protocol issue, more like an
> initialization problem.

Thank you Rafi for explaining this issue.  Wacom tablets have a
relatively long history, longer than the time that digitizers were
supported on Linux. The default HID mouse protocol was introduced due
to the history, which was required by Microsoft.... We can not change
history, can we :)?

Wacom kept their devices consistent protocol-wise over the years.
That's why even the modern digitizers default to the basic HID
protocol.

> Still I'm not quite sure.

I think you've got the point.

>>> This impact very much the decision of a Linux user wanting to buy wacom mt device.
>>
>> I think end users have already got the Linux support from Wacom.  Can
>> you share the specific issues that are preventing Linux end users from
>> buying a Wacom MT devices? If I didn't misunderstand your point here,
>> you are talking about Linux users, not Linux developers/hackers,
>> right?
>
> I think normal users wanting to use linux for MT are generally frustrated and
> confused at the moment.

Why do the end users get frustrated and confused? Can you share some
examples from an end user's (not developer's) perspective with me?

> Certainly the protocol between the wacom kernel driver
> and the wacom user space driver is way below the level that most are following.

By "most", do you mean most MT device drivers, or most end users, or
most developers? I am eager to bring the Wacom driver to the specific
level.

> As developers, I think we have a strong interest in seeing wacom migrate to a
> shared protocol.  But our interests are in pushing the functionality beyond
> what's currently available, and we just want to see everyone get equal access to
> features as we create them.  I think to convince Ping (if he weren't already),
> and the others working on the Wacom code, we'd have to show off some of what
> he's going to gain by switching, and at the moment, as a community, I don't
> think we are quite ready, maybe soon.

I do see the positive sign here. And, BTW, it is a she that you are
talking to :).  Don't worry, you are still taking to the same person.
So, it really doesn't matter to me. But, I don't want you to get the
fact from someone else and to be "puzzled" :).

> So, Ping, consider that I and others will have cool things to show you in the
> near future, and if you start migrating you will get some cool new functionality
> when the two waves of progress collide.  Also, I think Henrik's B protocol has
> some cool advantages which you've demonstrated considerable interest in.  So I
> gather you do intend to switch when its convenient?

Why not? Do you think I am that stupid :)? To be honest, I've been
waiting for the day for a while.

>>> Ping can you clarifie the situation of wacom based devices ?
>>
>> What else do you need me to clarify? I am all ears and I am ready to
>> clarify any thing that puzzles you.
>
> Will Wacom be moving to more than 2 finger support in the near future?  Do you
> already have some products on the market?

No, we do not have 2+ finger devices on the market and I can not tell
what the future would be. We'll see what it is when time comes.  Any
more questions :)?

Ping
--
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Wacom based devices and the mt kernel protocol.

[Rafi Rubin]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 06/20/10 20:31, Ping Cheng wrote:
> On Sun, Jun 20, 2010 at 12:22 PM, Rafi Rubin <rafi@seas.upenn.edu> wrote:
>>>> N-Trig based devices and Stantum ones are the most compliant,
>>>
>>> That's fair since they don't need to track the fingers.
>>
>> N-Trig doesn't have hardware tracking, Stantum does.
> 
> Sorry for my ignorance. I don't have much time to keep track of what
> other devices are doing. Keeping all of my own balls rolling is more
> than enough work for me :).
> 
>>>> so Wacom situation needs to be clarified specially considering they are HID
>>>> devices (meaning similar to the others devices which respect the
>>>> protocol).
>>>
>>> Well, it is not a pure HID issue.  It is more of how we want to
>>> support MT on Linux issue.
>>
>> Side note for Ikbel, its not clear the wacom devices actually stay HID compliant
>> when the full protocol is active, and the basic modes are really basic (and not
>> very useful). As my understanding of the way HID and usb stuff communicate in
>> general, I've been growing less convinced its a protocol issue, more like an
>> initialization problem.
> 
> Thank you Rafi for explaining this issue.  Wacom tablets have a
> relatively long history, longer than the time that digitizers were
> supported on Linux. The default HID mouse protocol was introduced due
> to the history, which was required by Microsoft.... We can not change
> history, can we :)?
> 
> Wacom kept their devices consistent protocol-wise over the years.
> That's why even the modern digitizers default to the basic HID
> protocol.
> 
>> Still I'm not quite sure.
> 
> I think you've got the point.
> 
>>>> This impact very much the decision of a Linux user wanting to buy wacom mt device.
>>>
>>> I think end users have already got the Linux support from Wacom.  Can
>>> you share the specific issues that are preventing Linux end users from
>>> buying a Wacom MT devices? If I didn't misunderstand your point here,
>>> you are talking about Linux users, not Linux developers/hackers,
>>> right?
>>
>> I think normal users wanting to use linux for MT are generally frustrated and
>> confused at the moment.
> 
> Why do the end users get frustrated and confused? Can you share some
> examples from an end user's (not developer's) perspective with me?

Sorry, I've been too focused on the n-trig users.  They have been frustrated,
and I've seen from the forums that some have been playing with customized
versions of both the kernel and wacom x driver to get your two finger gestures.

I'm also projecting my own frustrations, seeing the X mt interface still not
quite set, and very little in the way of applications that use mt at all.  We're
still quite a ways away from seeing the full potential of a more hands on desktop.

>> Certainly the protocol between the wacom kernel driver
>> and the wacom user space driver is way below the level that most are following.
> 
> By "most", do you mean most MT device drivers, or most end users, or
> most developers? I am eager to bring the Wacom driver to the specific
> level.

End users, they don't tend to care about the internal protocols as long as their
programs do what they want.

>> As developers, I think we have a strong interest in seeing wacom migrate to a
>> shared protocol.  But our interests are in pushing the functionality beyond
>> what's currently available, and we just want to see everyone get equal access to
>> features as we create them.  I think to convince Ping (if he weren't already),
>> and the others working on the Wacom code, we'd have to show off some of what
>> he's going to gain by switching, and at the moment, as a community, I don't
>> think we are quite ready, maybe soon.
> 
> I do see the positive sign here. And, BTW, it is a she that you are
> talking to :).  Don't worry, you are still taking to the same person.
> So, it really doesn't matter to me. But, I don't want you to get the
> fact from someone else and to be "puzzled" :).

Um, what can I say, the hazards of interacting with just text in a language
curiously lacking in proper androgynous pronouns.  Please forgive my ignorance
particularly culturally.

>> So, Ping, consider that I and others will have cool things to show you in the
>> near future, and if you start migrating you will get some cool new functionality
>> when the two waves of progress collide.  Also, I think Henrik's B protocol has
>> some cool advantages which you've demonstrated considerable interest in.  So I
>> gather you do intend to switch when its convenient?
> 
> Why not? Do you think I am that stupid :)? To be honest, I've been
> waiting for the day for a while.

Not at all, I'm just incompetent at subtle nudging.

>>>> Ping can you clarifie the situation of wacom based devices ?
>>>
>>> What else do you need me to clarify? I am all ears and I am ready to
>>> clarify any thing that puzzles you.
>>
>> Will Wacom be moving to more than 2 finger support in the near future?  Do you
>> already have some products on the market?
> 
> No, we do not have 2+ finger devices on the market and I can not tell
> what the future would be. We'll see what it is when time comes.  Any
> more questions :)?

Nope, I think I've put my foot in it enough for the evening :-)

Rafi
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkwevgQACgkQwuRiAT9o60+k9QCdEPogsLgVUDc7K/vqWNvjd/yS
IYwAoPyS0hOuqS51Dlft011EInnrx7+J
=mPjC
-----END PGP SIGNATURE-----

Re: Wacom based devices and the mt kernel protocol.

[Ping Cheng]

On Sun, Jun 20, 2010 at 6:19 PM, Rafi Rubin <rafi@seas.upenn.edu> wrote:
>>
>>>>> This impact very much the decision of a Linux user wanting to buy wacom mt device.
>>>>
>>>> I think end users have already got the Linux support from Wacom.  Can
>>>> you share the specific issues that are preventing Linux end users from
>>>> buying a Wacom MT devices? If I didn't misunderstand your point here,
>>>> you are talking about Linux users, not Linux developers/hackers,
>>>> right?
>>>
>>> I think normal users wanting to use linux for MT are generally frustrated and
>>> confused at the moment.
>>
>> Why do the end users get frustrated and confused? Can you share some
>> examples from an end user's (not developer's) perspective with me?
>
> Sorry, I've been too focused on the n-trig users.  They have been frustrated,
> and I've seen from the forums that some have been playing with customized
> versions of both the kernel and wacom x driver to get your two finger gestures.
>
> I'm also projecting my own frustrations, seeing the X mt interface still not
> quite set, and very little in the way of applications that use mt at all.
> We're still quite a ways away from seeing the full potential of a more hands on desktop.

As far as I understand,  this issue is not device driver or MT support
specific. It is universal  on Linux.

>>> Certainly the protocol between the wacom kernel driver
>>> and the wacom user space driver is way below the level that most are following.
>>
>> By "most", do you mean most MT device drivers, or most end users, or
>> most developers? I am eager to bring the Wacom driver to the specific
>> level.
>
> End users, they don't tend to care about the internal protocols as long as their
> programs do what they want.

Then, I still don't get what you meant by "way below the level that
most are following". I'd really like to understand it so I can work on
it.

Thank you,

Ping
--
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Wacom based devices and the mt kernel protocol.

[Mohamed Ikbel Boulabiar]

On Mon, Jun 21, 2010 at 7:30 AM, Ping Cheng <pinglinux@gmail.com> wrote:
> On Sun, Jun 20, 2010 at 6:19 PM, Rafi Rubin <rafi@seas.upenn.edu> wrote:
>>>
>>>>>> This impact very much the decision of a Linux user wanting to buy wacom mt device.
>>>>>
>>>>> I think end users have already got the Linux support from Wacom.  Can
>>>>> you share the specific issues that are preventing Linux end users from
>>>>> buying a Wacom MT devices? If I didn't misunderstand your point here,
>>>>> you are talking about Linux users, not Linux developers/hackers,
>>>>> right?
>>>>
>>>> I think normal users wanting to use linux for MT are generally frustrated and
>>>> confused at the moment.
>>>
>>> Why do the end users get frustrated and confused? Can you share some
>>> examples from an end user's (not developer's) perspective with me?

it would be no difference between users and developers behaviors, it
is just a matter of time, developers get frustrated some time before
normal users would be :)




>> Sorry, I've been too focused on the n-trig users.  They have been frustrated,
>> and I've seen from the forums that some have been playing with customized
>> versions of both the kernel and wacom x driver to get your two finger gestures.
>>
>> I'm also projecting my own frustrations, seeing the X mt interface still not
>> quite set, and very little in the way of applications that use mt at all.
>> We're still quite a ways away from seeing the full potential of a more hands on desktop.
>
> As far as I understand,  this issue is not device driver or MT support
> specific. It is universal  on Linux.

Maybe because not all of drivers are up-to-date to new changes happening.
The new protocol is a way to standardize the way all mt devices are handled.



>>>> Certainly the protocol between the wacom kernel driver
>>>> and the wacom user space driver is way below the level that most are following.
>>>
>>> By "most", do you mean most MT device drivers, or most end users, or
>>> most developers? I am eager to bring the Wacom driver to the specific
>>> level.
>>
>> End users, they don't tend to care about the internal protocols as long as their
>> programs do what they want.
>
> Then, I still don't get what you meant by "way below the level that
> most are following". I'd really like to understand it so I can work on
> it.

if 5 or 6 mt devices are supported in the new protocol and wacom is
not, this means a split in the support way. You've just told us the
point of how bad having different ways things are supported.




To summarize the discussion with you Ping, you will support the new
protocol as far as it will show a good reason for a switch yep ? But
what if this can make other devices look as they are better supported
in the near future with upcoming changes (even if wacom are better
technically) ?
This remembers me the days when ntrig was not supported in Linux (and
with no intention to be supported), then after the non-corporate
driver comes, they tried to publish their own.
http://www.ipetitions.com/petition/ntrig-on-linux/
--
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html