Re: Pointer Events on Chrome

[Rick Byers]

Hi Robert,

Thanks for reaching out.  /cc input-dev - a good public list to discuss this further on.

 

Pointer Events implementation is proceeding nicely, but there's a lot to do.  You can star http://crbug.com/471824 for major updates (though much work is happening in the other more detailed bugs it's marked blocked on).

I assume it's support for stylus on Windows that you're primarily interested in, right?  We've got some folks adding stylus support for Android and stylus support (including Wacom I believe) for ChromeOS, but we haven't yet schedule the work necessary for Windows.  We might need to update all of Chrome to use WM_POINTER (something we've avoided because we still have a large number of touch users on Windows 7), but we haven't really looked into it yet.

Rick

On Sat, Aug 22, 2015 at 9:13 PM, Robert Cohn <rober...@wacom.com> wrote:

Hello -

Is there a site or email group that I can monitor to see the progress of Chrome adopting Pointer Events?  The most recent postings I can find are in the March 2015 time frame at which time Google made a commitment to implement the API.

Thanks very much.

Best regards,

Robert Cohn

Software Developer

Wacom Technology

[Rob Bradford]

Hi Robert,

Progress with Wacom Tablet support on ChromeOS is going well, the data structure required to hold the data has now landed (http://crrev.com/344250) along with required "driver" code in ozone (http://crrev.com/345049). One bit that remains for stylus devices on ChromeOS (and which applies for some other Aura platforms too) is making the values from the ui::PointerDetails available to blink (http://crrev.com/1266643002)

In general for stylus support one of the main outstanding tasks is that blink needs to fire PointerEvents for mouse events (http://crbug.com/476571) 

Did you have a particular question in mind?

Rob

To unsubscribe from this group and stop receiving emails from it, send an email to input-dev+...@chromium.org.

[Robert Cohn]

Hi Rob –

 

Since Chrome dropped support for NPAPI, we’ve had a number of customers who had been using our NPAPI-based plugin being left without a viable solution.  Mostly, I’ve been alerting them to the fact that Pointer Events will eventually be available and can be used for their web apps. 

 

My question to Rick was to get some idea of when PE would be available on Chrome on Windows.  From his response below, it seems that it might be a while yet.  Is there a ballpark timeline for when Chrome support for PE will be available that I can refer to if anybody should ask?  That would also help direct our efforts to fill the NPAPI gap if necessary.

 

Thanks very much for getting back to me so quickly.

 

Best regards,

Bob

 

Robert Cohn

Tablet Software Engineering

Wacom Technology Corp.

360-896-9833 x172

rc...@wacom.com

[Rick Byers]

Hey Robert,

I didn't realize people were relying on an NPAPI plugin for Wacom, it sucks that they're now stuck - sorry!  Out of curiosity, can you point me at a couple example use cases / sites?  It'll help with our testing / prioritization to better understand the needs of your customers.

It's really hard to say when we'll ship pointer events - there are a couple tough outstanding issues with the spec that may take some time to get consensus on (shipping isn't even guaranteed if we can't resolve the issues).  But given the current rate and work to be done, I suspect we'll have a reasonably complete experimental implementation around the start of 2016 (mustaq@ may have a better idea).  At that point we could give early adopter users the option of turning on a flag (or maybe allow site developers to opt-in using our experimental framework), although it'll still have some bugs / incompatibilities with other browsers.

However, perhaps we can leverage your expertise to get the Windows stylus support done sooner?  I don't know much about Win32 low-level input myself.  I've kind of been assuming we'll need to switch completely to WM_POINTER (from the WM_MOUSE and WM_TOUCH events we use today) to fully support stylus - and I'm a little afraid of the churn that'll cause to all the windows mouse code in chromium.  But perhaps there's a more incremental approach where we could start listening to WM_POINTER messages just for pens, and leave all our existing mouse/touch code alone?

Rick

[Mustaq Ahmed]

Yes, we should have an experimental version ready by early 2016.

Mustaq

[Robert Cohn]

Hi Mustaq –

 

Thanks very much for the update.  Much appreciated.

 

As to which API to use, please remember that WM_POINTER is supported on Win 8 going forward, so this wouldn’t be a solution for Win7, Vista, XP.  And WM_TOUCH supports starts at Win7.

You may want to consider using one of the other APIs, such as InkCollector or RealTimeStylus.

 

Talk to you later.

 

Best regards,

Bob

[Rick Byers]

On Wed, Aug 26, 2015 at 11:44 PM, Robert Cohn <rober...@wacom.com> wrote:

Hi Mustaq –

 

Thanks very much for the update.  Much appreciated.

 

As to which API to use, please remember that WM_POINTER is supported on Win 8 going forward, so this wouldn’t be a solution for Win7, Vista, XP.  And WM_TOUCH supports starts at Win7.

You may want to consider using one of the other APIs, such as InkCollector or RealTimeStylus.

Yes, that's why we've avoided using WM_POINTER in the past.  Are these other stylus APIs considered deprecated in Win8+?  We're seeing a lot of Win7 -> Win10 upgrades and so are starting to thing we may be willing to drop touch support on Win7.  Do you have any data on what fraction of your customers are on Win8+?

[Robert Cohn]

Hi Rick –

 

I really don’t have explicit usage information for our tablets, but since Pointer Events will be used by all Chrome users, it’s really an “ecosystem” issue (if I can use that over-used term) for anybody’s pointing device.

This site: https://www.netmarketshare.com/operating-system-market-share.aspx?qprid=10&qpcustomd=0 shows that more that 60% of all users (Windows and Mac) are using Win7.  This is just one data point, but I think the important thing here is to note the small percentage of Win8, Win8.1 users (Win10 will probably gain a bigger share of that part of the pie).

 

My suggestion would be to not exclude Win7 users at a minimum in your planning (ie, support at least Win7 going forward), and to support Vista and XP if possible.

 

Best regards,

Bob

[Rick Byers]

Thanks Bob.

I almost certainly won't be practical for us to support/test stylus devices on XP (Chrome itself is slated to stop supporting XP by the end of the year), and the testing burden alone is probably not something we can justify ourselves.  We'll probably try to collect some data on what fraction of our users have stylus devices attached on various versions of Windows.

We certainly stiff have a ton of users on Win7.  But we've been watching the numbers shift specifically for touchscreen users, and for that group an increasing majority have Win8+.  But I can imagine that stylus is completely different, so we'll have to check before we make any decisions here.

If, by the way, you have an engineer who is interested in contributing Windows stylus support to chromium, we're happy to provide guidance and review patches.  Otherwise we will get to it :-).

Thanks,

  Rick

[Robert Cohn]

Hi Rick –

 

I totally understand about XP; same for Vista.  But I would urge that Win7 support be strongly considered.

 

As to helping out with the stylus support, I think we’re pretty much up to our eyeballs in project things right now (I’d actually love to assist, but my wife insists on the rather arcane idea that I get some sleep sometimes).

Anyway, if I can help out with stylus or touch questions when they arise, please don’t hesitate to send an email.

[Rick Byers]

Sounds good and understood - thanks Bob.  We'll let you know when we start working on this.  I've filed bug 526153 to track the Windows-specific work - star it to get updates on that in particular.

Rick