[Openicc] diagramming status

[Hal V. Engel]

On Wednesday 05 March 2008 12:52:45 edmund ronald wrote:
> I don't think it fair to say that Xrite is interested - it would be
> more accurate to say that *I personally* am interested in defining a
> platform-agnostic device interface, and Xrite and Barbieri like
> exploring the idea I presented, I have been talking to them and other
> vendors about some specs, and the hardware vendors are prepared to
> donate some hardware to further this effort.

I suspect that all of these vendors want to do business in this market segment since the user base is now significantly larger than the Mac user base and is growing at a very fast rate.   But they do not understand this market segment and are completely confused by it because it is so different from the closed systems they normally work with.   The main sticking point, as far as I can tell, is that they can't figure out how to work with us (IE. out in the open) while still remaining completely closed.  Of course, the realistic answer is that they can't at least not without it being very expensive for them in terms of support costs.

>
> This idea of open interface comes from the discussions we had earlier
> on list. It is unrealistic to hope that the companies will "get it"
> immediately regarding opening their hardware, 

There are a number of people here who have been talking to X-Rite and (before the merger) GretagMacbeth for several years about this.  For example, I first contacted GMB in late 2005 about this and I know Graeme has spoken with X-Rite managers at some length as well.    

I suspect that the merger is in part responsible for how slowly this has progressed (or should I say regressed) over that last two years.  Mergers typically disrupt things for several (typically 5 or more) years since it turns the focus inward as people fight for their jobs and/or for power and then fight to consolidate that power in the new organization.  My experience with mergers, having been through several very large ones and a number of smaller ones (even the small ones were bigger than the X-Rite/GMB merger), is that the fighting gets more intense and more protracted the higher up in the organization you go.  The closing of the device interface specs by X-Rite after the merger appears to me to be an internal power play rather then being based on well reasoned business principles.  

The sad part about this is that until the merger X-Rite had open interface specifications for their hardware that were readily available to developers, including open source developers.   X-Rite was universally considered to be an open source friendly company.   This was the case until Dec 2006 when they discontinued all devices for which there were interface specifications available.  It was not that long ago that X-Rite "got it" and I suspect that many inside of X-Rite still "get it".   I should add that GMB stopped "getting it" some time between when it started producing the SpectroLino and when the EyeOne devices were introduced.  So they have not gotten it for about 10 years now.

> and it's hopeless to get 
> the open source guys to agree to use proprietary vendor-supplied
> profiling software. 

Even if someone were to convince us that using proprietary software was the way to go the vendors are not supplying software that runs on our platforms.   In effect there is no such thing at this time as "proprietary vendor-supplied profiling software" available to us.  So this is not even on the table and the ball here is definitely in the vendors court not ours.  But you are correct that most users of open systems if given a choice between an open solution and a closed solution will choose the open solution even if the open solution has some issues.

> So my idea is to at least have clean and open 
> platform-agnostic and product-agnostic interfaces 

This would be a huge improvement and I think everyone here is in favor of this.  Of course, what the vendors view as a "clean and open platform-agnostic and product-agnostic interfaces" may not be considered open or clean or ... by anyone here.  

I was told about a year ago by higher ups in X-Rite that they would be creating a new cross platform (meaning *nix support) interface library for their devices that would be available about the middle of this year.  Have you heard anything about this?

> and then move things 
> a step further towards open drivers later on when the hardware guys
> see that they are making sales to open-source clients.

One of the things that I have seen when dealing with post merger X-Rite and GMB before the merger is that much of the policy there seems to be driven by the software side of the business.  The software tail waging the hardware dog if you will.   I suspect that if the hardware side of the business was in control that we would not be having these issues.    I would be glad if we were in fact dealing with the "hardware guys" but I don't think this is true at this time with respect to X-Rite.

>
> If anyone here wants to chat to me about it, I'd be very happy. The
> point of course is to create an instrument interface  definition that
> would be completely open and documented.

Again this is what we had until fairly recently from X-Rite and it would be good to get back to that state again.  The other vendors may be another matter.  ColorVision has not done this for a long time if ever and as far as I can tell neither have Barbieri or Avantes (Avantes did have an x86 binary blob Linux interface library for a while but it was not maintained and they pulled it from their web site).  If any of these other vendors can be convinced to make interface specs available then that would be a very good thing and I suspect that it would encourage other vendors to follow suit.

Hal
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.freedesktop.org/archives/openicc/attachments/20080305/77b8b43a/attachment.html