Fwd: GOK users?
|
View:
New views
6 Messages
—
Rating Filter:
Alert me
|
 |
|
||||||||||
|
|
Ang. Fwd: GOK users?
by Mats Lundälv
::
Rate this Message:
Hi Will, Ben, Steve and all, As the GOK development is linked to the AEGIS project ambitions this is of double interest to me, that is; in general and as part of AEGIS. How many users are there? I don't know of any single GOK user in Sweden, and I would suspect that the number around the world is probably is a very limited one. Why is that? In my view that is because GOK, while concept wise being a very advanced, state-of-the-art alternative OSK input alternative, still in practice appears more as a demonstrator software tool than a real alternative to the AT solutions offered on the Windows and Mac platforms for its potential target users. This is probably partly due to the problems in securing further maintenance since the original development. This in turn is depending on the lack of active user participation and feedback I suppose. This is in turn depending on the characteristics of the user group - users with severe physical limitations, often with additional neurological challenges, who are generally dependent on helpers/facilitators/professionals to set them up with a proper sytem tailored to their individual needs - ... and so on. This is tricky - not only from a technical development point of view, but also from a cultural one. The whole setup within the Oppen Desktop development environment is very depending on active participation of stakeholders - such as the blind and visually impaired technical savvy users driving and supporting the Orca development. That kind of user participation is hard to find for GOK. You have to reach the potential users mostly via their supporters, like parents, teachers, AT professionals etc. And it will be hard to persuade most of those to pay an interest in the Open Desktop minority environment for their users until there are tools that match current alternatives on the Win and Mac platforms - not only in basic technical functionality, but also in usability for local tailoring and support. So to get there, the development will have to go on in rather thin air, relying on a very limited group of people, hopefully with good references to users and tools on other platforms. I suppose the greatest chance to find users would be in communities, regions and countries where the policy is to use OSS platforms, and where there is a general shortage of resources for buying expensive AT solutions. In many of those areas we may have substantial language barriers making it hard to establish communication to get feedback. I suppose a symptom of the problem is the introduction of the OnBoard OSK in Ubuntu. GOK seemed to be perceived as being too "advanced and special" and not slick enough in appearance, so an alternative was introduced - which is not very well maintained (I've posted bugs a year ago about broken support for UTF-8 in the "Snippets" text there without any reaction ). This is actually an unfortunate fragmentation. Why not a set of simplified/basic layouts in GOK covering the OnBoard ambitions instead? Ideally, we should have all good forces focusing on the development and maintenance of one solid OSK solution for Gnome (and possibly other platforms) - covering well the wide range of needs - from basic standard QWERTY layouts - to advanced dynamic layouts, and including support for graphics to include AAC users. The basic functionality for all these needs is a set of rock solid input - navigation and selection - alternatives to cover different user demands. The OSK content and layout should be handled by a common layout manager. In the AEGIS discussions we have touched upon the alternative of moving GOK over to Java in order to give it a much broader user and maintenance base. The question is whether a Java implementation - a JOK - could be as effective and well integrated in the Gnome environment as GOK is and has the potential of being? What do you guys think? One thing that would greatly support participation in the short perspective were if new releases could be packaged directly for say the Ubuntu and OpenSolaris platforms. I myself find it hard to participate and give feedback as I would like to when I regularly get the message about where I can download the source code of the latest update. To get any wider participation for this user group you just cannot go on expecting stakeholders to download and compile sources. You should just be able to download and install the new binary directly - in parallell to the current official release installation - and run, test and give feedback! Please excuse this verbose reply! Cheers, Mats Mats Lundälv DART, reg.hab., Drottning Silvias barn- och ungdomssjukhus, SU och Specialpedagogiska skolmyndigheten Kruthusgatan 17, SE-411 04 Göteborg, Sweden www.dart-gbg.org , www.spsm.se tel (SPSM): +46 (0)10 47 35 513, tel (DART): +46 (0)31 342 08 04 mobil:+46 (0)705 298093, fax: +46 (0)31 15 58 29 E-post: mats.lundalv@... eller mats.lundalv@...
A chance to influence GOK on Screen Keyboard for Linux/Open Solaris Steve ---------- Forwarded message ---------- From: Willie Walker <William.Walker@...> Date: 2009/9/11 Subject: GOK users? To: Steve Lee <steve@...> Cc: Ben Konrath <ben.konrath@...> Hi Steve: It's been a while. Hope you're doing OK. I'm curious if you know of any GOK users out there? Ben Konrath is working on improving GOK for GNOME 3.0, and the direction he takes may depend upon how many GOK users there really are. Will |
||||||||||
|
|
Re: Ang. Fwd: GOK users?
by Steve Lee-3
::
Rate this Message:
Mats thanks for a very well considered response. I'll only make a couple of quick comments.
Why is that? In my view that is because GOK, while concept wise being a very advanced, state-of-the-art alternative OSK input alternative, still in practice appears more as a demonstrator software tool than a real alternative to the AT solutions offered on the Windows and Mac platforms for its potential target users. Agree and GOK is probably the worse affected of all bundled AT. So if we can crack it for GOK then we should crack it for all. So to get there, the development will have to go on in rather thin air, relying on a very limited group of people, hopefully with good references to users and tools on other platforms. We also need to think very carefully about how it is deployed. Peter Cudd said yesterday that he was congratulated at AAATE for doing this with Maavis and It makes great sense to me. Many facilitators cannot do techy setup and we have avoided that as much as possible. I suppose the greatest chance to find users would be in communities, regions and countries where the policy is to use OSS platforms, and where there is a general shortage of resources for buying expensive AT solutions. In many of those areas we may have substantial language barriers making it hard to establish communication to get feedback. so how can we identfy and engage them? I suppose a symptom of the problem is the introduction of the OnBoard OSK in Ubuntu. GOK seemed to be perceived as being too "advanced and special" and not slick enough in appearance, so an alternative was introduced - which is not very well maintained (I've posted bugs a year ago about broken support for UTF-8 in the "Snippets" text there without any reaction ). This is actually an unfortunate fragmentation. Why not a set of simplified/basic layouts in GOK covering the OnBoard ambitions instead? I saw OnBoard trying to be a simple OSK not switch, though that was added later.As you say it is a pretty inactive project, though until recently that was the same for GOK. I saw OnBoard having both common effects; fragmentation and waking up the incumbent (GOK) Ideally, we should have all good forces focusing on the development and maintenance of one solid OSK solution for Gnome (and possibly other platforms) - covering well the wide range of needs - from basic standard QWERTY layouts - to advanced dynamic layouts, and including support for graphics to include AAC users. The basic functionality for all these needs is a set of rock solid input - navigation and selection - alternatives to cover different user demands. The OSK content and layout should be handled by a common layout manager. Can't argue with any of that. In the AEGIS discussions we have touched upon the alternative of moving GOK over to Java in order to give it a much broader user and maintenance base. The question is whether a Java implementation - a JOK - could be as effective and well integrated in the Gnome environment as GOK is and has the potential of being? What do you guys think? Whatever the technical merits of Java (and political arguments) GNOME is settling on Python as the preferred language for AT devleopment. To my mind that is a good decision and we should not dilute the work on libraries such as pyatspi without good reason. I'm alos no sure of the status of the java bridge.
Agree that is very sensible, and only requires overnight builds or continuous integration. Perhaps something like Mozilla's technology could be used as David Bolter is now familiar with it (I expect). |
||||||||||
|
|
Re: Ang. Fwd: GOK users?
by Ben Konrath-3
::
Rate this Message:
Hi all,
I wasn't subscribed to the list when I replied to this message so I've include my original reply below. Cheers, Ben ++ Hi Mats, Thanks for your input; it's always nice to hear from passionate people like yourself :-). I've included a few comments below. On 09/17/2009 11:36 AM, Mats Lundälv wrote: > Why not a set of simplified/basic layouts in GOK covering the OnBoard > ambitions instead? Good point. Please file a bug about this against GOK in GNOME's bugzilla. > Ideally, we should have all good forces focusing on the development and > maintenance of one solid OSK solution for Gnome (and possibly other > platforms) - covering well the wide range of needs - from basic standard > QWERTY layouts - to advanced dynamic layouts, and including support for > graphics to include AAC users. The basic functionality for all these needs > is a set of rock solid input - navigation and selection - alternatives to > cover different user demands. The OSK content and layout should be handled > by a common layout manager. A reasonable goal but I think we need to take things one step at a time and come up with a realistic plan for GNOME 3.0. > In the AEGIS discussions we have touched upon the alternative of moving > GOK over to Java in order to give it a much broader user and maintenance > base. The question is whether a Java implementation - a JOK - could be as > effective and well integrated in the Gnome environment as GOK is and has > the potential of being? What do you guys think? As Steve mentioned, I think Python is better suited for this task but we should discuss all the options before we proceed. I encourage anybody interested to join the GOK mailing list to continue this discussions. > One thing that would greatly support participation in the short > perspective were if new releases could be packaged directly for say the > Ubuntu and OpenSolaris platforms. I myself find it hard to participate and > give feedback as I would like to when I regularly get the message about > where I can download the source code of the latest update. To get any > wider participation for this user group you just cannot go on expecting > stakeholders to download and compile sources. You should just be able to > download and install the new binary directly - in parallell to the current > official release installation - and run, test and give feedback! The GNOME project doesn't generally release binaries - that's up to the consumers of GNOME (the Linux distributions). For me, working through the details of making supplemental packages for various distributions would be a major distraction. But there's nothing stopping other people from doing this. Perhaps this should be perused at the distro level. Thanks again for your comments. I'm looking forward to further discussion on GOK's mailing list. Cheers, Ben |
||||||||||
|
|
First alpha version of Caribou
by Mats Lundälv
::
Rate this Message:
Hi Ben, Cheers,Congratulations to the first version of Caribou! I successfully installed and ran it on my Ubuntu 9.10 in VirtualBox (in WinXP). It seems stable and the direct mouse input - and output to active application seems to work fine. I made a customised Swedish keyboard layout with an additional row of keys to test non-English characters, and that worked fine too. The only character I've found so far that does not get output to the target app is the Euro sign € - for some reason. So your first goal for Dec 11 already seems to be well on the way. I'll get back in more detail on the GOK list ... Mats -----gok-list-bounces@... skrev: ----- Till: gnome-accessibility-list@..., gok-list@... -----<oats-sig-bounces@...> skrev: ----- Till: OATs Project Special Interest Group <oats-sig@...> |
||||||||||
|
|
Re: First alpha version of Caribou
by Ben Konrath-3
::
Rate this Message:
Hi Mats,
On 12/07/2009 08:46 AM, Mats Lundälv wrote: > Hi Ben, > > Congratulations to the first version of Caribou! > I successfully installed and ran it on my Ubuntu 9.10 in VirtualBox (in > WinXP). > It seems stable and the direct mouse input - and output to active > application seems to work fine. Great! Thanks for trying it! > I made a customised Swedish keyboard layout with an additional row of > keys to test non-English characters, and that worked fine too. Could you send me that layout? I would like to include additional layouts with caribou. > The only character I've found so far that does not get output to the > target app is the Euro sign € - for some reason. So your first goal > for Dec 11 already seems to be well on the way. Ok, that's probably a bug in the keysym library that I'm using. I'll look into it. Cheers, Ben |
| Free embeddable forum powered by Nabble | Forum Help |
