Re: Foaf-browser-app

On 10/6/11 8:23 AM, Henry Story wrote:
> On 6 Oct 2011, at 14:14, Kingsley Idehen wrote:
>
>> On 10/6/11 6:11 AM, Bob Ferris wrote:
>>> Hi Henry,
>>>
>>> On 10/4/2011 8:04 PM, Henry Story wrote:
>>>> While playing with the very cool Foaf-browser Android application [1], I noticed that there are a couple of things that we need to standardise to make that type of application seamless.
>>>>
>>>> As it requires having an Android phone which not everyone may have, I put together a little video of the Foaf-browser here
>>>>
>>>>     http://vimeo.com/30014844
>>>>
>>> This is an interesting representation of this Social Web Android application.
>>>
>>> +1 re. the UX/usability remarks
>> This tool is a good start, but UX/usability always struggles if the fundamental UI/UX bottleneck of Linked Data isn't addressed. The problem boils down to labels for URIs across TBox and ABox. Pulling it off requires a Linked Data platform that's able to use reasoning to harmonize labels across a variety of Annotation Properties. Doing this has to scale (as we demonstrate with 29+ Billion triples at: http://lod.openlinksw.com ).
> Kingsley, the usability issues I describe have nothing to do with reasoning.

I am referring to the need for labels replacing URIs i.e., automated 
generation of anchors for text when presenting relations that describe a 
subject.

This is the key to the matter. I've used the tool, and that's the only 
real UI/UX hurdle. They already figured out the value of QR Codes etc..

> They have to do with with focus on solving those problems. Some of those problems have to do with some missing information which we need to think about. For example the Foaf-browser is an address book that runs on a cell phone and is written in Java. It needs to be able to find the form to create a Secured WebID.

I think that's a separate product. Basically, a phone hosted WebID 
generator. We are making such a thing anyway. So stay tuned.

> There are usability problems that will come up with reasoning. But since we are all getting going, and we need to work together people should try to publish data that will require a minimum or no reasoning.

My points is that via a Web Service the reasoning burden is alleviated. 
The make retrieve a data object and its representation includes:

1. de-referencable name
2. a literal label.

The service does the work. The client just gets the data objects.

>   Then we are more likely to get other people on board, and work together as a community.

I am making the same point, but its kinda lost in translation .


>> We do offer an API into our faceted browser engine that enables anyone leverage this functionality. Thus, I need to double check that we actually surface what I've described since it took some serious engineering to pull off this solution in the manner I just described.
> We are speaking cell phones here. If you want to market your product here is an opening for you: do you have a version of virtuoso that runs efficiently on iPhones, which java cannot reach?

We can make a version of Virtuoso run wherever we choose if we deem it 
useful.

Kingsley
> Henry
>
>> Links:
>>
>> 1. http://virtuoso.openlinksw.com/dataspace/dav/wiki/Main/VirtFacetBrowserInstallConfig -- Virtuoso's DBMS hosted Faceted Browsing Service .
>>
>> Kingsley
>>> Cheers,
>>>
>>>
>>> Bo
>>>
>>>
>>>> where I also go into detail in some of the things that need to be improved, as I show how it works.
>>>>
>>>> 1. SEAMLESS STARTUP
>>>> ===================
>>>>
>>>> In order to allow seamless startup, there has to be a way for a user once he has installed the browser to find his home page.
>>>>
>>>> We discussed this yesterday on the WebID-XG incubator group meeting with Tim.
>>>>
>>>> a. getting to the home page
>>>> ---------------------------
>>>>
>>>> Clearly getting to the home page of the user is going to be important. That will require him to either:
>>>>
>>>>   - know his profile url
>>>>   - the company he is with to have implemented something like web finger, so that one could just ask him for his email address
>>>>   - ..
>>>>
>>>> We seem to have a few protocols to help here. What clearly was missing was the keygen equivalent for an application like that.
>>>>
>>>> b. keygen form pointer
>>>> ----------------------
>>>>
>>>>   We need  some way to get the equivalent of a keygen form in rdf. Then a non html browser could know where it needs to send an html POST of a keygen like response. Any ideas how to do this?
>>>>
>>>> 2. EDITING
>>>> ----------
>>>>
>>>> I suppose the RW community group have some idea of what is needed to allow such a tool to become an editor.
>>>>
>>>>
>>>> Henry
>>>>
>>>>
>>>> [1] http://code.google.com/p/mssw/wiki/Screenshots
>>>>
>>>>
>>>> You can run it in a emulator - very slow btw, but if you don't want to buy a phone it saves money.  Here is what I learned looking at it: The running stable release of the Address Book is
>>>>
>>>> http://code.google.com/p/mssw/downloads/detail?name=mssw-0.9.6-stable.apk
>>>>
>>>> The emulator can be found here:
>>>>
>>>>   http://developer.android.com/guide/developing/devices/emulator.html
>>>>
>>>> Once installed the tools/android  tool allows starts a GUI which allows one to create a virtual device. Perhaps one first has to install android 2.3.3 which is more like a phone, and create a virtual device for that. Then can launch the device. One installs the app on the virtual phone using once it is running
>>>>
>>>> ../platform-tools/adb install mssw-0.9.6-stable.apk
>>>>
>>>> (there may be a GUI way of doing that)
>>>>
>>>>
>>>> [2] this is true for creating WebIDs, for loving in and so on....
>>>
>>>
>>
>> -- 
>>
>> Regards,
>>
>> Kingsley Idehen	
>> President&   CEO
>> OpenLink Software
>> Web: http://www.openlinksw.com
>> Weblog: http://www.openlinksw.com/blog/~kidehen
>> Twitter/Identi.ca: kidehen
>>
>>
>>
>>
>>
>>
> Social Web Architect
> http://bblfish.net/
>
>
>


-- 

Regards,

Kingsley Idehen	
President&  CEO
OpenLink Software
Web: http://www.openlinksw.com
Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca: kidehen