- From: Kingsley Idehen <kidehen@openlinksw.com>
- Date: Mon, 17 May 2021 19:31:50 -0400
- To: public-rww@w3.org
- Message-ID: <091ddb11-229e-b313-0b96-f5cdb1df49e0@openlinksw.com>
On 5/17/21 11:44 AM, Timothy Holborn wrote:
>
>
> On Tue, 18 May 2021 at 01:09, Kingsley Idehen <kidehen@openlinksw.com
> <mailto:kidehen@openlinksw.com>> wrote:
>
> On 5/17/21 6:00 AM, Melvin Carvalho wrote:
>>
>>
>> On Mon, 17 May 2021 at 02:24, Kingsley Idehen
>> <kidehen@openlinksw.com <mailto:kidehen@openlinksw.com>> wrote:
>>
>> On 5/15/21 6:55 PM, Melvin Carvalho wrote:
>>>
>>>
>>> On Sun, 16 May 2021 at 00:09, Kingsley Idehen
>>> <kidehen@openlinksw.com <mailto:kidehen@openlinksw.com>> wrote:
>>>
>>> On 5/15/21 5:21 PM, Melvin Carvalho wrote:
>>>>
>>>>
>>>> On Sat, 15 May 2021 at 18:14, Kingsley Idehen
>>>> <kidehen@openlinksw.com
>>>> <mailto:kidehen@openlinksw.com>> wrote:
>>>>
>>>> On 5/14/21 3:26 PM, Timothy Holborn wrote:
>>>>> Sorry, re: clarifications,
>>>>>
>>>>> What did it do back in ~2011/2 when I first
>>>>> installed it? (Vs. now?). I can pull the dates,
>>>>> but you likely have them in your licensing
>>>>> server?? I was trying to do a POC via building a
>>>>> heritage capability as an initial usecase, at the
>>>>> time.... (supporting a Hysterical/ historical
>>>>> society)...
>>>>>
>>>>> Does it have more functionality since then? I
>>>>> assumed the answer was "yes" particularly given
>>>>> the status of "web payments" (pre credentials),
>>>>> way back then (before I ended up on the lists,
>>>>> something, I didn't consider would ever happen in
>>>>> my life, at the time)...
>>>>>
>>>>> Limitations linked to creating a knowledge banking
>>>>> framework. I started concepts in 2000
>>>>> ("information bank" or ibank) which progressed to
>>>>> "knowledge banking" circa 2011/2012 after doing
>>>>> some work "updating" old work, from mid 2010
>>>>> (indigenous application started 2009/10)..
>>>>>
>>>>> I think you're first "dataspaces" demo was
>>>>> 2007??? V.interested in "temporal web" /
>>>>> provenance solutions....
>>>>>
>>>>> Dignity enhancing web (vs. web slavery, or worse).
>>>>>
>>>>> Timothy Holborn.
>>>>
>>>>
>>>> Hi Timothy,
>>>>
>>>> Regarding Identity, Identification, Authentication,
>>>> and Authorization nothing has changed in Virtuoso.
>>>>
>>>> What has changed outside Virtuoso, via
>>>> complimentary tools and services that we provide
>>>> are as follows:
>>>>
>>>> [1] Browser Extensions for creating Private Keys,
>>>> X.509 Certificates, Identifiers (WebIDs and
>>>> NetIDs), and associated Profile Docs -- basically,
>>>> killing the headache left by predictable demise of
>>>> <keygen/>
>>>>
>>>> [2] Setting up WebID-TLS + Delegation from a
>>>> Browser so that the whole thing "just works" and
>>>> users aren't exposed to what they may perceive as
>>>> complexity re entity relationship type (and
>>>> associated semantics) triangulation
>>>>
>>>> The implications of the above are as follows, using
>>>> a Chromium or Mozilla browser:
>>>>
>>>> [1] You can create credentials using your browser
>>>> that are stored to an OS-provided Keystore (e.g.,
>>>> macOS Keychain) or PKCS#12 file
>>>>
>>>> [2] You can write data to a Data Space (e.g.,
>>>> OpenLink Data Spaces or Solid Pod) subject to ACLs
>>>> using WebID-TLS (with Delegation if so desired
>>>> i.e., kill off the UI/UX issues associated with
>>>> browser restarts since Person and their User Agents
>>>> have distinguished, but related identity)
>>>>
>>>>
>>>> Tools that demonstrate these capabilities include:
>>>>
>>>> [1] YouID <http://youid.openlinksw.com/>
>>>>
>>>> [2] OpenLink Structured Data Sniffer
>>>> <https://chrome.google.com/webstore/detail/openlink-structured-data/egdaiaihbdoiibopledjahjaihbmjhdj?hl=en>
>>>>
>>>> [3] OpenLink Structured Data Sniffer for Mozilla
>>>> <https://addons.mozilla.org/en-US/firefox/addon/openlink-structured-data-sniff/>
>>>>
>>>> As for Virtuoso, it hasn't changed bar adding
>>>> support for WebID-OIDC which enables compatibility
>>>> with Solid Pods for read-write operations via
>>>> WebDAV/LDP mounting functionality etc..
>>>>
>>>> Long story short, we are still waiting for everyone
>>>> to catch up :)
>>>>
>>>>
>>>> OpenLink certainly has been ahead of the curve
>>>>
>>>> I'm reminded of OSDB: https://osdb.openlinksw.com/osdb
>>>> <https://osdb.openlinksw.com/osdb>
>>>>
>>>> In particular this image:
>>>>
>>>> https://osdb.openlinksw.com/img/dastklohq01y.gif
>>>> <https://osdb.openlinksw.com/img/dastklohq01y.gif>
>>>>
>>>> This is the kind of thing I envisage as a next
>>>> iteration of the read write web
>>>>
>>>> The idea here being that each of those modular agents
>>>> are moving in time to a certain rhythm
>>>>
>>>> I dont think we can easily make something like in that
>>>> diagram today, aside from how the web already
>>>> operates. You visit a page, you might tweet it, or
>>>> share it, it gets indexed by a search engine etc. Yes,
>>>> it one way, but alot of centralization build in there
>>>>
>>>> What if the web had a more temporal set of heart beats
>>>> which the agents could be small, compact, modular,
>>>> robust. Also finite in nature due to block chains
>>>> being finite resources. Or as stated in paper trail
>>>> some teams collaborating or competing in different
>>>> contests.
>>>>
>>>> What is needed?
>>>>
>>>> - Tying read write agents to block chains using URIs
>>>> (so standardize a URI scheme to hook into a block chain)
>>>> - Ways to create fragments of a block chain that can
>>>> live as mirrored claims (so some schema)
>>>> - Ability to traverse chains in type, and data in time
>>>> - Ability to save the state of the agents, as well as
>>>> perhaps the logic, the code, the deployment (we have
>>>> VCS for this)
>>>> - Ability for state to evolve in time, so watching for
>>>> changes, for deployments
>>>> - Ability to identify agents (URIs) and described them
>>>> (Linked Data)
>>>> - Ability for agents to interact with one another, read
>>>> write verbs (e.g. PUT/POST/PATCH)
>>>>
>>>> All this can come from leveraging existing timestamp
>>>> servers, providing a heartbeat for multi agent read
>>>> write systems, largely gluing together the pieces we
>>>> already have
>>>>
>>>> Perhaps OpenLink can lead the way again here, and we
>>>> can devise a spec together. The aim is that gif
>>>> above. What tools can we use to get there?
>>>
>>>
>>> Hi Melvin,
>>>
>>> As you know, we are always happy to lead by example
>>> especially when specs are in place that offer critical
>>> foundation for interoperability. Personally, I believe
>>> that are a significant number of specs in place, hence
>>> our ability to quietly create the OpenLink Structured
>>> Data Bot Framework (OSDB).
>>>
>>> Going forward, we are currently looking at URIs and
>>> Blockchains which is an emerging and important frontier
>>> as you've already noted in your comments above.
>>>
>>>
>>> Excellent!
>>>
>>> So how advanced is the OSDB? What can it do? I've made a
>>> few bots before, and perhaps you'll agree with me on this,
>>> they are nice enough proof of concepts, but they are
>>> somewhat toothless. They lack robustness, and need
>>> maintenance. Can easily be turned on or off, and very much
>>> prone to race conditions. After all of this, they tend not
>>> to be all that useful.
>>
>>
>> OSDB can generate a REST-ful interaction console for any API
>> described using the OpenAPI standard or RDF (e.g., Actions
>> described using terms from the Schema.org Vocabulary).
>> Naturally, any OSDB instance is a proxy for interacting with
>> all the Actions that it has distilled from API documentation.
>>
>> OSDB was developed in anticipation of Siri and friends
>> becomes extensible via APIs. For example, simply giving Siri
>> new skills which are basically a collection of Actions.
>>
>> We are still waiting ...
>>
>>
>>>
>>> Let's give a test. Let's say I want to make a simple step
>>> counter. It hooks into my smart watch.
>>
>>
>> You simply need the counter to be documented using either
>> OpenAPI or RDF, that's it.
>>
>>
>>> It hooks into my phone pedometer, my treadmill, a bunch of
>>> stuff running at the same time.
>>
>>
>> Once the step above is completed you can integrate into any
>> device that has the notion of Actions and their execution.
>>
>>
>>> It then wants to store my data, and ensure that all devices
>>> can write to the store without conflicts. Also, importantly
>>> the store might go down in a DB or a pod or git, and it
>>> should just be able to come back up elsewhere, ditto the bot
>>> that is managing all of this.
>>>
>>> In your terminology, "it just works".
>>
>>
>> It will "Just Work" if the IoT devices understand Actions
>> distilled from API by way of documentation using OpenAPI or
>> RDF (e.g., using terms from Schema.org or other vocabs).
>>
>>
>>>
>>> So how close do you think we are to this, with your bots?
>>
>>
>> The "Bot" is OSDB is really about its ability to be
>> integrated into bots rather than being a bot itself per se..
>> It is a Bot capability enhancer, so to speak.
>>
>>
>>> This is the style of thing I'd like to spec with a supra
>>> operating system that offers web scale semaphores. That's
>>> what binding to a time stamp server gives you.
>>>
>>> So, what's required to do this?
>>
>>
>> So-called Smart Agents like Siri, Alexa, Google etc.. being
>> extensible using a common method e.g., the OpenAPI or RDF
>> standards. This hasn't happened yet, unfortunately. We even
>> assumed the API Economy folks (typically anti RDF) would at
>> least use OpenAPI (their own spec) but that hasn't happened
>> either :(
>>
>>
>> There are a lot of anti RDF folks out there.
>
>
> Yes.
>
>
>> We've always tried to encourage (semantic) web standards and RDF
>> in this group. There are more standards now emerging around
>> JSON(-LD) and schema.org <http://schema.org> is becoming a de
>> facto standard for the semantic web
>
>
> Yes, JSON is becoming the preferred format for data represented as
> entity relationship graphs. Like everything else, unbeknownst to
> RDF detractors, JSON and RDF are compatible since the former is a
> data representation format and the latter a abstract data
> definition language.
>
> We punctuate the comment above in our recent OSDS updates that
> transform both JSON and CSV to RDF deployed using Linked Data
> principles. Fundamentally, you click and "it just works!"
>
>
>>
>> Regarding Smart Agents instead of massive centralized personal
>> assistants, why dont we aim to create more decentralized and
>> distributed personal agents with declarative data store state
>> machines which operate on via standards (perhaps ones we create).
>>
>> Working with "small data" rather than "big data"
>
>
> Our world view has always been about "small data" rather than "big
> data" i.e., data access by reference rather than data copying
> which is fundamentally limited and inherently centralized.
>
>
>>
>> But working together
>>
>> Dont know much about OpenAPI, would it be a good inspiration for
>> a spec?
>
>
> It isn't widely adopted by its own supporters which indicates to
> me that it isn't "good inspiration for a spec".
>
> Logic as the universal conceptual schema is the spec, IMHO.
>
> RDF as a formalization of EAV is the abstract language for data
> definition informed by a universal conceptual schema, despite all
> the distractions from detractors.
>
> At OpenLink, we believe the specs are done. We are putting our
> energy into apps and services that demonstrate what's possible etc..
>
>
Hi Timothy,
> So then; the belief is that the ideological / social / geo-political
> issues; are simply about a lack of documentation due to the focal
> point of workers & the tasks assigned to them?
Personally, I don't think documentation is the issue. The issues as more
to do with politics, power, and control.
The Web unleashed power, despite its marginal exploitation at best, that
terrifies many.
>
> by all means, that's a far easiest problem to solve.
>
> I'm mindful; that the shape of a spear, is similar to the shape of a
> pen; and is in-turn similar, to the shame of a mouse-cursor...
>
> maybe the solution can be made, in 'one click'..?
More or less, once you subscribe to the deep powers of URIs :)
>
> Timothy Holborn.
--
Regards,
Kingsley Idehen
Founder & CEO
OpenLink Software
Home Page: http://www.openlinksw.com
Community Support: https://community.openlinksw.com
Weblogs (Blogs):
Company Blog: https://medium.com/openlink-software-blog
Virtuoso Blog: https://medium.com/virtuoso-blog
Data Access Drivers Blog: https://medium.com/openlink-odbc-jdbc-ado-net-data-access-drivers
Personal Weblogs (Blogs):
Medium Blog: https://medium.com/@kidehen
Legacy Blogs: http://www.openlinksw.com/blog/~kidehen/
http://kidehen.blogspot.com
Profile Pages:
Pinterest: https://www.pinterest.com/kidehen/
Quora: https://www.quora.com/profile/Kingsley-Uyi-Idehen
Twitter: https://twitter.com/kidehen
Google+: https://plus.google.com/+KingsleyIdehen/about
LinkedIn: http://www.linkedin.com/in/kidehen
Web Identities (WebID):
Personal: http://kingsley.idehen.net/public_home/kidehen/profile.ttl#i
: http://id.myopenlink.net/DAV/home/KingsleyUyiIdehen/Public/kingsley.ttl#this
Received on Monday, 17 May 2021 23:32:18 UTC