#LinkedData
Looking for a trusted partner to navigate the complex world of connected data? We've got you covered!
Our team of experts is dedicated to delivering reliable and scalable solutions for your data challenges. https://www.epimorphics.com/
Woot!
Another update for the #VirtuosoRDBMS (Open Source Edition) is here! Keep in mind that all my live demonstrations, whether they use #SPARQL or #SQL to fine-tune #ChatGPT, are also compatible with this edition (for example, the 100 billion+ #Uniprot instance-related demos at: https://community.openlinksw.com/t/using-sparql-query-templates-to-fine-tune-chatgpts-large-language-model-llm/3842/5).
#RDF #GraphDatabase #SPARQL #LinkedData #SQL #GraphQL #SemanticWeb #KnowledgeGraph
We've been supporting the #UKGovLD #LinkedData #Registry #OpenSource codebase for years, also with some funding from key user organisations. If you are a user of the code & have requirements that you'd like to discuss and help fund then get in touch. https://www.epimorphics.com/projects/registry-core/
Pretty amazing: using Linked Data, SPARQL Construct en LD-Framing to create IIIF Manifest files:
Learn how #TriplyDB users implement #IIIF support by relying on open #LinkedData standards: https://triply.cc/blog/2023-06-iiif/
In the #ise2023 lecture today we are going to continue our adventure in the #SemanticWeb with RDF, RDFS, RDFS semantics, and RDF Reification
https://drive.google.com/file/d/1hCfE5sFzk4iO-i4yaRzj3AOUJJC97For/view?usp=sharing
#knowledgegraphs #lecture #rdf #linkeddata #lod @fizise @KIT_Karlsruhe #stablediffusionart #artbot #creativeAI #spock #renaissance #dürer
📢 Today the first edition of my new weekly newsletter FAIR Data Digest was sent out!🎉 The 1st edition covered recent conferences and links to online keynotes that highlight the need for (formal) Knowledge Graphs. https://sven-lieber.org/newsletter #LinkedData #FAIRdata #DigitalHumanities
Over the last month we highlighted some of the #GovTech data services we run for the UK public sector.
From #Hydrology to #FoodAlerts via #HousePrices, #BathingWaters and #Registers take a look at: https://www.epimorphics.com/projects
If you'd like to know more or have a project idea that you'd like to chat through - then drop us a line https://www.epimorphics.com/contact/
I love what dbpedia brings to the world of linked open data. For example a great enrichment for https://data.digitopia.nl/id/p/p068435061
**Problem:**
#LLM based #SmartAgent solutions like #ChatGPT are fantastic natural language processors, but challenged by unpredictable guessing.
**Solution:**
Fine-tune using #SPARQL query templates and a #LODCloud #KnowledgeGraph like @Uniprot.
#SemanticWeb #LinkedData #HowTo #LLMs #Uniprot #Bioinformatics #Genomics #MolecularBiology #LifeSciences
Comparison Round 2:
This time around, I used the following prompt for the comparative challenge:
Describe the book with ISBN 9781610395694 and associated offers using JSON-LD.
Session Transcript & Analysis: https://community.openlinksw.com/t/comparison-google-bard-vs-microsoft-bing-chatgpt-vs-openai-chatgpt/3865#book-description-by-isbn-comparison-1
#SemanticWeb #LinkedData #JSONLD #LLMs #Bing #Bard #ChatGPT #Comparison
@david @mauve @neuralreckoning
absolutely :) browse around some wikidata pages and you'll also see heterogeneous data and provenance - that's sort of the name of the game with linked data. Wikidata is specifically very good with translation and source references. They are also (somewhat implicitly) good with tracing the evolution through time (same "view history" page as is present on wikipedia/mediawiki)
wikipedia/wikidata have a complicated relationship with surfacing the kind of dialogical process you're describing, even though they are of course built on it. Sorry for bombing you with more links but if it's useful i have written about that here: https://jon-e.net/infrastructure/#the-wiki-way
but i couldn't agree more how important it is to represent as a first-class part of the data. You're also right on that modeling documents is definitely less well-defined than genealogical relationships (eg. saying above how tedious it would be to manually annotate all the topics in some writing, totally right), and wikidata doesn't necessarily provide the perfect set of tools for doing that, but yes I plan to work on that problem after I get a draft of a p2p backend for sharing and communicating and negotiating over these kinds of packetized-graphlike information blobs <3
#SemanticWeb #LinkedData #WikiData
@mauve @david @neuralreckoning
Genealogy is a graph with people as nodes and relationships as edges! This is actually one thing that semantic web/linked data technologies are already really good at, so for example see the wikidata project on Genealogy.
Genealogy actually is the most common thing in their list of examples of visualizing wikidata.
So for example if you go here https://www.entitree.com/en/family_tree/Q4604 and keep clicking the "down" arrows, you can trace Confucius' family tree to the present day.
You can see how this works at the wikidata page - he has a child property, in addition to many others.
Accommodating growing/changing/incomplete/overlapping schemas is exactly what #SemanticWeb and #LinkedData technologies are good at - I explore that and the complex politics that surround these technologies, particularly corporate #KnowledgeGraphs in this piece: https://jon-e.net/surveillance-graphs/
specifically here and here though as I said it's one of the basic topics of the whole piece. That naturally contrasts with traditional relational databases, which, as u all are saying, can be brittle and difficult to refactor.
I also discuss the complex and fraught relationship between knowledge graphs and chat interfaces in that piece: https://jon-e.net/surveillance-graphs/#the-near-future-of-surveillance-capitalism-knowledge-graphs-get-chatbots
and spoiler alert automatically populated graphs are, uh, pretty bad, but KGs + #LLMs is the direction that the information conglomerates are headed in a much deeper and terrifying way than I have seen in the general discourse (that's why i wrote the piece in the first place i guess).
Interface design has been one of the major problems in the history of semweb/LD (in my opinion, next to the social systems and politics, it is the largest problem). I explore that history and the constraints a bit in a prior piece, here, and talk a bit about some potential roads forward here.
Ingestion is also a huge problem! and that is partially because most of the tooling has been designed for massive information conglomerates to guzzle down the semi-structured information on the web rather than for individual people to make sense of information they care about. So eg. existing tools like Neo4j or other graph databases are pretty dang tricky to get set up, but imo i'm pretty sure the target data format you're looking for is a graph of triples, and once that's in a serializable format like turtle or n-triples or whatever I'm pretty confident you won't "regret it" or that the effort will have been wasted. You might have to shop around a bit for the appropriate visualization/storage/query tech stack, but that's the neighborhood to look in. One example is wbstack which is intended to make hosting your own instance of wikidata (wikibase) more straightforward if you don't want to shove all your data in public onto wikidata.
Marking this public so i can search for it later because these questions come up frequently :)
Highlighting #GovTech data services: with the Food Standards Agency (FSA) and their Food Types Register project to showcase the FoodEx2 standard as dereferenceable linked-data - providing info about the composition of different food products
https://www.epimorphics.com/projects/food-types
#FoodTypes #FoodEx2 #LinkedData #URIs #DataStandards #PublicSector #Referenc
Sitt Min Oo, another IDLab colleague, is also presenting their #PhD at the #ESWC2023 PhD symposium! He is working on integrating streaming data with #RML and any-to-any data format #mapping!
#RDF #SemanticWeb #LinkedData #KnowledgeGraph @eswc_conf @pietercolpaert
My colleague at IDLab Jonni Hanski is presenting his #PhD at the #ESWC2023 PhD symposium! He mainly focusses on optimizing link traversal #query processing over #distristributed #LinkedData like #Solid
@eswc_conf @pietercolpaert
#RDF #decentralized #SemanticWeb #LinkTraversal
Paradigm change in data publication for the #humanities: from plain printed text to automated knowledge discovery and #ai. Eero Hyvönen is presenting „Plenary Speeches of the Parliament in #finland as #linkedopendata and data services“ at the #text2kg workshop at #eswc2023
Paper: https://drops.dagstuhl.de/opus/volltexte/2021/14544/pdf/OASIcs-LDK-2021-8.pdf
#PerplexityAI provides a cool productivity tool for interacting with the #GPT4 #LLM that enables me share my session using hyperlinks.
Courtesy of the service provided by perplexity.ai, the process is now reduced to:
1. Write prompt and execute
2. Share session link.
Example covering LLM fine-tuning examples using #SPARQL templates.
[1] https://www.perplexity.ai/search/6543ca49-14c6-4b44-89b3-61f0abf7b1f9?s=c -- exploring Recipes collated in our #URIBurner Knowledge Graph
Felicitas Löffler vom Thüringer Finanzministerium berichtet beim 1. Kongress zu offenen Verwaltungsarchitekturen vom Aufbau von Wissensgraphen, semantische Daten & Linked Data in der Verwaltung. Wow! #linkeddata #opendata #wikidata #openDVA #dcatap
My #Solid client and server are now ready for efficient access control demos on #BigData. Let me know if you are interested in the comments below.
I can demo with a server publishing N resources (in this case #LinkedData Event Stream (#LDES) data.
The client could need to make no more than N+2 requests:
1. Request 1 on a resource R returning a "401 Unauthorised"
2. a max of 2 requests to get the access control rules
3. from there on, N signed requests using HttpSignatures (when those all fall in the same container space)
Solid clients are essentially like Search Engine crawlers fetching data on the web, so they need to jump around from website to website. Having approx 2 requests extra per website for auth is very interesting in that scenario. Note: those 2 requests can be cached, so those may be only needed once over a long period of time.
The connection efficiency is possible by combining the following pieces:
* using the IETF's HTTPSig (a version from the beginning of the year)
* using default rules (part of the spec)
* caching of ACLs on the client
* the use of a "defaultAccessContainer" link header to reduce the number of requests.
The client is implemented in #Scala using #http4s, and the server uses #Akka . The libraries can be compiled to JS for use on #nodeJS frameworks too. Native is not far off either.
How to use #SQL query templates to fine-tune #ChatGPT.
https://community.openlinksw.com/t/howto-fine-tuning-chatgpt-using-sql/3862
All possible because of the underlying architecture of our multi-model #VirtuosoRDBMS equipped with native #HTTP support.
#SemanticWeb #SPARQL #DBMS #RDBMS #LinkedData #CDO #CIO #CTO #AGI
Next in our series on #GovTech data services, we highlight work with HM Land Registry on Standard Reports.
Our expertise in Data Modelling, Linked Data and Data Publishing helped create a key resource for property buyers and analysts.
See: https://www.epimorphics.com/projects/standard-report-builder
#GovTech #LandRegistry #StandardReports #PropertyData #OpenData #DataModelling #LinkedData #DataPublishing
#Google #Bard wins #SemanticWeb functionality precision test, compared to #Bing+ChatGPT and #OpenAI #ChatGPT (with #Web crawling enabled).
Prompt Example:
Describe https://developer.apple.com/wwdc23/3#thisEvent -- link that denotes upcoming #Apple #WWDC2023 event.
Today in the #ise2023 lecture we will start the chapter of #knowledgegraphs. Students will learn essentials of knowledge graphs packed into 4 lectures.
https://drive.google.com/file/d/1MteCSMzKngIPN2y5aJGsWTUmU8LFFVRH/view?usp=sharing
#semanticWeb #lecture #linkeddata #spock #creativeAI #stablediffusionart
Woot!
Another wonderful release of our modern multi-model #DBMS for operating on relational tables and entity relationship graphs.
BTW -- #ChatGPT, courtesy of its underlying #LLM and #ConversationalUI, is a great companion for utilizing this unique platform -- as recent posts demonstrate.
#LinkedData #SQL #SPARLQ #GraphQL #SemanticWeb #KnowledgeGraph
This week we continue our series of highlights of some of the #GovTech data services we run for the UK public sector.
#ReferenceData #OpenData #DataScience #LinkedData #DataStories
Comparative web page crawling exercise, taking into account the following factors:
[1] The emergence of a #SemanticWeb where web pages increasingly embed RDF-based structured data islands as enhanced metadata.
[2] The web crawling functionality provided by #ChatGPT, #Bing, and #Bard.
Next in our series on #GovTech data services, we highlight our work with HM Land Registry on the Price Paid Data (PPD).
Our expertise in Data Modelling and Linked Data helped create a valuable resource for property buyers and analysts:
https://www.epimorphics.com/projects/price-paid-report-builder
#PropertyData #PPD #OpenData #PricePaidData #DataModelling #LinkedData
I’ve finally been granted access to the #Web Browsing Plugin for #ChatGPT, enabling me to test Web Page lookups en route to #KnowledgeGraph generation.
Via https://github.com/mmmenno/nl-straatnaamvarianten zijn nu ook straatnaamvarianten (incl. schrijf-, transcriptie- en andere fouten) van Haagse en Goudse straten te vinden. #LinkedData
Celebrating the new Bathing Water Season in Wales with the Natural Resources Wales Bathing Waters service that we highlighted in a recent post in our #GovTech data services series
This week we continue our series of highlights of some of the #GovTech data services we run for the UK public sector.
From #WaterQuality, through #HousePrices and #ReferenceData.
#OpenData #DataScience #LinkedData https://www.epimorphics.com/projects
Derudas' PhD project developed, tested, rejected and redeveloped solutions for making 3D documentation of excavations more useful to archaeologists. Her final product, AIR (Archaeological Interactive Report), is built on #OpenSource #OmekaS code, #CIDOCCRM, and integrates #LinkedData.
More work is necessary, but great example of #FAIRdata implementation
#Archaeology #DigitalHeritage
https://omeka.ht.lu.se/s/reports/page/home
There's a powerful symbiosis between #ChatGPT a #SemanticWeb (public or private) that is not immediately apparent. This symbiosis exists because ChatGPT is inherently data-dependent, while a Semantic Web consists of boundless structured data constructed using machine-computable entity relationship types that are constructed from hyperlinks.
Here's a new #HowTo and live demos centric post about this issue.
We hope you had a lovely weekend, this week we continue our series of highlights of some of the #GovTech data services we run for the UK public sector.
Q: Can I fine-tune #ChatGPT using #SPARQL such that it basically becomes a #SmartAgent capable of querying across the massive #LODCloud #KnowledgeGraph, using natural language?
Yes!
Read: https://community.openlinksw.com/t/howto-chat-gpt-and-dbpedia-sparql-query-generation-from-natural-language-prompts/3823 -- which covers a basic example using #DBpedia (#Wikidata content in machine-computable and queryable form).
#SemanticWeb #LinkedData #JSONL #RDF #LLMs #DBpedia #VirtuosoRDBMS #JSONL
#ipld is yet another #Linkeddata format but for content adressable data in decentralised systems. DWN use IPLDs as a schema
https://www.youtube.com/watch?v=totVQXYS1N8
Another #SemanticWeb + #ChatGPT showcase:
Executing a #SPARQL query (informed by a template) without a plugin across the #LODCloud, courtesy of #LinkedData principles.
Wikidata Queries can generate timelines of when artworks owned by someone with some kind of link to a claim for restitution entered museum collections.
This query is filtered for after 1932 and for six museums only.
Try it!
https://w.wiki/6d48
#wikidata #histodon #digitalhumanities #linkeddata #museum #provenance #Sparql #timeline #art #arthistory #dataviz #jewish #history
A recurring #LinkedData / #RDF question: In German, French and many other languages, names for roles, occupations etc. encode the genus/(grammatical) gender, e.g. "Autor"/"Autorin" for "author". Lots of controlled vocabularies contain such roles. Is there a best practice to mark different labels (m, f, gender-neutral) in RDF, especially #SKOS?
It seems to me that #web3 types should be the strongest supporters and adopters of the #SemanticWeb (#LinkedData), because machine-readable facts reduce the need for (human) #oracles when settling #SmartContracts.
(And for some reason I suspect #crypto people may be the least interested in that kind of thorough web hygiene.)
@J12t ,
Sets up stage for understanding and appreciating the semantics associated with entity identifiers.
For instance, you ultimately end up with an entity relationship graph comprising a variety of handles that denote the same entity.
In a #SemanticWeb this is ground-zero, since relationship type semantics exist for this kind of identity reconciliation 😀
A friendly reminder that you can still complete the "Characterising the #IIIF and #LinkedArt communities" survey that I am doing for my PhD. --> https://forms.gle/Djvuf7HrrFqxXFxd8
Please fill in the survey if you are involved in either of these communities or if you have heard of those communities but aren't (yet) involved. 😃
#LOUD #LinkedData #culturalheritage #GLAM #CIDOCCRM
We want to give a shoutout to Stuart Williams for his incredible work on #LinkedData and #DataModelling projects with organisations such as the Environement Agency Food Standards Agency, The National Archives, BRGM in France, the JRC and many more.
Thank you for your dedication to advancing the field of #OpenData, Stuart! #DataStar 👏
@kidehen@fediverse.demo.openlinksw.com @atomicpoet @photocyte @helge,
Naturally, you can also interact with #ActivityStreams docs from my #ODSBriefcase hosted outbox using cURL.
Example:
curl -ILk https://fediverse.demo.openlinksw.com/DAV/home/kidehen/outbox/X4esBpMWizx717FT4THFvjR6f6o
#ActivityPub #DataSpaces #LinkedData #RWW #SemanticWeb #SocialMedia
@kidehen@fediverse.demo.openlinksw.com @atomicpoet @helge@mymath.rocks,@photocyte ,
Screenshots from the SPA that will soon be released to a #Github repo.
Note:
1. Post login it figures out the hyperlink that denotes the users #ActivityPub outbox
2. Authentication is multi-protocol courtesy of #OpenIDConnect which provides an open mechanism for loosely-coupling a variety of Identity Providers across a variety of protocols (including #OAuth)
3. #ActivityStreams objects input directly
Looking to partner on a data-driven project? Look no further! Epimorphics has a wealth of experience in data standards, #LinkedData, #DataIntegration and #DataDesign, and we pride ourselves on being great collaborators.
Let's chat! #DataCollaboration https://www.epimorphics.com/about/
our library is looking for a linked data expert: https://www.academictransfer.com/en/325443/linked-data-specialist/
Join our awesome #Maastricht with very lively #OpenScience #LinkedData community with various research groups that have been using Linked Data for at least a decade or longer
Active #SPARQL Query Services across the #LODCloud, courtesy of a #Wikibase query:
[1] https://tinyurl.com/2h7dze9u -- Tables
[2] https://tinyurl.com/2h7dze9u -- Bubble Chart
A SPARQL Query Service endpoint provides access to an #HTTP-based Query Language, Wire Protocol, and Negotiable Serialization Format combo that's fully understood by #ChatGPT.
New blog post about the newly released @youid 1.6.9 that now includes simplified Link In Bio Profile Doc generation without any cross-reference limits. Naturally, these docs embed a powerful Verifiable Credentials #KnowledgeGraph.
Another step towards pseudonymous #Identity #Authenticity that scales to both the #Internet and #Web.
Implementing the #LinkedData "follow your nose" handling fully requires, more or less, for every #ActivityPub system to handle data in a way that there are no patterns for and no real libraries supporting.
So to whatever degree it is practically _feasible_ for a single project, I think it is probably too big of a lift to ask _everyone_, regardless of tooling or experience, to behave that way.
But if you have a handshake—something we do have patterns around? You can have a little LD as a treat
One thing I've been chewing on as a thought experiment is "how do you evolve a protocol like #ActivityPub."
At its core I suspect some form of Server-Server handshake will be required.
At the moment there's a very strong "follow your nose" concept tied into some ideas around #LinkedData. Bluntly: I do not think this can be practically implemented due to a combination of constraints and requirements.
But a server-to-server handshake that says "this is who I am and what I do" could help.
@smallcircles @openlink @obale @J12t I will when its ready for public testing. Right now, its in the final stages of internal beta testing etc..
To get a feel, here are some links:
[1] #WebFinger handle -- @demo
[2] #Fediverse as part of a #SemanticWeb view -- https://fediverse.demo.openlinksw.com/describe/?url=http%3A%2F%2Ffediverse.demo.openlinksw.com%2Fdataspace%2Fperson%2Fdemo%23this&distinct=1 (you can follow-your-nose for deeper exploration by just clicking on a #hyperlink that denotes an item of interest).
OK here's another half baked #MastoDev idea for #LinkedData:
a URL can be used with a preceding hashtag to refer to some specific entity, whether it's a well-defined #RDF object or not:
eg.
Banana:
#https://wikidata.org/wiki/Q503
identifiers can be given shortnames like this:
#Banana::https://wikidata.org/wiki/Q503
or with wildcards, eg. for this podcast:
#CriminalPodcast::https://thisiscriminal.com/*
so then when I search #CriminalPodcast I also get all posts that use that URL. shortnames can be retained and reused either just by the account that declared them, or by the instance.
shortnames can be used like prefixes, so eg if I do:
#schema::https://schema.org/
#skos::http://www.w3.org/2004/02/skos/core#
I can then do
#schema:Person
#skos:sameAs
So then I can declare types like this:
#schema:PodcastSeries::CriminalPodcast
and make triples like this, eg. in case someone used a different hashtag:
#CriminalPodcast::skos:sameAs::Criminal
so then when you search a hashtag, at the top you would be shown all the links that have been declared for that and who declared them, so you could eg. browse to find more things with the same type, or more hashtags that are related to each other, etc.....
@jackgilmore Great!
Here's a query results page reusing your query (using generic #SPARQL) against our #VirtuosoRDBMS hosted #Wikidata instance.
Note, @datasniff is the tool used for query editing directly from the current query results page.
May we present: automatically recognised people, organisations, laws and terms in parliamentary debates! 🎉 This feature is still in an experimental stage, but you should totally check it out right now at https://de.openparliament.tv.
Nice #Wikidata use-case demo.
SeeAlso, using our #VirtuosoRDBMS hosted edition of Wikidata.
RT @EULawDataPubs: Cultural institutions in the #Netherlands are encouraged to publish data as #LinkedData and to assign standardised terms. @ennomeijers @KB_Nederland proposes the ‘Network of Terms’ to search Linked Data sources in a federative way
Join us at #ENDORSE2023
https://op.europa.eu/en/web/endorse?pk_campaign=ENDORSE2023_speaker&pk_medium=tw
🐦🔗: https://n.respublicae.eu/EU_opendata/status/1635046199597350914
Ok no I think the simpler way to do #ZenodoSocial is actually to just like generalize the types of messages you can send using #LinkedData / #JSONLD. I think one might call these #UserSchemas.
re: https://neuromatch.social/@jonny/110001265968620084
Make a schema that defines a type of message, eg with JSON Schema in RDF
Make an API endpoint that lets logged in accounts submit schema, as well as an {id} to use for a URL. if it is not already taken, that schema now can be found at instance.domain/schema/{id} and used in JSON-LD as a @context
Add a field to the API endpoints for posting statuses and fetching the home timeline that allows you to specify the schema that you're using when you post (ideally either local or by any resolvable URI, tho dealing with generalizable RDF parsing makes me gulp so maybe just start with JSON-LD schema). Don't deliver user schema messages to instances that don't support them or base masto accounts that have not specifically subscribed to them - see below. Alternatively just post JSON-LD from a base masto account and specify the type within the object.
Add a field to the instance metadata that specifies that you support user schema.
Backwards compatibility: generally, from a base masto instance one would make a designated alt that posts/receives user schema messages. Base masto accounts can specify an alt with profile metadata fields. User schema instances and other clients that use user schema know to resolve those fields and follow/query them. To follow user schema posts from a base masto account, one sends a structured DM to indicate that user schema posts should be DMed to it. clients know to expand threads to capture JSON-LD split across multiple posts
Make integrations with arbitrary other software that lets you swap structured data on your existing social infrastructure????? my first trivial example being to swap highlights and annotations in Zenodo, but that might be enough to pry more generalized linked data handling into masto without making a true plugin interface or completely rewriting it?????
@J12t "JSON-LD has benefits that have nothing to do with semantic web stuff, such as namespacing."
The following is ground-zero fora #SemanticWeb constructed from structured data deployed using #LinkedData principles.
## JSON LD Start ##
{
"@context": "https://schema.org/",
"@type": "Person",
"name": "John Smith",
"jobTitle": "Software Engineer",
"telephone": "(123) 456-7890",
"url": "https://www.example.com",
"sameAs": [
"https://mastodon.social/@jsmith"
]
}
## JSON LD End ##
There is a complete provenance trail for all the data in my #SemanticWeb example.
Examples plus associated screenshot.
[2] https://linkeddata.uriburner.com/fct/facet.vsp?qxml=%3C%3Fxml%20version%3D%221.0%22%20encoding%3D%22UTF-8%22%20%3F%3E%3Cquery%20inference%3D%22%22%20invfp%3D%22IFP_OFF%22%20same-as%3D%22SAME_AS_OFF%22%20view3%3D%22%22%20s-term%3D%22e%22%20c-term%3D%22type%22%20agg%3D%22%22%20limit%3D%2250%22%3E%3Ctext%3E%23SemanticWeb%3C%2Ftext%3E%3Cview%20type%3D%22text-d%22%20limit%3D%2250%22%20offset%3D%22%22%20%2F%3E%3C%2Fquery%3E&timeout=16000 -- 2,765 pages comprising items and Named Graphs (data sources) for "#SemanticWeb" text pattern
That's also obtainable via a #SPARQL query that includes Named Graph Identifiers in the solution output.
/cc @J12t @datasniff
@bobwyman @J12t Yes, a #SemanticWeb user-agent can handle that.
For example, the sponger #middleware module of our #VirtuosoRDBMS can do that.
Re Prospective Search, the focal point is variables in the body of a #SPARQL query that satisfy conditions expressed in relations.
A query solution can always optionally include document identifiers (a/k/a named graphs).
Example: https://tinyurl.com/4kwzvu28
@bobwyman @J12t The hashtag #LinkedData is denoted by the link https://mastodon.social/tags/linkeddata.
Add .rss to that link and you have:
https://mastodon.social/tags/linkeddata.rss -- and #RSS doc.
Semantically, the following holds true (albeit currently missing from the <head/> section) of https://mastodon.social/tags/linkeddata.
{
<https://mastodon.social/tags/linkeddata> xhv:alternate <https://mastodon.social/tags/linkeddata.rss>
}
Here's a very basic sponge (crawl) of content associated with #LinkedData from across Mastodon.
https://linkeddata.uriburner.com/about/html/https://mastodon.social/tags/linkeddata.rss
@bobwyman @J12t In addition, here are links to posts about the #VirtuosoRDBMS sponger module (our #SemanticWeb crawler) .
[1] https://medium.com/virtuoso-blog/what-is-the-virtuoso-sponger-middleware-about-and-why-is-it-important-293546b32a32 -- #VirtuosoRDBMS sponger (crawler) overview
This technology can crawl and merge data across links that denote hashtags across Mastodon instances, Twitter etc.. What it can't do is construct a canonical solution from an infinite universe of unknowns.
@bobwyman @J12t Prospective Search [1], is something that can be handled in a #SemanticWeb because the semantics of links (used to construct said #Web) enable a crawler to be fashioned within constrains (e.g., timeouts) while being informed navigationally by the semantics of relations denoted by links, courtesy of #LinkedData principles.
The above can happen within the context of a query or as an ETL-style background crawl.
Der #BullingerBriefderWoche stammt heute von Jakob Meyer aus #Basel und wurde am 8. März 1538, also vor 485 Jahren in Frühneuhochdeutsch verfasst. Der Zeitstrahl zeigt, dass Bullinger und Meyer sporadisch zwischen 1532 und 1539 in Kontakt standen. #digitalhumanities Zu Meyer findet sich nicht viel, aber sein Sohn (nehme ich mal an) hat eine Seite auf Wikipedia: https://de.wikipedia.org/wiki/Wolfgang_Meyer_(Theologe) Es wäre doch interessant, wenn es gelänge, diese Informationen zu verknüpfen. #LinkedData Meyers Grossvater wahr denn auch Bürgermeister von Basel & ein Unterstützer des Basler Reformators #Oekolampad, von welchem wir ebenfalls einen Brief in der Bullinger-Korrespondenz finden: https://bullinger-digital.ch/letter/10063. Zum Meyer-Geschlecht gibt es auch einen Artikel auf e-periodica: https://e-periodica.ch/digbib/view?pid=bzg-002:1925:23#110 Wir müssen diese Daten vernetzen!
I don’t know how any crawler solves the problem as described .
If I am interested in #LinkedData across instances that l encounter explicitly or serendipitously, our crawler can follow objects of owl:sameAs relations to wherever they lead, as part of a *timeout* constrained query solution pipeline or an explicit ETL-based crawl job .
The difference from conventional crawling boils down to an operation informed by the semantics of an owl:sameAs relation.
/cc @J12t
@bobwyman @J12t Query solution informed by inference can include crawling a #SemanticWeb as part of a query solution production pipeline, I've been demonstrating that for eons now.
Crawling a #SemanticWeb is a different kind of crawl informed by the semantics of relations denoted by hyperlinks, courtesy of #LinkedData principles 😀
I have written a slightly rambling #blogpost about building my band's new website https://sam.pikesley.org/blog/2023/02/18/massively-over-engineering-a-static-website/
@Edent @aslakr With JSON-LD contexts becoming ubiquitous due to their SEO capabilities and search engine integration, there’s hope JSON-LD can serve as a replacement for Open Graph.
JSON-LD has the benefit of being maintained by W3C, being an open standard, having significant amount of momentum, and serving a plethora of different use-cases, many of which haven’t been invented yet.
#wroBookMark
#LinkedData
https://www.w3.org/DesignIssues/LinkedData.html
+ [links on the semantic web](http://dig.csail.mit.edu/breadcrumbs/node/62)
+ [give yourself a URI](http://dig.csail.mit.edu/breadcrumbs/node/71)
+ [backward and forward links in RDF just as important](http://dig.csail.mit.edu/breadcrumbs/node/72)
https://www.w3.org/TR/activitypub/
Find Twitter or microblogging mentioned anywhere in this specification? This deals with #SocialWeb, protocol and #LinkedData extensibility 🙂
If someone wants to build their Twitter, the spec might be informing their tech stack.
the problem with #LinkedData technologies is that none of them work.
"here's a semantic web index, it's like a google for linked data"
https://sindice.com/index.html
Uh, what? No it's not. where do I search?
The only thing I see is API docs, surely that can't be right.
OK let's at least take a look anyway
https://sindice.com/api_subdomain/v3/search_fq_format_rdfa_fq_class_foaf_person_format_json.html
god almighty what on earth
"here's a tool for extracting RDF-like data from web pages"
https://github.com/apache/any23
OK, apparently it's a web service? can I try it out?
https://any23.apache.org/
no, not like that, not by downloading and compiling anything
halfway down the page, ok a live instance
http://any23.org/
(not a live instance)
I would love to see one linked data tool that isn't wikidata that a) tells me what it does, b) shows me a demo, c) lets me use it without having to read a whole book or compile some ancient blob of Java
ok so re-reading #IPFS paper and there are a few things I think in retrospect are undesirable about the #MerkelDAG spec. it's hard to parse them out as separable ideas because they depend on one another, but the main thing I think is how it conflates the structure of a metadata graph, the content of the graph, and the notion of authorship/identity.
In (basic) IPFS, each node contains some data and some links. the data is some unspecified binary blob, the links are all references to hashes of other nodes, and then the hash of all that identifies the node. There are some abstractions like flattened trees that can represent n-depth links, but that's the gist. I'm refreshing myself, so correct me where I'm wrong.
This makes traversing the graph expensive from a naive (cacheless) state- you have to fetch each node and parse its links serially, and since there isn't a notion of authorship except when used to sign a node, you might have to do the resolution process across a lot of the network instead of being able to say "ah ok this is from this identity so I should ask their neighborhood first"
Since the links are untyped, and because of the need for serial resolution, you can't really "plan" queries and move the query logic to the "edges" (in a networking, rather than graph parlance) of the network - the network resolution logic handles all that.
This structure also makes it so you can't "talk about" a node. A node contains its links. The links are directional, so I could make some statement about a node by pointing to it, but I can't, as a third party make a link under my identity, separate from the author and content of the node, that points from some object to another. That makes the network more like a hard drive than a social space.
Further, since links aren't typed, you have to move that metadata inside the node. This makes you need to re-hash each node more than you need to, and since "keys" for identifying different fields in the node aren't themselves links, you can't have any notion of "schema" where a term can be reused. So there isn't really a facility for being able to do graph queries like "find me this type of data whose field has this value" which restricts a whole huge range of possibilities too long to list here. This also makes knowing what the binary data inside a node is potentially impossible without out of band info, depending on how it's encoded. #IPLD and #Multiformats are intended to solve this, post-hoc.
I'll stop there for now, and save what I think could be a different model for later, but I am thinking along the lines of merging with #LinkedData #Triplets , encoding the notion of authorship into links (so that links can have an "utterance" rather than "fact" ontological status), a notion of container/contained for explicit block formation and metadata separation, and formalizing the notion of orthogonal Merkel DAGs to change the points where the content addressing happens to be able to have "graph subunits" that allow for cycles at a "complete" scope but for the purposes of hashing have no cycles. very much #WIP, still at conceptual stage haven't started writing spec yet.
For #ActivityPub you can arguably say that that also sort of applies.
Anyway, I don't have very strong opinion on whether or not AP should be a #LinkedData standard or not.
What I do know is that LD has this eternal struggle to gain adoption, and in the areas where its power lies, its complexity comes peeking around the corner. Making it very unloved in the broader developer community.
In theory LD is fabulous, leading to a magical wonderful semantic web, and in practice.. 🤔
In the context of responding to @helge #FEP proposal I bumped into a paper by Samuel M. Smith about the role of #LinkedData wrt the #W3C #VerifiableCredentials spec and thought about your earlier musings when exploring #ActivityPub and how #JSONSchema might have been used.
Sam's TL;DR is:
> The VC standard appears to be an adoption vector for Linked Data, not the other way around.
Might find it an interesting read:
https://github.com/SmithSamuelM/Papers/blob/master/whitepapers/VC_Enhancement_Strategy.md
And he claims the same applies to W3C #DID.
#Introduction
(after moving)
I'm a #Histodon based in Berlin (Humboldt-University) doing #DigitalHumanities.
Toots and boosts in English and German.
#AntiFascist, #queer and pro #OpenData and #OpenAccess: “Sharing isn’t immoral — it’s a moral imperative”
Aaron Swartz, Guerrilla Open Access Manifesto, 2008.
https://openbehavioralscience.org/manifesto/
My interests include #GIS, #SemanticWeb / #LinkedData, #TextCorpora —
'We have nothing to lose but our inexperience!'
My blog'ish website:
https://schoeneh.eu
@bobwyman @hrheingold FYI:
Installing the @datasniff browser extension offers direct parsing and browsing of the #JSONLD based #ActivityStreams graph i.e., just click and then follow-your-nose 😀
@evan @chrismessina The degree to which #Web20 companies sought to colonize the #Web never seizes to amaze me.
A hashtag is the fundamental unit of Web Magic, once understood. Basically, name things using hyperlinks (ideally, #HTTP variety) and connectivity magic happens, at Web-scale.
The #Fediverse itself is the latest demonstration of said magic, courtesy of #ActivityStreams 😀
Congratulations @noeldemartin for a great presentation! You are making #Solid accessible to people by showing real apps. For the #LinkedData developer ecosystem to improve, it is so important if devs get inspired by things and start to build stuff.
Here's a single #hyperlink that denotes the @anildash post that kicked off this thread.
http://linkeddata.uriburner.com/about/html/https/me.dm/@anildash/109753735569824671
That's a fully fledged data source name that delivers structured data to user-agents in a variety of negotiable formats.
Just click to explore 😀
/cc @atomicpoet
A cocktail of tools are required for fight against #phishing attacks, including an ability to de-reference and process an entity relationship graph comprising machine-computable entity relationship type semantics.
Please pass my response on to your external technical advisor since he knows exactly what I mean 😀
@tanepiper @GiorgiaLodi @pietervaneverdingen Wonderful!
Courtesy of our @datasniff browser extension, I can see how your #KnowledgeGraph is being integrated into #HTML docs published to your website.
Example page:
https://www.ikea.com/us/en/p/rengjord-built-in-dishwasher-90504572/
#SPARQL Query using that page as the Data Source Name (#DSN).
https://tinyurl.com/5bz5hxs4
#DataConnectivity #LinkedData #SemanticWeb #Web30 #UseCase #eCommerce
We (@openlink) are working on stuff from the #LinkedData and #SPARQL perspectives that include hooks into voice and text via CHAT-GPT shortcuts.
👩💻 Are you an expert in #MachineLearning & #LinkedData?
🗣️ And your superpower is explaining the interactions between computers and human language to the non-IT humans?
🔎 Check out this #vacancy and join #teamDIGIT! 👉 http://europa.eu/!bcQk66
alright I'm about to start working on some #LinkedData / #Graph-based p2p stuff, I am gonna spend some time digging deeper into #SPARQL, and I'm wondering if anyone can point me to any writing (blogs, papers, posts all good) on the good and bad from both a design and implementation POV. I can read the docs, but would like to learn the meta surrounding it from starry-eyed idealists and cranky engineers alike. Boosts for visibility welcome :)
@volkris @SrRochardBunson @atomicpoet,
Added an initial #Fediverse entry to our terminology glossary.
See: http://data.openlinksw.com/oplweb/glossary-term/Fediverse#this
Identifiers in this glossary are deployed using #LinkedData principles i.e., #Hyperlink based entity naming with in-built entity description document resolution -- in a variety of negotiable formats.
Yes!
Elon Musk and his crew (major Web 2.0 siloists and stalwarts) inadvertently blew up #Web20, moving attention to the burgeoning #Fediverse where the following are baked-in:
[1] Open Standards -- #ActivityStreams, #ActivityPub, #RSS, #HTTP, etc..
[2] #Identity and #SocialNetwork portability
[3] Loose-coupling and as a paradigm for application composition
/cc @atomicpoet @Mastodon
Cool!
BTW -- I just used #gptChat to generate a machine-computable description of "La Terminal" using content from its webpage.
Result, which you can just embed in said page using <scrip/> within <head>: https://tinyurl.com/ymsfbtue
/cc @Mastodon
Cool!
BTW -- I just used #gptChat to generate a machine-computable description of "La Terminal" using content from its webpage.
Result, which you can just embed in said page using <scrip/> within <head>: https://tinyurl.com/ymsfbtue :)
/cc @Mastodon
Here's a #screencast demonstrating how #gptChat generated accurate #RDF from a text snippet in a blog post about how to use the @openlink #ODBC Driver for #PostgreSQL on #macOS.
Note, GPT-Chat is integrated into a soon to be released edition of our @datasniff browser extension.
/cc @Mastodon
#SemanticWeb #UseCase #SmartAgent #Web30 #SPARQL #DataConnectivity #LinkedData
The FAQ text came from https://help.openai.com/en/articles/6783457-chatgpt-faq .
I visited openai.org and noticed that their #FAQ page doesn't include any modern #SearchEngine readable #metadata (the kind constructed these days using terms from the schema.org vocab). Naturally, this lead to a #gptChat exercise for QA'ing soon to be released @datasniff integration along the following lines.
Prompt text: https://tinyurl.com/4acvy9pz -- via FAQ text copy & paste
Response snippet: https://tinyurl.com/bdhh8hhk
@datasniff extracts & presents the generated #RDF for reuse.
There's every chance the actual website never gets completed because I've burned all my energy on all this #LinkedData
I have solved (hah!) this intermediate problem by building a bespoke #Python SSG purely for the #LinkedData
https://github.com/rawfunkmaharishi/data/blob/main/README.md