Hi all,
We are coming very fast to the end of January, and it means there's only one more build to share with you this month, and this build is available to you starting today.
Unlike the previous build which had only a change in the search syntax (albeit an important one!), in this build you'll have a unique chance to take a look at the behind-the-scenes of your projects information space - the ontology.
We are coming very fast to the end of January, and it means there's only one more build to share with you this month, and this build is available to you starting today.
Unlike the previous build which had only a change in the search syntax (albeit an important one!), in this build you'll have a unique chance to take a look at the behind-the-scenes of your projects information space - the ontology.
Getting Started:
- If you haven't signed up for the Zet Universe Insider Program, start here.
- If you are already have Zet Universe, the build is available now. If your build is older than 5786.2836, grab the full installer. Otherwise, you can go to Settings --> Update, and click on "Check Now" button to get it now.
- Total download might range from a few hundred kilobytes up to 11MB, depending on the number of updates you've applied already to your copy of the Zet Universe Insider Preview.
- Once the build is downloaded, Zet Universe will begin the installation. Once it will finish, it will ask you to restart the app.
WHAT'S NEW?
Some of you might have noticed that if you would go to Settings-->Options and choose "Show Views pane demo user interface", you'll see the new pane in the left sidebar, called "Views". The currently selected view is called "Zoomable Space", and there are hints that more views, like "Project Hearbeat" and "Mind Map" will be also available. However, this is just a mockup right now, as these views are not yet a part of the everyday builds.
Well, starting with this build, we are glad to introduce a second working view to Zet Universe, and, surprise, it won't be any of these two additional ones, as shown in the demo UI. Instead, this new view, called "Ontology", will enable you to look at the behind-the-scenes ontology of your projects. You can either click on the "Ontology" button in the "Views" area of the sidebar, or...
Or, you could open "Properties" of any of the items in the project space, and click on the entity's Kind to see its place in the ontology of the system:
Ok, now that you know how to get into this new view, let me give a short introduction into what it is all about.
ONTOLOGY AND KINDS
One of the founding principles of Zet Universe was and is the idea that it is an environment in which one can combine data from different places together, for a variety of purposes. A typical example could be use of this environment to gather intel on a particular subject.
Imagine having a need to analyze your potential competitors in a particular market; then, in Zet Universe you can collect various things you've discovered during your research: links to the useful webpages, PDF files, PowerPoint presentations, pictures, and other things. No matter what kind of data you have, you can see it all in one place, and you can organize it visually, no matter of its origin.
Imagine having a need to analyze your potential competitors in a particular market; then, in Zet Universe you can collect various things you've discovered during your research: links to the useful webpages, PDF files, PowerPoint presentations, pictures, and other things. No matter what kind of data you have, you can see it all in one place, and you can organize it visually, no matter of its origin.
In this screenshot, you can see an example of how such a project space could look like for someone collecting an intel on intelligent personal assistants in general, and Microsoft's Cortana specifically.
Some of the items here are captured web pages, others are PDF files, others are videos and images. You can also see several Sticky Notes that capture key thoughts of the user on the subject, placed near the items of interest. User can quickly find any of the items here by using her visual memory, but also by using a full-text search as Zet Universe indexes everything that is added into it.
In order for Zet Universe to understand all of these different items, it has to understand what makes them different from each other. One of the ways to categorize them is to use file formats (e.g., jpg, png, bmp - images, mp4, avi - videos, mp3, wma - audios, etc.), but sometimes file formats are irrelevant, and sometimes we want to have more specific items, like "people", and "organizations". This means we need to build some sort of a data model that brings a categorization of data into "kinds" as we call them, that brings also a structure (e.g., "photo" is kind of "media", and "media" is kind of "kind"), and brings other important things such as properties (e.g., each "kind" has "title"), and relationships (e.g., a "document" kind can have an "author"). In general, such data models are called ontologies.
Some of the items here are captured web pages, others are PDF files, others are videos and images. You can also see several Sticky Notes that capture key thoughts of the user on the subject, placed near the items of interest. User can quickly find any of the items here by using her visual memory, but also by using a full-text search as Zet Universe indexes everything that is added into it.
In order for Zet Universe to understand all of these different items, it has to understand what makes them different from each other. One of the ways to categorize them is to use file formats (e.g., jpg, png, bmp - images, mp4, avi - videos, mp3, wma - audios, etc.), but sometimes file formats are irrelevant, and sometimes we want to have more specific items, like "people", and "organizations". This means we need to build some sort of a data model that brings a categorization of data into "kinds" as we call them, that brings also a structure (e.g., "photo" is kind of "media", and "media" is kind of "kind"), and brings other important things such as properties (e.g., each "kind" has "title"), and relationships (e.g., a "document" kind can have an "author"). In general, such data models are called ontologies.
Ontology, by a definition, is a formal naming and definitions of the types, properties, and interrelations of the entities that really or fundamentally exist for a particular domain of discourse.
Generally, it is expected that any ontology is to be designed in such a way that features of its model closely resemble the real world.
There are many applications of ontologies in the information science; for instance, ontologies are one of the foundational elements of the Semantic Web, and ontologies are also an amazing way to capture and formalize our knowledge of the particular problem space or domain. In many ways, ontology could be also called a vocabulary of sorts, and, indeed, a well-designed ontology resembles an encyclopedia.
Speaking of the encyclopedia, I recall the fictitious Chinese taxonomy of animals described by the writer Jorge Luis Borges in his 1942 essay "The Analytical Language of John Wilkins".
There are many applications of ontologies in the information science; for instance, ontologies are one of the foundational elements of the Semantic Web, and ontologies are also an amazing way to capture and formalize our knowledge of the particular problem space or domain. In many ways, ontology could be also called a vocabulary of sorts, and, indeed, a well-designed ontology resembles an encyclopedia.
Speaking of the encyclopedia, I recall the fictitious Chinese taxonomy of animals described by the writer Jorge Luis Borges in his 1942 essay "The Analytical Language of John Wilkins".
Wilkins, a 17th-century philosopher, had proposed a universal language based on a classification system that would encode a description of the thing a word describes into the word itself -- for example, Zi identifies the genus beasts; Zit denotes the "difference" rapacious beasts of the dog kind; and finally Zita specifies dog.
In response to John Wilkins's proposal, Borges described an example of an alternate and rather funny taxonomy, supposedly taken from an ancient Chinese encyclopedia entitled "Celestial Emporium of Benevolent Knowledge":
The list divides all animals into one of 14 categories:
The list divides all animals into one of 14 categories:
- Those that belong to the emperor
- Embalmed ones
- Those that are trained
- Suckling pigs
- Mermaids (or Sirens)
- Fabulous ones
- Stray dogs
- Those that are included in this classification
- Those that tremble as if they were mad
- Innumerable ones
- Those drawn with a very fine camel hair brush
- Et cetera
- Those that have just broken the flower vase
- Those that, at a distance, resemble flies
As it was said in of the beautiful works on the ontology of the everyday things, though, this fictitious “encyclopedia” captures the potential hubris and arbitrary nature of many of our concepts of “what there is.” The encyclopedia is amusing and idiosyncratic but not very useful except as a psychological profile of the apocryphal compiler of the work.
However, having a well-defined vocabulary is a fundamental step in solving our goal of enabling Zet Universe to combine data from the different data sources, and it's a great opportunity for us to explain our approaching to building such a vocabulary in Zet Universe.
However, having a well-defined vocabulary is a fundamental step in solving our goal of enabling Zet Universe to combine data from the different data sources, and it's a great opportunity for us to explain our approaching to building such a vocabulary in Zet Universe.
ZET UNIVERSE'S DYNAMIC ONTOLOGY
Zet Universe's vocabulary starts with the definition of a Kind. Previously, in the Alpha timeframe, we called it a Thing, but as our goal is to make our ontology connected to the RDF and standard ontology language, OWL, we decided to rename this element to Kind, and make it a subclass of the standard owl:thing. Now, what is the Kind thing all about?
We live surrounded with things. These things can be physical ones coming from our environment, or they can be products of our imagination, or, finally, be digital objects created as elements of virtual world built by computer applications, websites and games. Things we deal with, starting from documents and other files in our folders, to users in Facebook and emails in our inbox, seem to be natural to us. We constantly make different interactions with these things, switching from physical world to a virtual one back-n-force, referencing to them in our endless collaboration with friends and work colleagues. The patterns of interacting with digital objects were brought from the physical world by Internet and computer pioneers, and now these patterns are highly interconnected as we constantly transfer our experiences between our real and digital lives.
Instead of a computer strict understanding of classes and instances, Kinds in Zet Universe are more familiar to the users; and as in the Metro Design Language, kinds are content.
We live surrounded with things. These things can be physical ones coming from our environment, or they can be products of our imagination, or, finally, be digital objects created as elements of virtual world built by computer applications, websites and games. Things we deal with, starting from documents and other files in our folders, to users in Facebook and emails in our inbox, seem to be natural to us. We constantly make different interactions with these things, switching from physical world to a virtual one back-n-force, referencing to them in our endless collaboration with friends and work colleagues. The patterns of interacting with digital objects were brought from the physical world by Internet and computer pioneers, and now these patterns are highly interconnected as we constantly transfer our experiences between our real and digital lives.
Instead of a computer strict understanding of classes and instances, Kinds in Zet Universe are more familiar to the users; and as in the Metro Design Language, kinds are content.
THE RATIONALE BEHIND THE KINDS
In general, file formats are the way files are distinguished from each other, but not every single piece of information is stored as a file, and we as people don't usually think in terms of file formats. The ontology of Zet Universe, as it was said above, is based on the things we interact with on the daily basis, and thus its kinds represent the categories of these things.
By bringing the kinds into Zet Universe, we explain it how to deal with different kinds of things; a kind definition explains Zet Universe how to show it (visually) in the Zoomable space; it explains how to name this kind (e.g., a "Photo"); it explains how to open items of this kind; it explains how this kind is related to the others.
By bringing the kinds into Zet Universe, we explain it how to deal with different kinds of things; a kind definition explains Zet Universe how to show it (visually) in the Zoomable space; it explains how to name this kind (e.g., a "Photo"); it explains how to open items of this kind; it explains how this kind is related to the others.
KINDS IN ALPHA AND IN INSIDER PREVIEW
In the Alpha timeframe, we had only three kinds:
In the Insider Preview timeframe (which is running right now), we have a significantly expanded list of kinds, including but not limited to:
These are the built-in kinds of our vocabulary. They are, in many ways, a continuation of our thinking circa the Alpha Timeframe (Topic is now a Visual Cluster, File is either a Generic File or a Media, etc.), and they were influenced by the kinds defined by Windows operating system (and mapped to the same file extensions as in Windows). This means that if you got used to search for "kind:photo" in Windows Explorer to look for pictures, the same query in Zet Universe will yield the same results.
However, although we've been inspired by the Windows Shell ontology, our ontology is not a try to explain and define everything, and it's not carved in stone.
- Topics,
- Files,
- Web Pages.
In the Insider Preview timeframe (which is running right now), we have a significantly expanded list of kinds, including but not limited to:
- Agent (Organization | Person | People Group | User)
- Calendar Entry (Appointment | Meeting)
- Communication (Call | Email | Instant Message)
- File Folder
- Generic File
- List
- Media (Document | Music | Picture | Video | Web Page)
- Project
- Visual Cluster
These are the built-in kinds of our vocabulary. They are, in many ways, a continuation of our thinking circa the Alpha Timeframe (Topic is now a Visual Cluster, File is either a Generic File or a Media, etc.), and they were influenced by the kinds defined by Windows operating system (and mapped to the same file extensions as in Windows). This means that if you got used to search for "kind:photo" in Windows Explorer to look for pictures, the same query in Zet Universe will yield the same results.
However, although we've been inspired by the Windows Shell ontology, our ontology is not a try to explain and define everything, and it's not carved in stone.
DYNAMIC ONTOLOGY
As it was stated above, our goal is to make it possible to integrate many different kinds of data from many different sources into a coherent whole that reflects how people naturally conceive of information.
To make it possible to reflect different people's ways of thinking about the information, we decided to make a compromise: although there is a default vocabulary of Kinds, both developers and users will be able to define and re-define the ontology on the fly.
Part of this functionality is already available in the product. For instance, if you use a Dropbox app within Zet Universe, your ontology will have a "Dropbox Folder" kind which is a child of a "File Folder" kind, and if you use Sticky Notes, your ontology will have a "Sticky Note" kind as a child of a "Media" kind, and so on. Both our own and third-party developers will be able to extend the default ontology with the new kinds.
Third-party Kinds have the same role of the building blocks of Zet Universe as the built-in ones. For example, Dropbox and alike folders can participate in the project space as external data sources, while Sticky Notes are items created within the project spaces, and so on.
In the future builds, you will be able to define your own ontologies so that Zet Universe could be better suited to your specific needs.
To make it possible to reflect different people's ways of thinking about the information, we decided to make a compromise: although there is a default vocabulary of Kinds, both developers and users will be able to define and re-define the ontology on the fly.
Part of this functionality is already available in the product. For instance, if you use a Dropbox app within Zet Universe, your ontology will have a "Dropbox Folder" kind which is a child of a "File Folder" kind, and if you use Sticky Notes, your ontology will have a "Sticky Note" kind as a child of a "Media" kind, and so on. Both our own and third-party developers will be able to extend the default ontology with the new kinds.
Third-party Kinds have the same role of the building blocks of Zet Universe as the built-in ones. For example, Dropbox and alike folders can participate in the project space as external data sources, while Sticky Notes are items created within the project spaces, and so on.
In the future builds, you will be able to define your own ontologies so that Zet Universe could be better suited to your specific needs.
ONTOLOGY VIEW
In order to see the ontology of Zet Universe, you can use our brand new "Ontology" view:
In this view, there are two tabs, illustrating the different aspects of the ontology browser:
- Items, and
- Properties.
ITEMS BROWSER
By using the Items tab, you can not only the hierarchy of the default Zet Universe's ontology, but you can also see actual items that represent the selected kind:
PROPERTIES BROWSER
By using the Properties tab, you can see the properties relevant to a selected kind.
CURRENT LIMITATIONS
It's important that as this is the first release with the Ontology View, there are many limitations. For instance, you can't see custom properties relevant to a specific kind. There is no way to browse relationships. There is no way to define new kinds, or new properties, or relationships. These features will be available in the upcoming builds and will be announced separately.
WHAT'S FIXED?
File Path is Too Long Exception leads to a fatal crash - starting with this build, you shall no longer experience this fatal crash. We've significantly improved checks for the very long file paths, so that your work experience with Zet Universe would not be degraded by such annoying errors.
SOME OF THE KNOWN BUGS
- Initial Office document upload to the OneDrive folder won't be picked up while Zet Universe is running. Zet Universe will pick up Office files uploaded to OneDrive after restart. This bug is being investigated, but there's no ETA at this point. We are sorry for this bug.
WHAT'S NEXT?
In the coming builds we will start working on the more scalable storage subsystem for Zet Universe, we will upgrade our plugin platform with the latest improvements from our Architect, Danila Korablin, and we will return to our work on the Dynamic Ontologies and the related Rules Engine, as we've defined that in our Product Roadmap. Of the other things, we will also continue working on the SDK, and we will start reaching out to the selected developers among Insiders to ask them to evaluate the SDK.
Thank you so much for being with us, for your continued support and participation!
As always please continue to give us feedback, suggestions, and problem reports via the UserVoice forum, use the Zet Universe Insiders Group on Facebook to connect with other Insiders for help and tips on problems you hit.
Thanks,
Daniel and the team
Thank you so much for being with us, for your continued support and participation!
As always please continue to give us feedback, suggestions, and problem reports via the UserVoice forum, use the Zet Universe Insiders Group on Facebook to connect with other Insiders for help and tips on problems you hit.
Thanks,
Daniel and the team
RSS Feed
