Quality of the file format doesn’t matter. And it doesn’t matter if a specific application is needed for working with the file format (or if there exists such an application at all).
It only matters if the file format is proprietary or open/free/libre.
With "excel", Tim Berners-Lee probably refered to the Excel Binary File Format, which is one of several ...
The format should fit the data structure not the need of the comsumer (or the one who uses the data).
If the data is semantically stored in a json structure you can easily export the data in different formats.
So my answer is: json
For level 1, any format is fine. Even a picture of data written on a sheet of paper is OK. That's because level 1 is only about web availability and license.
So, yes, any format: PDF, XLS, CSV, JPG, SQL, or even opaque binary blobs.
From http://www.w3.org/DesignIssues/LinkedData.html :
(1 star) Available on the web (whatever format) but with an open ...
There is no correct answer, but here's some pointers:
Its relatively painless to convert from JSON to CSV and vice-versa. what matters is the structure of your data. If its properly formatted, you should have very little problems converting/conforming to standardized formats.
Speaking of formatting, you should keep in mind that RSS is a great route to ...
I'm not sure I agree that you need to have any URI to demonstrate how open a set of data is. Either the data is open and shareable or it is not, the openness of the data is described by the openness of the data.
Let's say I publish a set of RDF linked data, as soon as it is published it is open to be consumed as a whole or in part, people can take which ...
The Open Knowledge Foundation (OKFN) published a study it did on government data portals around the world. A number of different variables (criteria) were used. I don't recall data formats as one of them, but if not you still might be able to reverse engineer from their study data.
I assume that your are talking about the 5-star Open Data from Tim-Berners Lee.
In this case, you can check this one about working with different kind of file types in Python.
3 star data is CSV and 4 star data is RDF.
From the linked answer:
Comma Separated Files (CSV) files can be a very useful format, because it is compact and thus suitable to ...
Just to add another format: iCalender (.ics) might be a nice add-on if the events are to be booked/bookmarked/scheduled by visitors interested in attending some/all of the events. Although the support for spatial data is limited to a plaintext field, it might be worth adding.
In December 2016, the Data on the Web Best Practices Working Group (Group Page, Group Charter) have published the Data Quality Vocabulary. This vocabulary is a meta-framework for frameworks that similar to one you need.
In the document, these points are relevant to your question:
Feel free to create your own framework; moreover, you rather should create ...
You can benchmark data vs. Open Data Census, you can use linting tools (CSV Lint, JSON Lint, etc.), and you can make them all into Data Packages.
I'm curious to why linked data is not a priority for you. Or rather, what your end goal(s) are.
You should create these URIs yourself, as well as mainteiners of all other DCAT catalogues.
Look at Data Quality Vocabulary, which is based on DCAT. Consider the following example from the section 6.7:
a dcat:Dataset ;
dqv:hasQualityAnnotation :classificationQA .
I would go with CSV. It will help your clients reach most of the audiences they interact with on a regular basis. There is no shortage of packages to read, arrange, and manipulate CSV files in languages like R and Python, which the data diggers will love. The less computer-literate will at least be able to open a CSV in a spreadsheet reader or text file ...
In my opinion CSV is more usable. It is older and supported by almost anything.
Is this data user will download? Stay on CSV and everyone can enjoy it, even with Excell o Libreoffice.
Is this data provided as web service? Json can be the choice.
Also consider format of the data. CSV has limitations. Array and complex things does not fit well ...
You can use HTTP-URI values from controlled vocabularies in the attributes of your WMS datasets, as properties in your WFS feature types and WCS coverages. You can additionally add links in your GetCapabilities response documents through metadata urls, and data urls... And finally you can make links in your metadata documents for your services and datasets....