SMW users survey results

From semantic-mediawiki.org

From January 10 to 28, 2013, an online survey was conducted of Semantic MediaWiki users, asking mostly about priorities for future development. 76 people took the survey. Below are the results.

Type of SMW usage[edit]

Which scenarios of using SMW are most important to you? Select one or more.

Response Number Percentage Graph
Public online community (public read and write) 33 43.4%
Wiki-based website (public read, restricted write) 44 57.9%
Organisational or intranet wiki (restricted read and write) 50 65.8%
Personal wiki (private use only) 16 21.1%

Development priorities[edit]

Below are a list of general areas which can be improved about Semantic MediaWiki, and a short list of specific improvements that might fall into each area. Please select one or more areas that you think the Semantic MediaWiki project should especially devote more effort to.

Answers have been rearranged in order of popularity.

Response Number Percentage Graph
Queries (support displaying linked properties like "?a.b", greater support for ORs and ANDs, more up-to-date display of data, free-text search features in queries) 53 70.7%
Forms (support for more than one free text area, easier maintenance of forms, more input types like a spreadsheet-style form input, better autocompletion on remote data) 44 58.7%
Access control (controlling what semantic data users can see, controlling which pages users can read and write based on semantic data) 39 52.0%
Searching and browsing (better interface for Special:Ask, new faceted search interfaces) 33 44.0%
Semantic Web and "linked data" (improved RDF triplestore storage, better integration of SPARQL queries, import and use of standard ontologies, more inferencing, querying of Wikidata data) 32 42.7%
Performance (faster query speeds, more scalability for storage and querying for large amounts of data) 32 42.7%
Other visualization (more calendar and charting options, more additional display formats) 30 40.0%
"Gardening" and automation (more tools to view data and data structure at a glance, modifying a large group of pages together) 29 38.7%
Mobile (better display and editing of SMW-based wikis in mobile-device web browsers, dedicated mobile apps for display (and possibly editing) of SMW data) 20 26.7%
Documentation (improving documention on the MediaWiki and SMW wikis; more books and other documents; more online screencasts and the like) 20 26.7%
"Semantification" (being able to query on additional data from MediaWiki and its extensions, like a wiki page's authors and "social" information like user comments and relationships between users) 18 24.0%
Bug fixes and testing (more emphasis placed on fixing bugs, more testing scripts to try to make bugs rarer in the future) 13 17.3%
Mapping (support for additional services like Apple Maps, set of points queried changes as map focus changes, better handling for lines and polygons) 9 12.0%
Marketing (greater attempt to get press coverage, marketing-style online videos) 7 9.3%

Interest in funding[edit]

For those items that you selected, would you or your organization be willing to provide effort or money to help in those areas? Checking "Yes" does not represent any commitment to provide support, but please be honest.

Response Number Percentage Graph
Yes 46 61.3%
No 29 38.7%

User experience[edit]

How much experience do you have as a user of SMW?

Response Number Percentage Graph
Advanced user (I have created many #ask queries/templates/forms/etc.) 56 73.7%
Casual user (I work with SMW and know many of its features) 16 21.1%
Beginner (I have hardly used SMW so far) 4 5.3%

Admin experience[edit]

How much experience do you have as a website administrator of SMW?

Response Number Percentage Graph
Experienced administrator (I have a lot of experience in running/configuring/upgrading SMW sites) 45 59.2%
Novice administrator (I have installed or configured SMW at least once) 26 34.2%
None (I have never installed or configured SMW) 5 6.6%

Free-form comments[edit]

The following are the free-form comments and suggestions people gave in the survey, subdivided by topic. In some cases, spelling and grammar were fixed for legibility.

Queries:

  • distinct results from queries
  • Provide the option to create relationships outside of a page. For example, sub pages could define relationships to parent page without the added clutter there. A use case could be a page with many synonyms for the page title - it would be nice to allow automated annotations of these synonyms into a sub page without having to overload the main page.
  • More possibilities for negation than just <prop>::!<value>
  • Better support for type:Date

Data storage: (was not a listed option, though it probably should have been)

  • handling for assigning query results as values to properties
  • New type:Duration that understands dd:hh:mm:ss.xxxxxxxxxx
  • to be able to have a template as a property type. I want a property to be either {{Done}}, {{Started}}, {{Canceled}} for example, but it is not possible.

Searching and browsing:

  • Make it easier to write queries. Consider using the SMW+ halo extension. It was super easy to visually design a query
  • Inline filtering is something that would be very welcome
  • inline filtering

Mapping:

  • I would really like is to be able to use any images as a layer in maps and not just the 'standard' google or openlayer maps.
  • complete custom maps (-> images as layers)

Other visualization:

  • I had to set a very high (virtual) limit for a query or result format for the timeline because there was a default limit without setting it manually and needet long time to find why before some dates were not shown
  • google rich format in calendar etc.
  • Pivot tables
  • more flexible charting
  • Support for projects would be nice.

Forms:

  • spreadsheet style input
  • to somehow be able to input data using a spreadsheet-like 'form'.

Semantic Web and "linked data":

  • Maybe in future having no wiki internal db and all data stored on triple store or from sparql endpoint?
  • Support for easier to read resource URIs for predicates.
  • Unified plugin for displaying RDF data from SMW (and other RDF sources) within wordpress, drupal and other software.
  • the ability to import ontologies.
  • Support for synchronization "text from data": 1/ the semantic store used by the wiki is updated using Sparql update 2/ Need support to update wiki pages! [ed.: in other words, be able to directly update the RDF triplestore and have the changes propagate back to the wiki]
  • A better compliance to semantic web standards.

Performance:

  • Performance is an issue! Displaying pages with long and intricate inline queries can take very long.

Documentation and support:

  • consistent legacy documentation
  • Open days where people can attend in person or online to go through their issue and problems.
  • Training courses for users.
  • I simply have difficulty in using and creating templates, forms and properties so training by going through a concept to realisation exercise would be great!

Import/export: (was not a listed option, though it probably should have been)

  • the ability to import data from a spreadsheet into your wiki and assigning properties to each row and column.
  • Better possiblities to import Data (CSV)
  • Export to e-books (PDF) etc. of collections of pages.
  • Getting data out of an restricted wiki e. g. via RSS/Atom-Feeds

Other:

  • Multilingual semantic wikis (localisation of properties, localised forms, integration with Translate)
  • maybe connect to opennlp or something like that?
  • Accessibility
  • PostgreSQL
  • Wiki Skins that work well with SMW
  • I recommend the formation of some form of steering group that could assist with strategy, driving an achievable roadmap and also improving quality.
  • it would be great if there could be an "LTS" SMW and the free software SMW+ features were formally supported rather than left as an awkward stepchild.
  • Bug free editing using CKEditor