A flavor for personal information management

Help / Discuss
#1

Hello everyone,

Several open source tools exist in the field of personal information management (PIM): they help us organize one’s notes, contacts, books, photos, lyrics, etc. I’m creating this post to discuss a possible XWiki PIM flavor, to get feedback on what it would mean for you, and to collect ideas about existing projects that you may find relevant in this area. I sketched a first design draft with several use cases. What do you think about it? Would it cover some of your needs? Are you using a personal information manager already?

Cheers,

Stéphane

#2

I’m adding the proposal directly to this discussion and removing the page that I had created on the design wiki for now since I had actually overlooked the feature proposal process.

Some tools or articles related to personal information management

Purpose of the PIM flavor

The purpose of this flavor is to support the organization, the processing and the sharing of personal information and services. It is meant to be used primarily by a single person, albeit also offering a secure read / write mechanism to others, hence the interest of having a collaborative platform.

Use cases

  • As a user:
    – I can write notes about the books and songs I like, my projects, my photos and videos, and I can draw relations between them. This network is a “personal web”.
    – I can retrieve the posts and documents that I created on sites such as Twitter or Google Drive, organize them uniformally and search across them globally.
    – I can create or install services harnessing my data, such as: organize a trip, gather information about a city, ask me questions, recommend a doctor and book appointment, etc.
    – I can grant secure input output to/from parts of my data and services and I can charge a fee for this.
  • As a project manager or consultant:
    – I can organize information related to the projects I’m involved in: participants, meeting notes, tasks, travel information, web pages, budget, pictures, drawings, etc.
    – I can create personal visual dashboards agregating data from multiple sources, helping me understand better my domain of expertise and the course of my projects.
  • As a researcher:
    – I can collect and annotate the articles I’m interested in and I can link these articles to pages containing author contact and profile, their labs and projects.
    – I can export this knowledge graph in standard bibliography formats such as bibtex.
    – I can write articles about my research experiments in a structured manner that ease the replay of these by others (electronic laboratory notebook).
  • As a teacher or as a student:
    – I can compose lectures / learning material from several data sources, keeping track of their origin and their license.
    – I can export these resources in various formats including static web site, epub, pdf, rdf statements, slides.
  • As a doctor or as a lawyer:
    – I can write confidential notes about my patients or clients, possibly offline to prevent any network breach.
    – I can get recommendations from the system regarding the most promising paths to follow in each situation, taking into account my personal knowledge and the one available from other sites.
    – I can run the data analysis in my own computing environment.

Data model

This flavor comes with an application defining an extensible data model. Each first level bullet below represents a class and each second level one a property.

  • Thing
    – is a: type
    – has name: string
    – has body: string (optional)
    – has Wikipedia id: string (optional)
    – has subtypes: all other types
  • Person
    – has name: computed string
    – has first name: string
    – has middle name: string
    – has last name: string
    – has maiden name: string
    – has nationalities: list of countries (optional)
    – is member of: list of groups (optional)
    – has webs: list of webs (optional)
    – has subtypes: contact
  • Contact
    – has email addresses: list of e-mail addresses (optional)
    – has phone numbers: list of phone numbers (optional)
    – has addresses: list of addresses (optional)
  • Location
    – has geo coordinates: point (optional)
    – has boundaries: polygon (optional)
    – has OpenStreetMap identifier: long (optional)
    – has subtypes: country, city, address
  • Opus
    – has authors: optional list of persons
    – has creation date
    – has subtypes: article, book, dataset, document, drawing, image, movie, object, piece of music, presentation, quote, song, software, video, web
  • Group
    – has subtypes: organization
  • Project
  • Issue
  • Event
    – has start date
    – has end date
  • Label
    – has text: string
    – has language: language
    – has translations: list of labels (optional)
  • Statement
    – has subject: thing
    – has predicate: predicate
    – has object: thing
  • Predicate
    – has label: label
    – has inverse: predicate
    – has scope: list of types
    – has range: list of types
  • Agent: a program performing a task harnessing my personal knowledge graph.

Some features

  • Most of content pages are always in edit mode for quick modification without the need to switch from view to edit mode.
#3

Let me start answering your question to kickstart the discussion! :slight_smile:

Sounds interesting but it’s a lot of work to develop and maintain :slight_smile: Are you proposing to do that?

Personally I don’t have the need but maybe I’m not the typical user or maybe I’ve not realized I need this :wink: FTM the only thing I use is a Blog on https://massol.net. And all my other ideas / proposals are put on *.xwiki.org.

Maybe it’s more useful for independent contractors or consultants especially if they are in the wiki domain since it allows for them to use the tool (for billing purpose, to store their knowledge, etc) and at the same time to demonstrate how it can be used for real work.

Generally speaking, I like your idea and your enthousiasm :slight_smile: It’s cool.

#4

Thanks for sharing your thoughts on this topic Vincent, and for your encouraging words :–)

My proposal would be indeed to develop such a flavor – I may understimate the workload though, time will tell… That’d be cool if it could be useful to others as well.

Regarding the usage : beside targeting consultants indeed, I’m wondering if the approach could meet also more day-to-day needs or not.

It became progressively quite common to let various services manage our personal data: for instance Gmail for contacts and e-mails, Evernote for personal notes and pictures, Dropbox for binary files, Deezer for playlists, Dashlane for online accounts, etc. It’s efficient because the services are cheap (if not free) and well designed. However, it has some downsides: searching across all this data and interlinking some elements is not easy; each service offers its own user experience (typically, a specific content editor), while a transversal one, customizable, might be nicer; the way the data gets potentially used commercially is not always transparent; the services are heavily centralized, which opens the door to risky breaches; it’s not necessarily clear what the data becomes in case of a heavy crash, etc.

On the other hand, it might be possible (but complex) to let users manage their data with the software and the hosting provider of their choice and enable case by case access to this data under certain conditions by third-party software. The powerful XWiki platform APIs would be helpful on this matter, beside offering advanced features for dealing with semi-structured data. To some extent, the approach shares some concerns with the ones expressed by Doc Searls et al. in the VRM Project. Just wondering here how the XWiki community feels about this, and if any progress along this path would be relevant to any of you.

#5

I stumbled upon a recent article promoting the personal web vision, which shares some concerns with the ones mentioned in this discussion:

Be Your Own Platform - Instead of relying on another platform to host your thoughts, with their specific branding guidelines and restrictions, you should be the platform you want to be. By making the social network revolve around your site, you’re keeping it within your grasp and built for the things you want.

Still believing XWiki as a platform is a good fit for this kind of approach.