Wardley Maps - online course now available!
  • Trainings
    • Online Mapping Course
    • Advanced Workshop
  • Community
  • Blog

User management is not novel!

28/12/2014

Comments

 
User management is not considered to be a novel act. Even if you are doing it for a very first time, there is a legion who did it before you, and there is no new value you could add. If you are trying to design and implement the solution you are actually reinventing the wheel, which is actually a cost of doing business.

My initial authentication relied on a simple library called passport.js, which was asking Google about Id representing a user.  This was sufficient to keep maps separated. But more advanced scenarios were unlikely to be easy to implement. Questions I was constantly asking myself were f.e.:
  • how to solve map sharing and inviting others to discuss the map? Google ID is not sufficient to do that.
  • how to inform users about planned outages?
  • how to inform users about UI changes to avoid confusion?
All those things sound ridiculously simple, but i had a feeling that I was wasting too much time for things that were not the core of my business. So i have decided to play the scenario represented on Figure 1 in order to either deliver user management as a service or use a service provided by others.
Picture
Figure 1: Adopting professional user management mechanism.
Yes, the Figure 1 contains the name of a solution that I have adopted. It turned out (to my disappointment, because somebody delivered the solution before me, and surprise at the same time, that it is already done and ready to use) that there are at least two companies delivering User Management As A Service.

  • Stormpath - created on March, 2012. $9.7M funding
  • Auth0 - created on September, 2014. $2.4M funding.

Stormpath turned out to be much more developer friendly and mature, and had a very reasonable free quota in addition, so here we are - the mapping tool uses Stormpath now.

Mapping is great. Without it, I would not even thought about looking for those providers without being aware of their existence!

Have you tried to look for services that should be already commoditized? What results did you get?
Comments

it is all about user needs

20/12/2014

Comments

 
My users want to create maps, so creating and managing maps are the needs that are first in my value chain, and I have promised to take reasonable steps to protect those maps.

Easier said than done. I had to update the database schema used for storing maps, because it was too simplistic and too limiting. This has been looming for like a month. Once the code was ready I had to put it in production. As you are aware, changing the database schema is not a simple task, and the best way to accomplish it in the pre-beta phase is to drop the database, deploy new code and start again with a clean desk.

I wonder whether somebody could put it next to reasonable steps (well, it all depends on the number of maps, their value and the work required to migrate the database), but what is most important is that it is against mapping principles.

User needs should be first.

Users have no need of doing the same work again.

The data migration is over. You should not have noticed anything but a short maintenance break.

Just in case there is something wrong with your maps - let me know via twitter.
Comments

Click-to-focus instead of hover-to-focus in the map editor

14/12/2014

Comments

 
Desktop and mobile are two entirely different philosophies, and subtle design decisions may seriously affect the user experience the tool can provide. One of such decisions was 'hover-to-focus' philosophy adopted for selecting map nodes and creating dependencies. Look at the figure 1 below. If the mouse enters the node, connection endpoints appear. This looks nice, but can be sometimes painful to use, as the underlying model is not perfect, and endpoints may disappear in the most undesired moment, f.e. when you were about to drag out a connection. Additionally, advanced node operations are pretty difficult to represent as moving mouse over the map causes a lot of things appearing and disappearing. It is very annoying.
Picture
Figure 1: Hover-to-focus approach.
But what is more important is that there is no such a thing like mouse cursor on mobile devices. So, after certain considerations I have decided to change the philosophy. Look at the Figure 2.
Picture
Figure 2: Click-to-focus approach.
The endpoints are not appearing when you hover over the node, but you need to click on it. You get immediate visual feedback - the node background turns yellow and endpoints are getting larger.
This change allows for further tool development, especially for context sensitive tips, and brings tablet version one step closer.
Comments
<<Previous
    Community blogs
    Tweets by @wardleymaps

    Categories

    All
    Analysis
    Book
    Community
    Feature
    Hidden Knowledge
    History
    Making Decisions
    Manual
    Map Examples
    Risk Mitigation
    Strategy
    Theory

    RSS Feed

    Archives

    February 2018
    January 2018
    December 2017
    November 2017
    October 2017
    September 2017
    July 2017
    May 2017
    January 2017
    August 2016
    February 2016
    December 2015
    October 2015
    September 2015
    August 2015
    July 2015
    May 2015
    April 2015
    March 2015
    February 2015
    January 2015
    December 2014
    November 2014
    September 2014
    August 2014
    July 2014
    June 2014

Researchers   About Us    Privacy Policy    Contact Us