Questions and Bugs: some first impressions

[David Bovill]

I've made one proposal, then was unavailable to participate - the discussion moved on to the second stage. I return to see what happened and how I can participate now...

First it is pretty damn hard to find my proposal :) Eventually, by looking at the page for the specific question, bringing up the voting table, scrolling down to my pseudonym, and clicking on my pseudonym in the table (which appears to do nothing but a refresh) - then scrolling way down to the bottom - I find my proposal.

Clicking on the full detail - there is a very nice full description of what happened and a very useful graph - which however lacks and visual clarity with regard to which my proposal was - the information appears to be there but it is very hard to see visually.

This brings me on to the main feedback - make more of the graphs. The graphs should be the primary form of navigation - not an end result to then interpret. To achieve this there needs to be some improvements in the visual components of the graphs, and some bug fixes, and navigation enhancements.

By the way where is the open source repo - I remember coming across is - but there is no link to it from the web site. There should be a link on every page to the wiki / issue tracker on GitHub or the equivalent?

The bugs I've spotted so far are the faulty links in the graphs to the issues - see http://vilfredo.org/map/1#proposal2398 There are also no links from author nodes to author pages.

The design feature I feel needs most work is in the area of proposal icons. There are some easy fixes, and some things that I believe we should go into in some more detail elsewhere. The problem is that currently it is impossible to view and track proposals across graphs as one blue square looks just like any other - this may make the graph easier on the eye in terms of clutter - but at the cost of meaning. We want to see what happens to a particular proposal first and only secondarily the overall argument. An easy fix would be to apply a randomly assigned blue colour to each proposal which would allow me to have a much better guess as to what happened to proposal 2398 - relying on me to be able to see the small text of the number, or having to blow the graph up to full screen does not work.

These are my first impressions - very promising, and a few quick fixes there that would dramatically enhance the user experience.

Also a note on process - you've also posted about open source methodology. I'll take a read - looks good from a technical perspective. It looks weak on the design issues with regard to the history or tools, community and open source. I feel it is more important that Vilfredo understand the soft aspects of community buidling espeacially how this relates to design, over the hard aspects of open source in terms of code.

With that in mind - the comments I make about design issues and bugs, should be dealt with through a process - and not ad-hoc. The process should be a process relevant to design - so it is not a good idea to use Vilfredo, nor is it a good idea to rely on the existing community as I suspect design is not a strong point. I'd recommend looking closely at user centred design, focus groups, mood boards, design sprints, and only afterwards look at what you code, and how you engage an open source community, and version control in the software development?

[Pietro Speroni di Fenizio]

Hi David,

thanks for writing up those points. Indeed there are a lot of bugs in the code. Also as we are moving from the PHP to the Python version it becomes less and less meaningful to clean up the old code. So we tend to simply turn off the less relevant features, and keep to the core, while moving to the new system.

But let's go through your mail, because not is a bug

On Monday, October 7, 2013 1:32:48 PM UTC+2, David Bovill wrote:

I've made one proposal, then was unavailable to participate - the discussion moved on to the second stage. I return to see what happened and how I can participate now...

First it is pretty damn hard to find my proposal :)

Indeed, and that's not a bug it's a feature. The point is that the discussion moved on and the system makes it much easier to see the proposal that passed the Pareto Front filter. Everything else is still available, but in the archive. A simple way to find it is. At the bottom of the page you will find a link to the history page. And in the history page there are all the proposals presented, who wrote it, and who voted for it. But it is all temporary, in the next version it might be different. We both know that the UI is what mostly needs renovation in VgtA.

 

Eventually, by looking at the page for the specific question, bringing up the voting table, scrolling down to my pseudonym, and clicking on my pseudonym in the table (which appears to do nothing but a refresh) - then scrolling way down to the bottom - I find my proposal.

Clicking on the full detail - there is a very nice full description of what happened and a very useful graph - which however lacks and visual clarity with regard to which my proposal was - the information appears to be there but it is very hard to see visually.

This is indeed a problem. We would like to make each proposal easily recognisable in the graph, but we do not know how. We are now thinking to let people fill in 4 short rows of 6 characters that will appear in the graph as the title. A bit like EU grant proposals ask for a short title to be used when addressing the proposal.

 

This brings me on to the main feedback - make more of the graphs.

Oh, yes!
 

The graphs should be the primary form of navigation - not an end result to then interpret.

To achieve this there needs to be some improvements in the visual components of the graphs, and some bug fixes, and navigation enhancements.

 

Absolutely agreed. 

 

By the way where is the open source repo - I remember coming across is - but there is no link to it from the web site. There should be a link on every page to the wiki / issue tracker on GitHub or the equivalent?

I recently changed the "about"  link. The new link sends to the wiki on github. It will take some time to go live as Derek is changing the code in some other ways in the meantime.

 

The bugs I've spotted so far are the faulty links in the graphs to the issues - see http://vilfredo.org/map/1#proposal2398 There are also no links from author nodes to author pages.

Yes, those are among the bugs we are not addressing right now. There should be a page for the user, and a page for each proposal. They are there but missing so many features it would be better if they were not there. 

 

The design feature I feel needs most work is in the area of proposal icons. There are some easy fixes, and some things that I believe we should go into in some more detail elsewhere. The problem is that currently it is impossible to view and track proposals across graphs as one blue square looks just like any other - this may make the graph easier on the eye in terms of clutter - but at the cost of meaning.

 

We want to see what happens to a particular proposal first and only secondarily the overall argument.

I don't understand, I am sorry. 

 

An easy fix would be to apply a randomly assigned blue colour to each proposal which would allow me to have a much better guess as to what happened to proposal 2398 - relying on me to be able to see the small text of the number, or having to blow the graph up to full screen does not work.

Making different colours is a possibility, but at the moment we are using colours to represent which one are the proposals you have voted for, and which ones are in the Pareto front.  As I said, we shall look for short Titles. That should make it easily understandable.

 

These are my first impressions - very promising, and a few quick fixes there that would dramatically enhance the user experience.

Also a note on process - you've also posted about open source methodology. I'll take a read - looks good from a technical perspective. It looks weak on the design issues with regard to the history or tools, community and open source. I feel it is more important that Vilfredo understand the soft aspects of community buidling espeacially how this relates to design, over the hard aspects of open source in terms of code.

Indeed this is very important. Unfortunately I am hardly the right person for that. If you want to get more involved with the community you are very welcome. For now I was very grateful for your messages :-)

 

With that in mind - the comments I make about design issues and bugs, should be dealt with through a process - and not ad-hoc. The process should be a process relevant to design - so it is not a good idea to use Vilfredo, nor is it a good idea to rely on the existing community as I suspect design is not a strong point. I'd recommend looking closely at user centred design, focus groups, mood boards, design sprints, and only afterwards look at what you code, and how you engage an open source community, and version control in the software development?

as I said, I am not the right person for that. I get a glimpse of what you are saying, but actually we need a team dedicated to user design. As soon as the API will be ready we shall start gathering the people.

Bear with us in the meantime