The UI is nice. The workflow is easy enough to understand. Unfortunately it just didn't work for me. Trying to automate (1) loading a specific page (2) ticking the first 3 check boxes (3) clicking a button (4) waiting for page refresh (5) clicking another button (6) clicking a confirmation dialog (7) waiting for page refresh (8) restarting the process again.

Headline: Very nice graphical interface approach for automation. but has a long way to go with functionality.

I work in IT and was using imacros extension which helped me a lot with automating server configurations using the browser.

As FF57 came out, I found Wildfire as alternative. but it still not ready for heavy automation.

I would like to stress that I'm not putting the review to compare products, but to show why I find it hard to accept this one.

I faced 3 main issues which gave me this conclusion:

1- Event recording. imacros was recording changes in page elements while wildfire is recording my input into the browser. as small example, lets say I open google.com and search for "wildfire".
imacros will record (open URL, focus on search bar, Entered value "wildfire",clicked on "search" button)
wildfire will record (open URL, left clicked at position x 5 y 3 , entered w, entered i, entered ........, left clicked at position x 7 y 5).

This way of recording is causing several problems, first, if i was recording with firefox window occupying half the monitor then simulated with full size window, it will not work (mouse positions are different). if I activate Sim while firefox page zoom options are different from recording, it will not work. If i use this script on different computer with different monitor size. it will not work. this did not happen in imacros as it looks for the element on page, not the position i clicked. this goes worse as I record more complex things as choosing from drop menu....etc

2- Workflow editor: the graphical approach is a great feature which makes it very user friedly. but it has 2 drawbacks in its current state

1) The lack of command line like editor. As an IT person. and a person who is looking for automation to save time. I find programing a script with text is a lot faster than using graphics. for example, typing "create node type input value xxxx" and copy paste that sentance 3 times then " connect node1-2 node 2-3 " or something similar is a lot faster than creating these nodes graphically, then configure the vlaues for each one then connecting them using the mouse.

For small tasks this time gap is niglegible, but as the amount of script modification grows, it becomes a nightmare.

I suggest you add an interactive "command line like" panel in the same workspace which changes the script and the graphical workflow. so now both the "easy to learn" graphical interface and "fast to script" command line interfaces are available in the same workspace of the workflow editor.

2) The graphical interface itself is stiff, the tools menu (add node, duplicate...) and properties window should be movable within the workspace. sometimes i zoom in (ctrl +) to focus on a part of the flow. but the tool menu and the icons overlap. simply put, it needs more polishing.

3- I couldn't find a comprehensive way of looping inside the flow. or at least, if it is there, I couldn't understand it. As automation software, I find the most important feature is (importing data from csv to webpages and vice versa) which mainly requires looping and interactions with csv files.

Again, This extension has the best graphical interface among the other automation extensions I saw. but it needs more functionality.

I liked the simplification but it has some issues. FIrst it opens new windows for intructions or to show you the workflow etc but instead of opening a proper window it shows a minimized windowd set to the side and you are only able to see the bottom right part of the window and you are only able to see it properly if you resize it by double clicking on the window border. An other issue I noticed is that when you run a simulation it takes waaaaaaaaaaaaay to much time to start it and the worst of all when recording the events they are either recorder unproperly or simulated improperly because there were mouse movements that didnt show clicks done that were not done and the addon instead of opening pages because a link was clicked or a bookmark etc was pressed it just redirects the current page to that url which is not the same.

There is also no proper way of leaving bug reports or suggestion listed on the amo page. The page listed there is just of their homepage telling you to download the addon.

EDIT: I have 1824x1036 screen slightly adjusted to reduce it a bit to fit my monitor which is actually a Polaroid tv using the nvidia control panel.