Dear Steve Ballmer
blog.mozilla.com
blog.mozilla.com
A joke, surely. For the Windows phone I'm seeing unnecessarily gigantic type set against irrelevant background images with no consistency amongst the other UI elements. The "head to head comparison" he links to is, itself, based on criticisms of Microsoft's UI by Edward Tufte.
Please explain. All I see is consistency in this UI style. And the background images look pretty relevant to me.
Non-stroked or unshadowed white text on busy photographic backgrounds isn't good for readability.
All I see is consistency in this UI style.
It's consistently sloppy, but I don't think that's what you meant.. :-) Perhaps pointing out the inconsistency was a bad choice of term. Really, it's just "bad." I'll use http://www.youtube.com/watch?v=j9F4QJK1wFs as a reference walkthrough of the interface for some examples.
On "still" screens (that is, not in the middle of scrolling or similar) you end up with the giant typography getting chopped up: http://skitch.com/petercooper/n46wk/w1 .. "pictur"? Or http://skitch.com/petercooper/n46w6/w3 "game"?
When displays like "pictur" and "game" are scrolled horizontally to show further content, the distance travelled doesn't match the position of the header at all. See here how the headline has moved slightly (now obscuring the "g" of "games") yet an entirely new style of view has appeared (with overly light text, no less): http://skitch.com/petercooper/n46ib/w4 - Everything seems to move at a different speed in relation to everything else on the Windows Phone. This is inconsistent and "feels" wrong.
Front page elements are varying sizes: http://skitch.com/petercooper/n46ws/w2 - some items even embed other items inside. And white on turquoise == poor readability.
If you watch the video above around the 11:00-11:30 mark (the "People" screen) you'll see contact boxes randomly changing, pictures popping up inside, etc. There's no indication that this should be occurring.
Another problem is that there's also no solid indication that there's anything to scroll to on the right, yet when the guy does so, there's a bunch of stuff over there. Using the truncated title text as a sort of "this page is wider than you think" indicator is a really crappy UI hack.
I haven't got time to go through everything that feels wrong about the Windows Phone UI, but I'll leave with the Tufte-derived view that the information density is also poor. In the demo of the photo section, the number of photos shown on screen at once is very low. On a touch device you can't go too nuts with filling the screen up, but Apple has struck a smart balance between too little and too much - Microsoft leans way too far towards whitespace and endless scrolling.
I believe this effect is mitigated when most/all of the offscreen elements are just more instances of the same type of thing as those already shown (example: repeating table views), and also when the offscreen portion of the elements are along a single axis only (i.e. you would only have to lengthen the device in a single direction to be able to see everything at once). Somehow this type of scrolling doesn't cause the same "looking through a keyhole" response in me.
Many of Palm's advertisements for webOS portray a similar effect (screens tapering into offside infinity on either side of the device) which seems to me likewise unattractive and poorly considered, though the actual experience of the device doesn't appear to be like that, and certainly isn't designed around it as a feature like WP7S is. Regardless, the ads emphasize complexity and overwhelming amounts of data -- to their unfortunate detriment, I expect.
This is just a weird way of pointing out that they don't have a NDK on the new Windows Phones, and if they did then Firefox would get ported (with some gratuitous plugs for ongoing work on competing platforms).
http://www.theonion.com/content/columnists/i_got_some_sweet_...
Obviously this could have been avoided by not having a personal blog on mozilla.com, or by not writing posts like this on it, or by not submitting it to the srs bzns HN crowd (which by and large can't take a joke)--but really, a more descriptive submission title probably would have been enough.
Is it that he doesn't think Firefox performance will be acceptable in XNA/Silverlight, or that he doesn't want to port any of the existing code to the CLR?