The best thing Firefox Browser Metrics is gonna do is put an end to paranoia. For years, people have complained about applications that “phone home”, supplying statistics. An example software that has been doing this for a long time (with opt-out first, then opt-in later) is MSN Messenger. And since Microsoft is such an easy target to attack, many immediately pointed and acted as if this was some kind of privacy intrusion.

Is it? I guess we shall never quite know, but I’ve always been on the (naïve?) side that no, this has nothing to do with privacy; it has to do with — just as claimed — improving the product in the long run, for the very people this technique is being performed on: the users.

The main thing to prevent misunderstandings in this type of information is transparency, and I can think of two major ways to make use of it:

One, describe what specific data you are collecting, and what you might be using it for. Mozilla clearly lists what type of information would be collected and how it could be used. For example, depending on how often a certain button is pushed, it can be determined how important people find it, how obvious it is for them to use, and how it could possibly be improved in the future.

And two, for those that don’t believe the above claim to be true, perform the collecting in an open format. Use something that’s pretty much plain text, such as XML or YAML, rather than a proprietary binary format, or, at the very least, describe that format. For example, Apple’s Software Update uses a rather simple XML-based format, and as such, it’s easy to reverse-engineer, such as here.

To summarize: show the what (what data is being collected?), the why (what would it be used for?) and the how (where is this data coming from?), and you can prevent a lot of privacy concerns.

Hopefully, this metrics program will work in that direction, showing that Microsoft’s (and others’) analysis of user behaviour is not “evil”.

Via OSnews: “Seamonkey beta improves on Mozilla legacy”

Can’t be too hard, as there’s barely an area that doesn’t desperately need improvement. In fact, why don’t we get rid of the outdated all-in-one concept, split it into multiple applications and redesign their UIs from scratch. Oh wait. I guess some people try really hard to make their lives hard.

In other news, some people are still hoping for AmigaOS4, while others still consider war a solution, not a problem.

…tell me that it sucks.

I know, I know. It’s in very early stages. It’s not fair to judge a project in early stages. Unfortunately, it’s not quite that simple.

1. It’s built on top of Mozilla Firefox, which they tout as a “feature”. Unfortunately, for Mac OS X users, it means that it interface is designed to suck. XUL on OS X just doesn’t cut it. Mac users rightfully demand carefully-crafted UIs with the user in mind. This isn’t. You can feel at just about any point in the UI — the top and bottom bars in the browser window, the contextual menus, the preferences window, heck, even the about box — that, during design, nobody bothered to test this on a Mac (or, some people did, but didn’t really know much about Mac UIs).
2. And don’t get me started, even, about the little fake applications that launch before the actual browser launches. Something appears in the Dock, starts bouncing, then disappears. Something else appears, then immediately disappears. Then, nothing happens. Just as you start thinking “maybe it crashed”, the actual program starts running. What’s up with this?
3. It also doesn’t use OS X’s Keychain. Why would I want to use the Firefox-style password manager when I have a perfectly fine, highly-encrypted, more flexible and application-independent password manager right in my system?

I like the basic concept of this, but as someone else noted somewhere (on digg, maybe), it would be much better as an extension to an existing browser.

Even if I did want to ditch my beloved browser for this capability (which I currently don’t, thankyouverymuch), why does it have to be Firefox-based? Don’t give me that cross-platform nonsense. The actual Flock code can’t be that complex that it’s so hard to port. If Skype can provide perfectly native applications for Windows, Mac OS X, Linux/Qt and Pocket PC, why can’t Flock?

Give me either an extension for Safari that does this (‘this’ being integration of del.icio.us — or alternatively Yahoo! MyWeb2 — , Flickr, tagging and blogging right into the browser), or, if it has to be, write my a browser that uses Gecko or WebCore, but write it natively. Aqua. Not some fake crap like XUL.

Thus, harsh as it may be, my judgement after just four minutes of using this piece of software is: it sucks. And I don’t expect it to suck much else in version 1.0 either.

Update: amon chimes in with a lot of good points, especially:

the FAQ clearly contains questions the developers would like the users to ask.

Indeed, reading the FAQ gives you the impression that the developers didn’t even for a single minute step down to think about what the users would like to know. Like, uh, “What is it?”, “Why should I care?”, “What advantages does this have?”. Nothing.

No matter where I look, all the results on Google seem to claim that AdBlock won’t work with Deer Park Alpha 1, the first pre-release of Mozilla Firefox 1.1.

But hacking it to work was easy as pie, and I’ve been using it successfully for about a week, so I fail to see where the problem is. I have only tried this on Windows (XP), but the code we’re about to edit is platform-independent, so it should work elsewhere as well. All you have to do is: Read the rest of this entry »

Via Zef Hemel Recent Links: Firefox 1.1 Preview at the creatively-titled Fedora Core Linux Blog. No dramatic changes, although in several regards, rendering performance has improved. Additionally, preferences take effect immediately now.

Certainly a move in the right direction.