UPDATE: I’ve already gotten a very thoughtful reply to this post – check it out in the comments below.
Lately I’ve been having some trouble with Firefox 3 – basically, when you click a link in some 3rd party program (that is, not in Firefox itself) and it tries to launch Firefox (because it’s not currently open), it hangs. Apparently, it’s hanging while trying to show the profile manager.
I use the Firefox Profile Manager to keep multiple profiles – I have one for when I’m working, one for me (personally), and one for Amanda – it’s nice because it keeps saved passwords, bookmarks, and so forth all separate, without all the hassle of full-blown Windows user profile switching. This way, I can exit the browser, get up, and Amanda can sit down and load her profile and her sites quickly.
I filed a bug on this, only to find out that it had already been reported (under a totally different description that no one in their right mind would have found without inside knowledge, but that’s neither here nor there) and fortunately it had been fixed and would be available in the next version of Firefox (3.1). The next version was not going to be released for a while yet, though, so I did some more digging on the issue.
After doing some digging and searching, I came across some disturbing comments:
Do not reopen this bug. Public opinion doesn’t really matter, it’s a design decision that has been made. (Comment 12)
Yikes. That’s really harsh – and not at all what I’d expect from an Open-Source project as high-profile as Firefox. In fact, it reeks of the same sort of developer arrogance that infected the Pidgin project not too long ago – and that ultimately caused a fork in the codebase over one very simple, minor UI change – but one that the user base and developers had very, very, very strongly held opinions on.
There’s also a lot of this sort of sentiment:
For windows XP and linux users, the typical users will presumably use XP login accounts to separate individuals.
Which, as I (and many other people) have pointed out, is a very narrow and incorrect view. Switching profiles in the browser takes a second. Switching user profiles in the OS can take much, much, much longer. And when all you want to do is separate browser profiles (not whole user profiles), why should you be forced to use the larger, more cumbersome method – especially when the more precise, quicker, easier one already exists?
Now, being incorrect is one thing – but a narrow view like this is bad, bad, bad. It’s this sort of thinking that results in UI disasters (having recently read The Design of Everyday Things has made me very cognisent of this sort of thing). When developers try to second-guess their users, they often get it (badly) wrong.
I don’t want to see this happen to Firefox. It’s not just about the profile manager – that’s just how it got started. I don’t want to see this sort of mentality infect the project, turning developers against their users (and vice versa). I agree with Jeff Atwood on the idea of “Strong Opinions, Weakly Held.” I can totally understand the technical reasons why the developers of Firefox would want to abandon the profile manager entirely – but being technically difficult isn’t an excuse that goes over well with users, who don’t care how complex something is on the inside – they just see how it works. And in this case, they’ve seen it work, used it, and like it. So there’s no excuse for trying to get rid of it, just because it has some difficult problems. Figure it out! Do something crazy! Re-work the entire profile system from scratch! After all, that was done with bookmarks in Firefox 3 – why can’t it be done with profiles?
Unfortunately I’m just a lowly user – I have no power within the Mozilla Firefox project (surprising, I know, for a community-driven, open-source project). I can only voice my opinion.
Which is just what I’ve done. Now, I just hope that the Firefox developers are still capable of listening…
UPDATE: In case anyone thinks I’m bashing Firefox or the developers here – I’m not. I’m just being… questioning… as part of keeping people on their toes. Sort of like how you should occasionally question your leaders. And if that doesn’t satisfy you, let me offer you this:
- If we shadows have offended,
- Think but this,—and all is mended,—
- That you have but slumber’d here
- While these visions did appear.
- And this weak and idle theme,
- No more yielding but a dream,
- Gentles, do not reprehend;
- If you pardon, we will mend.
- And, as I am an honest Puck,
- If we have unearned luck
- Now to ‘scape the serpent’s tongue,
- We will make amends ere long;
- Else the Puck a liar call:
- So, good night unto you all.
- Give me your hands, if we be friends,
- And Robin shall restore amends.
While I understand and appreciate the concern you’re bringing forward in this post, my feeling is that you’re way high up on the ladder of inference.
The Firefox developers are indeed very capable of listening, and have proven that they do so frequently and often. We empower “lowly users” (though none of us think of them that way) more than any organization I know, and give them the opportunity to voice their concerns and advocate strongly for their views.
Listening is not the same thing as obeying, though, which is where I think your argument goes astray. Yes, we’re listening, and we definitely know that there are a vocal set of users who rely on profile management for a set of their daily tasks. You may indeed be surprised to know that a *vast* majority of Firefox developers are in this subset as well, since we’re pretty constantly flipping between profiles for testing one version of the browser or another. What we have discovered, however, is that supporting multiple profiles can lead to some truly thrilling edge-condition bugs. The comment you decided to quote was from a bug to do with what happens if, say, a user creates a new profile and specifies that it is to be contained in c:\someimportantdir. Deleting that profile would delete all the contents of that directory! Now, we could spend a lot of time and effort trying to protect from that case (and indeed, we’d never get it fully right, as add-ons can lay down files in profiles with no way for Firefox to tell) but the decision was made that since profile manager isn’t meant to be used by average users, we weren’t going to worry about this (admittedly dangerous) use case where someone can shoot themselves in the foot.
Further, closing the door on that particular bug was a response to people behaving poorly and using improper Bugzilla ettiquette. This is another aspect about listening well: there is a cost to listening, and just because someone takes the time to write does not mean that they are owed a response.
Finally, we constantly revisit design decisions in the light of new information and changing contexts. Two years ago, with Vista and OSX taking over with Fast User Switching capabilities, there was a real expectation of one-profile-per-user. It’s not at all clear that was realized, and it would be wonderful if someone had better data than “everyone I know” to help us understand the current context. Supporting multiple profiles is not cost free. A good start would be to collect the passionate community around that need and generate a proposal for UI, or get the web to help create an Add-On to show us the experiences you’re trying to create. Yes, even you, a lowly user, can participate by being part of the solution, not just the identification of what you consider to be a problem. Work with us to convince us all that this is worth the effort and investment. I certainly think (especially with the deep personalization that we’re moving towards with things like the Awesomebar) that there are arguments to be made here.
I find it interesting that many people believe that because we are a community driven, open source project that means that we will gleefully fix all bugs reported. Often times, one man’s bug is another man’s feature, and to design well we must choose what we believe are the right experiences to focus on. Problem identification is easy; collaborating with the community to find solutions – that’s the hard part.
So yes, we’re willing and capable of listening. Are you willing and capable of participating?
Hmmm… well… “inference”… isn’t that what blogging is all about? 😉
Of course I understand that those bug quotes are from a slightly different bug – but my point was, that’s a dangerous sentiment to have. Oh, I understand the reasons (in the context of that bug), but the way I found that quote was through a discussion group on User Profiles being removed from the main UI. So there are people who see it this way.
And just to be clear – I’m a software developer, too. I understand the issues faced by this project better than the average user.
So, maybe this is more about a perception problem than an actual technical one. I hate to sound cheesy, but perhaps some “PR” is in order here, to explain things? Reading bug reports to find out why things were done can be quite difficult, sometimes even for me. (Sometimes parts of the conversation take place via email instead of bug comments, which is really hard to decipher.)
I’m glad to hear you’re willing and capable of listening. It’s very reassuring to hear you say so. As a user, it’s part of my job to occasionally doubt the motives of the project team, lest we become complacent. A little shake-up now and then results in more confidence after its been resolved.
And yes, I am more than willing and capable of participating – I do my best to dutifully file the best bug reports I can when I (rarely) find bugs. I fire up virtual machines to test new builds in new configurations. And I promote Firefox quite strongly (try visiting my site with IE and see what I mean).
And if I have come across as a little too strong, smug, or insulting – I apologize. But you must admit, the end result – your reply – is exactly the kind of reassurance I was after. So in the end, my post achieved the results I was after. 🙂