Where is the tool to import currect mozilla/netscape suite mails?
I remembered it's in 1.0 raodmap before.

Be sure to imitate Firefox's FTP structure for 1.0 final.

Also, off-topic kudos to whoever rigged up the Firefox product page to offer Firefox en-GB to visitors from Britain.

Hmmmm. Well, I haven't really fallen in love with any of the new features, but that's kinda to be expected- a mail client just isn't as "sexy" as a browser. It certainly seems rock solid thus far, at any rate. I do appreciate the Winstripe theme, nice to have the "birds" matching up again.

Hmmm- speaking of birds, now that its big brothers are grown up, is there going to be more weight behind Sunbird? Or is that just a permament side project?

Winstripe? It's not in my TB!

Am I the only one who feels that Tb just isn't as polished for 1.0 as firefox was?

Wait a second what's new on 1.0?

Everything you've listed up was already on 0.9......

inertia, no. It doesn't have software update or a Winstripe theme. It still uses the temporary Junk Mail Controls dialog from 0.3; the Install buttons in Themes and Extensions don't have an icon. It says "greater than" when it means "not less than". After inserting a smiley into a message, one can't add a subsequent space.

It doesn't have a Help file. I'll say that again - it doesn't have a Help file. F1 does nothing. "Thunderbird Help" leads to texturizer.net/thunderbird, which explains how to tweak obscure settings, but not how to send an email.

Thunderbird 1.0 is certainly a passable effort, but it's no Firefox.

Greg, I disagree. Thunderbird, to me, is as big an improvement to Outlook and Outlook Express, as Firefox is to IE. They're both small, fast, cross-platform, threat-resistent, themable, extendable, and come with a non-trivial number of features that I require that aren't available in any other email client. Thunderbird's junk mail controls alone make it light-years ahead of the clients I was using before it. The fact that it doesn't rely on IE and so I'm not going to get mutilated from virii and other nasties that infect and replicate via email is also a huge bonus. An integrated solution for handling my RSS reading is a big bonus, too. I think that Thunderbird 1.0 is definitely on par with Firefox 1.0 as far as a high-value user experience. And in my book, it's an equally important leap forward over it's predecessor, Mozilla Mail as Firefox was over Mozilla Browser.

...and the Thunderbird Pinstripe look is beautiful.

--Asa

Asa, I don't think anyone is complaining that Thunderbird isn't a very good mail client. It's excellent, and having come from Outlook Express at Tb v0.7, it was an enormous improvement. However, the point is that Thunderbird is still a bit rough around the edges when compared to Firefox when that attained 1.0.

The sudden acceleration of Thunderbird's version number didn't seem in keeping with the number of bugs that were being fixed, and as Greg pointed out, there are some pretty major omissions. I've encountered a number of serious bugs. The Thunderbird developers are as talented as the guys working on Firefox, but sadly there don't seem to be enough of them.

Asa, you really need to work on replying to concerns raised, instead of pushing other things. TBird 1.0 is not going to be as polished as Fx 1.0. That doesn't make it any less useful/awesome/kickass, but its still not as polished.

Lack of l10n-in-CVS (no official localized builds)
Lack of real Help on par with Fx Help.
Pinstripe may be great, but Qute is what the vast majority will be using.
Lack of software update is a bad thing.
Still uses wallet instead of password manager. I'm partly responsible for wallet, and all I want to do is kill it. Reading wallet code is pain.
There's a bunch of minor UI bits that bug me too, i.e. when you have an account selected in the tree view, Edit->Properties and Edit-Account Settings do the same thing. And they're next to each other. If you have a folder, Edit->Properties is Folder Properties. It'd be better to just disable the menu item than to have duplicate functions. Probably less code.

I don't know the current state of Tbird Bugzilla, but I still get weird things going on on a daily basis, and UI bits that just seem wrong to a veteran of the Firefox UI process. I've used Tbird since the first builds with the name and the version number (0.1a) so I obviously have learned to live with it and love it, but its just not on the same level. Given triple the hackers and another couple months to just fix polish stuff, we might be there.

Its probably the best mail client out there, but ignoring the fact that its not going to be at the same standard as Tbird is just not going to be productive.

Mike, I disagree :-) I don't think l10n in cvs should have even happened for Firefox. It cost us quite a bit of time and effort for not that much reward (all those localized builds would have happened, probably within a week or so and we wouldn't have spend hundreds of hours of qa and build time on it.)

Pinstripe and Qute look great. Their's nothing wrong with them at all. They're the most complete themes we've ever seen on a Mozilla email release. Qute wasn't replaced as the default theme because it sucked and Pinstripe for Thunderbird Mac kicks way more ass than Pinstripe for Firefox Mac.

Software update for Firefox is a joke. It's pretty much completely busted and I've suggested we disable it if we need to do a security update. Ben agrees, I think.

There's nothing at all wrong with wallet for Thunderbird. Users don't care about whether the code is painful for you to read.

There are a bunch of minor UI bits that bug me in Firefox (more than do in Thunderbird, actually.) I've got a list of at least a few dozen Firefox UI nits that are more serious than any Thunderbird nits I have.

The only item on your list that is at all interesting to me is Help. I'd love for our apps to have decent help. None of them do. When we write some good docs and integrate them into the OS help viewers (which each platform we run on conveniently provide us if we'll just use it) then I'll feel a lot better about all of our applications.

I really don't think Tbird is at any lower standard than Firefox. You're free to disagree but I think Tbird was actually ready well before Firefox was. It didn't see the initial destruction that Firefox saw and the only things that kept Tbird from releasing a few months ago, in my mind, were the late feature additions. Tbird has been a solid and well finished mail app for a long time. It's about time it shipped.

--Asa

The icons in this version are so much better than 0.9. They are a bit smaller and look more proffesional. Fantastic job!

"Software update for Firefox is a joke. It's pretty much completely busted and I've suggested we disable it if we need to do a security update. Ben agrees, I think."

IMHO, that sucks :( I thought security was high priority, but I guess it isn't THAT high a priority. I find that sad :(

> I'd love for our apps to have decent help. None of them do.

Sure. Seamonkey has a really detailed and complete help, and it's also written very well.

"I'd love for our apps to have decent help. None of them do."

Which I don't think excuses why TB doesn't even try to.
At least Firefox do, and while its help is thin, it covers pretty much the entire GUI.

Asa said: "I don't think l10n in cvs should have even happened for Firefox. It cost us quite a bit of time and effort for not that much reward (all those localized builds would have happened, probably within a week or so and we wouldn't have spend hundreds of hours of qa and build time on it.)"

We mustn't underestimate the reward we've got from _not_ getting lots of feedback from people around the world that their build of FF 1.0 sucked, because it was built with inconsistent or unusual build options or not properly optimised, etc. Having l10n in CVS means that all the Firefox builds we are distributing are of equally high quality. Given that only 30% of the web speaks English, I think it was well worth it.

You also need to count in the negative community effect there would have been from continuing to poorly cater for (or be perceived to poorly cater for) non-English languages.

But without l10n in CVS, won't we have a bunch of builds spread in various places that nobody will be able to find? Unless of course you're saying that for Thunderbird, Mozilla is going to disregard/change the trademark policy and procedures that localisers were wrestling with for Firefox. It's going to look rather silly if, while ensuring that unofficial Firefox localisations aren't linked to or hosted, unofficial Thunderbird localisations are hosted. If there are reasons why mozilla.org can't allow it for Firefox, that's fine. If there aren't, then what was the problem? (I understand that different people working on the projects have different views, but the localisers and others should still see a unified official position. Hopefully there will be some clarity on this soon...)

By the way, I'm not sure if you're saying that l10n in CVS isn't worthwhile in general, as it seems to me that it's a better way of doing things. I'm not sure it needed to happen on the branch for Firefox 1.0 though - seems it could have been introduced more smoothly if it had just been done on the trunk for 1.1.

Similar issues around Qute - it may not have been replaced because it sucked, but there was, apparently, an urgent need to replace it in Firefox for licensing and branding/visual-identity reasons. Again, if there wasn't an urgent need then why did it happen in a hurry, leading to a release with a version of Winstripe that the designers of it (not to mention the users) weren't happy with?

noname: There isn't much "new" on Thunderbird 1.0 - it's a fixed up and more polished version with the features that were in 0.9. I suppose you could consider 0.9 a beta or preview release of 1.0.

Sorry - that looks a bit unclear, the first part of my comment was a response mostly to Asa's comment, rather than the following comments.

Asa said: "I don't think l10n in cvs should have even happened for Firefox. It cost us quite a bit of time and effort for not that much reward (all those localized builds would have happened, probably within a week or so and we wouldn't have spend hundreds of hours of qa and build time on it.) "

I'll have to disagree with you 100% on this one. A l10n-team should not have to know how to compile Firefox/Thunderbird, create .xpi packs, etc., it should just need language/translation skills, and nothing else.

I too feel Thunderbird versions are being rushed out. Things are nowhere near polished. Adn there are tons of real sucky bugs. I still use it but overall the feeling is much more spaghetti code than FF, for example:
- newsgroup indexing is laughable - those files get screwed almost weekly. If I delete them to force reindex only messages still on the server are being indexed even though the old messages I wanted to save are still in the local file. But they don't get dispplayed.
- TB seems to forget preferences after being open for extended time. Things like don't mark my email as read, I already agreed on the email server having different sec cert than the domain (uses vendor one), stops checking some accounts, forgets passowrds... Restart fixes them all but it's annoying
- Unability to thread commands like pull email form one place while i work with another.
- Occasional 100% CPU usage even though it doens't seem to do anything
- menus and options not quite to stuff as michaell already mentioned.
- help sucks. Moving my data to a new profile with a new version of TB should be well documented. Overall never got any usefull help when needed.

Asa, l10n-in-CVS was long overdue, and anecdotally seems to have improved the quality of the builds significantly, since I've seen a lot fewer reports of oddly broken builds than for previous releases. And when you consider that there were less localized versions and any bug reports weren't official, so Bugzilla reporting wasn't being encouraged, that becomes a bigger change.

Qute IMO wasn't going to live up to the potential that Winstripe has, I installed it to compare what Firefox 1.0 would have looked like, and it just seemed kinda washed out and faded. The direction that Sunbird/Firefox have gone seems so much cleaner and visually pleasing than Qute, but that's just my opinion.

Software Update is essential for something professional. We need to make it work if its still broken.

These lists of technical shortcomings of T-bird are all very well, but it's ignoring the marketing problem. The fact is that Firefox has done well because people were becoming afraid of using IE, because of all the negative publicity on the security issue. There's not the same force driving a switch from OE to T-bird, so T-bird will have to be that much better to succeed. And it's not. For corporate users, no-one who has Outlook2003 on the desktop is going to give T-bird a second glance. I don't think downloads for T-bird will get to the high thousands in a few days, and those who do try it won't notice great improvements. Which is a pity, because there is a real gap for an effective email client, with a small footprint, but that does the essentials well.

W.

> I don't think l10n in cvs should have even happened for Firefox. It cost us
> quite a bit of time and effort for not that much reward (all those localized
> builds would have happened, probably within a week or so and we wouldn't have
> spend hundreds of hours of qa and build time on it.)

Please take a look at http://firefox.no/stats.html
"Not that much reward" is about 30% of all your downloads (in fact, more than this as the german stats have been reset). That would _never_ have happened if users weren't provided a build in their own language on the download page (which should have been translated as well but that's another problem). I beg you've never been involved in repackaging localized installers like we had to do in previous release, which really was a pain in the ass.

And in fact, I find your remark rather insulting for international users. Why wouldn't they deserve QA testing exactly?

Benoit, perhaps I should have stated that differently. I'll try here. Given how late in the development cycle the "official localized builds and simultaneous releases" plan was begun, and given how much of our resources it consumed, if I had it to do over, I would have said no and pushed that out to Firefox 1.1. Now that the costs in our Build Engineering and QA have been sunk, I'm glad that we have all of these localizations in CVS.

On the "not much reward", I'm talking about the reward above and beyond what we would have gotten had we done what we've done for every previous release in Mozilla's history -- get unofficial localized builds for French, German, and Spanish usually within a week or so after the official enUS release. I don't think that the (my estimate) 5-10% improvement in downloads that probably resulted from having "official and simultaneous" releases of a dozen localizations was necessary worth all of the work we had to do to get it to happen -- not to mention the wave of dissappointment that came to me and others via email from those l10n teams that were "close" but not quite ready so were held back on the night of release.

I'm all for localizations in CVS and coming off our our official machines. It gives me a much higer level of confidence in the quality of the build (not necessarily the quality of the translation) so it's definitely a good thing. Would I have signed us on for the work that was required had I known then what I know now about the schedule and the amount of work? I don't think so. I'd probably have said "let's shoot for 1.1."

--Asa

Thanks for the clarification. I just would like you to realize how all this work has certainly not been done in vain. Not at all.

AFAICT, our fr-FR release is an overwhelming success. Back in may, when I talked about Firefox, nobody knew nor wanted to hear what it was. And now, I can see it installed everywhere (friends, school, university) despite the lack of mainstream press coverage in my country (Belgium) and I really think it has something to do with the availability of our build on the Mozilla.org front page, and the fact they were automated. Another thing that we'd never been able to achieve is the simultaneous Mac release, as nobody in our core l10n team has a Mac.

Looking at the download numbers given above, we've done this month 50% more than in three months of 0.9.3. Given the difference between the numbers given there and the ones on spreadfirefox (probably including download.com and the like), I can only guess that they are in fact much higher than that.

So I can understand it was difficult, it was for everyone, but it's now done and was worth it.