The removal of the alternate stylesheet switcher

Petr Tomeš · Post by **Petr Tomeš** » August 25th, 2004, 1:32 am

http://forums.mozillazine.org/viewtopic ... 566#743566

Adam Hauner: "All extensions have localization issues, same with Mozilla Update site. Our Czech users want to get czech interface..."

Localization is very important for this project!

Vicne · Post by **Vicne** » August 25th, 2004, 1:57 am

Blake wrote:people here would do well to remember that 1.0 is not the end of the line.

That's frightening ! Do you mean you're gonna remove other features ? :-)

More seriously, I can accept the argument that a feature should be removed from a product if it's not useful to anybody in the targeted audience, but Style swithing and View-source definitely don't fall into this category.

I can accept the argument that a feature should be removed from a "extremely minimal build" if it significantly decreases the download size/memory requirements/performance, but I doubt these 2 features will (view-source is still available if I'm not mistaken, only the menu entry isn't, right ?). Numbers welcome.
However, a "not so minimal build" (or technically a bundle with preinstalled extensions if you like) should be available too. Firefox may be targeted at a large public, but future developers are in the large public now.

My opinion is that general features should never be removed "because they have bugs" (unless security is involved). If it takes time to fix these bugs, so be it, either leave the buggy feature or wait for the bugs to be fixed.

However, those features could be moved to a less visible place such as "Tools > Advanced > View source" or something like that, to avoid frightening the average user with terms he doesn't understand. I think that would make perfect sense.

Just my 2 eurocents.

Vince

stony3k · Post by **stony3k** » August 25th, 2004, 2:14 am

This is slightly off-topic since its about the "View Source" functionality. Currently if you visit a page with frames and wish to view the source of one of these frames, you can do that using This Frame -> View Frame Source. If this functionality is removed, then it will be difficult to view source in sites that use frames. I hope the Firefox developers have thought of this.

e-Gandalf · Post by **e-Gandalf** » August 25th, 2004, 3:02 am

Blake wrote:Comments to the effect of "there's no point in trying to disagree with the Firefox team" are just silly, given that I started this thread in the first place. The Firefox team listens to well-reasoned arguments that take reality into account.

Please, don't blind us with such words. What You're doing has nothing to do with "listening well-reasoned arguments". First of all You're making important decision that have impact on Firefox UI or functionality behind the scene. So instead of saying - "hey! we wonder to remove Alternate Switcher because it's buggy. If You can find any important problems with it - tell us about it" and hear the response and AFTER that make decision, you're making decision, realising it and somewhere in the middle (but near the end of process) the world hears about it. Thats great that we can now discuss the matter, but please, don't give us a feeling that we have any impact on the situation. The thing happened already.
Is it so hard to tell us about it a few days earlier? For example write us today what else will be removed from Fx 1.0? (yeah, bookmark notificator - bug 253478)

Blake wrote:Second, people here would do well to remember that 1.0 is not the end of the line. It is a single release, our first one ever to the public. It will certainly not be our last. The development cycle from Phoenix 0.1 to Firefox 1.0 has already been ridiculously long for a single release. We do not intend to make the same mistake Mozilla did in thinking that 1.0 has to be perfect in every regard. We are going to get to a solid, reliable 1.0, and then we are going to go full steam ahead with 1.5 and 2.0. And like all other software, we won't take 4 years getting there.

Sorry to say but impression around in my Mozilla community and a few others are completly different. Firefox is in the moment strong and very widely used product with a name which starts to be well known and a potential to grab the crown. Removing features that were in 0.9 will be very hard to explain by local communities to users and to journalists who'll write articles about Firefox. The key point is that Firefox 1.0 is not first public widely known version. I really don't think there will be any "jump" in market share after 1.0 release bigger than after 0.9. You're saing that development cycle was long. Formally yes, but somewhere around 0.8 Firefox (firebird) started to be stable, well known, highly promoted product with no difference to any other "stable, post 1.0" products. You can't force your way of thinking to everybody, because most of users will not ever hear it. The only change made by "1.0" release is this magic value which says "hey, i'm very stable and feature complete" - so users who still uses 0.8, 0.7 and every 0.9 will switch to 1.0 just to find it less usable.
I'm afraid that it will be very hard to regain those people with 1.5 release, and you'll wash your hands and work on 2.0 when tousands of people from communities will be fighting hard about what to say about 1.0 release to users, anwsering complicated questions and trying to spread a word.

Blake wrote:Changing stylesheets as a feature concept has not been removed from Firefox. We fully intend to revisit it in the 1.5 timeframe.

Please, give any ETA for 1.5 so those communities could tell something to users

Blake wrote: However, like offline, it is not yet of shippable quality, and it's not going to be. Those of you complaining how absurd it is that we're removing features this close to 1.0 are missing the point: it's precisely <i>because</i> we're this close to 1.0 that it's time for these features to go. They've been allowed residence in Firefox all this time because we weren't ready to ship, and there was a chance they'd get fixed up. But now we're revving up the engines, and they need to go.

Still can't agree. So it's ok to have some features in for years and remove it month before release because they're not fixed? How about bugs that could appear because of this change? Will they be fixed? And what if costs of fixing those bugs will be the same as a cost of fixing those features?
Why there is time to invent, implement and fix so many bugs with livemarks and there is not time to fix those about style switcher?

Blake wrote:Firefox 1.0 will have an impressive, stable and polished feature set that will suit a large number of web users. As we drive toward 1.5, 2.0 and beyond, we will continue to refine and extend our feature set. But it's time to release 1.0.

How can You call product "impressive feature set" in the timeframe of removing some of them? Hey - 0.9 has more features than 1.0! And i'm sure that more users will use Alternate Style Switcher or Offline Mode than LiveMarks which still ar not too usable (adding 15 feeds makes bookmarks looks ugly and very bloated - compare to Sage). How can You call it stable if You're adding completly new set of features less than month before release and removing other? I don't belive in miracles too much. So i'm pretty sure that this strong and very chaotic moves in the project will have bad impact on stability of Firefox 1.0 and a press it will get. (what i mean is that if you needed to remove it, you should made it ~2 months ago)

jrduncans · Post by **jrduncans** » August 25th, 2004, 4:17 am

I can see removing the Alternate Style Switcher from the status bar, as that's a pretty in-your-face place for a feature the Devs aren't happy with. However, I think it would be better to simply move it to the view menu. View->Alternate Style Sheets->(list of style sheets, including no style sheet). This would allow users to use the current functionality without the need for an extension, while limiting the impact on all those potential users that we wish to court. It certainly wouldn't be more confusing to these users than the fact that "Character Encoding" is on the view menu (seems to me that should be moved to Tools->Options->Advanced).

stylo~ · Post by **stylo~** » August 25th, 2004, 4:22 am

After 3 pages of posts, the summary appears to be:

90% vociferously opposed to removal for obvious and well noted reasons
5% indifferent because they don't use it
5% in favor *IF* there are nasty bugs (though none of that nature have been detailed and no one using it thinks there are any such bugs)
0% strongly in favor

>>The Firefox team listens to well-reasoned arguments that take reality into account.

Sounds like an overwhelmingly reasonable case for keeping it, which is what you asked for Blake and said you would listen to.

Fix the icon, tweak a little, and on to 1.0!

Mozcerize · Post by **Mozcerize** » August 25th, 2004, 5:13 am

There is no point keeping the Switcher option in the UI (presumably in the menu system) if its icon is removed from the status bar. Either it should all stay or it should all be factored out into an extension.

The whole thing that makes the switcher great in Fx is that it shows you at a glance whether alternate styles are available. (I've even gone so far as to restyle the icon so that it flashes a couple of times when I meet a page with alternate styles, in order to attract my attention to it.) Sticking alternate styles in a menu a la Opera is just ridiculous; who's going to look in that menu every time they visit a new site just on the off-chance that there's an alternate style?

My secondary point: the switcher is a strong candidate for that all-important soundbite-friendly bit of candy to bait users to actually act on the security scares that they've heard about IE. The rant below tries to explain this - stop reading now if you're not interested! ;-)

Moving such an innovative feature is a grave mistake. (Actually, innovative is hardly the right word, because alternate styles via stylesheets is patently the way the web should always have been, and this has been recognised by W3C for years.) Fx has little to compete with IE now. Both now have pop-up blocking. Tabbed browsing sounds completely pointless to those who have never used it (though most seem to come to love it once they've tried it for a while). Security scares don't seem to be enough to make people come to Fx; but probably they've heard about other browsers; maybe they need something extra as a bit of bait to persuade them to make that change which they know is sensible but suspect is a boring chore. What is this bait in the case of Fx? Realisitically, it can only be something in the main UI that'll they'll notice and use for comparisons with IE. And the ability to change the style of their favourite sites at will might just entertain them enough to give Fx a go.

And having it there in the principal UI of a mainstream browser will encourage more site developers to give alternative style sheets a go.

But put it in an extension? Well that defeats the purpose if we want stylesheet switching to be something that normal users do. Extensions are for advanced users: make it as easy as you like to install them, and people still won't bother checking out the lists of available extensions. Hell, I'm someone who takes a big interest in my browser's UI and capabilities, and even I struggle to be bothered to read that list of 138 extensions to see if there's anything useful there.

But if stylesheet switching is going to be relegated to a geeks-only pastime, then let's hope that at least it does get made into a decent extension.

Post by **Nitin** » August 25th, 2004, 5:22 am

stylo~ wrote:After 3 pages of posts, the summary appears to be:

90% vociferously opposed to removal for obvious and well noted reasons
5% indifferent because they don't use it
5% in favor *IF* there are nasty bugs (though none of that nature have been detailed and no one using it thinks there are any such bugs)
0% strongly in favor

How very scientific!

stylo~ wrote:>>The Firefox team listens to well-reasoned arguments that take reality into account.

Sounds like an overwhelmingly reasonable case for keeping it, which is what you asked for Blake and said you would listen to.

While I cant speak for Blake, I think posts like yours are precisely the ones that were'nt included in the "well-reasoned arguments".

DIGITALgimpus · Post by **DIGITALgimpus** » August 25th, 2004, 6:06 am

kstahl wrote:That being said, I basically agree with what TheOneKEA said. The AltCSS tool should stay in. It's small, unobtrusive, and the bugs are not really apparent unless you know already what you are looking for.

Agreed.

Perhaps the best solution, if these bugs are really so bad, is to put a simple dialog on the first time you change styles that says:

"this is an experimental feature as Firefox becomes compliant with CSS2 specifications. There may be bugs in this feature"

Removing it this late in the game isn't wise. Many bloggers (a big userbase of firefox, and big advertisers) are addicted to this feature.

It's like going back to making all bars men only. You loose the audience of people looking for women. Leaving only the drunks behind.

glazman · Post by **glazman** » August 25th, 2004, 6:08 am

blake: we're not missing the point at all. But you are totally missing ours : we do feel such a feature is a must-have for a first very visible release. Speaking of stability and quality, this feature has been IN for THREE YEARS, so please explain why it suddenly becomes incomplete and/or broken today, and why a CSS 2 conformance criterium is gone.

poningru · Post by **poningru** » August 25th, 2004, 6:19 am

Can someone from the Firefox dev team please put up a well-reasoned argument negating the points that were raised here.

mdakin · Post by **mdakin** » August 25th, 2004, 6:23 am

I think disabling a feature because of its lack of completeness is ok for a first time release. I was against removing "view source" but, removal of buggy CSS switcher for Firefox 1.0 seems to be the right thing to do. It can be enabled after the major release.

I guress there were two options,
- fix style switcher, offline browsing then release 1.0 late
- Disable buggy features and release Firefox 1.0 earlier

If I dont have enough time and everybody is waiting impatiently for a *clean* 1.0 release I would go for the second choice.

glazman · Post by **glazman** » August 25th, 2004, 6:26 am

Again, please explain how and how deep the Style Switcher is buggy...

whistleway · Post by **whistleway** » August 25th, 2004, 6:32 am

As much as i love it, I am sure, whoever is aware of it, is already a mozillazine member. As long it can be obtained as an extension, or something like that, any developer who is rooting for it, is insane.

When your grandma knows stylesheets, then, it's ready to give it a prominent location. Till then, it is one of the "confusing bananas".

Xout

Will Rickards WT · Post by **Will Rickards WT** » August 25th, 2004, 6:59 am

You need to put it back because:
The bugs aren't serious enough and most are enhancement requests. People who use firefox now actually use this feature. Maybe most of them are web developers but even so that is one user group you want to support. 1.0 is likely to get a lot more downloads and wider deployment. You want to expose these features to those people. These kind of features make firefox the better browser from standards support viewpoint.
I'll address each bug listed in Asa's second comment to bug 253722.
bug 216537: alternate stylesheet selection is not sticky
This is an enhancement, shouldn't even be a reason to remove it for 1.0. The basic functionality works fine.
bug 253332: Implement Page Style > None
This is an enhancement, shouldn't even be a reason to remove it for 1.0. There may or may not be a valid patch for this already.
bug 216424: Can't use stylesheet switcher from keyboard
Minor bug, it isn't necessary to use the feature. It would be nice but it surely isn't a good enough reason to remove the UI from 1.0.
bug 224250: Selecting "No Theme", then reselcting "Basic Theme" doesn't reapply styles
Probably a legitamite bug, but a minor nit. It can be worked around probably by reloading the page or some other method.
bug 220649: No Theme Selection Not Available on CSS sites that don't offer alternate stylesheets
This is an enhancement, shouldn't even be a reason to remove it for 1.0.
bug 83663: Alternate style sheet setting not stored [AltSS]
This is an enhancement, shouldn't even be a reason to remove it for 1.0. Similar, probably identical implementation, to the sticky bug above.
bug 220975: Style sheet selector does not display "No Theme" for pages without a base stylesheet (but shows Basic Theme)
I think this is an enhancement. At very best a minor bug. It is partially the page author's issue.

So no serious bugs out of all of those. Show me something that justifies removing the feature.

The removal of the alternate stylesheet switcher

not a big deal