Mozilla Firefox 2 Alpha 1 Available

Mini-Geek writes "Code-named Bon Echo, the first Alpha of Firefox 2.0 is now officially available. You can download it at ftp.mozilla.org. From the article: 'Here are some new features in Bon Echo Alpha 1 that require feedback: Changes to tabbed browsing behavior, New data storage layer for bookmarks and history (using SQLlite), Extended search plugin format, Updates to the extension system to provide enhanced security and to allow for easier localization of extensions, Support for SVG text using svg:textPath'"

Comments

But...

[Onymous Hero]

Will it use less memory than 0.x / 1.x ??

Re:But...

[fermizhang]

It's a bad experience swithing between 1.5 and 2.0 a1...seriously...

Re:But...

[Adult film producer]

I hope so, check out Ben Maurer's latest blog entry, near the bottom he talks a bit about this. Actually, the latest entry is quite informative about the new memory mapping features in the latest kernel.

Re:But...

[hkgroove]

Yes, but you have to think in Russian for it to be efficient.

Re:But...

[bunratty]

The problem described looks like Bug 259672.

Re:But...

[Pneuma ROCKS]

That's why it's called an alpha developer build. It's barely beginning active development. You cannot expect it to be stable. You cannot even expect it to be safe. This is not a beta or release candidate. It's meant only for testers and active developers. Use with extreme caution.

Re:But...

[Isotopian]

Dammit. I knew I should have kept Clint Eastwood around for something.

Re:But...

[temcat]

I do, but it's still not :-(

SQLite

[kevin_conaway]

We all know that Firefox has had (and still has) a lot of memory issues. Will embedding a database in memory help or worsen these issues?

I haven't used SQLite, can anyone with experience using it please comment?

Re:SQLite

[Ginger Unicorn]

I believe SQLite is a flat file database system, which means it is very lightweight and simple. I suppose it's like having an Access database inside Firefox. There aren't massive overheads.

That is all just a guess, i dont know for sure

Re:SQLite

[G)-(ostly]

I forsee no problems. It's a surprisingly minimal addition to a software package, and the problems with Firefox's memory management are very likely in unrelated modules.

"SQL" engines tend to evoke images of hulking software packages like PostgreSQL, SQL Server, and Oracle, but those things do an awful lot more than the typical desktop app needs, and the SQLite engine is much, much simpler in order to meet that lesser demand.

Re:SQLite

[LurkerML]

http://www.sqlite.org/

The website says 250KiB fully configured. That is tolerable, i think.

Re:SQLite

[smooth wombat]

We all know that Firefox has had (and still has) a lot of memory issues.

We do? Funny, I've been running FF since the 0.8 days (Phoenix) and have never had any memory issue. In fact, I've never had any issue other than one mini-crash which forced me to use a default profile until I pulled up my old one. Further, I've installed FF on several different systems, including W98, and not one of those systems has ever had a memory issue.

Looking at the FF boards it appears the issue is not so much with FF but the multitude of extensions that people think they need to install.

Re:SQLite

[Dr. Evil]

The issues appear when you use more than one major app on the machine on a regular basis. They're not too bad now, but they were horrible several years back, and it's not just becuase machines have more RAM.

http://www.google.com/search?hl=en&q=firefox+%22wo rking+set%22&btnG=Google+Search&meta=

Re:SQLite

[bperkins]

Mu.

The memory usage problems have been related to the image cache. (I've heard that this is often caused by an old version of the adblock extension)

Using SQLite to store profile information will probably have little impact the memory usage problems people see.

Re:SQLite

[CastrTroy]

Same experience for me. I swear is has to do with some extension that people are running. I only run flashblock and Web developer.

Re:SQLite

[bunratty]

All browsers have lots of memory issues. They also all have security problems, they crash under lots of different situations, have many kinds of CPU use problems, and thousands of other kinds of bugs. What else is new?

I don't think embedding a database will noticably impact memory usage. The most noticable change will be that your bookmarks, cache, and other parts of your profile will not be corrupted or lost nearly as easily. The dataloss problems SQLite will fix are much more severe than the memory problems some people expereience with Firefox. Firefox is using too much memory? Just restart. Firefox lost your bookmarks? Tough sh*t!

Re:SQLite

[smittyoneeach]

You believe falsely. (Possibly you're thinking of BerkelyDB). http//sqlite.org has full disclosure, even describing the virtual machine and opcodes that drive this little ~300kb, mostly-ANSI '92 compliant wonder.

Combine a scripting language for end-user forms, and you've got everything MS Access wishes it could be.

Get Hipp. Get SQLite.

Re:SQLite

[Jjeff1]

I have firefox 1.5.0.1 on windows xp with latest adblock and filterset G updater, nothing else.

I've noticed that web pages that refresh themselves cause a run-away memory situation. Specifically the win32 MRTG package from open innovations causes FF to use huge amounts of memory. It auto refreshes graphs I think every 10 seconds. If I leave a graph up on screen and leave for the weekend, FF will be using 1.8 GB memory when I come back on Monday. I've been unable to find out if this is a known problem or not, so I've not submitted this as a bug.

Re:SQLite

[ameoba]

I have to wonder how much is gained by moving bookmarks and prefs from plain text files (HTML & JS) to SQLite's binary file format. It's not like their bits of data that need complex searches done on them.

Re:SQLite

[dlZ]

I have the same issues with one of my web mail clients. Gmail doesn't cause any serious issues (takes ages to up memory usage to anything noticable,) but one of my work based accounts, running mailEnable, will ramp up memory usage over an 8 hour period to 700+ megs. The thing is, it does it on my SUSE 10 x86_64 notebook and on another Windows XP SP2 notebook, but not on my desktop, which is also running SUSE 10 x86_64. The notebooks are just running adblock, while my desktop has that plus Grease Monkey, Forecast Fox, and a few other extensions I'm sure I'm forgetting.

Re:SQLite

[Luthair]

Until a recent release, FireFox could use a ton of memory if you were visiting large pages (ie, phpMyAdmin). Devs have said previously that Firefox only cahced the most recent 8 pages, however in my experience allocated memory continued indefinately and I often saw it using 500mb+

I doubt it was extensions, I only run GMail Notifier and AdBlock.

Re:SQLite

[digidave]

FWIW, I am using yesterday's trunk build and there doesn't appear to be any memory problem at all. With several Firefox windows open each with a few tabs the firefox-bin process is using 65 MB RAM. This is on Ubuntu Dapper.

Re:SQLite

[LiquidPaper]

Hope this means Thunderbird will get it too. mbox sucks!

Re:SQLite

[bunratty]

I doubt it was extensions, I only run GMail Notifier and AdBlock.

AdBlock did have severe memory leaks, and may still leak quite a bit of memory. Have you upgraded to the very latest version? All it takes is one bad extension to leak memory like a sieve.

Re:SQLite

[MooUK]

Adblock itself has had, and still does have, memory leak problems. The original maintainer shows no inclination to deal with them. It's generally recommended to use Adblock Plus, which has fixed most of these problems and also has useful new features, such as whitelisting. The maintainer of Adblock Plus has also shown strong interest in debugging and fixing any problems, including memory leaks.

There's a few problems that can cause leaks in FF itself which have been fixed in the main trunk. Almost all of those fixes are supposed to be included in 2.0.

Re:SQLite

[MooUK]

See my reply to the parent poster; it may help you too.

Re:SQLite

[xant]

Any word on whether filterset.g updater works with plus?

Re:SQLite

[bjpirt]

It's probably more efficient than the current state of storing history / bookmarks. At least this was designed to rapidly query data. With the amount of bookmarks I have now compared to 5 years ago, this will be great.

Will it support adding metadata to the bookmarks a la spotlight so that I can finally rid myself of the limiting hierarchical organisation I have now?

Re:SQLite

[mwilli]

I installed FireFox on my parents computer running Win2K. They complained the other day about not being able to add or remove bookmarks. I looked in the profile and there were tens of thousands of copies of the bookmark file, for some reason. I removed them and all works fine now, but not sure why this would have happened.

Re:SQLite

[Tezkah]

Yes, here are the instructions for using with Adblock plus.

Re:SQLite

[Jherek Carnelian]

Yes it does, been using it myself for months.

Re:SQLite

[mod-e-rate]

I don't know if this happens with everyone. But if i type something like http//adsasdasdads into the Fx address bar and press enter it takes me directly to the microsoft homepage!! Anyone face this (problem!) before?

But it was fun watchin where fx took me when i typed http//iamevil and http//donoevil :-)

Re:SQLite

[MooUK]

It works perfectly, including entries that take full advantage of Plus' additional features.

Re:SQLite

[zerblat]

Firefox already includes a database, mork, which is used for e.g. storing history. The problem with mork is that it's a completely braindead format (it's text based, but definately not human readable), it's practically unmaintained and it's almost imposible for third party programs to read.

So, this will make all the data that Firefox stores accesable to others, and hopefully all the mork-related bugs will just disapear.

Re:SQLite

[towsonu2003]

Looking at bugzilla instead of forums will help you to find memory bugs that are there. The fact that you're not getting it is irrelevant.

Re:SQLite

[jacksonj04]

Oh please... anyone up for a Google Desktop / Spotlight integration plugin?

Re:SQLite

[dlZ]

I'll have to check it out later. I just put the alpha on the SUSE notebook, and am using it at the moment. So far it's working great, so I'm going to play around with it. And flash now works on the notebook (always has on the desktop.)

Re:SQLite

[uncl_bob]

Happens to me too .. don't know why!

Re:SQLite

[bofkentucky]

you imported your homepage from your IE settings.

Re:SQLite

[Bambi Dee]

Well, Microsoft is Google's top result for "http" for some reason and the Firefox address bar doubles as a "I'm feeling lucky!" search box.

Re:SQLite

[Goalie_Ca]

I think SQL-Lite is a horrible idea. I would have loved for Firefox to use Xbel for the bookmarks format just as konqueror and others. It's a simple, standard, xml file format for bookmarks. Would make my job of synchronization much easier.

Re:SQLite

[Luthair]

I would have been using the most recent version of Adblock at the time. This isn't much of an issue now, a recent release of Firefox which noted a fix of memory usuage eliminated it.

My intention was ti simply point out that it had been a problem.

Re:SQLite

[G)-(ostly]

And it would make the primary task of a bookmark tool - data insertion and retrieval - much more complicated with a significant performance penalty for large sets of data.

XML is not for data extraction or storage, it's for data description, primarily to simplify the task of exporting, importing and merging data through disparate systems (as with synchronization jobs).

The ideal solution would be to extend the bookmark tool to allow it to export XML content from the current data store for people such as yourself.

Re:SQLite

[Asztal_]

1.8GB, quite impressive. I myself only managed 1.5GB, although it only took me around 30 minutes to reach.

Re:SQLite

[HolyCrapSCOsux]

Why do people use adblock? Isn't that what the hosts file is for?

If you don't know what I'm talking about, the Hosts file lines in Windows\system32\Drivers\etc\
think of it as a blacklist.
Windows won't allow those sites to connect to you, thus, No ADS!
If you are thinking, golly, that's alot of typing, my hosts file is 421k. You can copy paste from others off the internet.
that's one less process, smaller footprint, and speeds up browsing somewhat, as the ad connections aren't made so the crap isn't loaded.

Re:SQLite

[Sepodati]

One feature is regular expressions to block certain patterns from some sites, and allow other things though. Not sure how much that's used by people, though.

I think a good HOSTS file followed by a small adblock for specifics that you come across would be pretty efficient. Anyone know how efficient is Adblock when it's got a large list?

---John Holmes...

Re:SQLite

[Bloater]

Or rather to link the bookmark handling code with a small commandline wrapper and a small xml read/writer lib.

Re:SQLite

[baadger]

When compiled with MS Visual Studio 2005 on the default options, the latest version of SQLite comes out to be a 68kB DLL, or 132kB when VCR is compiled in.

Re:SQLite

[baadger]

...by VCR (The MS C runtime) 'compiled in' I mean statically of course, so you don't need msvcr80.dll, which is ~600kB, as a dependency.

Re:SQLite

[JSmooth]

Excellent, glad to hear you have no memory leaks running FF without extensions. The primary reason why I choose to use FF is because of the extensions. If I didn't like/need the FF extensions I would run Opera. Opera beats a plain jane FF install on pretty much every count.

Re:SQLite

[WWWWolf]

I haven't used SQLite, can anyone with experience using it please comment?

I use amaroK, which uses sqlite for storing its stuff, and for all I care it behaves just like a normal desktop app to me. I guess adding the thing to Firefox won't make it any worse either.

Re:SQLite

[sapgau]

Adblock allows you to block specific file types (i.e. gifs or flash) or you can block the hole site like a hosts file.

Sometimes you need the finer blocking because the original site you are viewing will also host it's own ads. This happens a lot with newspaper sites, you want to block their ads without blocking also the pictures related to a story you are reading.

Also the tool is integrated in Firefox so you don't have to switch applications or editors.

Once you start using it you'll notice that you can concentrate on what you're reading and when you read something without adblock you feel you want to cry.

Re:SQLite

[WWWWolf]

Why do people use adblock? Isn't that what the hosts file is for?

Two reasons:

Granularity. Specifically, the ability to kill stuff based on not only the host name, but other components of the path.

Flexibility. Regex matching http://hostname/ad/ or http://ad.hostname/

more fun than adding every one of those to your hosts file.

Re:SQLite

[larry bagina]

mbox has inherent problems, race conditions, locks, file corruption, etc. For networked use, maildir is better. For local use, a db-backed solution is better.

Re:SQLite

[1110110001]

Get Hipp. Get SQLite.

For everyone not getting the joke (hope it was meant as joke). SQLite is written by D. Richard Hipp.

Re:SQLite

[sb]

> Why do people use adblock? Isn't that what the hosts file is for?

No. Here's an excerpt of the hosts manpage from a Debian system:

HISTORICAL NOTES
              RFC 952 gave the original format for the host table, though it has
              since changed.

              Before the advent of DNS, the host table was the only way of resolving
              hostnames on the fledgling Internet. Indeed, this file could be created
              from the official host data base maintained at the Network Information
              Control Center (NIC), though local changes were often required to bring
              it up to date regarding unofficial aliases and/or unknown hosts. The
              NIC no longer maintains the hosts.txt files, though looking around at
              the time of writing (circa 2000), there are historical hosts.txt files
              on the WWW. I just found three, from 92, 94, and 95.

> If you don't know what I'm talking about, the Hosts file lines in Windows\system32\Drivers\etc\ think of it as a blacklist.

No, think of it as what it's meant to be -- a way to "advise" the resolver library so as to override the DNS, add host aliases or do name-based address lookup in the absence of DNS, which is useful in very small networks. It works like this because the resolver tries the hosts file first by default, though on Unix systems you can change this.

> Windows won't allow those sites to connect to you, thus, No ADS!

Wrong. Windows will see an incoming packet's source IP address. It doesn't need a name to reply, so at no point will your hosts file come into play. A program could try to resolve the address and log messages or control access based on the hostname. That address will not be found anywhere in the hosts list so again no effect.

> If you are thinking, golly, that's alot of typing, my hosts file is 421k.

Puke.

> You can copy paste from others off the internet.
that's one less process, smaller footprint, and speeds up browsing somewhat, as the ad connections aren't made so the crap isn't loaded.

Yes, the crap isn't loaded, but connections *are* attempted to whatever address is in the host entry. If you have 127.0.0.1, your browser will try to connect to localhost -- that's your own machine. This will fail even if you do run a webserver, because the requested page will most likely not exist.

So, from your point of view, you add domains to the hosts file and they go away. Under the hood though, the system has to do a whole lot of crap before the request will fail. Contrast this with an adblocker, which does patern matching on a URL and simply ignores it if it's on the blacklist.

Can you see now why your 421K hosts file is a horrible kludge? :-)

Really?

[wampus]

So is this any different than the Firefox 2 alpha that wasn't released yesterday?

Re:Really?

[brunes69]

This one appears to be an officially released alpha.

Re:Really?

[Denyer]

Yeah. This isn't a dupe, which we could probably do with a tag for on the article...

It's still just an alpha though.

Re:Really?

[Uber Banker]

On this one I guess they won't get their knickers in a twist because the announcement is 'official'. Witness the comments in the last story a couple of days ago:

"Are we really about to release an alpha where adding a bookmark silently fails (330052), bookmarking all tabs doesn't work (330929), and autocomplete is useless (330125, 330126)?"

Or my personal favourite

"The nightlies are now branded 2.0 alpha because... well, for some odd reason they like to brand their CVS builds before things get released, to make sure the act of rebranding breaks nothing. IIRC that actually hit them way back and they got scared."

Nice. So call it something, release it into the public (as is required by the license) and insist it doesn't exist. For a product to bear a new name means it is newly branded product, thus a nerly branded product has been released.

Of course it could be a genuine name trick or list of genuine problems, but I don't know why so many people went bitching in the previous story when the 'official release' is so soon after. Both were headlined Alpha. Cliques wanted to restrict the glory I guess.

Re:Really?

[garcia]

So is this any different than the Firefox 2 alpha that wasn't released yesterday?

Yes, this one is officially released so they get *two* days of hype about an alpha release.

Re:Really?

[bunratty]

Nice. So call it something, release it into the public (as is required by the license) and insist it doesn't exist. For a product to bear a new name means it is newly branded product, thus a nerly branded product has been released.

So you would rather they not rebrand the browser until the very last patch, at which time they are completely committed to release it to the public without testing? Frankly, I don't see the point. Why not rebrand the product and allow it to be fully tested before release?

Re:Really?

[Uber Banker]

It was released. That was the point. And then all the excuses came because the big and aspiring names in the Mozilla foundation did not get the credit of first post. Surely the credit of the browser is more important than ego stroking?

Re:Really?

[bunratty]

No, Firefox 2 alpha was not released until Mozilla said it was released. What was reported as a release before was a nightly build. The distinction is that if the build is not officially released, there might be more fixes introduced in the final offical version. This distinction is important because perhaps at the last minute a serious crash might be fixed. And we don't want tens or hundreds of thousands of users download a build with a serious problem, do we? Maybe you do...

Re:Really?

[Uber Banker]

OK. I put orange juice in my shop window. Someone walks in and says they'd like some orange juice. I say there is no orange juice, its just apple juice. I shuffle the bottles around, empty them and refill them with the exact same contents plus what ever else came to mind over a fraction of a minute making absolutely no pereeived difference to the consumption of the orange juice, but not I put a sign in the winder. The same customer comes in and says "so what's the difference of the sign being in the window, can I now buy orange juice" and I reply "of course, not I have fully exercised my ego". The customer says "why did you shout at me before about buying orange juice" and I reply "because I'm a cliquey arsehole - the fact I put a sign in the window with a product description was not reason enough to expect the product".

Oh, and as I understand it is an alpha build, so "tens or hundreds of thousands of users download a build with a serious problem" is not taken as a given? Millions of Firefox versions have been downloaded with a serious problem so far.

Re:Really?

[TheDormouse]

You visit your favorite music or DVD dealer on a Monday afternoon. An employee is sorting through a carton of shipment.

"Hey is Good Night and Good Luck out?"

"No, it comes out tomorrow. New music and DVD release are almost always on Tuesdays."

"But you're holding it in your hand! Just sell me one."

"It doesn't come out until tomorrow."

"Dude, I'll pay you double for it."

"If I sell it to you today, the company will get fined and they'll stop shipping us new releases a day early so we can sell them when we open on Tuesday."

"But...uh..."

"Meet me out back after 5pm. Bring cash."

Re:Really?

[bunratty]

Your analogy fails because it doesn't take into account the version number. Just like the orange juice in your example, Bon Echo existed before the alpha version was announced. Anyone wanting Bon Echo could have downloaded it. No one was saying that Bon Echo didn't exist, just that it wasn't the final alpha version. In effect, it was a release candidate.

Getting a Firefox Alpha

[gurutc]

takes me back to the good old days when it was new, fresh, and charmingly not yet seemingly perfect, but so much the best choice!

Re:Getting a Firefox Alpha

[AKAImBatman]

(raises eyebrow) Getting an Alpha takes you back? Pff. I used to have a cron job that would download and compile Mozilla Nightlies every night on my Solaris box. It would automatically back up the current version just in case the new version didn't work in the morning. Every morning it was a new and wonderful experience to see how stable Mozilla would be today, if it would even run, and if there were any new features.

Back then we didn't have no "Alphas". We had semi-stable code snapshots called "Milestones" and we liked it that way! ;)

Re:Getting a Firefox Alpha

[m50d]

Alphas. Wonderful machines, years ahead of their time. If only they'd succeeded, ah, *lovingly pats his ev5*

What's this about a web browser?

Re:Getting a Firefox Alpha

[bcattwoo]

Well, back in my day we would download "Hourlies" to our abacuses and sit around wondering "WTF is the Internet?!"

Re:Getting a Firefox Alpha

[AKAImBatman]

single-handedly (I infer from your post) keeping the project alive

You inferred incorrectly. I was merely a user of Mozilla who wanted an alternative to Netscape 4 and IE for Solaris. (The latter of which didn't work, but DID screw up my CDE profile.) I wanted to contribute back the binaries from my nightly build (since the project didn't have anyone doing it at the time), but my machine was owned by my work. I simply didn't feel comfortable donating resources that weren't mine to give. :)

Re:Getting a Firefox Alpha

[gurutc]

Actually, your downloads counted towards the project activity. So you did contribute through your consistent and measurable interest in the latest binaries. And I meant the thanks, take the compliment already.

Re:Getting a Firefox Alpha

[fak3r]

I was one of the authors of getmoz [http://getmoz.mozdev.org/%5D that would do just that, dnld the latest, backup your old copy, install the new, migrate bookmarks, etc. I *think* I got started with Mozilla sometime during M20 (? can't recall if this is the right milestone or what ?) Back then one night would reveal a bunch of changes that would/would not be there the next day...a fun time!

Re:Getting a Firefox Alpha

[AKAImBatman]

Why would bcattwoo run a web browser on his abacus when he has no idea what the Internet is? That does not make sense! But more important, you have to ask yourself: What does this have to do with this case? Nothing. Ladies and gentlemen, it has nothing to do with this case! It does not make sense! Look at me. I'm a slashdotter defending a major browser vendor, and I'm talkin' about an abacus! Does that make sense? Ladies and gentlemen, I am not making any sense! None of this makes sense! And so you have to remember, when you're in that jury room deliberatin' and conjugatin' the Emancipation Proclamation, [approaches and softens] does it make sense? No! Ladies and gentlemen of this supposed jury, it does not make sense! If bcattwoo run a web browser on his abacus, you must acquit!

The defense rests.

Re:Getting a Firefox Alpha

[jonbritton]

Cron? You had cron!? Back in my day, we'd pay a few dollars to that neighbor boy who ain't quite "right" and make him bring back the damned CVS snap on floppies!

Actually, my day was the mid-nineties. In my day, we'd just throw out the old computer and buy one pre-installed with the newest milestone release, twelve offers for dialup providers and a Windows background so hideous that it made 7 out of 10 eyes bleed.

Re:Getting a Firefox Alpha

[obender]

I really don't understand why parent was modded Funny. I compile from the trunk on a daily basis and I see nothing unusual about that. At least not among slashdotters.

I hope they don't change the tabs too much

[caluml]

I've always wondered why bookmarks don't sort themselves by most often used to least recently. Maybe it will happen now. But the changes to tabbed browsing behaviour - hmm - I hope that means something like memory optimisation and not making it more like the tabs in Konqueror. Blech.

Re:I hope they don't change the tabs too much

[Tim C]

Personally, I'd rather the bookmarks were sorted in alphabetical order. Hopefully moving to an SQLLite backend will enable users to set up their own ordering preferences as they see fit.

Re:I hope they don't change the tabs too much

[night_flyer]

Open Bookmarks, right click and Sort By Name

Re:I hope they don't change the tabs too much

[laa]

Personally, I dislike all kind of autosort behaviour like the office feature of showing only recently used commands. Many (most?) people remember the positions of items and choose before reading the actual bookamark. If the sorted in some random way (like most recently used) then each item has to be read -- and you can't assume that people with a collage degree are able to read :)

Re:I hope they don't change the tabs too much

[Trigun]

I prefer my bookmarks sorted based upon distance from my house. It takes a while on ip2location, but at least I know that if I ever have to set a bag of poop on someone's doorstep, how far I'd have to travel.
But I can see both alphabetically and by most used being valuable. Good thing Firefox lets you manage them however you want, just not automatically.

Re:I hope they don't change the tabs too much

[ILikeRed]

You don't need a database for sorting algorithms (think gnu sort), but what this will almost certainly do is complicate backup and transfer of bookmarks. I really can't understand what is wrong with a simple text file. Do they not see all the issues Microsoft has because of their registry format??? This is NOT a speed or sorting issue. (I could care less about the history, but don't think that will help anyone other than some possible edge cases there either.)

This will also almost certainly kill any chance of reusage of bookmark data by other programs - which could be a really inovative area if the barrier to entry is kept low. They need to read the Art of Unix Programming.

Re:I hope they don't change the tabs too much

[stony3k]

I do agree with you in spirit, but as long as they support exporting the bookmarks to a text or html format, we should be fine.

Re:I hope they don't change the tabs too much

[caluml]

Is that a one-off sort, or do newly added bookmarks go in to the right place?

Re:I hope they don't change the tabs too much

[digidave]

SQLite databases are a single file, so backing up your bookmarks should be the same... copy the file.

I wonder if there is a password on the database. If not it could lead to bookmark injection attacks from other programs you install (that could also happen with the bookmarks.html file) like happens with IE. I think this is a great opportunity to create a bookmarks file that can't be altered unless you type in a password. Perhaps the SQLite password can be set during install and altered from within Firefox's preferences. It could be a keyring password that also gives you access to your saved site passwords, sort of like Opera and Konqueror.

Re:I hope they don't change the tabs too much

[parkrrrr]

Unless you propose to ask the user for that password every time they want to do something with their bookmarks, how is having it going to prevent other programs from creating bookmarks? It's just an extra step: read password from wherever Firefox stores it, write to bookmark file.

Re:I hope they don't change the tabs too much

[Tweekster]

The registry is garbage because it is sooo huge with no standards and just crap all over the place... A database format for a specific set of items does make a lot of sense. I honestly dont see why for the bookmarks system but thats besides the point. a lot of cases it makes sense for (for when you want to search much more complicated data)

Re:I hope they don't change the tabs too much

[Doctor Crumb]

bookmarks don't sort themselves because that violates the rule of least surprise. Bookmarks should never re-sort themselves, just as menus should not hide "infrequently used entries", and just as files should not automatically move themselves to other folders.

A much better solution would be to have "sort alphabetically" as an option under "Manage bookmarks", where the user is the one who decides when and if they get sorted.

Re:I hope they don't change the tabs too much

[Anonymous Coward]

This will also almost certainly kill any chance of reusage of bookmark data by other programs - which could be a really inovative area if the barrier to entry is kept low.

This radically lowers the real barrier to entry. SQLite provides a simple standard interface (a subset of SQL) to manipulate data. You no longer have to parse data out of a flat file, and then keep your parser in sync with the format of a file which will certainly change over time.

Databases are only intimidating to people who don't understand them. I've never met a developer who, after learning what databases can do, didn't go absolutely crazy for them.

Re:I hope they don't change the tabs too much

[rsadelle]

I've always wondered why bookmarks don't sort themselves by most often used to least recently.

Good God, no. I use my bookmarks all the time, and I know where things are based on where they are in the list. This would just make it impossible to find anything.

Re:I hope they don't change the tabs too much

[Ligur]

I really like that idea, you wouldn't happen to know if there's some extention that does that exact thing would you?

Re:I hope they don't change the tabs too much

[gopalarathnam_v]

Its not only about sorting, with a database a lot more things are easier like for example, querying based on the fields. In this case, the places UI groups the history by title, web site, etc, and also providing a good search.

And SQLite is really better since its a standalone database.

If you think this is a performance improvement, its not because SQLite is not suited well for more writes, but less writes, more reads.

Re:I hope they don't change the tabs too much

[Solarahawk]

This is not a registry kludge, but a means of managing bookmarks as individual entities in a lightweight DB layer. The key advantage I see is the ability to cleanly synchronize changes to/from the bookmarks DB with a server or other PC. Although there are BM synch extensions available for FF, due to its implementation of the bookmark file, these solutions are jerryrig fixes. The original netscape bookmark file was not designed to allow this effectively.

Although an XML-format file with altered functionality in the browser code could fulfill this need fine, the DB will open a much greater field of opportunity for creative programmers.

So I, for one, am excited by this functionality.

Re:I hope they don't change the tabs too much

[Anonymous Coward]

They aren't doing this just to get sorting algorithms. This actually improves things because right now there are a number of different formats that Firefox stores data in. Look at your bookmarks.html(HTML), cert8.db(Berkeley DB), formhistory.dat(mork... don't ask). Using sqlite reduces the amount of code for accessing all of these formats and provides some degree of uniformity. Getting data from sqlite is pretty easy(and much easier than the current situation) because we have dozens of language bindings and tools to do it with.

Yes I read the arguments against this is in AUP, but Firefox is an application that runs completely counter to most of what's in there. Firefox is never going to be a Unix application following the advice in AUP. It wants to be an operating sytem(or platform if you prefer), and not just an application.

Re:I hope they don't change the tabs too much

[richwklein]

The big reason for using SQL as bookmark storage is that hyistory and bookmarks have been merged. A bookmark now consists as a set of annotations to a url in history.

Re:I hope they don't change the tabs too much

[jdeluise]

Or spell apparently.... unless you were talking about art students, in which case I understand..

Re:I hope they don't change the tabs too much

[cduffy]

This will also almost certainly kill any chance of reusage of bookmark data by other programs

Not at all. SQLite is extremely easy to use -- it has bindings for major scripting languages, and trivial queries can be run on the command line. I use the Python bindings in a number of my minor scripts, and it has frequently resulted in a massive performance improvement (as opposed to using flatfiles and writing the data-munging and analysis code myself).

Re:I hope they don't change the tabs too much

[buraianto]

It seems to me that the analog of changing the format of a flat text file would be to change the database schema. I can't think of any reason why a file format would change any more or less often than a schema change. It is probably easier to adjust to a schema change though, considering a change to the schema would probably just require an updated query as opposed to rewriting the parser. But it still would require changes.

Re:I hope they don't change the tabs too much

[digidave]

Same way KDE's wallet or the Gnome keyring work: ask the user once for each session in which it was used. Maybe read-only access to the bookmarks should be allowed without the password.

I just remember that so many apps would drop their own bookmark files into IE's bookmark folder. Every forum was filled with questions like "How do I get rid of the ICQ bookmark? It keeps coming back after I delete it."

Re:I hope they don't change the tabs too much

[caluml]

I could care less

So why don't you?

Re:I hope they don't change the tabs too much

[John Whitley]

Some of the tabbing issues being addressed are actually big fat bugs, such as what happens when you open too many tabs in one window (which I do all the time... :-P ). E.g. see Bug 221684: When opening too many tabs you can't move to them with the mouse ("X" button and tabs overlap) for a prime example. This case simply wasn't handled at all gracefully in the GUI -- tabs just "run off" of the right side.

Tho I am curious to play with the alpha to see what other changes might be in store. Who knows, maybe I'll get to go do some damage in Bugzilla? 8-)

Re:I hope they don't change the tabs too much

[WWWWolf]

I really can't understand what is wrong with a simple text file.

Oh, there's nothing wrong with a "simple text file". It's just that there's like six different kinds of "simple text files" used by Mozilla. One of them is not "simple" either, it's really damn stupid, actually.

And Mozilla certainly doesn't use "simple text files". The moment you throw XML in it, it turns non-trivial. And as for Mork mentioned above, well, I challenge you to come up an "Art of Unix Programming" solution that's better than this.

Plus, as everyone who has programmed anything knows, "simple" text files, ultimately, aren't.

This will also almost certainly kill any chance of reusage of bookmark data by other programs

Au contraire, this will make reuse of bookmark data much simpler. Just load up your sqlite driver in your favorite scripting language and do a few SQL commands. Perl, Ruby and Python are already well supported.

Re:I hope they don't change the tabs too much

[Suppafly]

I've always wondered why bookmarks don't sort themselves by most often used to least recently.

Because that would be annoying. It's just like the smart menus in office that are constantly hiding features that you need but don't use everyday.

Re:I hope they don't change the tabs too much

[Silverlock]

That's an interesting comment because my boss said to me yesterday, "Why doesn't it put all my bookmarks in alphabetical order? It's stupid."

The current method of putting them in the order they were added may not be what you want, but the alternative is not obvious. Perhaps they could add all of the above behaviours as options?

Re:I hope they don't change the tabs too much

[random735]

that's great, but when i want to go hack up my bookmarks, i don't want to download the latest sqllite bindings for perl...i just want to write some regexes to hack the file.

also, occasionally i find myself needing a bookmark from my home machine while i'm at work..currently i can ssh in and look at my bookmarks file to find the bookmark... with this, that's a nogo.

i agree with the OP, this is less accessible, regardless of available programming libs... you're not going to beat the accessibility of a text file, and there is not a speed/transaction/etc issue that needed to be solved.

even IBM's websphere moved from using a DB for config, to XML on the filesystem.

Re:I hope they don't change the tabs too much

[RobertLTux]

what i would like to see is a "do something sane folder" where if you copy a file to it 1 if its an archive its uncompressed (with folders and top folder named for the archive file) 2 if its music move to the media folder 3 if its a known registered file move to where it belongs 4 if its a document move to folder 5 otherwise do nothing

Re:I hope they don't change the tabs too much

[tepples]

Because in practice, "I could care less" is short for "I could barely care less".

Re:I hope they don't change the tabs too much

[cduffy]

that's great, but when i want to go hack up my bookmarks, i don't want to download the latest sqllite bindings for perl...i just want to write some regexes to hack the file.

Well, it's not like you *need* the Perl bindings -- you can use the SQLite command-line tool to dump the database contents to stdout and then use awk or something to filter through that. Same thing works for reading through your bookmarks from home. Sure, it's an extra tool you need -- but it's one that your modern distros will have already packaged.

you're not going to beat the accessibility of a text file, and there is not a speed/transaction/etc issue that needed to be solved.

With the text file, load time is going to vary linearly with number of entries. With SQLite, it may not be constant time (I don't know), but it's *much* more flat.

Anyhow, why not go track down the Bugzilla entry associated with the feature and find out what the reasoning behind implementing it was, rather than just blindly calling it needless?

That's all?

[Anonymous Coward]

Seriously - that's all the new features? How does that warrant a 2.0 label and not a 1.8? Firefox has been pretty innovative or good at putting great features together that Opera and Microsoft haven't done (yet), but now it seems IE7 has caught up in so many ways, but Firefox 2.0 will be just a minor, incremental update. Hell, bigger changes have gone in the post 1.0 releases. Come on...

Re:That's all?

[Pneuma ROCKS]

It will be much more than that when Firefox 2 actually makes it out to the world. This is a very early build and according to the Roadmap, it will be released near the third quarter of 2006. I'm guessing it will actually be a little later than that. I also found this Feature Brainstorming page, which seems to be closer to what's being planned for 2.0. I see a lot of new stuff.

SVG support

[caluml]

I hope more browsers end up using SVG. There are some very nifty uses that can be made of it - an example of which is the porn database - http://pdatabase.dyndns.biz (how's it going, John? :) )

Re:SVG support

[geminidomino]

Not a very good example, since it's in closed beta.

I was looking to see what you could do with SVG. Honest.

Re:SVG support

[caluml]

Indeed. Send the guy an email :)

Re:SVG support

[gr8_phk]

I'm hoping they'll have SVG animation for 2.0. For me personally, just constant rotations would do the job. There are other important things though. At least the status for animation features isn't all red any more. I suspect some of them will fall rather quickly once they have a couple done.

Re:SVG support

[Jonathan Watt]

2.0 is *mainly* a release for reworking the UI. Only a few hand picked SVG changes are making it into this release, and SVG declarative animation will definately not be one of them. Mozilla Firefox won't get the majority of the SVG (or other engine) improvements until it's updated to the current development version of the Mozilla Platform/Gecko. That's scheduled for Firefox 3.0 in the first half of 2003, not Firefox 2.0.

Re:SVG support

[bigpat]

2.0 is *mainly* a release for reworking the UI. Only a few hand picked SVG changes are making it into this release, and SVG declarative animation will definately not be one of them. Mozilla Firefox won't get the majority of the SVG (or other engine) improvements until it's updated to the current development version of the Mozilla Platform/Gecko. That's scheduled for Firefox 3.0 in the first half of 2003, not Firefox 2.0.

2003? scheduled? I think we are a bit behind then?

Seriously, more and more SVG animation support would be great. I am on the SVG developers mailing list and SVG development is very very active these days. Better and better support for SVG would mean that richer graphics, animations and applications based on open standards would be available natively for firefox users. I know in many people's minds AJAX techniques have meant less need for SVG as a Flash killer. But there are still areas such as graphics scalability for different displays (cell phones) and very rich set of animation features combined with the possibility of inline animations without need for a plugin that make SVG support in firefox very desirable. Also, fuller SVG support in firefox could help drive firefox marketshare and will certainly help us get to a open standard for Vector graphics and animations faster and away from any remaining reliance on Flash.

Re:SVG support

[Jonathan Watt]

Oops! I meant 2007 of course. :-)

I don't think you need to convince anyone at Mozilla of the need for better SVG support. The issues holding things up are technical issues and a lack of contributors. There's only one guy (from IBM) working on it full time and two or three contributors working on it on and off as their spare time allows.

So Far So Good ....

[abhinavmodi]

Works fine on Windoze even after 2 hours .. No crashes or memory hogs. In addition, it is co-existent with Firefox.

Re:So Far So Good ....

[Stevyn]

Using it on Linux. None of the extensions were compatible, that's okay for now. The only improvement I noticed was that sound in flash actually works now. It's nice to visit youtube.com and actually hear audio.

Other than that, similar interface but the tabs organize themselves now. Also, the ACID2 test doesn't render properly.

Re:So Far So Good ....

[SimonH_1978]

In addition, it is co-existent with Firefox.


Kind of . . . it's disabled all of my extensions, even when I start FF instead of Bon Echo.

Re:So Far So Good ....

[Nevenmrgan]

Installed on Win, co-exists with 1.5 peacefully.

Yay for the Safari/Camino-style tabs with Close buttons right on the tab. On a 21" monitor, it was weird flying over to the top-right corner to close a tab that was in the top-left.

The new bookmark manager is unfinished, but I frankly don't use it much anyway. Seems ok, just not exciting.

I am excited about SVG, though. Play with it here: http://www.croczilla.com/svg/samples/

Re:So Far So Good ....

[_xeno_]

As you're going to learn if you keep that up, never, ever use the same profile with two different versions of Mozilla or Firefox. They'll gang up to trash it.

What you'll want to do is start up Bon Echo using "-profilemanager" (either by editting the shortcut or through the command line) and create a new profile. It's highly recommened to create a new profile whenever testing prerelease Firefox builds anyway since they're more likely to trash your profile than the final releases.

I haven't tried it yet, and I can't remember where it remembers which profile to use, but you may need to select the appropriate profile to use for each version of the browser whenever you start it.

However, unless you're willing to risk losing your profile, you really should be using a different profile for Firefox 1.5 and Bon Echo.

Re:So Far So Good ....

[cortana]

This is a stupid failing on Mozilla's part. It is not acceptable to tell people to start with a fresh profile over again when they upgrade their browser!

Re:So Far So Good ....

[_xeno_]

Not really - upgrading is fine, it should (operative word there...) upgrade older versions to the new version of the profile. However, by upgrading the old profile to the new version, it necessarily loses compatibility with previous versions of Firefox.

So when running both Firefox 1.5 and Firefox 2.0, you need to keep the profiles separate, as Firefox 2.0 will likely save data in a format Firefox 1.5 doesn't understand. As long as you never want to back-rev your version of Firefox, you can safely use the same profile.

Obviously it's a good idea to back up your profile prior to upgrading... fortunately the Profile Manager won't do that for you so that it's a giant pain in the ass to do. Wait...

what's really new?

[scarlac]

TFA doesn't say anything about new exciting features. I wonder what made them decide it to be 2.0 alpha instead of 1.6? Was it just so that they could reach the planned milestone?
I read something about they were trying to optimize the renderengine, so it could support cairo and have hardware acceleration... no promises was made, but they expected it to be in 2.0 (correct me if I'm wrong).

I guess the more comprehensive changelog (which isn't available yet) will reveal some more interesting changes - perhaps some nice performance enhancements?

Re:what's really new?

[Denyer]

Have a read of the list of things not working yet. There seem to have been some major changes to the underlying code.

Re:what's really new?

[bunratty]

New features planned for Firefox 2.

ACID 2

[Agelmar]

For those who are wondering - the 2.0 alpha build renders the ACID 2 test exactly the same as Mozilla 1.7.12. (http://www.webstandards.org/files/acid2/test.html #top)

I don't personally think that the ACID 2 test is the be-all end-all test, but I know the question will be asked, hence the post.

Re:ACID 2

[digidave]

Grab a nightly trunk build and it's quite a bit closer to the reference rendering. I believe the trunk is what's going to be Firefox 3. It has the newer Gecko.

Re:ACID 2

[bunratty]

There's a new reflow factoring branch off the trunk where the work for Acid2 is taking place.

Re:ACID 2

[nonpareility]

Firefox 2.0 is based off the 1.8 Gecko branch, just like Firefox 1.5 was. 1.5 uses 1.8.0, 2.0 will use 1.8.0.1, 3.0 will use 1.9. There shouldn't be much difference in terms of rendering pages between 1.5 and 2.0.

Re:ACID 2

[bunratty]

Close, but not quite right. Firefix 1.5 uses Gecko 1.8.0 and Firefox 1.5.0.1 uses Gecko 1.8.0.1. Firefox 2.0 will use Gecko 1.8.1.

Re:ACID 2

[From A Far Away Land]

"but I know the question will be asked, hence the post."

The more common /. question for a new product is:
"Oh yeah, but does it run Linux?"
And the answer is:
In Soviet Russia, Firefox 2.0 alpha runs Linux!

Re:ACID 2

[JemalCole]

Isn't that because Moz 1.7.12 is based on the latest version of Gecko? Sience FF1.5 is based on an older version, this is actually an upgrade.

Re:ACID 2

[Sycraft-fu]

I think it's probably a fairly worthless test when you get down to it, overly synthetic tests usually are. It reminds me of a test that was real popular back on the now defunct r3mix.net, a site devoted to MP3 encoding, specificly working on the LAME encoder. I can't remember the name of the test, but basically it was very loud low frequency drum-like sounds followed by high frequency clicks. Turns out, this was really problematic for MP3 encoders. Well because they had such trouble with it, it was seized upon as a good test, and work was done into improving performance on it... Except it turned out that often happened at the expense of normal music encoding. Changing it in such a way it could deal with the oddities of the synthetic test made it such that it didn't work as well for it's oringal intended purpose.

IMO, the important question for a browser is can it render the kind of HTML you are likely to find on the net well. That includes broken, incorrect HTML. This idea that "well if all broswers mandidated good HTML, sites would fix it" is bunk. People are lazy, they make mistakes, sites will have broken code. The ability to render that well is an asset, just like it's an asset to be able to render complex code that uses cutting edge HTML features.

So I don't really care how FF ends up working on the Acid test, what I care about is pages looking good when viewed with it, which they do in almost all cases.

Re:ACID 2

[spitzak]

Though I agree that working only to pass the Acid test is probably not a good idea, it is somewhat different than the MP3 encoding test. The MP3 encoding is lossy.

The Acid test is more like a test for a lossless encoding format. If the lossless encoder fails the test then the encoder is not working, because it is not lossless. However if the programmers worked on the Acid test exclusivley it is quite likely the result will encode that fine but fail on other data (or be really slow on that other data).

Non-compliant HTML is perhaps like music that exceeds the dynamic range that the lossless encoder does, or something like that (I know nothing about sound). The lossless encoder should do as well as it can with that, but there is no reason why that should make it fail to be lossless with legal data.

A correct browser would render the Acid test perfectly, no matter whether the programmers paid attention to it or not.

Re:ACID 2

[metamatic]

I can't remember the name of the test, but basically it was very loud low frequency drum-like sounds followed by high frequency clicks. Turns out, this was really problematic for MP3 encoders. Well because they had such trouble with it, it was seized upon as a good test, and work was done into improving performance on it... Except it turned out that often happened at the expense of normal music encoding.

Yeah, but I listen mostly to techno.

And no, that's not just a joke. Some techno is extremely hard to MP3-encode well--pure square waves cause havoc. In fact, the work on LAME that went on on r3mix.net is probably why Warp ended up encoding their catalog with LAME to get the quality they needed for bleep.com.

Re:ACID 2

[AeroIllini]

IMO, the important question for a browser is can it render the kind of HTML you are likely to find on the net well. That includes broken, incorrect HTML.

But the Acid 2.0 Test was designed to test this, as well. There are several purposefully incorrect CSS statements in the Acid 2.0 Test that any standards-compliant CSS rendering engine should correctly ignore or gracefully degrade in a prespecified way, according to the spec.

You're right, passing the Acid 2.0 Test is not an end-all. Passing the Acid 2.0 Test does not imply perfect CSS standards compliance, but having perfect CSS standards compliance implies you will pass the Acid 2.0 Test. It is simply a goal to strive for, and should certainly not be the only goal you have for reaching compliance.

Re:ACID 2

[Sycraft-fu]

But the question is prefect compliance at what cost? Unfortunately with rendering engines you often have some tough choices to make. You have to support old code, broken code, and new code all at the same time. Sometimes supporting one 100% right can imply breaking another. I don't want 100% CSS2 support that nobody uses if it means I find older pages that I actually visit are broken. The "well they should fix their code" argument doesn't hold water with me. I want to see them now. I want the widest amount of support possible for what I encounter. If that happens at the expense of some esoteric CSS2 processing, so be it.

Re:ACID 2

[labratuk]

That's not really a comparabale situation.

Lossy encoding is always a compromise. It's completely qualitative. By improving one area you will usually degrade other areas. The art is choosing where the optimum point in the compromise lies.

That's not the case with the ACID test. It's all based around adhering to a published standard. Any pages which break because of changes made to pass ACID are broken because they are invalid HTML/CSS. Alternatively it's due to coding errors made by the programmers when trying to fix the problems.

Re:ACID 2

[AeroIllini]

Non-compliant HTML is perhaps like music that exceeds the dynamic range that the lossless encoder does, or something like that (I know nothing about sound).

Non-compliant HTML is code that does not conform to the spec. If you wanted to talk about it in musical terms, it would be like a music file that was recorded with a variable sample rate*, but it was given a wav file header that specifies a fixed sample rate of 44.1kHz, as per the spec. This file is NOT a wav file, since it does not meet the spec, but it declares itself as one anyway. If this file is fed to a lossless encoder that assumes it's a wav file based on it's header, and the encode-decode process does not produce an exact copy of the original, then it's not the encoder's fault. The file is mal-formed. Many files are declared by a webserver to be HTML files (and thus proporting to conform to the HTML standard) when they are in fact not HTML files, since they don't conform to the spec. If they are not processed properly by an HTML renderer, then it's not the renderer's fault.

Anyone who claims that renderers should correctly render improperly formatted files is simply too lazy to write well-formed files, and thus their opinion does not matter. Should I demand that my car run properly on Kool-Aid(tm) instead of gasoline, even though Kool-Aid(tm) clearly does not conform to the automobile fuel spec?

-------

*Perhaps 96kHz during complicated sections, and 16kHz during silence or pure tones, with the range adjusting throughout the file. I leave the technical details as an exercise for the reader.

Re:ACID 2

[AeroIllini]

You have to support old code, broken code, and new code all at the same time.

No, I don't. If it's not HTML, then my HTML renderer does not have to render it properly. And unless it conforms to the HTML spec, it's not HTML. Period.

Old code is different from broken code, if it's compliant with an old standard. Hopefully the newer standard would include provisions for backwards compatibility (which HTML does), and thus old code becomes a non-issue for fully compliant renderers, because it's still valid. But I still have no obligation to properly render broken code, ever.

Sometimes supporting one 100% right can imply breaking another. I don't want 100% CSS2 support that nobody uses if it means I find older pages that I actually visit are broken.

Only if the spec is poorly written, which is not the case with HTML/CSS. CSS2 is fully backwards-compatible with CSS1. If a browser fully supports CSS2, and the pages you visit are still not rendered properly, then the pages are broken, not the renderer. And if all browsers supported CSS2 (well ... if IE supported CSS2, since the rest of them already do) then people would use it. There are some very powerful features in there.

The "well they should fix their code" argument doesn't hold water with me. I want to see them now. I want the widest amount of support possible for what I encounter. If that happens at the expense of some esoteric CSS2 processing, so be it.

Because you are a lazy person who would rather patch the symptom than fix the problem. What would the internet be like if every ethernet card manufacturer supported a slightly different protocol, and some cards couldn't even talk to each other? We have standards for a reason. This is no different. Just because Microsoft has ignored the published HTML spec (which has existed since before Microsoft even decided to write an internet browser) as part of their "embrace, extend, extinguish" activities, it does not mean that MSHTML is the standard. Please keep in mind that the only reason that compliance is even an issue is that Microsoft purposefully breaks compliance in the interest of keeping people using their technology. If that didn't happen, then everyone would be using the same standard (just like TCP/IP in networks or gasoline in cars) and it would be totally transparent.

Re:ACID 2

[tepples]

Should I demand that my car run properly on Kool-Aid(tm) instead of gasoline, even though Kool-Aid(tm) clearly does not conform to the automobile fuel spec?

If most filling stations in the area where you plan to operate your car have sugar drink and not gasoline, then it would be a good idea to have your car converted to run on sugar drink in addition to gasoline. Likewise, there are too many web sites that have sugar drink (malformed HTML or broken IE 5 box model CSS) instead of gasoline (conforming HTML and W3C box model CSS).

Re:ACID 2

[tepples]

Any pages which break because of changes made to pass ACID are broken because they are invalid HTML/CSS.

Given an end user and a web site that uses broken HTML/CSS, which of the following will the end user choose to view the web site?

1. A web browser that gives an error message that the site fails validation
2. A web browser that gives a kinda-sorta usable rendering of the broken HTML/CSS

Re:ACID 2

[Sycraft-fu]

No it's because I am a practical person. I'm not a web developer, I odn't have HTML code to make sure it's compliant, i'm a web user. I go to websites, I want to see them. I don't care if the person who made them did it properly or not, I just want to see a usable site. If they forgot to close a tag or something like that I don't acre. I'm not interested in spending time tracking them down and trying to get them to fix their problem, I just want the information and/or files I need and then to go about my business.

All the advanced formatting and styles are cool and all, but if those are eschewed for a browser that works better over all, I'm all for it.

Re:ACID 2

[AeroIllini]

If most filling stations in the area where you plan to operate your car have sugar drink and not gasoline, then it would be a good idea to have your car converted to run on sugar drink in addition to gasoline.

That's very practical of you, but unfortunately, CSS rendering does not work that way. To continue the car analogy, gasoline and sugar water are mutually exclusive fuels. Pouring sugar in the fuel tank of a standard internal combustion engine will gum up the works, and designing an engine to run on sugar water will mean it won't run on gasoline. One fuel requires vapor explosions, and the other would require chemical reactions.

Likewise, the broken CSS in Internet Explorer is mutually exlusive to spec CSS. Take, for example, the box model. In the spec, the "width" property is the width of the content area, and the border, padding, and margins are extra. In IE, the "width" property is the width of the entire box, border, padding, and margins included.

Take this statement:

#box {
        width: 100px;
        padding: 10px;
        border: 1px;
        margin: 10px;
}


In W3C browsers, the box will be 100 + (2*10 + 2*10 + 2*1) = 142px wide, with 100px for the content, according to the spec (http://www.w3.org/TR/CSS1#formatting-model).
In IE, the box will be 100px wide, with only 100 - (2*10 + 2*10 + 2*1) = 58px for the content, because IE ignores the spec.

Browsers cannot support both; they are mutually exclusive. The browser cannot tell whether the webmaster wanted "width" to mean the entire box, or just the content area. It has to assume one and use it. If the browser chooses the wrong one, then the website is "broken." But it's not the browser's fault, as long as the browser rendered correctly according to spec.

This is not a systemic problem: the spec works. This is Microsoft being a bunch of bastards and refusing to follow the spec.

The car won't run on both sugar and gasoline; it has to choose one. Asking browsers to support both box models is asking them to read the webmasters' minds, and don't think Firefox has an extension for that yet. Of course, the workaround now is to have the filling stations stock both sugar and gasoline, and dispense the correct one based on the type of car you drive. But it would be easier for everyone if all the car manufacturers just supported gasoline instead of their own brand of sugar fuel.

Tabbed browser update complaint

[MMC Monster]

One complaint about the close buttons on the tabs:

The close button itself sucks. Take the one from the Firefox 1.5 Mac theme. It's much nicer.

Also, the button should be grayed out (or invisible) unless the mouse is on the tab bar.

Re:Tabbed browser update complaint

[evil agent]

Meh, I think close buttons on tabs are kinda worthless. You can close a tab by just middle-clicking it.

Re:Tabbed browser update complaint

[szembek]

The close buttons are greyed out unless you mouse over them. I think this is really the best solution because you can close tabs without having to switch to the tab first. Although I do like the point somebody made about middle clicking them to close them, I just tend to forget about that third button.

Re:Tabbed browser update complaint

[joeljkp]

Not on Linux (at least not with middle-click-URLs enabled).

Re:Tabbed browser update complaint

[ben there...]

I click my thumb mouse button, which is programmed to CTRL-W. I can't live without a 12-button laser mouse now.

Re:Tabbed browser update complaint

[mattkinabrewmindspri]

Not if you've got middle-click assigned to Exposé, which is common among Mac users. Also not if you don't have middle click, which is also common among Mac users.

I would prefer the close tab button to be on the right side of the tab, though. Tabs have relatively little space, and in that situation, I think the favicon is more important because it helps differentiate between tabs, so I think the favicon should stay on the left.

As for closing tabs quickly, I would recommend using the Tab Clicking Options extension, which I have set to close tabs on double-click, and the All-in-one Gestures extension, which lets you close tabs by making an "L" gesture.

Both of those take less effort than middle-clicking, and double-clicking on a tab to close it is more consistent with opening a new tab by double-clicking on the tab bar.

Re:Tabbed browser update complaint

[xtieburn]

Close buttons on tabs is the worst idea ive ever seen. (Well thats a bit of an exageration but itll stress my point.)

You open up say seventy tabs on a spree of web page opening. Your tabs are now no bigger than the close button that is on them. Suddenly you find your losing tabs to accidental close button hits and you have no idea which have gone. You have to sift back through all the information to find what was missed.

All of this for what reason? Is it much faster than just hitting a close button thats off to the right of the bar? Is it faster than middle clicking? Nope.

Now Firefox has done some things to make the feature work better. It doesnt shrink tabs too far and the x button vanishes when you have too many open. This minimises the problem. Doesnt get rid of it and when its a pointless feature why even bother. Opera (my current choice of browser) isnt so leniant. Itll shrink tabs down ad infinitum.

While they were creating these great close buttons why did they decide that the fact when you open too many it just vanishes off the bar wasnt a bit more of a priority? (Im also uncertain if firefox has fixed its tendency to forget the address of web pages that dont open correctly. I shouldnt have to have an extension for something so basic.)

The best tabs in my opinion remain IE7 beta's. A little ironic and also I know people will disagree. Do keep in mind before laying in to me that I do still use Opera or Firefox as an overall package. I just think there tabs are quite poor. (Opera with its terrible collapse order and close buttons, Firefox with its vanishing off screen and no open tab button.)

How id make them. Have the new page button and layout of opera, the collapse order of firefox and the simple buttons and ability to scroll through tabs off screen of IE.

Youd think with three competitive browsers one of them would at least give me the option to have all that.

Re:Tabbed browser update complaint

[brianlj]

"Opera (my current choice of browser) isnt so leniant. Itll shrink tabs down ad infinitum."

Take a look at Tools|Appearance^Toolbars-Wrapping. In there, you'll find:

No wrapping
Wrap to multiple lines
Show extender bar

HTH

Re:Tabbed browser update complaint

[xtieburn]

Yes I know about that. Wrap to multiple lines will flood half my screen the extender bar gets horrendously cluttered and goes off the bottom. Neither is a match for simply cycling through the pages youve opened.

wrapping has potential if it shrunk the tabs to a degree first but it only takes about 5 in order to go on the next line.

Re:Tabbed browser update complaint

[brianlj]

Yes, I must confess that I don't like 'Wrap' either, but the extender bar simply pops a menu down, so what's cluttered about that?

And, I have to say, any system of tabs that doesn't (a) automatically shrink the tabs to the minimum size needed to display the Title and (b) reduce their width even further by takling the 'x' off inactive tabs, isn't really a decent tab system.

Tell me... does IE do that?

Re:Tabbed browser update complaint

[xtieburn]

IE does shrink tabs down to a point. Once there small enough the last or the first tab will cycle through them.

It doesnt have x's on its tabs it uses middle click or the close button at the far right.

This is from IE7 Beta 7.0.5112.0 There are more recent versions but I believe the same system is intact.

Re:Will Firefox 2.0 support the latest standards?

[lithvanguard]

Acid2 Test? http://www.webstandards.org/action/acid2/

XForms support?

[VP]

Is there a plan to add XForms support to Firefox, or will they be waiting for XHTML 2.0?

Re:XForms support?

[VP]

Why don't you use the plugin http://www.mozilla.org/projects/xforms/

Because the use of XForms that I have in mind requires that no additional plugins or other software installs are required for the user. I can require the browser, but it has to be a vanilla install. Think non-technical users with pretty much non-existant IT support.

Re:XForms support?

[tepples]

Because the use of XForms that I have in mind requires that no additional plugins or other software installs are required for the user

What is it? If it involves residential end users, then they are used to downloading and installing plug-ins for rendering SWF, Windows Media, RealMedia, and the like.

Firefox 2

[56ker]

I just hope that the greater prevalance of Firefox leads to a greater number of sites supporting it. I've had problems with some sites telling me my version of 1.5 needs to be upgraded to an earlier version!!! The site in question was the Comedy Channels's website. To many website designers seem to still design for IE only or use version checking to serve different pages. People should stick to writing valid HTML code that works across all browsers instead of making their websites unusuable for those who don't use IE.

Re:Firefox 2

[56ker]

Well as someone who has designed sites for money - it is an issue as 1/3rd of the visitors aren't using IE these days including some of the clients.

Why shouldn't people write valid HTML code? It's not that difficult to correct the bugs. I admit that say designing a WAP version for mobile phones may be beyond some sites but valid HTML should work just as well on any OS (short of bugs in the program the person is using to view it).

It's when websites deliberately tie you down to IE that it annoys me (although generally it's IE/Firefox these days). It's also a search engine issue as they don't like being served different pages based on the user agent to "normal visitors". Hence why we have so much spam in Google. Firefox has got a momentum of its own - and Microsoft is on the wane.

Re:Firefox 2

[56ker]

Yes but the site shouldn't say you have an earlier version and insist you "upgrade" to an earlier version of the browser. We should write code in a way to cope with future changes. Yes browsers have bugs - but that's not the issue here. Your solution is to work around the bugs by serving different pages (based on useragent/version). My point is that it shouldn't be required (or necessary). If the person has a buggy browser they should upgrade it or apply a bugfix.

Re:Firefox 2

[Ckwop]

I just hope that the greater prevalance of Firefox leads to a greater number of sites supporting it. I've had problems with some sites telling me my version of 1.5 needs to be upgraded to an earlier version!!! The site in question was the Comedy Channels's website. To many website designers seem to still design for IE only or use version checking to serve different pages. People should stick to writing valid HTML code that works across all browsers instead of making their websites unusuable for those who don't use IE.

The problem is that the browsers support standards so badly that even a fairly modest design requires CSS hacks galore to get good cross-browser support. You really can have a CSS file that validates exactly with no warnings, a HTML file that validates exactly with no warnings that renders differently in every available browser. In fact, ACID-2 is prime example of this problem.

The only clean solution is to detect the user-agent string and to send a different CSS file to each version of the different browsers you want to support. I don't really advocate browser hacks because the resultant CSS files are harder to maintain and prone to break on future releases.

It's a real bitch and it makes developing web-sites less of a joy than it should be.

Simon.

Re:Firefox 2

[Threni]

> Well as someone who has designed sites for money - it is an issue as 1/3rd of
> the visitors aren't using IE these days including some of the clients.

I put it to you that IE accounts for a lot more than 66% of web site accesses.

> Why shouldn't people write valid HTML code?

There's no moral requirement that they do. There's no financial incentive to do so. I'd prefer it that they did, because I use Firefox, but that's irrelevant - I'm not paying these guys to design websites, and the people who do generally don't care about Firefox or any other niche browser.

> Firefox has got a momentum of its own - and Microsoft is on the wane.

Any day now, right?

Re:Firefox 2

[ben there...]

I'm not paying these guys to design websites, and the people who do generally don't care about Firefox or any other niche browser.

Ah, but the people who design websites do care about Firefox. 25% of w3schools.com's visitors use Firefox to visit their site, twice as many as the percentage reported by web statistics organizations across other sites.

They should care, and they often do care. Whether designing for accessibility is budgeted by their bosses is another matter.

Re:Firefox 2

[drinkypoo]

The problem is that the browsers support standards so badly that even a fairly modest design requires CSS hacks galore to get good cross-browser support.

I disagree. The problem is that CSS failed to take the fact that all browsers would not be identical into account during the design phase and as a result we have to rely on hacks (or browser capability detection, or browser detection, or...) in order to get CSS to do what it is supposed to do.

Even a drunken microsoft employee could have foreseen that CSS support would be a minefield.

Re:Firefox 2

[Threni]

> They should care, and they often do care. Whether designing for accessibility is
> budgeted by their bosses is another matter.

They should care, and they often do care, but they usually *don't* care. And whether not their bosses think it's important is not another matter - it's *the* matter which is responsible for the problem (of stuff not working on Firefox). As soon as it becomes worth the while of bosses everywhere to have their web monkeys spend however much extra time it takes to have their sites work on Firefox they'll do it. Until then it's just a water-cooler discussion.

Re:Firefox 2

[naelurec]

Why shouldn't people write valid HTML code? It's not that difficult to correct the bugs.

What are you talking about? It is a PITA to fix the bugs. I do XHTML/CSS design work (table-less designs) and spend a LOT OF TIME fixing the designs to look ok in Internet Explorer. It absolutely sucks. The design looks great in Firefox, Mozilla, Safari, Konqueror, Links/Lynx, etc but IE will effectively crap on the design. Most of the time making it work in IE (or atleast look presentable) can consume up to *half* the design time for a site. Needless to say, spending twice as much time on a design just to work around bugs is not something most customers want to pay for.

Ultimately it *can* be done and I have done quite a lot of sites that look acceptable across browsers and do fully validate. However, this does come at the cost of CSS utilization (lowest-common-demoniator) and design compromise.

Re:Firefox 2

[iabervon]

Of course, there's no reason you should expect that a site will render the same in every browser. In fact, no site renders the same in Firefox 1.5 when other people run it and Firefox 1.5 when I run it, because I insist on a relatively narrow window that's taller than it is wide. A well-designed site will render differently in my browser, but still look good and have the intended characteristics. Further out, a site has to generate a vastly different effect on a system with large print for the vision-impaired or a screen-reader for the blind.

Re:Firefox 2

[cortana]

I've yet to see a site that doesn't work on non-IE browsers and that has any worthwhile content.

Re:Firefox 2

[Maltheus]

I find a lot of internal corporate websites only work for IE. Training course, etc. Perhaps not worthwhile, but I do have to reboot (into Windows) if I want to keep my job. The more corporate the website, the more I find flash over substance. A lot of our vendor webtools are like this too.

Re:Firefox 2

[brianlj]

"As soon as it becomes worth the while of bosses everywhere to have their web monkeys spend however much extra time it takes to have their sites work on Firefox they'll do it"

They don't care now because the issues just aren't phrased correctly.

Easiest way to put it is: "Okay, so do you mind the company taking a 5% drop in turnover? Every 20th customer walking through that door there, we're gonna kick them in the ass and tell them to go somewhere else. That's okay, right?"

Re:Firefox 2

[cratermoon]

You really can have a CSS file that validates exactly with no warnings, a HTML file that validates exactly with no warnings that renders differently in every available browser.

Pixel perfectionist website designers get no sympathy from me.

Version inflation

[Ulrich Hobelmann]

ok, some features most users won't even notice, and that deserves a bump to 2.0...?

Well, Slackware did it. FreeBSD did it.

Even NetBSD did it.

I'm waiting for Mac OS 11.

Re:Version inflation

[ivoras]

Um, just for clarification sake, FreeBSD has never skipped a major version number. Going from 5.5 to 6.0 doesn't count as version bloat. (see here if you have an hour to spare: http://www.levenez.com/unix/history.html). If anything, the BSD projects went backwards: from 4.4BSD to (Free|Net)BSD 1.0.

Also, how exactly did "NetBSD did it"? As is stated here: http://www.netbsd.org/Misc/history.html the version numbering is clean.

Re:Version inflation

[Troglodyt]

hmm, mac os is at 10.5 soon, wonder what they'll do after 10.9...

Re:Version inflation

[Ulrich Hobelmann]

I didn't mean skipping numbers, I merely meant quickly pushing up version numbers. FreeBSD was 4.X for very long, and then after only very few years at 5 it turned 6.X.

NetBSD was 1.X for VERY long, turned 2.0 with multithreading support (and SMP IIRC?), and now is already at 3.0. Don't ask me why...

The same is true for Firefox, where I can't see why it's not just 1.6.

Re:Version inflation

[Ulrich Hobelmann]

The same as GNOME? Go to 10.10 maybe. Probably Mac OS will reach version 11 before that, though. I'm really curious what Apple will bring in the future, especially since Leopard (10.5) is scheduled to be released at the same time as Vista (early '07).

Worth the jump in major numbering?

[PoprocksCk]

I've been toying around with the new alpha, and it has some interesting additions. But heck, the changes made do not warrant a jump in major version numbering in my books. But I guess that's because I'm used to how version numbers are in the Free Software world, where a jump in a major version number usually means there was a rewrite, or ABI was broken in favour of some fundamental changes.

I'm definitely not seeing that here with Bon Echo.

Not that this is a bad thing -- heck, I'm as much against featuritis as the next guy. But frankly I see less change here than from 1.0 to the Deer Park alphas.

IMHO the #1 thing the guys should have focused on for the 2.0 release was to make Firefox a XULRunner application.

Re:Worth the jump in major numbering?

[danpsmith]

It's an _alpha_ it's not even a beta. What's wrong with you? Alpha doesn't even mean that all features are intact yet. Or at least it shouldn't.

Re:Worth the jump in major numbering?

[bill_mcgonigle]

But I guess that's because I'm used to how version numbers are in the Free Software world, where a jump in a major version number usually means there was a rewrite, or ABI was broken in favour of some fundamental changes.

Come now, Mozilla breaks its API's with every .1 release. That's most of the reason Mozilla development is so slow to progress, both for bitrotted patches and Extensions that are slow to get updated.

libstdc++

[calzplace]

Note that most FC4 machines out there will need the compat-libstdc++-33 package for the libstdc++.so.5 library. Just an FYI. :-)

FF Extensions Contest

[Kranfer]

I find it very strange that the winners of the recently posted FF Extensions contest do not work. The extensions that is. I like this alpha of FF 2 but I wish I still have the extensions / Themes I had before still working.

Portable version just posted

[Anonymous Coward]

A portable version of this build was just posted for those that want to use this with a separate profile:
http://www.cybernetnews.com/?p=417

How soon to version 3.0?

[TheSkepticalOptimist]

I mean, Mozilla used to be slow and steady, now they are firing out updates on .5 increments.

Is this good or bad? I think Firefox will end up becoming bloated and bug ridden just like IE if they keep up this kind of product update cycle. Firefox 1.5 hasn't even been out for 6 months and they are previewing version 2.0.

While I do think that some open source projects move a long at a pace that make snails impatient, I have found that this quick turnaround for FireFox versions isn't beneficial in the long run. I have found there to be more problems in each new version, and I have stopped using Thunderbird for several problems that haven't been addressed yet (such as opening up the wrong email when you click on a header).

I think Mozilla should slow down a bit, or at least go back to the .1 version increments. If they are just trying to drive up the version number to match I.E.'s 7.0, then they will find that Firefox performs about as well as I.E. 7.0, or even less so considering it took Microsoft 10 years to get there.

Re:How soon to version 3.0?

[bunratty]

How does what the size of the version increment deteremine how bloated or bug ridden software is? Anyway, Firefox 2 was available as a "preview" as soon as Gecko 1.8 branched off the trunk last year. The day after any stable branch is created, a new trunk build is produced that is the first preview of the next stable version. Is there some kind of problem with doing that?

Re:How soon to version 3.0?

[4D6963]

I mainly agree with you, the incrementation goes too fast. One purely psychological thing about this is that major releases lose their impact if you keep on incrementing that fast. I mean what happened to Firefox 1.1, 1.2, 1.3 and 1.4? Why jump directly to 1.5? It seems that it's to say that it's half way to 2.0, but I don't think this is a good idea

Psychological effects aside, I've been using Firefox since 0.9, and since then I have been complaining about both the memory usage (although it seems that a great part of that is considered a disableable feature, but when disabled you still have some big memory usage) and the lack of solution for reaching tabs that are out of scope (although third-party extensions such as Tab Mix Plus brought a solution to it, a Mozilla-made solution wouldn't be unwelcome)

Re:How soon to version 3.0?

[Kjella]

So how exactly is version numbering related to the speed of development? Linux has been moving along in 0.0.1 increments for over two years now, yet most have been complaining about how much they've added to the kernel between increments. Debian's got a higher version number but I haven't seen anyone complaining about their rapid pace. Version numbering is either a) plain bookkeeping, similar to build numbers, b) some sort of interface/stability indicator or c) marketing, trying to create a perception of how fast it's evolving. I think Firefox squarely ends up on c). Announcing the first alpha 1.5 years after their last major version isn't break-neck speed to anyone except OSS geeks who're used to entire overhauls being 0.1 releases.

Re:How soon to version 3.0?

[anaesthetica]

I think the release numbering has more to do with the logic of post-1.0 branding. Now that Firefox has reached 1.0+ status it's supposed to be a public, finished product. Major corporations don't do point releases for each iteration. Big round numbers inspire trust in the non-computer savvy, which is the market that Firefox is trying to break into now. Let's face it, Firefox 2.0 sounds much more reliable for n00b or corporate use than Firefox 0.9.8.13.1. It may ruffle some feathers in the open-source community, but such is the price we pay for Firefox playing with the big boys now.

Re:How soon to version 3.0?

[SharpFang]

The tab access problem is a big and pity one, and an example of a bad approach some (many) open-source projects take: there are many solutions to the problem, all with their own drawbacks and advantages, and people keep arguing on which one is the best, but as result none is implemented.

This particular bug has several 100's of comments but people just can't decide on which of perfect, advanced hypersolutions to pick and as result not even a very simple and basic one exists.

Same as with Autosave in Gimp. People can't agree how should the Autosave work, in regards of filetypes, multiple projects, diskspace conservation, performance, accessiblity, security etc etc and as result Gimp has no autosave at all and if it crashes, all your work is lost. (despite it was planned to be included in 1.3, and now it's nearing 2.4!)

Some things are just 'not perfect enough to be included' and preferred to be left out instead of included in non-perfect form.

Re:How soon to version 3.0?

[jsoderba]

If the roadmap holds up (and it usually slips a few weeks/months), Firefox 2.0 will be released in September or so, 10 months after Firefox 1.5, I don't see how that's too quick.

Re:How soon to version 3.0?

[4D6963]

This particular bug has several 100's of comments but people just can't decide on which of perfect, advanced hypersolutions to pick and as result not even a very simple and basic one exists.

The smart way is the Tab Mix Plus way, give people the choice (like, horizontal scrolling or vertical scrolling, how many rows, etc..).

Some things are just 'not perfect enough to be included' and preferred to be left out instead of included in non-perfect form.

If so then firefox should not load in memory (since its implementation of the use of memory is very far from perfect, I wonder what kinda stuff they can hide in 300 MB of RAM tho..)

Re:How soon to version 3.0?

[spauldo]

Or d) when the code breaks its own API.

Libraries on UNIX systems do this, in general terms: major functionality/API change -> major version increase, minor feature change -> minor version increase, bugfix -> point number increase. Not every library follows that, but most do.

Arguably, given the way data storage is being merged into a single format that's different than any of the previous formats supports a major change. There was mention of changes in the way extensions were handled, so if extensions have to be rebuilt/modified for 2.0 then that also supports a major change.

Other people have said that the next major update to gecko will be in 3.0, which definitely deserves a major number.

All this is conjecture, and you're right on the idea that version numbers are basically pulled out of the developers' asses. Only the mozilla guys could say for sure.

Re:How soon to version 3.0?

[Maltheus]

It's unfortunate that that is true. I'm more inspired by smaller numbers (0.9 - 2.0). I find larger numbers are an indication of institutionalization. A product with a large number doesn't take chances because it's been around too long (kind of like old people). A smaller number indicates the new kid on the block, the cutting edge, nimbleness etc. Lot's of little sub-versions indicate continual progress to me.

Also, I see large leaps in version numbers as a desperate move. It says to me that you don't have a lot going on, but here's a nice shiny new number to base a BS marketing campaign on. I'm not saying this about Firefox per se, just in general. I stay in the 1.0.0 - 1.9.9 range unless there is a fundamentally leap in application functionality or complete incompatibility with the previous versions. Too many major version jumps and you soon have to resort to gay-ass names like Vista to describe your product.

I also find that I'm more likely to upgrade to a minor version, sooner than I would to a major version. Anything that ends in .0 screams buggy to me. It took me about a year to upgrade to java 1.5.0 because of that (glad I finally did).

Re:How soon to version 3.0?

[anaesthetica]

I agree, and I find that generally version 3.0 is when most given applications begin to jump the shark and lose out to their nimbler competitors.

Re:How soon to version 3.0?

[anaesthetica]

If you look at the release roadmap it looks like they should not have bumped 1.1 to 1.5 and thus 1.5 to 2. The change between 1.5 and 2 is going to be nearly all bugfixes and interface changes, but no underlying upgrade. Then, we'll all-of-a-sudden leap from 2.0 to 3.0 in six months, with the major change being the Gecko upgrade. IMHO, the Gecko upgrade should have signalled the shift from the 1.x series to the 2.x series. The change in Gecko from 1.5 to 2.0 was only from 1.8 to 1.8.1. Silly. That six month leap from 2.0 to 3.0 is not legit.

What is Bon Echo & FF2 Alpha does not update

[JavaManJim]

What is Bon Echo anyhow? Yet another version of Firefox? Why does Firefox ignore Firefox 2 Alpha? Or is Firefox 2 Alpha now another dead armadillo with its legs in the air and flys buzzing around like Mozilla 1.7.11? I click on Firefox 2 Alpha help\check for updates and get the message "there are no new updates available".

Disgruntledly,
Jim

Re:Will Firefox 2.0 support the latest standards?

[malsdavis]

I just tried the new alpha release with the test you mention.

It almost works, but not quite unfortunatly, A couple of lines are out of place.

Still, it is far closer than say Internet Explorer 6.

Browser dreams

[MadFarmAnimalz]

I'm not all that enthusiastic about yet another iteration of Firefox... It's my primary browser and I do like it, but it will never be the browser that I would regard as the ultimate.

I envision a web browser which is the browser equivalent of Linux; a collection of simple programs performing very specific and narrowly defined tasks, all working through clean APIs or protocols. The HTML rendering being split off entirely, the javascript in its own library, image rendering separate, cookie management, security features, history management, bookmarks display, etc. Ideally, the various parts would be so simple that the barriers to development would be lowered drastically resulting in the organic rise of alternatives in the various segments; imagine having a flamewar over which js rendering plugin/library were better!

Extensions are not the solution by far. The functionality decentralization necessary to realize the vision of a browser like this far exceeds what the design idea behind extensions was.

Firefox will never be this. The only thing I've seen which might be salvaged into some sort of semblance of this vision is Kazehakaze, though that remains to be seen (I'm not sure you can even hotswap html rendering in Kazhakaze; I've never managed to keep it from crashing for long enough to test).

Re:Browser dreams

[pilkul]

No. If you drastically reduce dependencies, you also have to reduce integration and efficiency. That means a much less smooth experience for the user. Playstation emulators follow a model similar to what you're suggesting (separate plugins for input, sound and video) and everbody hates them.

Re:Browser dreams

[Cl1mh4224rd]

I envision a web browser which is the browser equivalent of Linux; a collection of simple programs performing very specific and narrowly defined tasks, all working through clean APIs or protocols. The HTML rendering being split off entirely, the javascript in its own library, image rendering separate, cookie management, security features, history management, bookmarks display, etc. Ideally, the various parts would be so simple that the barriers to development would be lowered drastically resulting in the organic rise of alternatives in the various segments; imagine having a flamewar over which js rendering plugin/library were better!

Ahh, yes... because everyone will love choosing which of the 100 Firefox distros will be right for them...

Re:OT: Your sig

[bcattwoo]

Reducing this, we end up with [E]/[L]. However, the dimensions of the Newton are ([M][L])/([T][T]), as the units are kgms^-2. (From F=ma, F is in N, m is in kg, a is in ms^-2)

But [E] can also be expressed as force times distance, i.e. Newton-meters, which when divided by distance obviously returns a measurement of force.

In other words, your equation is wrong :)

Looks like you were wrong, Mr. Smarty Pants.

Screenshots

[MagPulse]

Here and here.

Getting FF Extensions To Work

[mopslik]

I find it very strange that the winners of the recently posted FF Extensions contest do not work.

The usual reason for extensions "not working" is that the extension creators usually specify a maximum compatible version in the manifest. Quite often this is something like 1.5.*, as this is (was) the latest series for some time now. Naturally, this would exclude 2.0.

Try opening up the XPI file in your ZIP program, and change the maximum supported version in the INSTALL.RDF file, and see if the extension works. In most cases it does.

Re:Getting FF Extensions To Work

[Yer Mom]

Or use the Nightly Tester Tools to force them to install anyway, which involves a whole lot less fiddling about :)

My favourite bug...

[GeekDork]

Bug 9458 (referrer block for links from slash), "Implement inline-block in layout" hast its 7th birthday coming up.

Re:My favourite bug...

[porneL]

There's more in only-gecko-doesn't-support-that -basic-thing-for-years-now category: meet lack of soft hyphen support [9101].

Re:OT: Your sig

[c_forq]

Dude, look at the speed. There is an implied flux capacitor in that equation. My memory is a little fuzzy as to if a DeLorean is implied though...

Re:Javascript debugger?

[SharpFang]

The javascript debugger is an extension (it's up to the author to catch up with current releases) and as the author put it, is "officially unloved" so don't expect support from the core crew :)

Re:OT: Your sig

[hkgroove]

The DeLorean doesn't matter as much as if it's running off of Plutonium or Mr. Fusion.

Re:So basically ...

[PhysicsPhil]

They're changing features that work great now (tabbed browsing) and adding a whole bunch of features that the vast majority of end users really don't care much about (new data storage layer for bookmarks and history, extended search plugin format, updates to the extension system to provide enhanced security and to allow for easier localization of extensions). How Microsoft-esque...

If by Microsoft-esque you mean that version 1 has the features to keep 99% of the user base happy, you're absolutely right.

As far as the "average user" is concerned, what features is Firefox actually missing right now? It renders webpages, keeps bookmarks, has tabs and stores webpage passwords. That's enough for the vast majority of the world's users.

But would you prefer that the development team declare victory and stop coding? The Firefox team could stop development today and Joe User would be happily surfing with Firefox version 1 for many years to come. Any new developments are going to be for that last 1% category, because everyone else is happy.

SQL Bookmarks- overkill and overcomplex

[HighOrbit]

What was wrong with the html bookmarks file? Was it broke? Did it not work?

Granted that SQLite has a small footprint ,but not as small as a flat ascii file like html or xml. And why add complexity? It is very nice to be able to "export" a bookmarks file by just coping it or by opening a text browser and cutting and pasting into an email. I understand that SQLite's storage is also just a flat file, but is it in a commonly understood human readable format like html or xml? Or do I have to learn to parse a whole new file layout?

Re:SQL Bookmarks- overkill and overcomplex

[iamlucky13]

I was wondering that myself. Perhaps they have plans for some sorting features that would benefit from entries with information stored in seperate fields.

Re:SQL Bookmarks- overkill and overcomplex

[boa13]

You can simply use SQLite to read the data. The standard implementation comes with a proof-of-concept command line utility called sqlite which you can use to perform SQL queries on whatever SQLite database you want. You can probably script it, if not, sqlite is available wrapped in *plenty* of scripting languages (e.g. pysqlite for Python) which will allow you to develop tools to work on the database.

Surely, for the individual who just wants to copy some URLs from a text file, things have gone worse.

But for everything else, things have gone much better! It is now much easier to develop innovative tools on top of the bookmarks and history databases. History which, by the way, was almost totally unexploitable before that. Queries, stats, etc. People can go crazy! :)

Re:SQL Bookmarks- overkill and overcomplex

[Teddy Beartuzzi]

This change really sucks for those of us who use the bookmark file as our home page. :(

Gecko version

[jonasj]

1.5 uses 1.8.0, 2.0 will use 1.8.0.1, 3.0 will use 1.9

Almost:

1.5 uses 1.8.0, 1.5.0.1 uses 1.8.0.1, 1.5.0.2 will use 1.8.0.2, etc.

2.0 will use 1.8.1.

3.0 will use 1.9.

HTML validity is not the issue

[Antiocheian]

Valid HTML, either because it validates on a validator against a popular SGML doctype (such as various W3C proposals) or because it conforms to a standard set of tags will not solve the problem.

The problem is usually due to use of platform specific scripting commands to serve content. You can write perfectly valid HTML for Firefox that fails to work on IE and vice versa.

Re:Will Firefox 2.0 support the latest standards?

[jsoderba]

It will not. 2.0 is about non-rendering features.

Because of the huge changes going on in the Gecko rendering engine the Gecko team needs more time to work on it. 3.0 with the new Cairo-based Gecko 1.9 is scheduled for Q1 2007. See the Mozilla Wiki for more information.

How easy to retrieve SQLlite data in case of crash

[cexshun]

This situation just happened yesterday. A client's PC went down hard. I popped out the hard drive and tossed it into my machine. I was able to copy the bookmarks file, mailbox files, etc from their Mozilla software quite easily with a single cp command. Their new PC arrives, and some quick drag and drops has everything back to normal.

Now with this SQLlite layer, will I be able to do this just as easily? I pray that Thunderbird never decides to go this route!

Re:How easy to retrieve SQLlite data in case of cr

[frankm_slashdot]

well... heres some opinions from the "probably pulled out of my ass dept."

if thunderbird stored mailboxes in a db... it would probably be faster with a lot of messages...
if a computer crashed with sqllite'd history/bookmarks/email... and youre tech savy enough to copy the old style bookmarks over... youre probably savy enough to copy over the db files...

what you should be hoping for is that they go the route which best helps the project and the users... this may or may not be that route, but we wont know till we go down it.

</rant friendly="yes">

Faster? Smaller? Better?

[xavdeman]

Instead of making Firefox a subject of creeping-featurism, how about optimizing the code? Keeping it simple? I really don't see the need for a SQL style system just to manage my bookmarks and history (not that I even have the history function enabled). Just focus on making it start fast, and solve the memory management problems. I can't have 15 tabs open without a significant slowdown.

Re:Faster? Smaller? Better?

[spauldo]

This is streamlining, as least as far as the SQLite conversion goes.

They can toss the parsers for mork, html bookmarks, javascript preferences, and whatever else for a unified data layer. SQLite is made for this sort of thing, and it's optimized to be small and fast.

Basically, they're not really adding anything with this, they're just taking a lot of stuff they already had and going with a better solution. Best of all, SQLite is a separate project so they don't have to maintain the code, which lessens the burden on the dev team.

It'll also make it easier for 3rd party utilities to manipulate firefox data, since SQLite has all kinds of bindings for various scripting and non-scripting languages. I expect some really cool stuff to come out of all this, and improvements in existing 3rd party utilities since they won't have to rely on handbuilt parsers anymore.

Re:How easy to retrieve SQLlite data in case of cr

[Leffe]

Why do you think things would get harder?

SQLite stores its DB in one file, the previous system stored the bookmarks in one file.

Nothing has changed (except that the history is also stored in the bookmarks file, so you won't have to bother copying is specifically).

I pray that Thunderbird decides to go this route!

Fantastic - I can now write in spirals

[Forget4it]

> Support for SVG text using svg:textPath'" This mean that we can now publish texts along arbitrary paths. Searchable spiral poetry here we come! Compose in Inkscape - render in Firefox. For example [Cut and paste]

<?xml version="1.0" standalone="no"?>
<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN"
"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd" >
<svg width="12cm" height="3.6cm" viewBox="0 0 1000 300" version="1.1"
xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
<defs>
<path id="MyPath"
d="M 270.18634,82.608696 C 270.37894,82.840915 269.84742,83.007795 269.71822,82.996945 C 268.86959,82.925674 268.74445,81.803663 269.00939,81.18963 C 269.65841,79.685416 271.66302,79.596056 272.90121,80.35703 C 275.16538,81.748562 275.20055,84.922795 273.75419,86.910501 C 271.47392,90.044227 266.88857,89.999872 264.03872,87.707426 C 259.93187,84.403819 260.0792,78.194605 263.36056,74.378756 C 267.81264,69.201518 275.83942,69.473029 280.71836,73.873616 C 287.05778,79.591495 286.6427,89.615704 281.00202,95.649329 C 273.90745,103.23815 261.71672,102.66154 254.44134,95.667338 C 245.5198,87.090595 246.27474,72.57296 254.72988,63.97261 C 264.88962,53.638395 281.88727,54.587527 291.8925,64.605969 C 303.71643,76.445511 302.55801,96.070078 290.87822,107.55721 C 277.26558,120.94531 254.87258,119.56319 241.82909,106.12772 C 226.80467,90.651803 228.42435,65.35361 243.70653,50.68159 C 261.13312,33.950778 289.46906,35.821338 305.83972,53.038376 C 324.34513,72.500521 322.21079,104.00295 302.97336,122.1405 C 281.39306,142.48699 246.59883,140.07638 226.62783,118.7354 C 204.37535,94.956407 207.07436,56.748181 230.59994,34.878775 C 256.65623,10.656815 298.39776,13.65604 322.22912,39.445284 C 348.48268,67.85574 345.17174,113.24725 317.04156,139.10274 C 286.20171,167.44881 237.04601,163.8149 209.10542,133.26819 C 178.60714,99.925234 182.57498,46.893408 215.61228,16.807923 C 251.53065,-15.901217 308.54841,-11.588687 340.83749,24.011808 C 375.81511,62.576566 371.14735,123.6881 332.9124,158.23845 C 291.63152,195.5412 226.32027,190.50789 189.45192,149.56848 C 149.76829,105.50292 155.1773,35.887755 198.88998,-3.3544033 C 245.80768,-45.473804 319.82929,-39.679143 361.50021,6.8744969 C 406.10945,56.71076 399.91934,135.23977 350.45812,179.39359 C 297.63787,226.54599 214.50199,219.95078 167.81191,167.51634 C 118.06379,111.64764 125.07358,23.806825 180.54589,-25.472146 C 239.52665,-77.86785 332.1691,-70.43414 384.08888,-11.860175 C 439.18337,50.295371 431.31652,147.83488 369.57801,202.44673 C 304.18576,260.29056 201.65499,251.98149 144.30046,187.01635 C 83.65735,118.32624 92.417595,10.711206 160.67067,-49.436005 C 232.71901,-112.92776 345.51023,-103.70749 408.49953,-32.105955 C 474.88871,43.360318 465.19969,161.41853 390.18987,227.29876 C 311.24668,296.63363 187.83168,286.46722 119.01221,207.99 C 46.683932,125.51155 57.336246,-3.3490338 139.33934,-75.155507 C 225.41075,-150.52443 359.80475,-139.37778 434.64551,-53.790996 C 513.1018,35.930468 501.4525,175.9448 412.22482,253.86663 C 318.7968,335.45658 173.07611,323.29637 92.026882,230.37118 C 7.2574529,133.18067 19.936719,-18.331549 116.61546,-102.55404 C 217.62384,-190.54833 375.01194,-177.34194 462.45312,-76.854164 C 553.71724,28.026952 539.97573,191.37447 435.62397,282.07934 C 326.81578,376.65785 157.42608,362.37333 63.412908,254.10313 C -34.524138,141.31402 -19.688764,-34.199876 92.55379,-131.56555 C 209.37719,-232.90491 391.09658,-217.51093 491.85857,-101.24256 C 596.64366,19.668036 580.68337,207.6735 460.3362,311.87525 C 335.28597,420.1491 140.91452,403.61489 33.22994,279.13639 C -78.575364,149.89451 -61.459685,-50.922153 67.202237,-162.1323 C 200.68734,-277.51141 408.02782,-259.80677 522.80591,-126.90977 C 641.80084,10.869728 623.49981,224.81199 486.3165,343.20009 C 344.19184,465.85253 123.57045,446.94775 1.5306108,305.42728 C -124.82072,158.90709 -105.30489,-68.470206 40.603176,-194.20319 C 191.56889,-324.29449 425.77828,-304.16035 555.24551,-153.81463 C 689.11755,1.646224

Re:Fantastic - I can now write in spirals

[zero0w]

Interesting, but it's awefully slow to render in Bon Echo.

Inkscape has much better performance in rendering this SVG poetry you posted here.

Design for FF, typically works in IE

[beemishboy]

Most web people I know design for Firefox and then check IE for any weirdness. That seems to be much better than the reverse. The only exception would probably be certain CSS tags that IE has yet to support.

Re:How easy to retrieve SQLlite data in case of cr

[cexshun]

Thanks, that's why I asked. I did not know.

Re:How easy to retrieve SQLlite data in case of cr

[bwilson]

You can copy the very obviously named bookmarks_history.sqlite just the same way. Not surprisingly, this file contains the profile's bookmarks and history. It might also have a corresponding .journal file (if the previous run crashed) which you'll need. Did you think there would be some kind of secret storage that can't be copied?

To answer some other people's question: there will probably be some kind of bookmarks.html export in the final version if you want to just back up the bookmarks or want to transfer the bookmarks somewhere else.

Re:OT: Your sig

[Ginger Unicorn]

Type the left hand side of my equation into google.

then email google and tell them their calculator thingy doesnt work.

only someone else who owns a delorean may criticise my signature. :p

Is the download manager still broken?

[edxwelch]

I mean it currently doesn't resume downloads across sessions. In other words when there is only 10 bytes left of your 1GB download and you accendently shutdown ... tough titty, you have to start over. In opera the download manager work perfectly.

Same holds true with JavaScript

[Phil John]

Develop in Firefox, try in other browsers, fix any wierdness.

Re:OT: Your sig

[Disavian]

Mr. Fusion... any relation to Mr. Clean?

Extension Security?

[klenwell]

I looked at the Mozilla page. It states:

Updates to the extension system to provide enhanced security and to allow for easier localization of extensions

but doesn't elaborate.

I've always wondered about extension security. I looked into it a bit a while ago and came to the conclusion that it relied on the integrity/reputation of the developer and the vigilance of the community.

Maybe somewhere here knows the answers to these questions or can shoot back a link:

Are there any safeguards within the browser architecture itself that would, say, prevent an extension that blocked pop-ups from logging keystrokes or otherwise acting as spyware?

How are extensions reviewed before being linked from the Mozilla developers' page? Is that the job of the comments section at the bottom of the page?

Can extensions be unsafe?

Tom

Known Issues

[Anonymous Coward]

From TFA:

History and Bookmarks:

• There are significant issues with the user interface, including the history/bookmarks manager, the personal toolbar, the bookmarks menu, the bookmarks add/properties dialog, and livemarks. Many operations cause assets or other warnings, not everything updates properly, some operations don't work or are disabled.
• Can't export to bookmarks.html.
• Livemark loading locks up the browser.
• No sidebar-like functionality yet.
• Viewing all history is slow if you have a lot.
• All bookmarks with the same URI will have the same title/properties.
• The first run may take a few seconds to import the data from Firefox 1.0/1.5. There is no progress UI for this.

Extensions:

• When moving to Bon Echo from an earlier version of Firefox, some of your Extensions and Themes may be disabled. This is not an issue, but it may appear to be one (hence its listing here). For rationale, see "Extension and Themes" above.

Re:OT: Your sig

[Pdj79]

Possibly, but I do believe Mr. Fusion is a second cousin to Mr. Coffee on their father's side...you know...Fry Daddy.

Extensions

[roman_mir]

It is going to be quite annoying time for the extension/theme developers, the things will change every few days, users who are early adopters will be all like "why isn't your extension fully working with this alpha revision 1.5.0.14835 yet?" I know I am not touching this thing until it is at least beta.

System Requirements

[g3n0m]

I don't know why, but the system requirements page states that:
For running on Mac OS the minimum requrements are:

Operating Systems

* Mac OS X 10.2.x and later

Minimum Hardware

* Macintosh computer with a PowerPC G3, G4, or G5 processor
* 128 MB RAM (Recommended: 256 MB RAM or greater)
* 4 GB hard drive space

I don't know what you need the 4 GB of hard drive space for (52MB for both windows and linux)...
Link:
http://www.mozilla.com/firefox/system-requirements /

For those who want multiple versions of Firefox

[zero0w]

If you want to try out the latest 2.0 alpha version of Firefox without affecting your current 1.x installation, that they can run along each other, then you can check out this homepage for further detail (I have tried it, the tricks works for Linux version as well):

Running multiple Firefox versions concurrently
http://www.jeroencoumans.nl/journal/multiple-firef ox-versions

Features vs. Plugins

[Locarius]

What I'd really like to see the devs too is to move a large chunk of the FF featureset to a default plugin that is enabled by default. That way, users who want to run a stripped down version of FF can just remove the plugin and have an extremely light browsing experience. The users who want to run the full-featured Firefox simply run it out of the box.

Re:Features vs. Plugins

[Brother Dysk]

Considering that FireFox started life as a light version of Mozilla, I find that outrageously funny - just look how far they've drifted from their original goal!

Re:What is Bon Echo & FF2 Alpha does not updat

[jhermans]

BonEcho is the codename for Firefox 2, just like DeerPark was for Firefox 1.5. They're not using the name Firefox, otherwise people will claim various bugs and half-implemented features for Firefox, ignoring that it is an ALPHA release.

Re:But...use Konequeror

[moro_666]

sure the idea behind konqueror's "freedom" is quite nice, qt and trolltech :)

anyway, i sometimes use konqueror for simpler stuff, but as soon as the dhtml/js gets complicated, i have to go for mozilla because konq. just mocks things up. must admit tho, from memory usage and gui speed, konq. beats the ass of mozilla.

phoenix was nice and shiny, why in the world did they have to change it into this "item" that it is right now ? firefox is not really that much faster or lightweight than the mozilla-browser from the mozilla suite anymore.

i still sometimes find myself using galeon instead of the 2 mentioned above, it's faster than mozilla and has the compatibility with html/js of mozilla since it runs the same gecko, just without the slow gui :p

epiphany was also quite nice, but too "gnomefied", you couln't really configure much from "inside", had to run the gnome conf each time which sometimes wasn't such a bright idea from inside kde.

Don't forget Opera and Safari

[wanorris]

You can still encounter weirdness in other browsers even if it works in Firefox and IE. Opera and Safari are the most important other browsers to check -- along with IE 5, which still shows up on the browser usage charts.

Depending on your target audience, you may want to test other browsers as well, such as Konqueror (Linux), NetFront (handhelds), or v4 IE & NN.

Re:OT: Your sig

[jahknow]

At this moment Mr. Sparkle is disrespectin' dirt.

works well on OS X

[Quevar]

I've been using it on OS X and it seems faster than previous versions. Renders everything just fine. I'll be looking forward to the final version.

Previously, I'd favored Safari over Firefox, but this version might start to change my mind, unless Apple comes out with Safari 3....

Where is MNG support

[seb42]

Bugzilla Bug 18574, restore support for MNG animation format and JNG image format, has this guy wanting to use Firefox for research to view MNG movies of bio-molecules. But this beta does not open a mng file for me. Maybe it could be included in the next beta.

Re:Where is MNG support

[Glenn R-P]

MNG support is at http://mngzilla.sf.net/
includes binaries and patches. We were
tossed out of bug #18574 in December.

Embeddable/Extended Browser Component

[fncll]

I just wish that development would resume on the Firefox Embeddable/Extended Browser Component (or whatever the heck it's called) that would allow it to be used for internal browsing with third party programs-- it stinks being limited to IE's engine for internal rendering in development apps, feed-reading apps, etc.

What about 1.6?

[Hangin10]

What happened to 1.6?
1.6a1 is the version available on the developers page/nightly builds...

What version comes out next?

Re:md5 sig

[MacJedi]

I cannot quite tell if you are being serious or not with your md5 signed sig, but i have not been able to verify it. Are you including html formatting?

Re:can middle click, open new window?

[PieSquared]

I'm not sure how I did it, but middle click opens a new window for me in version 1.5 The only option I can find regarding this feature is in tools:options under tabbed browsing:tab focus. I'd suggest seeing if that works for you in 1.5, and not trying out an alpha release at this point.

Re:What is Bon Echo & FF2 Alpha does not updat

[JavaManJim]

Ok, thanks jhermans. Good rating too. I wish that the Firefox team would update Deer Park Alpha 2 instead. But I cannot change the direction the wind blows nor the direction of BonEcho.

Thanks again,
Jim

Don't ruin tabbed browsing!

[Trogre]

See http://slashdot.org/comments.pl?sid=180746&cid=149 56122.

In essence, don't ever put little 'close' icons on every tab. It just takes up tab real estate and makes it too easy to accidentally close tabs when navigating. Other programs like Lotus Notes have suffered from this problem.

I've heard a rumour that, sadly, the GNOME terminal might be going down this path.

Re:Don't ruin tabbed browsing!

[aok]

Will it be themeable?

I do like the current way...close button on the top-right or else middle-click to close.

But I'm not sure if I have a problem with the new way...although I guess it takes away space for longer titles.

Database vs Mork

[Richard_J_N]

I am delighted to see this. Some of the mozilla stuff still uses Mork, which is truly and utterly horrid. I recommend reading this delightful code by Jamie Zawinski, which has a brilliant rant about it:

Re:Database vs Mork

[ILikeRed]

Wow, I had not realized it was already such an awful, stupid mess. With that in mind it is definately a step forward, but I still think everyone would be better served with the simplicity of a text file. The data formats some people use... I mean I can understand it from a first year CS student, but wow.

Re:Database vs Mork

[Richard_J_N]

I like text files too [ grep :-) ] - and for bookmarks, maybe that's sane.

But bear in mind you can bookmark a group of tabs, and it no longer looks quite so simple. There's quite a lot of metadata: the URL, the user's name for it, the sort order, last accessed date etc.

The place this *really* matters though is the mozilla-mail address book. Have you ever tried to fix that? change one character, and the whole thing won't even parse.

Anyway, sqlite is easily used by other applications - and that's what really matters. OK, I can't grep it anymore, but on the otherhand, I don't need to learn the file format.

Re:What is Bon Echo & FF2 Alpha does not updat

[jhermans]

Updating Deer Park alpha 2 ? Why ? Deer Park was already released as Firefox 1.5 in november. The newer builds (both the 1.8.1 branch where FF 2 came from, and the trunk that will become FF 3) still carried the name until recently. There's no codename for Firefox 3.0 yet ; The Ocho was used for a while, but that's too close to Bon Echo.

False sense of security

[spaceturtle]

The only way to good security is to not give untrustworthy applications write access to all your files. There are far worse things they could do than add a few bookmarks, E.g. adding key loggers or (Sony) rootkits.

Ideally all applications would be run through something like the Principle of Least Authority shell which limits the applications so they can only access those files they actually need to function.

XPI = ActiveX

[Kagami001]

Extentions and the XPI installer sytem can execute any arbitrary action they feel like on the account Firefox is running in. They can contain executable native code and launch it. They're Firefox's version of Browser Helper Objects and ActiveX.

Re:XPI = ActiveX

[klenwell]

Have there been any documented cases of malicious extensions?

Re:md5 sig

[AeroIllini]

I cannot quite tell if you are being serious or not with your md5 signed sig, but i have not been able to verify it. Are you including html formatting?

Well, it's not the real hash, if that was what you meant by "serious." But it's not really a joke, either.

I put that statement in my sig to provoke thought about possible recursive message digests. Given a message m, a hash function h(), and a hash string k, is it possible to find a k such that h(m+k)=k? What about finding a k such that h(m1+k+m2)=k?

I'm not very well-versed in the mathematics of cryptography, so I can't answer that question in an abstract way, and in my very rudimentary searching in Google and Wikipedia, I can't find any information about it. Since I'm an engineer, my solution would probably be to set up an iterative program to brute-force a solution, using some popular root-finding method, but it would take forever, and there's no guarantee that it would work for every possible m, which is why I haven't done it yet.

Re:OT: Your sig

[RobertLTux]

check out the movie "back to the future" (its a trilogy btw) basically a delorean was used to mount a time machine and 1 if the TM (called a flux capacitor) was on 2 and the car was currently traveling at 88mph+ then the car was sent to the time period displayed on the second? row of the control panel in BttF 2 Mr fusion was a power unit mounted in the delorean.

Re:Bunratty is a Firefox troll.

[bunratty]

many problems that exist in Firefox and no other browser... There are major memory management and CPU hogging bugs the Mozilla developers have not been able to fix

Yeah, it's not like Opera has major memory management problems or CPU hogging problems or anything, right?

My point is not that these problems are okay — they're not. If you find any of these serious problems in any browser, report it right away so the problem can be fixed. My point is just that there's no reason to keep harping on the fact that browser X has problem Y, and acting as if other browsers don't also have a few issues of their own with problem Y.

"Sorry, our site doesn't come in HTML."

[tepples]

If it's not HTML, then my HTML renderer does not have to render it properly.

What if your employer refuses to provide conforming HTML? What if the only bank with ATMs in town refuses to provide conforming HTML? What if your government or a public utility refuses to provide conforming HTML?

Re:"Sorry, our site doesn't come in HTML."

[tepples]

Switching to an unregulated fuel source would be idiotic except in highly unusual & postapocalyptic circumstances.

The Web is "highly unusual & postapocalyptic circumstances."

Tab X's Not as Bad as You Think

[slateX]

From reading through replies here, one would think that X's were added to every tab, wasting a ton of space and ruining Firefox FOREVER! Disclaimer: I can't install the Beta release because it's only provided as an .exe, but I was able to grab the latest nightly, which should be approximately the same. If this behavior is not the same as in the release, the behavior of the latest nightly should at least be informative. In the latest nightly, it is true that in the situation of 6 or so tabs or less there is an X on every tab. But once I open more than 6 tabs, there is only an X on the currently active tab. Since there is no longer a static tab on the right side of the tab row, real-estate wise, this is a wash. I just middle click anyway, I was prepared to be peeved by the loss of room on the tab row.

firefox broken/dead since 1.0 please fix.

[bobs666]

I posted a valid question and was Moderation -1 100% Redundant Hello, McFly, I am the only one asking, and no valid answer yet exists. So one more try.

Can Firefox 2.0 have middle click to open a new window?

I use firefox 0.8 since every time I down load a new version it seems to only open new tab screens. I hate Tabs. And I don't want to left click and pull down.

I got two answers, one said use the right menu. wrong answer. the other said use the menu tools:options well that was moved to edit:pref... in pre 1.0 versions. no help yet. Then I was moded down, like some old trash.

Don't mod me down, I realy want an answer. ... or I guess firefox is become a wast. ... "move along nothing to see here.

The other wast are people miss using there mod points. asking questions, perhaps questions other might ask, is a valid way to /.

Quirks vs. standards mode by the DOCTYPE

[tepples]

designing an engine to run on sugar water will mean it won't run on gasoline.

Unless you use two engines. Mozilla Firefox in fact does have both a quirks mode and a standards mode.

Asking browsers to support both [broken and correct code] is asking them to read the webmasters' minds, and don't think Firefox has an extension for that yet.

Both Mozilla Firefox and Microsoft Internet Explorer base use of quirks mode vs. standards mode on the presence and content of the document's <!DOCTYPE declaration. Analogy would be different shaped nozzles that plug into the gasoline tank or the sugar tank: round for gasoline or oval for sugar.

Of course, the workaround now is to have the filling stations stock both sugar and gasoline, and dispense the correct one based on the type of car you drive.

Another way of doing it is through a proxy that converts one form of fuel (or markup) into another at the refinery (or web server). Sugars can be converted into ethanol, and a gasoline engine can easily be modified to run on mixtures of gasoline and ethanol.

Re:Quirks vs. standards mode by the DOCTYPE

[AeroIllini]

Both Mozilla Firefox and Microsoft Internet Explorer base use of quirks mode vs. standards mode on the presence and content of the document's <!DOCTYPE declaration.

Yes, and that only deals with HTML. I was discussing CSS (and I noticed you deftly avoided talking about the boxmodel issue I raised, choosing instead to discuss conversion of sugar to ethanol. Since the analogy is no longer valid and is obviously confusing the issue, I will stop using it). The differences between W3C HTML and non-spec HTML are not mutually exclusive; basically, non-spec HTML adds tags that are not in the spec. (Mozilla recently did this with the addition of the <canvas> tag, but they are working with the W3C to make it part of the next spec; Microsoft does no such thing when they add tags. Both Microsoft and the Mozilla Foundation are members of the consortium, but only one's products actually follow the specs. Hmmm.) The analog with the broken box model would be if the <font> tag affected the typeface of text in W3C browsers, but in Microsoft IE it created a text box much like a <div>. These two usages would be mutually exclusive renderings of the exact same definition, and would effectively render the tag broken in one of the two browsers. If they both followed a spec instead of making up their own definitions, these things wouldn't happen.

And for the record, all of your "broken" websites that you demand be fixed right now are only broken because of either CSS rendering inconsistencies such as the aforementioned box model, or they are using JavaScript methods that are not part of the official JavaScript spec and are exclusive to IE. No matter how crazy nasty your HTML is, there are no mutual incompatibilities between versions, and quirks mode is possible. This is not true for CSS and JavaScript.

Re:Quirks vs. standards mode by the DOCTYPE

[tepples]

And for the record, all of your "broken" websites that you demand be fixed right now are only broken because of either CSS rendering inconsistencies such as the aforementioned box model, or they are using JavaScript methods that are not part of the official JavaScript spec and are exclusive to IE.

Or if the server is programmed to be a discriminazi. Or if the site relies on an interactive <object> for which the only compatible plug-in is an ActiveX control.

So how do we convince owners of web sites to fix their broken CSS, JavaScript, user agent sniffing, and multimedia playback?

Re:Quirks vs. standards mode by the DOCTYPE

[AeroIllini]

So how do we convince owners of web sites to fix their broken CSS, JavaScript, user agent sniffing, and multimedia playback?

At this point, as a general web surfer, the only thing you can do is evangelize. When you find a website that does not have standards-compliant code, drop the webmaster a short note pointing them to the Webstandards.org FAQ, and perhaps (politely) point out where in their code they are non-compliant. Please note that there is a very big difference between "non-compliant" and "poorly designed." Flash, which is considered very poor form in web design, is compliant with the W3C spec as a valid <embed> object, and it does have its uses.

Unfortunately, user agent sniffing is a necessary evil, until such time as a certain unnamed browser (*cough*IE*cough*) decides to support webstandards properly. But since all that happens server-side, it should be trivial for a webmaster to impliment, and thoroughly transparent to the user. The way I see it is this: the webmaster can identify all the browsers which do not support web standards, and code a seperate stylesheet specifically for them (currently only IE). Everyone else gets the standard, and compliant, stylesheet. Hacks should not be used, since they rely on known bugs and could break with updates (such as in IE 7, which fixes bugs webmasters use to hack the broken box model in IE, but does not fix the box model). Many of these broken box model issues can be fixed with extra markup as well (nested, empty <div>s), which is not wrong per spec, but certainly clutters the page and goes against the intention of CSS.

Get out there and tell people why web standards are important. Standards of any kind are only useful if people adhere to them.