
The first release candidate for Firefox 1.5 has just been released. Exciting times! I’m thinking about trashing my default profile (gasp!) and starting fresh for 1.5. Recreating my profile will take some time, but I think it will be worthwhile in the long run.
In semi-related news, I hope to have Googlebar Lite 3.1 available very soon (I’m only waiting on one more translation), so 1.5 RC 1 users can continue using GBL. CoLT should be upgraded soon as well.
According to Asa, this Wednesday might be “ship day” for the first Firefox 1.5 release candidate. Once it’s released, I think I’m going to replace 1.0.7 with RC1 as my primary browser. The beta releases weren’t nearly as stable as I wish they had been (I unsuccessfully tried switching to them shortly after their release), but the latest nightly builds have seemed very stable. All of the new features and bug fixes in 1.5 make it awfully hard to pass up.
With Firefox 1.5 moving ever closer to an actual release, I am doing my best to keep Googlebar Lite up to date. The nightly trunk builds recently changed the maxVersion for installed extensions (it’s now 1.5). The bad news is that the latest version of GBL (3.0.1) only supports the 1.5 beta builds (maxVersion of 1.4). The good news is that I am working on Googlebar Lite 3.1 which will fix this problem. Other changes in this upcoming release include Google Maps search in the context menu, an updated about box (giving credit to all the translators), and other miscellaneous changes. I am aware that the Swedish translation is corrupt, and I hope to remedy that situation as well.
One of the great new features of Firefox 1.5 is the vastly improved extension manager. An article over at the MozillaZine Knowledge Base details how to use the new extension manager system to do dynamic extension development. I’ve given it a try with Googlebar Lite, and I can only say that I am truly amazed! I can now easily edit the XUL and JavaScript for Googlebar Lite and see the results in real time (no more re-packaging and re-installing). This change in my development cycle has streamlined things quite a bit, so I hope to have GBL 3.1 available very soon.
Here’s a small but handy Firefox tip for “safekeeping” your bookmarks. It also lets you share your bookmarks across multiple profiles!
You can use this trick in multiple profiles, allowing them to all point to the same bookmarks file. Additionally, it helps to safeguard against possibly losing your bookmarks if your profile becomes corrupt.
Scott Berkun used to work for Microsoft on the Internet Explorer web browser. So if there is anyone around who can appreciate the machinery behind the web browsing experience, it’s him. Scott recently switched to using Firefox as his primary web browser (his article explains why he made the change). But more interestingly, he provides several things he sees wrong with Firefox. Several of his points are right on target (there is some broken-ness in the Firefox world), and several of his points are a little off the mark (a few of his problems can be solved with extensions).
But Scott has class. He recently posted a followup article, clarifying a number of questions raised by the first one. In it, he admits that after he heard from a number of Firefox users, he saw that some of his views warranted another look. And he also (correctly) points out that there is still inherit broken-ness in the browsing world as a whole. Thanks for an insightful post, Scott. And here’s to your switch to Firefox: you won’t regret it.
A day or two ago, I replaced my primary Firefox install with the 1.5 Beta release. This afternoon, I went back to 1.0.6. As excited as I am about the upcoming 1.5 release, this beta simply has too many problems to switch to it full time. Too many errors crop up in the JavaScript console, too many extensions and themes do not yet work with it, and its overall stability is questionable (the beta crashed on me minutes after installing it for the first time). I think I might go back to running the nightly branch builds on my “nightly” profile. Had I been thinking clearly at the time, I would have used this profile with the beta install. Unfortunately, I contaminated my normal profile instead. So now I’m having to build a new one. *sigh*
Such is the price of progress.
Almost all of today and a better part of last night were spent developing the next incarnation of Googlebar Lite. In doing so, I’ve run into a bit of a versioning dilemma. The changes that have been made in this next release are rather large, and seem to warrant a major version number jump. But I don’t want to do something corny, much like Firefox has done with the 1.0 » 1.5 transition (hey - let’s just call it Firefox 2.0, OK?).
So, do I make a giant leap to 3.0, or a small step to 2.3? At least for now, I’m leaning towards the former.
Asa Dotzler reported yesterday that Firefox 1.5 Beta 1 is now available to download. Apparently, the beta has revved the maxVersion for extensions to 1.4. As such, I will no doubt need to check that Googlebar Lite still works and get a new version out the door (with updated version support).
This is an exciting time for Firefox users! The Beta release of 1.5 includes a host of new features and thousands of bug fixes, making the best browser in the world even better. Improved accessibility will also help those who make use of screen-readers, making their lives much easier.
According to site statistics, my tutorial on creating Firefox toolbar extensions has become rather popular again (it originally surged in popularity after Asa Dotzler commented on it in his blog). It has shown up on a number of web sites recently, for reasons I do not fully understand. Regardless, I’m always appreciative of such linkage (so keep it up!). The more people that can benefit from the article, the better.
It has been quite some time, however, since I have actually updated the tutorial. I hope to remedy that in the next week or two. Since writing the article, I have improved my knowledge of Firefox toolbars a great deal (as evidenced by my Googlebar Lite extension). I think it might be interesting to add some information on how to move a toolbar item around, how to get notification that the customize toolbars process has ended, and other useful tidbits. I also intend to eventually zip up the article itself, so that people can view it offline. Good times.
Recently, the Mozilla team changed the upcoming Firefox 1.1 release to version 1.5 instead. This got me thinking about version numbers in general, and how silly they are. I have gotten trapped in this kind of thing before, so I’m no less guilty than the next guy. Here’s what I think is wrong with version numbers:
Actual Numbers are Too Vague
As they are, version numbers are a little too vague. How is 1.0.3 different than 1.0.3.1? Well, we put in a teeny-tiny change that didn’t warrant the 0.0.1 number bump. So why didn’t you name it 1.0.3.0.0.0.1 if the change was so tiny? Well, it wasn’t that tiny. Sheesh.
Leading Zeroes
This is one place where I have been caught before. Typically, a leading zero (as in 0.8 or 0.9.1) indicates that the software is in a test or “pre-release” phase, and that is just fine. But there are far too many abuses of this numbering scheme. Firefox extensions like ForecastFox or Web Developer (both of which use leading zeroes as of this writing) are far too mature and stable to warrant the leading zero. Go to a 1.x release already, dammit!
Version Numbers Too Long
The maximum length of any version number should only ever be 3 positions long: “X.Y.Z.” That’s it. If you want some sort of date stamp for nightly builds, have that as a separate item; not as a part of the version number. I should not have to see versions like 1.03.45.1200.129393.120202 any more. And non-sensical stuff, like Adblock’s version of “0.5 d2 nightly 39″ should be outlawed from the planet. Please, use numerals only.
Arbitrary Number Change Decisions
This kind of stuff irritates me, but I’m also guilty as charged. The Firefox team feels like the changes they’ve made are “worth more” than a 0.1 version number bump. I heartily agree. But this 0.5 version bump is a joke. Just bump the major version number! Go directly to 2.0 - do not pass Go and do not collect $200. What’s so bad about that? The next release, if it’s just as big, can be 3.0. And then we can go to 4.0 if necessary. Everyone is so scared of their version number becoming too “large”. Think about it: how many programs out there are version 12.0 or version 29.3? Not many. Once we reach double digits, we feel like we have to come up with a fancy name (like “XP” or “2005″ or whatever else). The AutoCAD folks had it right years ago. “Buy AutoCAD 14.0 today,” they would advertise. Now it’s simply AutoCAD 2006. How cheap. What happened to AutoCAD 20.0? That’s just as “cool” sounding.