You're thinking of minor releases. During the 1.0 series, sometimes extensions were broken or disabled going from 1.0.n to 1.0.n+1. This was fixed in 1.5 by adding the extra minor version number and declaring that that number would not change the extension API, and by allowing extensions to specify wildcards (i.e. 1.5.*). So going from 1.5.0.n to 1.5.0.n+1 rarely if ever broke or disabled an extension. (I think there was one release which broke something, and they issued a fixed version the next day.)
Now, going from 1.0 to 1.5, or 1.5 to 2.0, is a bigger leap, and one might reasonably expect change in functions on which extensions rely. So, all the old extensions are marked as being compatible with the 1.5 series, not with 2.0. A lot of extensions have been tested with 2.0 and either marked as compatible or updated to become compatible. A lot haven't.
First, I hate how it defaults to a fixed location (~/Desktop, which means nothing to my wm), and you have to press a button to "Browse for other folders."
Do you use a new profile every time you launch the browser? 'Cause that's one of those settings you change once and never have to touch again, like setting your home page.
Yes. Just download the Linux version from Mozilla.com and install it.
Hey, you asked whether it was available for Debian, not from Debian....
Re:Close button in the active tab. Argh !
on
Firefox 2.0 RC2 Review
·
· Score: 3, Insightful
Blame it on conflicting usability studies -- or maybe conflicting usability goals.
Close buttons on the tabs are good from a discoverability standpoint. A close button on the end is good from a clicking-in-the-right-place standpoint.
Firefox has traditionally given discoverability a high priority.
Re:Is the big fat memory leak fixed?
on
Firefox 2.0 RC2 Review
·
· Score: 5, Informative
Yes and no.
Because there is no big fat memory leak. There are a whole bunch of little ones that add up. They've fixed a lot of them. They fixed a bunch of 'em in the 1.5.0.x series, and a bunch more in 2.0.
I doubt they've got everything, but 2.0 should have less of a memory problem than 1.5.
I've been using the Firefox 2 betas and RCs since beta 1. It is, overall, better than 1.5, but there's no "gee whiz!" factor this time through. (Though I'll admit inline spell-checking is quite nice!)
Of course, as a web developer, I'm really looking forward to Firefox 3, which will be built on Gecko 1.9 and should have some good improvements to the rendering engine. (Firefox 2 jumps from Gecko 1.8 to 1.8.1 -- minor changes only.)
Oh, yeah, on the extensions issue -- admittedly I don't use very many, but most of the ones I rely on have been updated by now. At this point I'm mainly waiting for the HTML Tidy-based validator.
Basically, as long as your version of Windows is supported, the version of IE that came with it will also be supported. Keep in mind that Microsoft considers service packs to be different "versions" for purposes of their lifecycle policy, and older SPs drop out of support before the newer ones do.
So IE6 will get security patches as long as Windows 2000 SP4 and Windows XP SP2 are still supported. Of course, the patches might not work on older versions like Win2k SP3 or WinXP SP1.
think you are confused as to what this guy is saying. The judge ordered ICANN to take Spamhaus' domain off the DNS system. If people put the IP for spamhaus.org in their hosts file, they will still be able to use it.
Their website, sure. But taking down spamhaus.org will also make the SBL inaccessible. Putting the IP of their web or mail server in your hosts file isn't going to help, because you have to look up a different hostname for each IP you test. Putting it resolv.conf might, but then you have to deal with sending all your DNS queries to spamhaus or waiting for timeouts.
Really, there are three ways to access the SBL if spamhaus.org gets taken out:
Someone could set up a mirror at another domain.
If you run a DNS server, you could arrange to retrieve a full copy of the zone (via rsync, for instance) and mirror it locally.
If you run a DNS server, you can set it up to forward all requests for *.spamhaus.org to their DNS servers.
All of these are more complicated than putting an IP in your hosts file.
Um... you are aware of how Spamhaus's list is distributed, right?
You convert the IP address of the server you're trying to check into a host name, such as W.X.Y.Z.sbl.spamhaus.org, then do a DNS lookup on that hostname. The result you get indicates whether the original IP is liste or not.
Trust me, you don't want to put 4 billion records in your hosts file!
Its funny how the best tool for the job theory goes out the window when it comes to Firefox.
That depends on "the job." For Firefox 1.x and Opera 7-8, if "the job" required WYSIWYG text editing or heavy AJAX functionality, then Firefox was the best tool. If "the job" involved highly advanced CSS, then it would have been Opera.
Don't assume everyone uses the web the same way you do. That's the same mentality behind the "Oprah sux b/c it don't have no extenshuns" attitude.
My guess: They don't want to see headlines saying "IE7 delayed again!" if they find some last-minute problem.
As for the deluge of incompatible users, keep in mind that the summary is incorrect, and it's not going to be an automatic update until next month (the exact wording is "a few weeks after it's available for download"). So that should limit the initial problems to early adopters.
Incidentally, IE7 RC1 has been out since August. I trust you've been working on updating your apps for the last six weeks, and just need time to finish?
I've never understood why the guys at Firefox can't (or won't?) fix something as simple and serious as memory leaks.
You are aware that they've fixed memory leaks in nearly every point release since 1.5, right? (Check the release notes.) And that they've fixed more memory leaks in 2.0?
It's not a matter of fixing a couple of big leaks. It's a matter of tracking down a bunch of little leaks and fixing each of them.
MS won't be pushing it out to automatc updates until a few weeks after it's been released for download. I'm guessing November's round of patches.
Even then, it looks like even the "automatic" installation will still be interactive. So instead of just installing it, it'll bring up a screen saying "You can get an updated Internet Explorer! Do you want it?"
I was dreading the inevitable process of trying to get a new CSS design working in IE 6; but hopefully now I don't have to:)
Ah, if only!
Even with Microsoft pushing it out via automatic updates next month, I expect it will take at least a year before IE6's marketshare declines to the point where we can ignore it.
Consider:
People who don't update (and don't have automatic updates set up).
People who are still on Windows XP Service pack 1
People who are still on Windows 2000 (still common in business), or on Windows 98/Me (yes, they're still out there!)
We can get some of the people on outdated OSes to switch to Firefox or Opera, but the ones who don't update are also unlikely to install third-party software that (as far as they can see) just duplicates something their computer can already do.
For that crowd -- and for the offices running Win2k + Microsoft apps across the board -- we're going to need to wait for people to replace their computers entirely.
The good news is that, in a matter of days or weeks, Internet Explorer 6 will officially be the new Netscape 4: a relic with declining relevance. If only it would decline fast enough!
I would guess not. IIRC, only "critical" updates are available to systems that haven't passed WGA, and this is a "high priority" update. It looks like even the automatic update next month is only going to be semi-automatic (it'll offer the download, but require user confirmation).
On top of that, even after it does become available through automatic updates -- which will most likely be in the November patch cycle, given that it's "a few weeks after" the October release, you can block the update (at least for now).
That's where recommendation systems come in -- the "other people who purchased XYZ also bout QWE" systems. If you have a broad enough sample size, you can tell a lot about someone's tastes by what else they're watching, listening to, etc.
This of course doesn't work for people who purchase different categories of stuff from different vendors (beyond the obvious "what's available") -- for instance, someone who buys only tech books on Amazon, but reads a ton of general fiction that they purchase at brick-and-mortar stores and check out from the library. IN that case, Amazon will get the mistaken impression that they mainly read tech books.
Anyway, you can take the data from recommendations, ratings, etc. and categorize people by what type of shows, books, or music they prefer -- at which point you can already target advertising for other media (that's what recommendations are, after all) -- plus you can correlate that data with demographic surveys to target other types of advertisements.
When Netscape started dying, people didn't have a lot of choice for browsers. There was Internet Explorer (cost: $0) and Opera (cost: annoying ads).
Depends on when you think Netscape started dying. At the time I gave up on Netscape Communicator, Opera was still 30-day trialware, with a cost of $39. (Fortunately I still qualified for the student discount at the time.) The ad-supported mode didn't appear until version 5, in December 2000.
I think that if Opera had had a non-ad, basic freeware version, they would probably have captured a lot of the people who now use Firefox. And the ad-version could very well hang over them in the minds of some people.
I'll definitely agree with that. The only problem is that I don't think there was another business model in place at the time that would have enabled them to sink money into a free product. IE had Microsoft's entire software empire. Netscape had AOL/Time Warner's media empire. Opera had... Opera. They had some versions of their embedded browser at the time, but that would've been it. The deals with search engine placement that fuel both Opera and Mozilla these days probably wouldn't have ben viable at the time.
Posting from Mozilla/5.0 (Macintosh; U; PPC Mac OS X; en) AppleWebKit/418.9 (KHTML, like Gecko) Safari/419.3
which is Safari 2.0.4
Not that you can tell that from the UA string. I simply don't understand why Safari insists on using a build number instead of the actual version number. Use the build number for the WebKit part, but for the Safari part, why not use the actual version number?
Back on topic, I'd guess those stats are taking the build number and dividing by 100, hence "Safari 4.x."
Slashdot Mirror
You're thinking of minor releases. During the 1.0 series, sometimes extensions were broken or disabled going from 1.0.n to 1.0.n+1. This was fixed in 1.5 by adding the extra minor version number and declaring that that number would not change the extension API, and by allowing extensions to specify wildcards (i.e. 1.5.*). So going from 1.5.0.n to 1.5.0.n+1 rarely if ever broke or disabled an extension. (I think there was one release which broke something, and they issued a fixed version the next day.)
Now, going from 1.0 to 1.5, or 1.5 to 2.0, is a bigger leap, and one might reasonably expect change in functions on which extensions rely. So, all the old extensions are marked as being compatible with the 1.5 series, not with 2.0. A lot of extensions have been tested with 2.0 and either marked as compatible or updated to become compatible. A lot haven't.
Do you use a new profile every time you launch the browser? 'Cause that's one of those settings you change once and never have to touch again, like setting your home page.
Yes. Just download the Linux version from Mozilla.com and install it.
Hey, you asked whether it was available for Debian, not from Debian....
Blame it on conflicting usability studies -- or maybe conflicting usability goals.
Close buttons on the tabs are good from a discoverability standpoint.
A close button on the end is good from a clicking-in-the-right-place standpoint.
Firefox has traditionally given discoverability a high priority.
Yes and no.
Because there is no big fat memory leak. There are a whole bunch of little ones that add up. They've fixed a lot of them. They fixed a bunch of 'em in the 1.5.0.x series, and a bunch more in 2.0.
I doubt they've got everything, but 2.0 should have less of a memory problem than 1.5.
I've been using the Firefox 2 betas and RCs since beta 1. It is, overall, better than 1.5, but there's no "gee whiz!" factor this time through. (Though I'll admit inline spell-checking is quite nice!)
Of course, as a web developer, I'm really looking forward to Firefox 3, which will be built on Gecko 1.9 and should have some good improvements to the rendering engine. (Firefox 2 jumps from Gecko 1.8 to 1.8.1 -- minor changes only.)
Oh, yeah, on the extensions issue -- admittedly I don't use very many, but most of the ones I rely on have been updated by now. At this point I'm mainly waiting for the HTML Tidy-based validator.
Sorry to hear it. Good luck with the damage control (and, more importantly, convincing the vendor to fix their product!)
Sorry, didn't see your other reply. Must've been posted while I was typing.
Short answer: yes. Long answer: depends on what version of Windows you're using.
Basically, as long as your version of Windows is supported, the version of IE that came with it will also be supported. Keep in mind that Microsoft considers service packs to be different "versions" for purposes of their lifecycle policy, and older SPs drop out of support before the newer ones do.
So IE6 will get security patches as long as Windows 2000 SP4 and Windows XP SP2 are still supported. Of course, the patches might not work on older versions like Win2k SP3 or WinXP SP1.
Their website, sure. But taking down spamhaus.org will also make the SBL inaccessible. Putting the IP of their web or mail server in your hosts file isn't going to help, because you have to look up a different hostname for each IP you test. Putting it resolv.conf might, but then you have to deal with sending all your DNS queries to spamhaus or waiting for timeouts.
Really, there are three ways to access the SBL if spamhaus.org gets taken out:
All of these are more complicated than putting an IP in your hosts file.
Um... you are aware of how Spamhaus's list is distributed, right?
You convert the IP address of the server you're trying to check into a host name, such as W.X.Y.Z.sbl.spamhaus.org, then do a DNS lookup on that hostname. The result you get indicates whether the original IP is liste or not.
Trust me, you don't want to put 4 billion records in your hosts file!
On the plus side, that might convince the judge to rethink the order.
That depends on "the job." For Firefox 1.x and Opera 7-8, if "the job" required WYSIWYG text editing or heavy AJAX functionality, then Firefox was the best tool. If "the job" involved highly advanced CSS, then it would have been Opera.
Don't assume everyone uses the web the same way you do. That's the same mentality behind the "Oprah sux b/c it don't have no extenshuns" attitude.
My guess: They don't want to see headlines saying "IE7 delayed again!" if they find some last-minute problem.
As for the deluge of incompatible users, keep in mind that the summary is incorrect, and it's not going to be an automatic update until next month (the exact wording is "a few weeks after it's available for download"). So that should limit the initial problems to early adopters.
Incidentally, IE7 RC1 has been out since August. I trust you've been working on updating your apps for the last six weeks, and just need time to finish?
You are aware that they've fixed memory leaks in nearly every point release since 1.5, right? (Check the release notes.) And that they've fixed more memory leaks in 2.0?
It's not a matter of fixing a couple of big leaks. It's a matter of tracking down a bunch of little leaks and fixing each of them.
MS won't be pushing it out to automatc updates until a few weeks after it's been released for download. I'm guessing November's round of patches.
Even then, it looks like even the "automatic" installation will still be interactive. So instead of just installing it, it'll bring up a screen saying "You can get an updated Internet Explorer! Do you want it?"
Ah, if only!
Even with Microsoft pushing it out via automatic updates next month, I expect it will take at least a year before IE6's marketshare declines to the point where we can ignore it.
Consider:
We can get some of the people on outdated OSes to switch to Firefox or Opera, but the ones who don't update are also unlikely to install third-party software that (as far as they can see) just duplicates something their computer can already do.
For that crowd -- and for the offices running Win2k + Microsoft apps across the board -- we're going to need to wait for people to replace their computers entirely.
The good news is that, in a matter of days or weeks, Internet Explorer 6 will officially be the new Netscape 4: a relic with declining relevance. If only it would decline fast enough!
Hasn't IE's marketshare increased every summer for the last few years, wth Firefox and company catching up again in fall?
I would guess not. IIRC, only "critical" updates are available to systems that haven't passed WGA, and this is a "high priority" update. It looks like even the automatic update next month is only going to be semi-automatic (it'll offer the download, but require user confirmation).
On top of that, even after it does become available through automatic updates -- which will most likely be in the November patch cycle, given that it's "a few weeks after" the October release, you can block the update (at least for now).
But what is "prime time" when the show is distributed via download-on-demand?
That's where recommendation systems come in -- the "other people who purchased XYZ also bout QWE" systems. If you have a broad enough sample size, you can tell a lot about someone's tastes by what else they're watching, listening to, etc.
This of course doesn't work for people who purchase different categories of stuff from different vendors (beyond the obvious "what's available") -- for instance, someone who buys only tech books on Amazon, but reads a ton of general fiction that they purchase at brick-and-mortar stores and check out from the library. IN that case, Amazon will get the mistaken impression that they mainly read tech books.
Anyway, you can take the data from recommendations, ratings, etc. and categorize people by what type of shows, books, or music they prefer -- at which point you can already target advertising for other media (that's what recommendations are, after all) -- plus you can correlate that data with demographic surveys to target other types of advertisements.
Depends on when you think Netscape started dying. At the time I gave up on Netscape Communicator, Opera was still 30-day trialware, with a cost of $39. (Fortunately I still qualified for the student discount at the time.) The ad-supported mode didn't appear until version 5, in December 2000.
I'll definitely agree with that. The only problem is that I don't think there was another business model in place at the time that would have enabled them to sink money into a free product. IE had Microsoft's entire software empire. Netscape had AOL/Time Warner's media empire. Opera had... Opera. They had some versions of their embedded browser at the time, but that would've been it. The deals with search engine placement that fuel both Opera and Mozilla these days probably wouldn't have ben viable at the time.
Not sure why it doesn't show up with find-as-you-type, but it's still there if you hit Ctrl+F.
Not that you can tell that from the UA string. I simply don't understand why Safari insists on using a build number instead of the actual version number. Use the build number for the WebKit part, but for the Safari part, why not use the actual version number?
Back on topic, I'd guess those stats are taking the build number and dividing by 100, hence "Safari 4.x."