Yeah, I submitted a similar story to Slashdot about a year ago, and as others have pointed out it goes back at least as far as WW2 and morse code operators.
The best part is the bandwidth throttling, back to 1994 dial-up speeds. I was looking at this yesterday, and it was weird to watch the interlaced GIFs load line by line. (Remember how Netscape used to have a LOWSRC attribute for images, so you could specify a low-res version that could be loaded quickly and displayed while it tried to download the massive, whopping 50K full image?)
A flashback to the way I first encountered the web.
Of course, it's probably even slower today, now that it's linked here.
The only that i would like to say is that Acid3 nuked really fast. At least the Acid2 ghosted the browser market for at least 2 years (correct me if i am wrong). Acid3 passed in a less that a month. It seems at last that the Acid3 wasn't so hard at all. Someone didn't do really good job (kidding). I think you're missing something. We've got only two browsers that are close to passing Acid3, both of them only in development builds. Acid3 won't reach endgame until Opera, Safari, Firefox and IE all ship releases that pass it.
Initial progress on Acid2 was just as fast, with internal builds of Safari passing in just 2 weeks. A final release came later that year, with Opera following the next year. As for Firefox and IE, it was a matter of priorities. Mozilla decided not to upgrade the rendering engine in Firefox 2, so all the Acid2-related changes got pushed back. Microsoft, as far as I can tell, wanted to get IE7 out as quickly as possible and target those issues that had people jumping ship. That meant things like tabs were more important than improved rendering, though at least we got some improvements out of it.
It's worth remembering that Opera's install base is only tiny when looked at as a percentage of overall web use. I'm not sure about current statistics, but a year ago they were at 10-15 million desktop users -- not counting a huge install base of Opera Mobile on phones, and millions of active Opera Mini users.
For comparison, the average U.S. state has around 6 million people. By the 2005 census, only 4 states -- California, New York, Texas and Florida -- had more than 15 million.
Admittedly, not all 15 million people are likely to be visiting your specific website (unless you're Google or MySpace or something), and it really does come down to how many Opera users are in your potential audience, but 15 million doesn't seem quite so small as 1%.
Firefox 3 is late in the beta cycle. They should be focusing on stabilization right now, not bunches of changes to site rendering.
Similarly, all this work on Opera is not likely to show up in 9.5, for exactly the same reason. It'll probably show up in Opera 10 (or 9.6, or whatever they end up calling it).
If there's anywhere to work on Firefox 4 stuff, that would be where I'd look for Acid3-related changes.
The reason Acid2 hasn't improved matters so far is that only ~6% of the browsers in use pass it. We need Firefox3 and IE8 to hit the streets. More importantly, we need them to thoroughly supplant IE6, IE7, and Firefox2.
Once that happens, developers will be able to count on those features being available. It's a bit tricky to do that when 25-50% (depending on where you look) of your audience is still running IE6.
The test criteria:
To pass the test, a browser must use its default settings, the animation has to be smooth, the score has to end on 100/100, and the final page has to look exactly, pixel for pixel, like this reference rendering. The reports from Opera indicate that they've got all 100 subtests, but still have additional issues with the remaining criteria:
Our latest internal build (screenshot below) scores 100/100 and renders the test almost perfectly! We have some work to do still, but we expect to have that taking care of shortly
There you go - it installs new stuff that wasn't asked for. I take it you missed the word required in the sentence you quoted. Safari isn't required for iTunes, or vice-versa. There is no reason that updating Safari should install iTunes. There is no reason that updating iTunes should install Safari. Neither yum nor apt, nor any reasonable GUI wrapper for them, would do so if you just told it to update.
And in my GUI package manager, updating and installing is both handled in the same window. As I said, I don't have a problem with them being in the same window. But I'll bet your package manager makes it clear which ones are which, right? And it doesn't just pre-select every uninstalled app in the active repositories, does it?
How about notification? Does it pop up a window on the desktop to tell you that new packages are available? If so, does it pop up only when there are updates that apply to your system, or does it pop up every time any package gets a new version?
If you open it up and tell it to update your system, without any further customization, does it only update what you have, or does it install every single app available for your distribution?
Apple's updater is only like a Linux package manager if the Linux package manager makes no clear distinction between new installs and updates, installs all available software when you tell it to update what's on your system (unless you carefully deselect the packages you don't want), and (if it does notification) pops up a notice every time any program has a new version, whether you have that program installed or not, even if you told it to ignore the previous version of that program.
That doesn't sound like any package manager I've ever used on Linux.
Wait. I can both update existing software, and install new packages from my Linux package manager. Exactly the same way Apple does. I don't really see the problem. Why is it evil when Apple does something, but OK under Linux? It's not the same. If you do, for instance, "yum update" or "apt-get upgrade" it will only install updates to software you have currently installed, plus anything that's required by the new version. If I don't have Firefox installed, and I update my distro-provided copy of Thunderbird, it's not going to try to install Firefox along with it (unless the dependencies are messed up).
What Apple is doing is the equivalent of treating "yum update" as a command to install every package in the active repositories. It just happens that they only have a few pieces of software in their repository -- Safari, iTunes, QuickTime, and the updater itself (maybe one or two others) -- so it's not as unwieldy as it would be on a Linux distro.
I disagree with sulfur's opinion that they should use separate apps for update & install -- as I mentioned above, I think they can be in the same app as long as they're clearly separated and the default does the sensible thing, which is to only update what's already on your machine.
What are the reasons that MS doesn't open up Windows Update so that 3rd party suppliers can add hooks to update their own software? Wouldn't this allow me to setup preferences in Windows Update so that I could be notified and asked to approve any items that wanted to add themselves to my update queue? Sure would be nice not to have to worry about uninstalling the myriad of auto-updaters just to avoid giving large chunks of my system resources to Adobe, Apple, etc... Personally, I think this would be fantastic, but then, like you, I'm looking at it from the perspective of a Linux user who can update everything through one update system, including stuff from third-party repositories.
My best guess? They don't want to risk being held liable for broken/hacked/malicious third-party updates. I've seen third-party drivers on the service, but that's it, and I expect Microsoft is vetting them in some way before putting them on the list.
Why is it a bad thing that most computer users aren't paying attention? Because it makes them more susceptible to fraud and malware. For example, if someone puts up a fake bank site to collect financial info, it's the phisher's fault. But a user who is informed and/or looks at it critically is less likely to fall for the scam.
But aside from that, I think your statement is 110 per cent wrong. Leaving aside the issue with math, are you saying that you think users who don't pay attention aren't going to get surprised by things like this, or that most computers are paying attention, or that it's a good thing that most computer users don't pay attention to what they install on their computers?
Mac users: prefer Firefox over Safari (I don't know if Opera is available for it...) It is, and has been for several years, but I think it's only just getting to the point where the Mac version is really usable. I saw it go through the same process when they ported it to Linux: it took 2 or 3 major releases before the Linux version of Opera achieved parity with the Windows version in terms of stability and performance.
then I saw that it gets downloaded whether you wan tit or not unless you hit cancel. Not quite. You can disable it -- the problem is that it's selected by default, and you have to be paying attention to notice that something unfamiliar is in there. Since people have been trained by years of "Keep your system up to date so you don't get hacked/infected/etc!" to accept all updates, a lot of them are going to just accept that update without realizing that they allowed it to install a new program.
And it's not just Safari. It's iTunes as well. If you have QuickTime or Safari (it's been in beta on Windows since last summer), but not iTunes, the updater will offer you iTunes -- preselected -- every time a new version comes out, and call it an update. It's only become an issue now because most people using Apple Software Update on Windows were using it for iTunes. Since Safari was in beta until recently, the only things the updater offered were iTunes and QuickTime -- things that were already on most users' machines.
The last time I downloaded QuickTime I had to hunt through Apple's site to find the standalone version. You mean the one on the QuickTime Download Page? When I look at that page (which I got to in two clicks from the front of Apple's website: Downloads, then the Download button under QuickTime) on a Windows box, I see two options: QuickTime+iTunes, or QuickTime.
I call bullshit on Mozilla. Microsoft forced IE 8 on me. I did not have a choice. Apple offered me Safari and I turned them down. Considering that IE8 is only out as a download-it-deliberately beta, I doubt Microsoft forced it on you.
And if you meant IE7, there's a difference: Unless you went to a great deal of effort to remove it, an older version of Internet Explorer was already on your computer, so it actually is an update to software you already had. It's not as if Microsoft installed IE7 on a Mac or a Linux box in such a way that someone who was not paying attention (i.e. most computer users, unfortunately) could get it by accident.
Actually, in regards to QuickTime and iTunes, it doesn't simply show you iTunes. The ONLY update available for QuickTime is "QuickTime + iTunes". Not always. I've seen updates for just Quicktime show up in the list.
After I tried using systems default browser (Safari) as my only browser instead of 3rd party and ended up downloading Firefox 2 because some large site required it for extra needed function (Firefox'es sponsor too) I think Mozilla CEO should be the last to talk about "pushing browsers to people". How is the fact that you deliberately went out, downloaded and installed a browser in response to a website requirement/incompatibility anything like an updater automatically installing software for people who aren't paying attention?
It's an updater for all Apple software on Windows. So whether you're trying to update QuickTime, iTunes, or Safari, it will show you new versions of all three. The real problem is that it automatically enables "updates" for items that you don't already have installed, making them opt-out instead of opt-in.
And no, it doesn't make Safari the default browser.
1. Make all not-yet-installed software unchecked by default, so you have to opt into it (keeping actual updates checked by default) 2. Clearly label, probably by putting a separator and header in the middle of that list, which software is an update to what's on your machine and which software is another offering that Apple wants you to install.
That, and make it possible to ignore a product, instead of just a particular install. My Windows box at work has Safari and QuickTime for web development purposes, but it keeps telling me to "update" iTunes. I can tell it to ignore the item, but every time a new iTunes version comes along, it asks again.
I've been using the nightly builds for a couple weeks now, and they're flagged as beta 5... I figured beta 4 had been out for a while already.
For what it's worth: I'm very impressed with what I'm seeing of Firefox 3 so far. It's faster, uses less memory, and I really like the new address bar features, and the bookmarking. (It has tagging built into the bookmarks now.)
I've been using the nightly builds for a couple weeks now, and they're flagged as beta 5... I figured beta 4 had been out for a while already.
As I understand it, they branch the code when they're ready to produce a beta. So there's a set of nightlies still on the trunk, which will eventually become Firefox 3, and there's a set of nightlies that will become beta 4. That way it's possible to refine beta 4 a bit, and still keep working on issues that are targeted at beta 5 and beyond.
So the plain nightly builds are after the branch for beta 4, and working toward beta 5.
I think the point is that they *did* know that this particular vulnerability affected Opera and took their time about telling them.
Did they? I asked about the nature of the communication, and no one seems to be willing to say whether it was a case of "Oh, we just noticed this affects you" or "By the way, we knew this affected you two weeks ago and we're just telling you now." The former is bad timing, but if that's the case I don't see anything wrong with Mozilla's behavior. On the other hand, if they did sit on that knowledge for however many weeks, then it's irresponsible.
But no one "in the know" has been willing to reveal what, to me, is the critical detail.
Slashdot Mirror
The article mentions "Mozilla," so I'd assume the actual page rendering uses Gecko.
Yeah, I submitted a similar story to Slashdot about a year ago, and as others have pointed out it goes back at least as far as WW2 and morse code operators.
Slashdot could join in by reviving the OMG Ponies theme. Pink is close enough to magenta, right?
The best part is the bandwidth throttling, back to 1994 dial-up speeds. I was looking at this yesterday, and it was weird to watch the interlaced GIFs load line by line. (Remember how Netscape used to have a LOWSRC attribute for images, so you could specify a low-res version that could be loaded quickly and displayed while it tried to download the massive, whopping 50K full image?)
A flashback to the way I first encountered the web.
Of course, it's probably even slower today, now that it's linked here.
Yeah, but in the past they've started at midnight GMT.
Remember OMG Ponies?
At least the Acid2 ghosted the browser market for at least 2 years (correct me if i am wrong).
Acid3 passed in a less that a month.
It seems at last that the Acid3 wasn't so hard at all. Someone didn't do really good job (kidding). I think you're missing something. We've got only two browsers that are close to passing Acid3, both of them only in development builds. Acid3 won't reach endgame until Opera, Safari, Firefox and IE all ship releases that pass it.
Initial progress on Acid2 was just as fast, with internal builds of Safari passing in just 2 weeks. A final release came later that year, with Opera following the next year. As for Firefox and IE, it was a matter of priorities. Mozilla decided not to upgrade the rendering engine in Firefox 2, so all the Acid2-related changes got pushed back. Microsoft, as far as I can tell, wanted to get IE7 out as quickly as possible and target those issues that had people jumping ship. That meant things like tabs were more important than improved rendering, though at least we got some improvements out of it.
It's worth remembering that Opera's install base is only tiny when looked at as a percentage of overall web use. I'm not sure about current statistics, but a year ago they were at 10-15 million desktop users -- not counting a huge install base of Opera Mobile on phones, and millions of active Opera Mini users.
For comparison, the average U.S. state has around 6 million people. By the 2005 census, only 4 states -- California, New York, Texas and Florida -- had more than 15 million.
Admittedly, not all 15 million people are likely to be visiting your specific website (unless you're Google or MySpace or something), and it really does come down to how many Opera users are in your potential audience, but 15 million doesn't seem quite so small as 1%.
Firefox 3 is late in the beta cycle. They should be focusing on stabilization right now, not bunches of changes to site rendering.
Similarly, all this work on Opera is not likely to show up in 9.5, for exactly the same reason. It'll probably show up in Opera 10 (or 9.6, or whatever they end up calling it).
If there's anywhere to work on Firefox 4 stuff, that would be where I'd look for Acid3-related changes.
The reason Acid2 hasn't improved matters so far is that only ~6% of the browsers in use pass it. We need Firefox3 and IE8 to hit the streets. More importantly, we need them to thoroughly supplant IE6, IE7, and Firefox2.
Once that happens, developers will be able to count on those features being available. It's a bit tricky to do that when 25-50% (depending on where you look) of your audience is still running IE6.
How about notification? Does it pop up a window on the desktop to tell you that new packages are available? If so, does it pop up only when there are updates that apply to your system, or does it pop up every time any package gets a new version?
If you open it up and tell it to update your system, without any further customization, does it only update what you have, or does it install every single app available for your distribution?
Apple's updater is only like a Linux package manager if the Linux package manager makes no clear distinction between new installs and updates, installs all available software when you tell it to update what's on your system (unless you carefully deselect the packages you don't want), and (if it does notification) pops up a notice every time any program has a new version, whether you have that program installed or not, even if you told it to ignore the previous version of that program.
That doesn't sound like any package manager I've ever used on Linux.
What Apple is doing is the equivalent of treating "yum update" as a command to install every package in the active repositories. It just happens that they only have a few pieces of software in their repository -- Safari, iTunes, QuickTime, and the updater itself (maybe one or two others) -- so it's not as unwieldy as it would be on a Linux distro.
I disagree with sulfur's opinion that they should use separate apps for update & install -- as I mentioned above, I think they can be in the same app as long as they're clearly separated and the default does the sensible thing, which is to only update what's already on your machine.
My best guess? They don't want to risk being held liable for broken/hacked/malicious third-party updates. I've seen third-party drivers on the service, but that's it, and I expect Microsoft is vetting them in some way before putting them on the list.
And it's not just Safari. It's iTunes as well. If you have QuickTime or Safari (it's been in beta on Windows since last summer), but not iTunes, the updater will offer you iTunes -- preselected -- every time a new version comes out, and call it an update. It's only become an issue now because most people using Apple Software Update on Windows were using it for iTunes. Since Safari was in beta until recently, the only things the updater offered were iTunes and QuickTime -- things that were already on most users' machines.
And if you meant IE7, there's a difference: Unless you went to a great deal of effort to remove it, an older version of Internet Explorer was already on your computer, so it actually is an update to software you already had. It's not as if Microsoft installed IE7 on a Mac or a Linux box in such a way that someone who was not paying attention (i.e. most computer users, unfortunately) could get it by accident.
It's an updater for all Apple software on Windows. So whether you're trying to update QuickTime, iTunes, or Safari, it will show you new versions of all three. The real problem is that it automatically enables "updates" for items that you don't already have installed, making them opt-out instead of opt-in.
And no, it doesn't make Safari the default browser.
IMO, all Apple has to do to solve this is:
1. Make all not-yet-installed software unchecked by default, so you have to opt into it (keeping actual updates checked by default)
2. Clearly label, probably by putting a separator and header in the middle of that list, which software is an update to what's on your machine and which software is another offering that Apple wants you to install.
That, and make it possible to ignore a product, instead of just a particular install. My Windows box at work has Safari and QuickTime for web development purposes, but it keeps telling me to "update" iTunes. I can tell it to ignore the item, but every time a new iTunes version comes along, it asks again.
It's been a standard joke in my family to describe anything "new and improved" as "with extra cheese!"
Odd that it's actually appropriate in this case.
For what it's worth: I'm very impressed with what I'm seeing of Firefox 3 so far. It's faster, uses less memory, and I really like the new address bar features, and the bookmarking. (It has tagging built into the bookmarks now.) I've been using the nightly builds for a couple weeks now, and they're flagged as beta 5... I figured beta 4 had been out for a while already.
As I understand it, they branch the code when they're ready to produce a beta. So there's a set of nightlies still on the trunk, which will eventually become Firefox 3, and there's a set of nightlies that will become beta 4. That way it's possible to refine beta 4 a bit, and still keep working on issues that are targeted at beta 5 and beyond.
So the plain nightly builds are after the branch for beta 4, and working toward beta 5.
Let's see if this works:
. . ._____Beta4 . . ._______Beta5
____/_______________/__________________
. . . ^nightlies start calling themselves beta5pre
Treat the dots as empty space; it's the only way I could get it to not collapse the spaces together. (What's so wrong about allowing the PRE tag?)
Did they? I asked about the nature of the communication, and no one seems to be willing to say whether it was a case of "Oh, we just noticed this affects you" or "By the way, we knew this affected you two weeks ago and we're just telling you now." The former is bad timing, but if that's the case I don't see anything wrong with Mozilla's behavior. On the other hand, if they did sit on that knowledge for however many weeks, then it's irresponsible.
But no one "in the know" has been willing to reveal what, to me, is the critical detail.