Slashdot Mirror
IE7 Leaked
lju writes "IE7 has been leaked according to pcpro. From the article: '...last Friday it was revealed that a build of the new browser - version 5299 - along with numerous screenshots, was available online.' "
← Back to Stories (view on slashdot.org)
IE7 Screenshots
GetOuttaMySpace - The Anti-Social Network
It's been available for some time now legally. I believe you have to be a member of MSDN. I'm using it right now on my work PC. It's like IE 6 with tabs and with some rough edges. The build I'm using doesn't add "http://" for you, which is a bit annoying but might not be a bad security precaution. Tabs pretty much work like FF.
EvilCON - Made Famous by
Its not a big deal, its been available to MSDN members for quite a while. I haven't updated this machine in a month or more and IE on here shows 7.0.5112.0 (Vista Beta 1 5112)
Wow it sure does take long for news to filter through, I would stay clear of it though as I nearly had to reformat as it broke AVG and numerous other things. Even when its released I will stick with firefox.
If look at the security hole listings for IE, there are exactly 3 buffer overflow problems found in the entire history of the browser. Not exactly indicative of insecure coding. Look at Firefox for comparison.
Most of the IE security holes are cross-site scripting BS and things that require stupid users to work properly. And of course when running as non-admin, all the security exploits are irrelavent.
here
The original seems to be slashdotted.
No, that's the price for using a largely un-optimized application. Compared with IE, Firefox is like the 'new kid on the block'. Around for some time now, but not as long as IE. And probably a lot less men-hours of development effort, compared to what must have been poured into IE over the years. Firefox has reached 'feature-complete stage', but it's not heavily optimized or in a 'mature' state, where there's only small improvements left to do.
Does it matter? Not much, Firefox is plenty useful NOW. And what would you rather waste: megabytes of your computer's memory (only till you close the app), or countless hours of your (human) time, cleaning out malware that leaked in? I'll take the 'waste memory' option anytime.That isn't the job of glibc, its the job of the OS. At program termination, the OS knows what memory an app used and simply marks those pages (in RAM and the pagefile) as unused.
glibc is in charge of memory during the running of the program. WHen you call new or malloc, it takes memory from a pool to satisfy the request. If it doesn't have enough, it requests additional pages from the OS. When you call delete or free, it takes that memory and puts it back in the pool. What I'm not sure is if glibc returns memory to the OS if the pool gets big enough. If it doesn't, the program isn't really leaking memory, but memory use will never go down.
I still have more fans than freaks. WTF is wrong with you people?
I think it's actually somewhat "difficult" to install ActiveX anymore. Since their updates in SP2, you have to click on the information bar and explicity state that you want to install this ActiveX (and it gives you a warning about how it can be unsafe).
I mean, I'm the first one to admit that IE has been very crappy in the past... but IE6 SP2 (other than the damn rendering, of course, but there's hope for IE7)does a pretty good job of being secure.
No, a memory leak is when an application allocates memory but doesn't free it when it's done with it. For example, load up a document from disk to memory, and when closing it, forgetting to deallocate the used memory.
All memory the application uses should be cleared by the operating system at the time the program is terminated. Otherwise, we'd be in a deeply sticky situation when some applications abruptly die for whatever reason...
Firefox has not been particularly awful at leaking memory; I've not looked at memory usage but around 1.0 series the thing was pretty slow when left open overnight... I restart Firefox every day and I'm happy with the operations =)
Actually, don't download it from Mozilla, cause it's not-free software. The binaries are under an EULA, not the MPL or the GPL. Non-free artwork, too.
Join the Free Software Foundation
At least on Windows, people have a habit of opening the task manager , look at memory usage and go "OMG !!!!"
Whats really showing in task manager are the WORKING SET, and thats a poor way to tell how much memory an application has currently allocated and are using.
Look at "private bytes" instead, Working set is a soup of memory curently in use or free:d during the whole application life time (until its trimed by the OS).
Working set will continue to grow on a Windows machine until its trimed due to other process needing the memory.
So 77 MB in Working set, big deal, its perfectly normal for such an application to cycle trough that amount of memory.
http://www.intellipool.se/ - Intellipool Network Monitor
I've found the ietab extension for Firefox to work fairly well. Windows Update has a tendency to take focus on occasion, but it's technically possible to do both. I'm curious to see if/how they fix the focus problem in IE7.
warning: This post is likely to contain gobs of dripping sarcasm. Consume at your own risk.
Adblock leaks memory like a sieve and everyone blames the memory leaks on Firefox. The latest nightly supposedly fixes the problem.
What a fool believes, he sees, no wise man has the power to reason away.
Well, I suppose you are mostly correct.
BUT, that being said, sometimes leaks aren't from within the company. Sometimes they are from partners, reviewers, or beta testers who've been given early access to the product, but who have no legal right to give it to anyone else. . . but do anyhow.
Heap defragmentation in languages with pointers kind of sucks, though. The classic Mac OS memory manager did this was *really* aimed at not wasting any memory -- it could move chunks of memory around and even free chunks of memory that you had marked as unimportant. It turned out to be really difficult to program bug-free software for something like this.
Any program relying on (nontrivial) preemptive multithreading will be buggy.