> I don't understand why Mozilla has ANY form of > disk caching built in in the first plac
Because an external cache does not provide enough control (eg the ability to fail to load a file and throw an exception if it's no longer in cache, which is critical to prevent repeats of POST requests).
The svg people need to land this mythical branch of theirs that fixes all sorts of stuff on the trunk first... And the licensing issues with libart need to still be resolved (assuming this branch uses libart).
Hmm.. by all the metrics I have here, 1.5 is smaller than 1.3 (by close to a meg of binary size) and faster (about 20% faster pageload, actually, see the pageload tests on the SeaMonkey tinderbox).
You know, this comment is an insult to all the work that has been happening over the last 6 months. Not all that much has been "free floating" as you put it; many of the core modules were either already owned by non-Netscape people or are still owned by people working on Mozilla (though no longer as Netscape employees).
I suggest taking a look at the actual CVS checkin logs next time before making statements like that.
Oh, one more thing. Konq is nowhere close to rendering "real-world" HTML like Mozilla did two years ago. Safari is closer, and maybe some of that will migrate back to Konq eventually....
> Well, I'd say that's a pretty big contribution. > Proportional to the number of people employed, > in fact.
Not necessarily. Paying people is just a way of keeping score. In the end, economic development depends on producing goods and services that consumers actually want. That is, you could "employ" everyone at digging holes and filling them in, and "pay" them, but if that results in less goods and services then the pay they get does not matter -- quality of life will suffer.
So in fact, advertising has worth only insofar as it enables consumers to get the goods and services they want.
> If you're like me, and you highlight the text you > want to replace with what is in the clipboard
That works fine. It's if you highlight the text and want to replace it with what's in PRIMARY that you get screwed. CLIPBOARD (where text goes when you highlight it and then hit Ctrl-C or use the "Copy" menu option in any sane modern Linux app) behaves just like the Windows clipboard (pasting happens with Ctrl-V or "Paste", of course). Note that old KDE versions are _not_ sane; KDE3 has it fixed. PRIMARY is more like drag-and-drop but you're allowed to let go of the mouse button and rearrange some windows before you have to drop.
> They're using 2 adjacent channels - which leaves 9 > for others to use
Actually, no. They're using 5 and 6, which overlap with 2, 3, 7, 8 if the device is operating perfectly (the channel bands just overlap by design). If the device actually has out-of-band leakage as the article claims, the could easily be wiping out all 11 channels.
> zionist racists drove a large number of > palestinians from their homes in 1948... by making said Palestinians religious leaders loudly incite them to leave "to clear the way for the victorious Arab armies"? Impressive feat, I would say.
"They" being the slashdot editors? If you ask any actual Mozilla developer, they will tell you that the switch will happen when Firebird and Thunderbird are ready. I very much wonder where Timothy got his "this is supposed to be the last one" BS.
"contrib" means they are contributed by people OTHER than mozilla.org. So they are not existing because no one has built them yet. Are you volunteering?
> I doubt Mozilla, Opera, Konq, etc are fully > standards-compliant either.
Of course not. For example, Mozilla has known bugs with margin-collapsing on blocks nested inside of floats.
And Opera has known issues with esoteric parts of CSS2 as well.
IE/Windows has a completely broken implementation of the "height" property in CSS. So when I set "height: 10px" on a block, IE makes it whatever arbitrary height it wants to instead of 10px.
So while all have their bugs, the point is that the IE bugs in CSS are much easier easier to trigger and much more likely to completely screw up an attempt to use the technology.
It's like comparing a Model T with a Ferrari and saying "well, neither can hit 600mph, so both are slow." The premise is correct, but the conclusion is not...
For Mozilla, let's not forget that MSVC++ (Windows Mozilla) produces code that's about half the size and far faster than the code produced from the same source by g++ (Linux Mozilla).
In fact, going from g++ 2.9 t g++ 3.2 for Mozilla meant a 10% across-the-board speedup in all operations. Testing with current CVS g++ shows further speedups, bringing Linux Mozilla much closer to Mozilla on Windows.
You're talking about the PRIMARY buffer. CLIPBOARD supports #1 from your list, and if you run 'xclipboard' supports #2.
PRIMARY is basically like drag and drop except you drop explicitly, not when you accidentally let up the mouse button.
Oh, both CLIPBOARD and PRIMARY support tagging data with the type. It's just that most apps are dumb and don't do it. But Mozilla does. So does OpenOffice.
> plus the death toll of the Second World War > (maybe another 8 million
Sorry, but that number is wildly off. The death tolls for WWII were, approximately:
USSR -- 20 million (more like 21, really, but
there is some overlap with the 6 million
figure you already counted) Germany -- 7 Million Poland -- 3 million (6, but the other 3 million
were Jews whom you've already counted) UK -- 388 thousand
And just military deaths(to ensure no overlap with the 6 million number quoted):
This is 32 million already. I'm not counting Japan or China, of course, nor the 500 thousand Greek deaths (because I don't know what the military/civilian breakdown was), nor the 1.4 million Yugoslav civilian deaths (again, because I'm not sure how big the overlap with the 6 million + 2 million is).
As for sweeping in, that would have been quite impossible if the Red Army had actually marched against Europe to the Channel. It was hard enough as it was; doing it against 7-10 times the troops (the Germans had something like 15 divisions in France on D-day; the Red Army was over 100 divisions on the same day; the divisions involved are of comparable strength) would have been suicidal.
It's 50 million _phone_numbers_, not 50 million households. Note that there is one landline for every two people in this country, and that there are more cell phones than there are land lines. Therefore, the only logical conclusion is that many households have multiple landlines (not to mention a cell phone or three).
Slashdot Mirror
> I don't understand why Mozilla has ANY form of
> disk caching built in in the first plac
Because an external cache does not provide enough control (eg the ability to fail to load a file and throw an exception if it's no longer in cache, which is critical to prevent repeats of POST requests).
> margins around HR tags don't work properly now.
Actually, in 1.5 was changed to work correctly; before that it was horribly broken. If there is a problem, make sure you file a bug. Have you?
It's pretty simple to understand... Just look at the checkin list for the relevant timeframe (http://bonsai.mozilla.org/cvsqueryform.cgi).
;)
Lots of under-the-hood stuff that you may not see, but 1.6 is about 10% faster than 1.5 at rendering web pages.
The svg people need to land this mythical branch of theirs that fixes all sorts of stuff on the trunk first... And the licensing issues with libart need to still be resolved (assuming this branch uses libart).
3 grand divided by $50 is 60 hours. So if it would take more than about two weeks of work to do it, it's just not worth the money...
Hmm.. by all the metrics I have here, 1.5 is smaller than 1.3 (by close to a meg of binary size) and faster (about 20% faster pageload, actually, see the pageload tests on the SeaMonkey tinderbox).
I'm surprised you're seeing the opposite...
You know, this comment is an insult to all the work that has been happening over the last 6 months. Not all that much has been "free floating" as you put it; many of the core modules were either already owned by non-Netscape people or are still owned by people working on Mozilla (though no longer as Netscape employees).
I suggest taking a look at the actual CVS checkin logs next time before making statements like that.
Oh, one more thing. Konq is nowhere close to rendering "real-world" HTML like Mozilla did two years ago. Safari is closer, and maybe some of that will migrate back to Konq eventually....
> But I have to wonder whether Mozilla requires
> the huge programming push that it has needed two
> years ago.
If you ever want it to support things like CSS2.1 (with its major change of how style computation and inheritance work), then yes.
If you ever want it to support CSS3 Selectors, then yes.
If you ever want to be able to read layout code without losing your sanity, yes.
If you want to have multiple apps using the same shared code (Firebird/Thunderbird/whatever), then absolutely.
The list goes on and on.
> Well, I'd say that's a pretty big contribution.
> Proportional to the number of people employed,
> in fact.
Not necessarily. Paying people is just a way of keeping score. In the end, economic development depends on producing goods and services that consumers actually want. That is, you could "employ" everyone at digging holes and filling them in, and "pay" them, but if that results in less goods and services then the pay they get does not matter -- quality of life will suffer.
So in fact, advertising has worth only insofar as it enables consumers to get the goods and services they want.
> If you're like me, and you highlight the text you
> want to replace with what is in the clipboard
That works fine. It's if you highlight the text and want to replace it with what's in PRIMARY that you get screwed. CLIPBOARD (where text goes when you highlight it and then hit Ctrl-C or use the "Copy" menu option in any sane modern Linux app) behaves just like the Windows clipboard (pasting happens with Ctrl-V or "Paste", of course). Note that old KDE versions are _not_ sane; KDE3 has it fixed. PRIMARY is more like drag-and-drop but you're allowed to let go of the mouse button and rearrange some windows before you have to drop.
> They're using 2 adjacent channels - which leaves 9
> for others to use
Actually, no. They're using 5 and 6, which overlap with 2, 3, 7, 8 if the device is operating perfectly (the channel bands just overlap by design). If the device actually has out-of-band leakage as the article claims, the could easily be wiping out all 11 channels.
> zionist racists drove a large number of ... by making said Palestinians religious leaders loudly incite them to leave "to clear the way for the victorious Arab armies"? Impressive feat, I would say.
> palestinians from their homes in 1948
> Finally, can you guarentee that it'll run on a
;)
> decent number of browsers?
Up to this point the answer was "yes".
Invading neighbors is hardly domestic policy... That's foreign policy.
Developer previews are at http://webperso.easyconnect.fr/danielglazman/compo ser/composer++.html if you care....
> the file extension concept is wrong, but it's how
> the world works, damnit.
No. That's how the Windows world works. There is more to the world than Windows.
> They've been saying that for a while,
"They" being the slashdot editors? If you ask any actual Mozilla developer, they will tell you that the switch will happen when Firebird and Thunderbird are ready. I very much wonder where Timothy got his "this is supposed to be the last one" BS.
"contrib" means they are contributed by people OTHER than mozilla.org. So they are not existing because no one has built them yet. Are you volunteering?
> I doubt Mozilla, Opera, Konq, etc are fully
> standards-compliant either.
Of course not. For example, Mozilla has known bugs with margin-collapsing on blocks nested inside of floats.
And Opera has known issues with esoteric parts of CSS2 as well.
IE/Windows has a completely broken implementation of the "height" property in CSS. So when I set "height: 10px" on a block, IE makes it whatever arbitrary height it wants to instead of 10px.
So while all have their bugs, the point is that the IE bugs in CSS are much easier easier to trigger and much more likely to completely screw up an attempt to use the technology.
It's like comparing a Model T with a Ferrari and saying "well, neither can hit 600mph, so both are slow." The premise is correct, but the conclusion is not...
For Mozilla, let's not forget that MSVC++ (Windows Mozilla) produces code that's about half the size and far faster than the code produced from the same source by g++ (Linux Mozilla).
In fact, going from g++ 2.9 t g++ 3.2 for Mozilla meant a 10% across-the-board speedup in all operations. Testing with current CVS g++ shows further speedups, bringing Linux Mozilla much closer to Mozilla on Windows.
You're talking about the PRIMARY buffer. CLIPBOARD supports #1 from your list, and if you run 'xclipboard' supports #2.
PRIMARY is basically like drag and drop except you drop explicitly, not when you accidentally let up the mouse button.
Oh, both CLIPBOARD and PRIMARY support tagging data with the type. It's just that most apps are dumb and don't do it. But Mozilla does. So does OpenOffice.
Don't bash X for bugs in xterm and company.
> plus the death toll of the Second World War
> (maybe another 8 million
Sorry, but that number is wildly off. The death tolls for WWII were, approximately:
USSR -- 20 million (more like 21, really, but
there is some overlap with the 6 million
figure you already counted)
Germany -- 7 Million
Poland -- 3 million (6, but the other 3 million
were Jews whom you've already counted)
UK -- 388 thousand
And just military deaths(to ensure no overlap with the 6 million number quoted):
Yugoslavia: 300 thousand
Romania: 500 thousand
France: 350 thousand
Austria: 350 thousand
Italy: 300 thousand
This is 32 million already. I'm not counting Japan or China, of course, nor the 500 thousand Greek deaths (because I don't know what the military/civilian breakdown was), nor the 1.4 million Yugoslav civilian deaths (again, because I'm not sure how big the overlap with the 6 million + 2 million is).
As for sweeping in, that would have been quite impossible if the Red Army had actually marched against Europe to the Channel. It was hard enough as it was; doing it against 7-10 times the troops (the Germans had something like 15 divisions in France on D-day; the Red Army was over 100 divisions on the same day; the divisions involved are of comparable strength) would have been suicidal.
> That number is most likely made up
It's 50 million _phone_numbers_, not 50 million households. Note that there is one landline for every two people in this country, and that there are more cell phones than there are land lines. Therefore, the only logical conclusion is that many households have multiple landlines (not to mention a cell phone or three).
Thunderbird+Firebird will use more memory than the appsuite up until they are sharing a single GRE.