Two camps. One camp says, "To hell with Netscape users." The other camp says, "This is an example of poor cross-platform testing, or the designer didn't give a damn about the audience."
You're missing the third camp - Use standards, and screw any browser that can't keepup with many year old standards like HTML 4 camp.
As it stands, M16 can't render standard HTML4 entities that Lynx copes with. Like the HTML 4 quoting entities I used above.
Building an HTML4 renderer is one thing. Building a Web browser that supports Javascript, CSS (1 and 2), XML and DOM (*especially* DOM 1 and 2) is a totally different ball game.
But it does support HTML 4 and CSS *completely*, something which Mozilla still does not do as of M16. If I ran MacOS, I'd take those standards over ECMAscript any day.
I fear that AOL's management has cold feet about the open-sourceness of Mozilla, and has been sticking with NS4 because they are unable to make up their mind about what is worse- the open-source developer community or just letting Microsoft win the browser war once and for all.
I don't see a lot of substance to this idea, although it's been bought up more than once - if AOL were allergic opening the source to the community, it hardly seems likely they would have bought NaviServer and released it under the GPL. AOL have been on the GPL bandwagon for core parts of their infrastructure since before there was a bandwagon.
I think the biggest problem is that they felt they could not just build a *browser* but rather it has to be a "Web browsing desktop environment"
Preach on, brother. Does the world need a browser with an xterm, IRC client, and all the other crap that's in the M16 release? Not as much as it needs a functioning browser!
but Netscrape 4.x completely botched the interpretation of the style information
Amen. Netscape is a festering sore on the HTML world, with a complete inability to render a huge chunk of the HTML 4 standard.
What really ticks me off is that Mozilla is still not much better. Point a browser at this test to see, or try using the HTML 4 standard quoting entities, and watch M!6 go for a loop.
I'm just plain curious as to why the cheating is such a big thing. In a game like half-life, particularly Counter-strike, you can pretty easily tell when someone's cheating, and noone is impressed.
Actually, one of the reasons cheating is such a big thing is that because the best players are as good as bot-enhanced players, if not better, so top players start getting accused of cheating. Hell, not even top players - I know good players who have to endure accusations of cheating because the numbnuts they play can't believe people can be good without cheating.
The other reason is that people who cheat are often sufficinetly socially deficiant that they don't care that they're spoiling it for other people are are held in contempt - they want to win, and if winning means preventing others from enjoying themselves, they aren't particularly concerned.
do not want my local net admin/nazi to control *MY* e-mail
Aww, poor little snookums. Here's a suggestion - run up rackspace.com, order a server, and configure it howyou please. Run your open relays, accept your spam, do whatever you damn well please (assuming you're actually capable of running a system). The only freedom you get on someone else's system is the freedom they decide to give you.
Wow. I wish I was a smoker with Emphysema right about now.
This has to rank as one of the most stupid things I've seen not only on salshdot, but anywhere on the net, and the net is a famously vast repository of stupidity.
Try spending some time with people with emphysema. It's one of the least pleasant ways you can die, and one of the least pleasant ways for friends and family to have to watch someone go.
You are correct that trademarks do expire over time. However, they expire if they are not defended - and the widespread use of Olympic in so many contexts makes it almost certain that, barring special statute protection a la the US, the IOC would lose trademark status if anyone fought them sriously.
Oh to be a multimillionaire happy to drop a few bucks on doing just that...
It's amazing how many uninformed/.ers assume copyright == patent == trademark in any IP dicsussion.
True, but you should be careful lest you perpetuate misinformation yourself.
Except, of course, you will be regarded with opprobrium by such luminaries as George Bush Sr, who referred to a political opponent as a "card carrying member of the ACLU". Anyone would think ACLU members were terrorist subversives...
It boils down to: you only have to release the source to what you choose to distribute. If you choose only to distribute the unmodified pages, that's all the source you need to distribute.
The GPL - and Stallman himself - are quite clear that you aren't forced to relase stuff you hack on for your own use, even if the output (eg, gcc generated binaries, HTML pages sent over HTTP) is sent out to people.
These guys have a TLS wrapper for existing sendmail installations.
So I jumped the gun a bit on BSDLed sendmail having TLS - it will RSN, or you can use a wrapper. On the upside, I was also wrong about zmailer, who apparently have TLS now. Encrypted linux-kernel anyone?
I think you need to read a little more closely. STARTLS extensions are described in an RFC which is implemented by:
Sendmail
Exchange
qmail
PostFix
There are plenty of MTAs that don't have support (e.g. Exim, Zmailer), but the above cover a huge chunk of the installed MTA base. The main challenge to TLS is an better key exchange mechanism, which is the current main impedement.
Regarding Dell being forced to waive patent claims as a result of participation in an industry stantdards forum. I wonder if this could also be applied to Frauenhofer's (sp!) patent on Layer 3 MPEG Audio encoding, or Apple's patents relating to MPEG4 encoding?
Because Microsoft products use a different encoding to iso-8859-1, which is standard in the UNIX world - and, indeed the standard to which most moders specs are written. Windows encoding has several extra characters encoded in undefined areas, which contain not-so-smart quotes (the typeographers' quotes).
The shouldn't be a problem, because when translating to HTML, one should either use non-smart quotes (' and "), or one of the HTML entities (“, ‘, <q>) that deal with this problem. Instead, MS products use &# entities, specifying numbers which are meaningless on non-Windows encoded systems.
Of course, it isn't all Microsoft's fault. While Lynx and IE 4 & 5 both recognise and render the various quoting entities and markup correctly, stupid old Netscape, and worse yet, Mozilla don't, even in their latest incarnations. (Netscape 4.72 not doing so is bad enough, IMO, but mozilla has no excuse, since the extneded entities are part of the HTML 4 standard. So much for claims gecko is a standards-compliant rendering engine).
At any rate, the most correct thing to do would be to use dumb old typewriter style quoting in HTML. But Microsoft products don't do that, sadly.
Getting subselects right requires that (a) the query optimiser in the DB works well and that (b) the person formulating the queries knows how to produce good SQL.
Usually, it's the latter, not the former that falls down. Speaking as a SQL cavemen, I usually hand of complex queries involving suub-sub-queries to our DBAs to do. Why? Well, Oracle sure as heck ain't crap at doing them. In fact, Oracle can do a complex sub query in less time than it takes to fetch a query, process it, submit another query based in te first, etc. The difference is that a well-formed query will be orders of magntiude faster than a poorly formed one.
Overall, performing multiple queries instead of subqueries is almost always slower, and imposes heavier loads on your front end and database boxes than simply sriting a good piece of SQL. The problem is writing good SQL, because it's really easy to write something very slow indeed. And rarely ever the database engine's fault.
This is my concern - overclocking is a great thing to play with if you know what you're doing and don't mind the risks involved. With the hype surrounding overclocking and tools like this, there will be more people who have bought into the "overclocking is more for free at no risk" hype. Some of who will be dissapointed when they do reduce the lifespan of components, start getting intermittent crashes, and so forth.
Yeah. Isn't this all so much better when they were a natural monoploy! Opening the domain market to competition and allowing the invisible hand to guide us has worked wonderfully, hasn't it!
Except instead of claiming he will make the game to end all games Real Soon Now, he explains he plans on having definite release dates of quality games and expects to have to build a reputation.
In other words, a realistic business plan, with out an excess of hype and bullshit. Fancy that!
For that matter, foreign companies operating in the US operate under US law - unless the US has started offering extraterritorial status to businesses.
Slashdot Mirror
My bad, I meant this page.
You'll not only see M16 doesn't render correctly, but that IE 5 does better in some areas.
Well, let's see. There's this one, for starters. Or, for that matter, any piece of text using HTML 4 quoting entities. Like this one. Which Lynx and IE 4+ all render just fine.
You're missing the third camp - Use standards, and screw any browser that can't keepup with many year old standards like HTML 4 camp.
As it stands, M16 can't render standard HTML4 entities that Lynx copes with. Like the HTML 4 quoting entities I used above.
But it does support HTML 4 and CSS *completely*, something which Mozilla still does not do as of M16. If I ran MacOS, I'd take those standards over ECMAscript any day.
I don't see a lot of substance to this idea, although it's been bought up more than once - if AOL were allergic opening the source to the community, it hardly seems likely they would have bought NaviServer and released it under the GPL. AOL have been on the GPL bandwagon for core parts of their infrastructure since before there was a bandwagon.
Preach on, brother. Does the world need a browser with an xterm, IRC client, and all the other crap that's in the M16 release? Not as much as it needs a functioning browser!
Amen. Netscape is a festering sore on the HTML world, with a complete inability to render a huge chunk of the HTML 4 standard.
What really ticks me off is that Mozilla is still not much better. Point a browser at this test to see, or try using the HTML 4 standard quoting entities, and watch M!6 go for a loop.
The standard Netscape acroread and flash plugins work just find under M16 here (that is, they don't crash M16 more than they do Nav).
Other than UI differences, I fail to see any problem.
Actually, one of the reasons cheating is such a big thing is that because the best players are as good as bot-enhanced players, if not better, so top players start getting accused of cheating. Hell, not even top players - I know good players who have to endure accusations of cheating because the numbnuts they play can't believe people can be good without cheating.
The other reason is that people who cheat are often sufficinetly socially deficiant that they don't care that they're spoiling it for other people are are held in contempt - they want to win, and if winning means preventing others from enjoying themselves, they aren't particularly concerned.
Aww, poor little snookums. Here's a suggestion - run up rackspace.com, order a server, and configure it howyou please. Run your open relays, accept your spam, do whatever you damn well please (assuming you're actually capable of running a system). The only freedom you get on someone else's system is the freedom they decide to give you.
Dude, not especiially funny.
This has to rank as one of the most stupid things I've seen not only on salshdot, but anywhere on the net, and the net is a famously vast repository of stupidity.
Try spending some time with people with emphysema. It's one of the least pleasant ways you can die, and one of the least pleasant ways for friends and family to have to watch someone go.
You are correct that trademarks do expire over time. However, they expire if they are not defended - and the widespread use of Olympic in so many contexts makes it almost certain that, barring special statute protection a la the US, the IOC would lose trademark status if anyone fought them sriously.
Oh to be a multimillionaire happy to drop a few bucks on doing just that...
True, but you should be careful lest you perpetuate misinformation yourself.
Except, of course, you will be regarded with opprobrium by such luminaries as George Bush Sr, who referred to a political opponent as a "card carrying member of the ACLU". Anyone would think ACLU members were terrorist subversives...
Yup. Looking back, I can see why people would have parse errors. I wan't very clear.
It boils down to: you only have to release the source to what you choose to distribute. If you choose only to distribute the unmodified pages, that's all the source you need to distribute.
The GPL - and Stallman himself - are quite clear that you aren't forced to relase stuff you hack on for your own use, even if the output (eg, gcc generated binaries, HTML pages sent over HTTP) is sent out to people.
Having gone back and re-read a bunch of information, I can tell you:
So I jumped the gun a bit on BSDLed sendmail having TLS - it will RSN, or you can use a wrapper. On the upside, I was also wrong about zmailer, who apparently have TLS now. Encrypted linux-kernel anyone?
I think you need to read a little more closely. STARTLS extensions are described in an RFC which is implemented by:
There are plenty of MTAs that don't have support (e.g. Exim, Zmailer), but the above cover a huge chunk of the installed MTA base. The main challenge to TLS is an better key exchange mechanism, which is the current main impedement.
Regarding Dell being forced to waive patent claims as a result of participation in an industry stantdards forum. I wonder if this could also be applied to Frauenhofer's (sp!) patent on Layer 3 MPEG Audio encoding, or Apple's patents relating to MPEG4 encoding?
Because Microsoft products use a different encoding to iso-8859-1, which is standard in the UNIX world - and, indeed the standard to which most moders specs are written. Windows encoding has several extra characters encoded in undefined areas, which contain not-so-smart quotes (the typeographers' quotes).
The shouldn't be a problem, because when translating to HTML, one should either use non-smart quotes (' and "), or one of the HTML entities (“, ‘, <q>) that deal with this problem. Instead, MS products use &# entities, specifying numbers which are meaningless on non-Windows encoded systems.
Of course, it isn't all Microsoft's fault. While Lynx and IE 4 & 5 both recognise and render the various quoting entities and markup correctly, stupid old Netscape, and worse yet, Mozilla don't, even in their latest incarnations. (Netscape 4.72 not doing so is bad enough, IMO, but mozilla has no excuse, since the extneded entities are part of the HTML 4 standard. So much for claims gecko is a standards-compliant rendering engine).
At any rate, the most correct thing to do would be to use dumb old typewriter style quoting in HTML. But Microsoft products don't do that, sadly.
Getting subselects right requires that (a) the query optimiser in the DB works well and that (b) the person formulating the queries knows how to produce good SQL.
Usually, it's the latter, not the former that falls down. Speaking as a SQL cavemen, I usually hand of complex queries involving suub-sub-queries to our DBAs to do. Why? Well, Oracle sure as heck ain't crap at doing them. In fact, Oracle can do a complex sub query in less time than it takes to fetch a query, process it, submit another query based in te first, etc. The difference is that a well-formed query will be orders of magntiude faster than a poorly formed one.
Overall, performing multiple queries instead of subqueries is almost always slower, and imposes heavier loads on your front end and database boxes than simply sriting a good piece of SQL. The problem is writing good SQL, because it's really easy to write something very slow indeed. And rarely ever the database engine's fault.
This is my concern - overclocking is a great thing to play with if you know what you're doing and don't mind the risks involved. With the hype surrounding overclocking and tools like this, there will be more people who have bought into the "overclocking is more for free at no risk" hype. Some of who will be dissapointed when they do reduce the lifespan of components, start getting intermittent crashes, and so forth.
Yeah. Isn't this all so much better when they were a natural monoploy! Opening the domain market to competition and allowing the invisible hand to guide us has worked wonderfully, hasn't it!
Except instead of claiming he will make the game to end all games Real Soon Now, he explains he plans on having definite release dates of quality games and expects to have to build a reputation.
In other words, a realistic business plan, with out an excess of hype and bullshit. Fancy that!
For that matter, foreign companies operating in the US operate under US law - unless the US has started offering extraterritorial status to businesses.