Because they don't want to use something they didn't write if they can help it, because they don't want to give it any attention if it hasn't already gotten lots and they need to counter it, and because if they do that, they can't run off and do "intellectual property" lawsuits later on.
Does the Print to PDF feature of OS X usually result in documents having proper internal bookmarking and hyperlinks?
Not normally. Safari supports persisting some links over to PDF (if you were to PDF-ify apple.com, the "bottom four" blobs don't get links, but everything else does; I suspect this is a bug). Printing a TextEdit document with a link in it doesn't retain the link or the link style. And bookmarks/TOC generation isn't being done in any case.
Serious PDF generation will require adjustment, but take Save as PDF for what it is - neat use of the printing system and a good pack-in that suffices in most cases.
Reasonable. Especially that the "Web 2.0" bubble will be smaller than last time. Most stuff this time around is actually useful - not everything is useful *enough* (file under: potato-peeler-with-ajax-well-that's-useful), but it's still better.
No. The Pirate Bay (and every other BitTorrent tracker) distributes.torrent files containing just hashes and checksums and other metadata. They don't distribute whatever copyrighted material or other material you end up with if you run the torrent file through a BitTorrent client - it never lands on their servers at all.
Nothing truly useful about the web went down in flames with the dot-com bust, it was all based on the fact that people wanted to shoe-horn the rules of their existing media into web content. Or they thought that their crackpot business plans would finally gain acceptance and work simply because it now included running a web site, and web sites were hot, new and cool.
There's some deep zen shit behind the 1) Collect underpants; 2) ???; 3) Profit!!! parallel from South Park. Sometimes what's new and dazzling makes people go completely blind with regards to what's a rational idea or not.
If Dvorak's point that this cycle will repeat itself over and over againhas some validity, it's that the CD-ROM multimedia iteration was reasonably similar in structure. First an idea, then it growing, then it being adopted by lots of people, and then finally it being abandoned because their implementation of the idea just wasn't that good. It just didn't end that bad since it had mostly fizzled out by 1997, before computers rose to real ubiquity with well-entrenched, high-bandwidth Internet connections. And the precursors and survivors of the multimedia boom are the adventure games (especially the Myst/Riven end) and the Flash animations, just as the useful web sites and applications are what was before and came out of the bubble.
By CD-ROM, I think he means the "interactive", "multimedia" "games"/apps/"experiences" of the mid-90s, hailed by certain powers-that-be as the things that would get us to run out and buy CD-ROM drives in droves. The full screen genre preludes to Flash (in fact, a great deal were made with Shockwave) that never were useful.
I can't say whether it was a bubble or not, but I know I'm glad no one, except for perhaps some cell phone/PDA software autorun presentations, still uphold that particular art form. With this in mind, he's not all full of crap.
As for the industry going towards more or fewer bubbles, I have no idea. On one hand, the industry is stabilizing and maturing (if by "maturing" you mean "big companies can upsell other big companies on ridiculous systems no one needs") so more and more jobs are guaranteed. On the other hand, there's still technical evolution and still wrong-headed venture capital, so there will always be costly software projects that fail (and software projects *do fail*, more than half of them, regularly). On the gripping hand, there are people who know way more about this industry than I do and even they can't say which way it'll go.
By multiple standards, I don't mean GSM vs UMTS, I mean things outside of a single standards roadmap. The Swedish mobile market is healthy because of two big local phone companies (Sony Ericsson, half Swedish and Nokia, Finnish; there's also Swedish minor act Neonode) that both on average make reasonably good phones; because of wide availability to almost any phone brand whose phones will work with the Swedish networks; because of a single standards track (GSM and UMTS) and easy portability; because of reasonable nation-wide coverage for any carrier (not just Carrier X works in cities A, B and C); because of the way you can go out and *just buy a phone* without the carrier nonsense in any non-carrier mobile phone store and then go home and insert your SIM card and it will *just work*; because you don't need to pay for receiving SMS messages; because you can send and receive as many SMS messages as you want and pay as you go along; because pre-paid and proper monthly plans are both offered and treated seriously by every carrier.
Everyone in Sweden - absent small children, pets and people that have actively chosen to not use a mobile phone for various reasons - have a mobile phone. It's not mandatory, yet everyone have them. Everyone. Just like regular phone connections. This is the sign of a well-functioning free market. I don't expect the US carriers to get this. They have too much to lose, they'll have to change in dramatic ways to stop shafting their consumers and help spreading the technology. Don't get me wrong, Swedish carriers are also getting around to putting bullshit restrictions on "operator phones" saying which apps you can install; a practice introduced in Sweden by multi-national carriers Vodafone and 3.
That's the point. By cancelling the crap games, both the new vein of reaching out to non-gamers and the "old"/"pre-existing" vein of serving existing gamers could be fueled by fresh blood. Mario's already in way too many titles. The last Mario-related title that I played that was anywhere near New Super Mario Bros was either Super Mario Land 2 on Game Boy or the original Yoshi's Island on SNES - not a sign of being rehashed for 20 years. The character, yes; the exact game, by way of retro re-releases, yes; the fresh new titles in the same vein, hell no. If Mario was overexposed for being in a series of well-developed platform games carrying the SMB(1)-SMW rule to its logical conclusion and occasionally starring in the odd Mario Kart and Super Smash Bros game, I'd have no problem with it. But that's not the case.
Even if I'd like to see two good new original Mario titles per year instead of 10 new completely irrelevant crap titles starring Mario, I'd of course also be happy to see new non-Mario titles in the same spirit. Things like Pikmin is existence proof that there can be new successful - and I hate myself for saying this word - "franchises". Nintendo is doing interesting stuff with new games and branches, they're just focusing most of it on their new audience. They'll need to realize that their existing audience wants some of that as well, and hopefully they have, and just don't have something to show us yet.
Yes, and Retro Studios makes the Metroid Prime series, an unnamed studio makes Brawl and HAL made Melee and the original SSB. It doesn't matter that internal Nintendo studios don't make the games (Hudson is a Konami subsidiary, HAL is a second-party studio and Retro Studios was once a second-party studio but are now a first-party studio by being wholly-owned). My point was that whoever is developing Nintendo's crap titles (and I'm not including Metroid or Brawl here just by virtue of mentioning them for comparison) could be making other games instead.
Nintendo's mostly on the right track. What kind of games are they pumping out these days? Games like Wii Sports, Wii Fit and Brain Age, designed to attract *anyone*. Games like Metroid Prime 3: Corruption, Zelda: Phantom Hourglass and Super Smash Bros Brawl (and to a lesser extent perhaps Super Mario Galaxy) designed to appeal to folks that are already gamers. And games like Mario Party 8.
The only error I find in Nintendo's plans is that they should decapitate their pure money-makers (the "Mario Party 8" leg) and redistribute this brainpower evenly among the other two remaining legs. No one gets religion about Mario Party 8 and the "Core" gamers would get very happy to get a bigger piece of the action. I've never seen Reggie brag about how proud he is over Mario Party 8 - why not just drop it?
By being a listing/index/catalog of all books with references to where to get them instead of being a site dedicated to reproducing the source material of stuff in the public domain, perhaps?
Apple may be a for-profit company but at this level (below iPods and DVDs and DRM), Apple has been opening up. Remember the discussion about the TPM module that vast chunks of Slashdot were convinced was part of a future evil plan to lock everything up with a software update? They only ever shipped it in a handful of models. Remember the discussion about the Intel Darwin sources missing XNU, the kernel component? They opened that back up. Remember the fuss about KHTML having difficulties making anything out of the WebKit contributions because they were often Mac OS X-centric? From what I understand, Apple has recently turned WebKit more toolkit-neutral. They have started shipping several of their components (like Bonjour/zeroconf and launchd) under the Apache license rather than their own APSL, and are making available their CalDAV server under the Apache license as well.
I realize that it's healthy to remain skeptic, or at least not assume that the company will forever stay open to the extent that it already is open. But give Apple some credit where due. I don't think it's very likely that Apple will lock down CUPS in the future, since they have their own printing architecture that they're pushing primarily. Apple has been more BSD than FSF/GPL, and it does seem likely that they just want to avoid GPL 3, which is more, uh, religious than is GPL 2. And if Apple do end up locking down CUPS - why not start by forking their last release and *then* start talking about how Apple is chair-throwingly evil?;)
(Additionally, as a result of these kinds of acquisitions, Apple has a bunch of people who are all about open source onboard, like Jordan Hubbard and Dave Hyatt. I think any theoretical upcoming closedowns would meet strong resistance within the organization.)
Right, but no one is expecting Apple to open up its cellular network communication API. Just the rest of the stuff, so we can build real apps for the phone itself.
I'm hoping Apple will open up the iPhone to third parties sooner or later, but being able to run existing Java midlets is low on their list, and probably for good reason. That's a 160dpi display. You want to aim for a 9-pixels-high checkbox using your finger on that sort of screen? It's not like Java's well-positioned for resolution independence either.
Java apps would stick out like a sore thumb, would work entirely differently and wouldn't actually be able to take advantage of the features people are buying iPhones for. My guess is that they're working on a native SDK for later on.
The bundling is awful. The only technically required bundling is QuickTime with iTunes, since iTunes depends on QuickTime. At least now it's fairly public what you get - I remember when you had to hunt around for the "QuickTime only" link. Those kinds of tricks aren't just Microsoft-bad, they're Real-bad.;)
Safari for Windows is a blessing for web developers. Up until early June, three of the four most used browsers were available on Windows (IE, Firefox and Opera), but the third most used (Safari) wasn't. The more browsers are available and popular on Windows, the more people will finally understand that "standards-compatible" doesn't mean "works like IE". Building for standards, checking in each browser and then doing horrible hacks you wish you didn't have to do to make it work in IE is a better way than the old and broken way: building for IE, checking in the other browsers and sighing about the other browsers not being standards compliant. (I wish I had a nickel for every time someone gave me that crap.)
I'm a Mac OS X user. Firefox is great on Windows, but on Mac OS X it's sticking out like a sore thumb, and it's much slower than the other alternatives. My primary browser is OmniWeb, which uses a variant of WebKit and offers and pioneered some interesting functionality like site-specific settings, a vertical list of tabs with thumbnails, workspaces where sets of windows and tabs are persisted. Even if OmniWeb is an odd choice - it costs money! my god! I must be a complete moron! - almost no one I know use Safari because of the wide ecosystem of good browsers, like Firefox, Camino (a Cocoa app embedding Gecko), Shiira (an alternative WebKit browser) and OmniWeb. Safari has never been considered really good against this background, but it's starting to turn competent in 3.0. Inline Find, draggable (and de/re-attachable) tabs and something as simple as asking when you quit and have tabs open and finally, only took them four damn years, AppleScript tab support means Apple has done a lot of basic tackling and is really listening to people beyond gluing on RSS support and working on WebKit alone. I had almost given up hope.
Because the US mobile phone market is completely wack. Coverage sucks and is highly varying and there are competing standards. Not competing carriers, competing *standards*. GSM and CDMA.
They could sell it without SIM locking, and by Bob I hope they do it in Europe, but you'd only ever reach half of the market. My guess is that it's least suspicious to just do it the way things are usually done in the US - just tie it to a carrier. Nevermind the free market and stuff like that.;)
It probably *is* available only to carriers, but wanting your own phone's location is a legitimate request, and I don't see a huge problem in the phone being able to request the estimated position. A service that allowed you to get your current position was launched in Sweden several years ago and I didn't see any mention of infrastructure upgrades on the carrier's part.
It is possible to reasonably pinpoint your position using GSM localization. The accuracy sucks (one order of magnitude worse than GPS at best), but it's more likely to be close enough in areas where you'd want to use maps - unless you are lost in the woods, that is.
The DOM - or something like it - has to be constructed by the browser even when no scripting is involved whatsoever. Specifying it does not do any harm.
I have to say that I don't pretend to know all the background of the noscript case in particular, but if you believe that the explanation given to the special noscript XML treatment is bullshit, why would it even be in there to begin with? There's no reason effort should be spent shooting down XML in an asinine way without needing to adjust for either backwards compatibility, current established browser behavior or what the XML standard says - don't you think you're missing something? And if you're 100% sure you've got the right answer in this case, why not point out exactly what's wrong to the WHATWG folks themselves?
I have a tough time thinking that the canvas tag or the XMLHTTPRequest object would be invented in an HTML specification - yes. This is paving of cow paths - specifying things that are already widely implemented. The W3C, in their infinite wisdom, are *also* specifying things like it. There's already an XMLHTTPRequest specification.
All I'm advocating is that those writing XML write valid XML. Again, nowhere have I suggested parsing HTML using xmllint, we were discussing XHTML and I proceeded to answer your question in that context after explicitly noting your change of topic.
I did not catch this nuance. I apologize.
However, it did sound like you were shooting down anything that was *not* XHTML as valid backing for a new HTML standard. There's a W3C HTML working group now, and on their mailing list are two kinds of requests: people wanting their own tag or people wanting to adopt the Web Apps draft standard as a starting point. Am I amiss in thinking a third alternative - another good way for HTML forward - will not arise for the forseeable future? Why not provide constructive critisism for the HTML5 branch going forward instead of saying that they're all nuts?
I think I should say that I do not want XHTML dead - although I believe that XML's parsing sematics are holding back its tenability, because not everyone produce valid code all the timewithout XML libraries, and very few text editors have XML libraries. Like you say, if you can produce good XML, then XHTML may be better for you than HTML. I also do not think the Web Apps spec has got everything correct. Like you, I do not want the web to just turn into an application API. However, I do want the parts that are already there - like forms - to stop sucking. At over 120 printed pages, it's hard to agree with everything, but according to the browser guys themselves, this level of detail is what they want.
Our stances on whether XML's strictness is good or bad *for the web* is obviously different. I do think it's useful to be strict when handling strictly-formatted documents, as most XML documents are. HTML documents, beyond html, head and body tags, don't really have a strict structure to follow, just rules about how the tags will be nested, and therefore I think it's a really different situation. I will point to Mark Pilgrim's "Thought experiment" to explain my side, but I think it's a waste of our time to continue to debate this if your stance is set in stone. (As an aside, that article ended up converting John Gruber - the creator of Markdown, which you mentioned - to Pilgrim's side.)
There's also a philosophical side: *writing* HTML this far has been about firing up Notepad and a web browser. (Uploading and hosting is a different matter, but I'm talking about writing.) Requiring xmllint to write HTML isn't the web I want.
XHTML 1 is HTML 4 reformulated in XML syntax. By definition, it's backwards compatible. And no, I mean that HTML5's parsing is backwards compatible. Run a page, any HTML page, through an HTML5 tokenizer. HTML will come out right unless you are relying on browser-specific bugs, and there are allowances for XHTML void-tags with the / embedded. Google ran this tokenizer on a billion web pages to scrape them for attributes and tags. It works.
Regarding the XML 'noscript' point, the spec offers up this note: "All these contortions are required because, for historical reasons, the noscript element causes the HTML parser to act differently based on whether scripting is enabled or not. The element is not allowed in XML, because in XML the parser is not affected by such state, and thus the element would not have the desired effect." Does this mean you actually would like to be able to write invalid XML?
Whether you like it or not, DOM manipulation is being used in tandem with HTML, and it needs to be defined somewhere. The spec is actually called "Web Applications", and it describes HTML5 and, additionally, the DOM that should be used to manipulate it and that's used to store it internally. It's also important to remember that the spec is for browser implementors as well as authors! You need not be proficient in the parsing or tokenization or DOM manipulation of HTML5 to write in it, just as you don't need to be proficient in the meaning of uppercase 'SHOULD' or 'MAY' as defined in RFC2119 to be able to use XHTML. HTML5 itself is still a markup language, just as ever, but it's not just markup that needs to be defined in the spec.
At the end of the day, the browser vendors that care about standards sat down and said "this is getting messy, what do we need to codify?" and put it in the same spec with HTML5. CSS2 and CSS3 are well-defined. JavaScript is well-defined as ECMAScript. The font tag is deprecated and irrelevant and doesn't help your argument one whit.
Slashdot Mirror
Because they don't want to use something they didn't write if they can help it, because they don't want to give it any attention if it hasn't already gotten lots and they need to counter it, and because if they do that, they can't run off and do "intellectual property" lawsuits later on.
Not normally. Safari supports persisting some links over to PDF (if you were to PDF-ify apple.com, the "bottom four" blobs don't get links, but everything else does; I suspect this is a bug). Printing a TextEdit document with a link in it doesn't retain the link or the link style. And bookmarks/TOC generation isn't being done in any case.
Serious PDF generation will require adjustment, but take Save as PDF for what it is - neat use of the printing system and a good pack-in that suffices in most cases.
Reasonable. Especially that the "Web 2.0" bubble will be smaller than last time. Most stuff this time around is actually useful - not everything is useful *enough* (file under: potato-peeler-with-ajax-well-that's-useful), but it's still better.
No. The Pirate Bay (and every other BitTorrent tracker) distributes .torrent files containing just hashes and checksums and other metadata. They don't distribute whatever copyrighted material or other material you end up with if you run the torrent file through a BitTorrent client - it never lands on their servers at all.
Nothing truly useful about the web went down in flames with the dot-com bust, it was all based on the fact that people wanted to shoe-horn the rules of their existing media into web content. Or they thought that their crackpot business plans would finally gain acceptance and work simply because it now included running a web site, and web sites were hot, new and cool. There's some deep zen shit behind the 1) Collect underpants; 2) ???; 3) Profit!!! parallel from South Park. Sometimes what's new and dazzling makes people go completely blind with regards to what's a rational idea or not. If Dvorak's point that this cycle will repeat itself over and over againhas some validity, it's that the CD-ROM multimedia iteration was reasonably similar in structure. First an idea, then it growing, then it being adopted by lots of people, and then finally it being abandoned because their implementation of the idea just wasn't that good. It just didn't end that bad since it had mostly fizzled out by 1997, before computers rose to real ubiquity with well-entrenched, high-bandwidth Internet connections. And the precursors and survivors of the multimedia boom are the adventure games (especially the Myst/Riven end) and the Flash animations, just as the useful web sites and applications are what was before and came out of the bubble.
By CD-ROM, I think he means the "interactive", "multimedia" "games"/apps/"experiences" of the mid-90s, hailed by certain powers-that-be as the things that would get us to run out and buy CD-ROM drives in droves. The full screen genre preludes to Flash (in fact, a great deal were made with Shockwave) that never were useful.
I can't say whether it was a bubble or not, but I know I'm glad no one, except for perhaps some cell phone/PDA software autorun presentations, still uphold that particular art form. With this in mind, he's not all full of crap.
As for the industry going towards more or fewer bubbles, I have no idea. On one hand, the industry is stabilizing and maturing (if by "maturing" you mean "big companies can upsell other big companies on ridiculous systems no one needs") so more and more jobs are guaranteed. On the other hand, there's still technical evolution and still wrong-headed venture capital, so there will always be costly software projects that fail (and software projects *do fail*, more than half of them, regularly). On the gripping hand, there are people who know way more about this industry than I do and even they can't say which way it'll go.
Here's why: Multiple standards. Poor coverage. Tight phone-operator binding.
By multiple standards, I don't mean GSM vs UMTS, I mean things outside of a single standards roadmap. The Swedish mobile market is healthy because of two big local phone companies (Sony Ericsson, half Swedish and Nokia, Finnish; there's also Swedish minor act Neonode) that both on average make reasonably good phones; because of wide availability to almost any phone brand whose phones will work with the Swedish networks; because of a single standards track (GSM and UMTS) and easy portability; because of reasonable nation-wide coverage for any carrier (not just Carrier X works in cities A, B and C); because of the way you can go out and *just buy a phone* without the carrier nonsense in any non-carrier mobile phone store and then go home and insert your SIM card and it will *just work*; because you don't need to pay for receiving SMS messages; because you can send and receive as many SMS messages as you want and pay as you go along; because pre-paid and proper monthly plans are both offered and treated seriously by every carrier.
Everyone in Sweden - absent small children, pets and people that have actively chosen to not use a mobile phone for various reasons - have a mobile phone. It's not mandatory, yet everyone have them. Everyone. Just like regular phone connections. This is the sign of a well-functioning free market. I don't expect the US carriers to get this. They have too much to lose, they'll have to change in dramatic ways to stop shafting their consumers and help spreading the technology. Don't get me wrong, Swedish carriers are also getting around to putting bullshit restrictions on "operator phones" saying which apps you can install; a practice introduced in Sweden by multi-national carriers Vodafone and 3.
That's the point. By cancelling the crap games, both the new vein of reaching out to non-gamers and the "old"/"pre-existing" vein of serving existing gamers could be fueled by fresh blood. Mario's already in way too many titles. The last Mario-related title that I played that was anywhere near New Super Mario Bros was either Super Mario Land 2 on Game Boy or the original Yoshi's Island on SNES - not a sign of being rehashed for 20 years. The character, yes; the exact game, by way of retro re-releases, yes; the fresh new titles in the same vein, hell no. If Mario was overexposed for being in a series of well-developed platform games carrying the SMB(1)-SMW rule to its logical conclusion and occasionally starring in the odd Mario Kart and Super Smash Bros game, I'd have no problem with it. But that's not the case.
Even if I'd like to see two good new original Mario titles per year instead of 10 new completely irrelevant crap titles starring Mario, I'd of course also be happy to see new non-Mario titles in the same spirit. Things like Pikmin is existence proof that there can be new successful - and I hate myself for saying this word - "franchises". Nintendo is doing interesting stuff with new games and branches, they're just focusing most of it on their new audience. They'll need to realize that their existing audience wants some of that as well, and hopefully they have, and just don't have something to show us yet.
Yes, and Retro Studios makes the Metroid Prime series, an unnamed studio makes Brawl and HAL made Melee and the original SSB. It doesn't matter that internal Nintendo studios don't make the games (Hudson is a Konami subsidiary, HAL is a second-party studio and Retro Studios was once a second-party studio but are now a first-party studio by being wholly-owned). My point was that whoever is developing Nintendo's crap titles (and I'm not including Metroid or Brawl here just by virtue of mentioning them for comparison) could be making other games instead.
Nintendo's mostly on the right track. What kind of games are they pumping out these days? Games like Wii Sports, Wii Fit and Brain Age, designed to attract *anyone*. Games like Metroid Prime 3: Corruption, Zelda: Phantom Hourglass and Super Smash Bros Brawl (and to a lesser extent perhaps Super Mario Galaxy) designed to appeal to folks that are already gamers. And games like Mario Party 8.
The only error I find in Nintendo's plans is that they should decapitate their pure money-makers (the "Mario Party 8" leg) and redistribute this brainpower evenly among the other two remaining legs. No one gets religion about Mario Party 8 and the "Core" gamers would get very happy to get a bigger piece of the action. I've never seen Reggie brag about how proud he is over Mario Party 8 - why not just drop it?
By being a listing/index/catalog of all books with references to where to get them instead of being a site dedicated to reproducing the source material of stuff in the public domain, perhaps?
Which is why open standards would be of particular importance. I mean, right?
Apple may be a for-profit company but at this level (below iPods and DVDs and DRM), Apple has been opening up. Remember the discussion about the TPM module that vast chunks of Slashdot were convinced was part of a future evil plan to lock everything up with a software update? They only ever shipped it in a handful of models. Remember the discussion about the Intel Darwin sources missing XNU, the kernel component? They opened that back up. Remember the fuss about KHTML having difficulties making anything out of the WebKit contributions because they were often Mac OS X-centric? From what I understand, Apple has recently turned WebKit more toolkit-neutral. They have started shipping several of their components (like Bonjour/zeroconf and launchd) under the Apache license rather than their own APSL, and are making available their CalDAV server under the Apache license as well.
;)
I realize that it's healthy to remain skeptic, or at least not assume that the company will forever stay open to the extent that it already is open. But give Apple some credit where due. I don't think it's very likely that Apple will lock down CUPS in the future, since they have their own printing architecture that they're pushing primarily. Apple has been more BSD than FSF/GPL, and it does seem likely that they just want to avoid GPL 3, which is more, uh, religious than is GPL 2. And if Apple do end up locking down CUPS - why not start by forking their last release and *then* start talking about how Apple is chair-throwingly evil?
(Additionally, as a result of these kinds of acquisitions, Apple has a bunch of people who are all about open source onboard, like Jordan Hubbard and Dave Hyatt. I think any theoretical upcoming closedowns would meet strong resistance within the organization.)
Right, but no one is expecting Apple to open up its cellular network communication API. Just the rest of the stuff, so we can build real apps for the phone itself.
Nonexistant, since it's not supported.
I'm hoping Apple will open up the iPhone to third parties sooner or later, but being able to run existing Java midlets is low on their list, and probably for good reason. That's a 160dpi display. You want to aim for a 9-pixels-high checkbox using your finger on that sort of screen? It's not like Java's well-positioned for resolution independence either.
Java apps would stick out like a sore thumb, would work entirely differently and wouldn't actually be able to take advantage of the features people are buying iPhones for. My guess is that they're working on a native SDK for later on.
No, the iTunes Store does *not* use WebKit. http://weblogs.mozillazine.org/hyatt/archives/2004 _06.html#005666
;)
The bundling is awful. The only technically required bundling is QuickTime with iTunes, since iTunes depends on QuickTime. At least now it's fairly public what you get - I remember when you had to hunt around for the "QuickTime only" link. Those kinds of tricks aren't just Microsoft-bad, they're Real-bad.
Safari for Windows is a blessing for web developers. Up until early June, three of the four most used browsers were available on Windows (IE, Firefox and Opera), but the third most used (Safari) wasn't. The more browsers are available and popular on Windows, the more people will finally understand that "standards-compatible" doesn't mean "works like IE". Building for standards, checking in each browser and then doing horrible hacks you wish you didn't have to do to make it work in IE is a better way than the old and broken way: building for IE, checking in the other browsers and sighing about the other browsers not being standards compliant. (I wish I had a nickel for every time someone gave me that crap.)
I'm a Mac OS X user. Firefox is great on Windows, but on Mac OS X it's sticking out like a sore thumb, and it's much slower than the other alternatives. My primary browser is OmniWeb, which uses a variant of WebKit and offers and pioneered some interesting functionality like site-specific settings, a vertical list of tabs with thumbnails, workspaces where sets of windows and tabs are persisted. Even if OmniWeb is an odd choice - it costs money! my god! I must be a complete moron! - almost no one I know use Safari because of the wide ecosystem of good browsers, like Firefox, Camino (a Cocoa app embedding Gecko), Shiira (an alternative WebKit browser) and OmniWeb. Safari has never been considered really good against this background, but it's starting to turn competent in 3.0. Inline Find, draggable (and de/re-attachable) tabs and something as simple as asking when you quit and have tabs open and finally, only took them four damn years, AppleScript tab support means Apple has done a lot of basic tackling and is really listening to people beyond gluing on RSS support and working on WebKit alone. I had almost given up hope.
Because the US mobile phone market is completely wack. Coverage sucks and is highly varying and there are competing standards. Not competing carriers, competing *standards*. GSM and CDMA.
;)
They could sell it without SIM locking, and by Bob I hope they do it in Europe, but you'd only ever reach half of the market. My guess is that it's least suspicious to just do it the way things are usually done in the US - just tie it to a carrier. Nevermind the free market and stuff like that.
Time Machine is once-a-day backup of modified files. ZFS is ongoing versioning.
In Korea, only old people use that joke.
It probably *is* available only to carriers, but wanting your own phone's location is a legitimate request, and I don't see a huge problem in the phone being able to request the estimated position. A service that allowed you to get your current position was launched in Sweden several years ago and I didn't see any mention of infrastructure upgrades on the carrier's part.
It is possible to reasonably pinpoint your position using GSM localization. The accuracy sucks (one order of magnitude worse than GPS at best), but it's more likely to be close enough in areas where you'd want to use maps - unless you are lost in the woods, that is.
Yes. It is second only to the henious contraction that is the word 'blog' itself.
"A delayed game is eventually good, a bad game is bad forever." - Shigeru Miyamoto
The DOM - or something like it - has to be constructed by the browser even when no scripting is involved whatsoever. Specifying it does not do any harm.
I have to say that I don't pretend to know all the background of the noscript case in particular, but if you believe that the explanation given to the special noscript XML treatment is bullshit, why would it even be in there to begin with? There's no reason effort should be spent shooting down XML in an asinine way without needing to adjust for either backwards compatibility, current established browser behavior or what the XML standard says - don't you think you're missing something? And if you're 100% sure you've got the right answer in this case, why not point out exactly what's wrong to the WHATWG folks themselves?
I have a tough time thinking that the canvas tag or the XMLHTTPRequest object would be invented in an HTML specification - yes. This is paving of cow paths - specifying things that are already widely implemented. The W3C, in their infinite wisdom, are *also* specifying things like it. There's already an XMLHTTPRequest specification.
I did not catch this nuance. I apologize.
However, it did sound like you were shooting down anything that was *not* XHTML as valid backing for a new HTML standard. There's a W3C HTML working group now, and on their mailing list are two kinds of requests: people wanting their own tag or people wanting to adopt the Web Apps draft standard as a starting point. Am I amiss in thinking a third alternative - another good way for HTML forward - will not arise for the forseeable future? Why not provide constructive critisism for the HTML5 branch going forward instead of saying that they're all nuts?
I think I should say that I do not want XHTML dead - although I believe that XML's parsing sematics are holding back its tenability, because not everyone produce valid code all the timewithout XML libraries, and very few text editors have XML libraries. Like you say, if you can produce good XML, then XHTML may be better for you than HTML. I also do not think the Web Apps spec has got everything correct. Like you, I do not want the web to just turn into an application API. However, I do want the parts that are already there - like forms - to stop sucking. At over 120 printed pages, it's hard to agree with everything, but according to the browser guys themselves, this level of detail is what they want.
Our stances on whether XML's strictness is good or bad *for the web* is obviously different. I do think it's useful to be strict when handling strictly-formatted documents, as most XML documents are. HTML documents, beyond html, head and body tags, don't really have a strict structure to follow, just rules about how the tags will be nested, and therefore I think it's a really different situation. I will point to Mark Pilgrim's "Thought experiment" to explain my side, but I think it's a waste of our time to continue to debate this if your stance is set in stone. (As an aside, that article ended up converting John Gruber - the creator of Markdown, which you mentioned - to Pilgrim's side.)
There's also a philosophical side: *writing* HTML this far has been about firing up Notepad and a web browser. (Uploading and hosting is a different matter, but I'm talking about writing.) Requiring xmllint to write HTML isn't the web I want.
XHTML 1 is HTML 4 reformulated in XML syntax. By definition, it's backwards compatible. And no, I mean that HTML5's parsing is backwards compatible. Run a page, any HTML page, through an HTML5 tokenizer. HTML will come out right unless you are relying on browser-specific bugs, and there are allowances for XHTML void-tags with the / embedded. Google ran this tokenizer on a billion web pages to scrape them for attributes and tags. It works.
Regarding the XML 'noscript' point, the spec offers up this note: "All these contortions are required because, for historical reasons, the noscript element causes the HTML parser to act differently based on whether scripting is enabled or not. The element is not allowed in XML, because in XML the parser is not affected by such state, and thus the element would not have the desired effect." Does this mean you actually would like to be able to write invalid XML?
Whether you like it or not, DOM manipulation is being used in tandem with HTML, and it needs to be defined somewhere. The spec is actually called "Web Applications", and it describes HTML5 and, additionally, the DOM that should be used to manipulate it and that's used to store it internally. It's also important to remember that the spec is for browser implementors as well as authors! You need not be proficient in the parsing or tokenization or DOM manipulation of HTML5 to write in it, just as you don't need to be proficient in the meaning of uppercase 'SHOULD' or 'MAY' as defined in RFC2119 to be able to use XHTML. HTML5 itself is still a markup language, just as ever, but it's not just markup that needs to be defined in the spec.
At the end of the day, the browser vendors that care about standards sat down and said "this is getting messy, what do we need to codify?" and put it in the same spec with HTML5. CSS2 and CSS3 are well-defined. JavaScript is well-defined as ECMAScript. The font tag is deprecated and irrelevant and doesn't help your argument one whit.