I'm not that impressed by iChat AV, since real time communication removes the one reason I like IM, but I digress.
I agree that it's not exactly a replacement for IM per se, but it's a much more practical person-to-person video-conferencing solution than I have seen, and with audio only and a broadband connection, it's free phonecalls at the margin.:-)
HTML has some beautiful properties that PDF doesn't have
Pff. Name one.
Why stop at just one?
HTML can be rendered on text browsers.
HTML can be rendered more appropriately on any device with an @media rule, not just as a printed page.
You can hide (or re-arrange, with XSLT) content with html, even dynamically.
In general, HTML styled with CSS provides a way to adjust presentation to meet the present need.
Well-written HTML provides at least some notion of semantic mark-up, and I'll stop there for now.
In fact, since you can render any HTML document AS a PDF document, I'd say PDF is a perfect superset of HTML.
Pff. Only if you just *look* at the *rendered* output. You might have noticed that a lot of content in an HTML document is not visible in the rendering. If I'm scraping a screen or spidering the web or doing any number of other things, I'll come up with situations where *most* PDF renderings of a document you'd see in practice are not going to be as useful. Really.
don't you suppose the W3C committe had a clue when they introduced @media rules and the page-break-before/after properties?
Are you kidding me? The W3C are a bunch of masturbators who introduce things with NO concept of whether they're what the world wants or needs or not. Have you READ the standard? It's a joke.
Wait, so you're claiming that page-break properties *aren't* the kind of thing that real world users want? Uhh...I've got news for you.:-) Moreover, yeah, I have read some of the standards (there are way more than one, as I know you know). No, they ain't page turners, but neither are most standards.
So we have a lot of nice parks and pools in town, but the only way to find their hours of operation on the web is to download a bloated PDF of the schedule.
Uh? Nobody said we should replace HTML with PDF, dipshit. I said you should use PDF for things that need to end up on the printed page, not HTML.
Uh...in most cases PDF is an ouput format that you end up with, not what you "use" to create content. (I say most cases because there are lots of things you can do with fillable forms in PDF, but that's a different story.) The great thing about HTML is that you don't need to care (and might not even need to know) where things are going to "end up". If I've got CSS support that includes even just a little control over page-breaks, and @media rules to render for the printed page or the screen, I've just handled probably 90% of most people's casual printing from the web needs without having to mess around with PDF at all.
In any case, I'm done responding to foul-mouthed Anonymous Cowards who really don't know as much as they think they do. It's way more boring than even the W3C specs.
Those things DO come at a price. The price begins at $1999.00 for the 1.6GHz G5, or $799.00 for an eMac.
I do agree with almost everything you say, but felt the usual geek urge to correct every micro-error.:-) In particular, I think you're falling into the "I can build a bottom of the barrel system for way cheap!" error.
The eMac does begin at $799, but that version of the eMac has only a CD-ROM (rather than a far superior combo drive), only an 800 MHz G4 (rather than a 1 GHz G4) and an undersized hard disk (40 GB; and I can't believe I just called that undersized but in this day and age...). Yeah, it exists, but the $999 version is clearly the cheapest "feature complete" one. You also really, truly want the swivel/tilt stand, and that's $59 if you have to pay full price. This is not to say that the eMac is a bad machine; I think it's a *superb* value for the price. But that price is more like $1000 than $800.
In a similar vein, the cheapest ($1999) G5 box actually has some subtle differences in specs from the wildly cool and high tech 1.8GHz and dual 2.0GHz box that might be worth mentioning. In particular, the low-end model lacks PCI-X slots, and has a slightly different memory arrangement (both slower, and less expandable). This might still be a good buy for many people, but it isn't just the 1.8GHz model with a slower processor, less memory and a smaller hard disk.
On the other hand, the pricing of these is exquisite marketing. Yes, you could buy the cheap eMac, but you really want the $1000 eMac...and then wouldn't it be nice to have a superdrive, too? Yes, you could buy the $2000 G5 system, but the $2400 model is really where the coolest high tech starts, and then for just $600 more, you can get a machine that is much more than 25% faster...
I think the marketing niches Apple is setting up now are
pretty clear. On the desktop, for $1000, you get an eMac.
For $2000, you'll get a maxed out iMac (I'm betting these will soon get faster processors). And for $3000, you'll get a G5. Way more rational than some of their older model line-ups.
The adobe 'rep' Greg Gilley, VP of Engineering for Adobe, actually said, "...on some areas of PhotoShop we have seen performance that's more than double of any other machine that we have ever seen."
OK, so the comments I was talking about were the ones on the G5 promotional video. There, Bruce Chizen makes the comparison explicitly to previous Macs.
Use PDF for documents that should be readable both on screen and on paper. If you are not using the proper document format, you can hardly blame your browser.
I think you're missing something here. HTML has some beautiful properties that PDF doesn't have, and I don't
think that just because you sometimes want page breaks *not* to occur right after an H1 header when you print something is *not* sufficient reason to ditch perfectly reasonable web standards and go to pdf. I mean, don't you suppose the W3C committe had a clue when they introduced @media rules and the page-break-before/after properties? A lot of silly uses of PDF would go awy if people understood style sheets and if browsers could take a hint about where to break pages from time to time. Really. So we have a lot of nice parks and pools in town, but the only way to find their hours of operation on the web is to download a bloated PDF of the schedule. By your logic, this is reasonable, but I think it's just nuts.
Now, I'm not prejudiced against PDF; I think it is the best and most appropriate way to put out material with very heavy formatting or archival permanence, (like a journal article reprint). For my stuff, PDF is definitely a premature optimization.:-)
Re:Safari is off to an excellent start
on
Safari 1.0 Released
·
· Score: 2, Interesting
So, for example, Safari provides very good support for @media style rules, but (oddly) doesn't support things like the "page-break-before" property or @page {size: landscape}.
Spoken like a true... well, fucking idiot, actually.
Nobody cares about whether Safari implements all of the fiddly little bits and pieces that make up CSS11 or whatever. What people care about is the user experience.
Uh...frankly *I* don't care about whether Safari implements all of the fiddly little bits of CSS2. Armenian style list items are a nice egalitarian thought, but I don't care if they ever get implemented.
But, believe me, if you had a bunch of undergraduates trying to print out your lecture notes (or a fresh copy of the syllabus, or...) then you would consider page-break-before and size: landscape to be jolly well a part of the user experience. Mozilla gets one of these right but not the other, as does IE. If I could rely on *both* of these working reliably, then life would be much nicer for me (just add a one-liner to my style section to get excellent print out) and for my students.
And, again, one *great* thing about Safari is that it *does* support @media rules really well. So I can make warning text bright red (say) on the screen but switch to bold italic on the print out...stuff that's useful and trivial to do if it's well supported. Doing page-breaks right would add on to a useful feature. Implementing XSLT style sheets in the browser...I'm not so sure.
The one big mega-feature I'd like to see supported, by the way, is SVG. Why? Because a lot of vector graphics look better as...vector graphics. And Illustrator will export stuff that way. But I didn't bring it up at first because SVG alone would be a feature that would justify a whole new major version-number. Page-breaks are a Software Update bug fix.
Somehow I don't think Linux will be left in the dust.
1. There's nothing preventing Linux from running on the new G5 hardware. With Linux, AMD's x86-64 will also be an option.
More to the point, IBM is almost certainly going to release low-price Linux stations using what Apple calls the G5 but whan IBM (it's maker and master) calls the 970.
What I think is actually more important here in the long-run is that *Wintel* will have some problems since the current windows OS is planned to be static until 2005, while Intel faces the challenges you mention. I think the very safest prediction you can make about desktop market shares and relative market shares is that Wintel's will be going down. How much, how fast, and how this affects Apple are additional questions.
Apple turned off hyperthreading in the Dell precision machines, and disabled SSE2. These are modifications you're gonna notice using photoshop, so those benchmarks say nothing.
As somebody else pointed out, *Veritest* turned off hyperthreading and SSE2 when doing the SPECmark tests. I have not found (but would love to know about) any similar tweaking in the application benchmarks.
Another reason that I am suspicious about this claim for the application benchmarks is that they had reps from Adobe and Wolfram Research on-hand saying very nice things about the Mac, even though I do not see any particular reason for them to with to favor it over their Windows versions of the same thing.
Actually, I think the rep comments were quite revealing about this. They were glowing recommendations...but not quite as glowing as Apple would have been alone. So,
in the case of Adobe, all the rep would say in the promotional video was that the G5 was (so far) twice as fast as any previous Mac rather than twice as fast as any Dell. A big, meaningful difference, since the "Finding Nemo" poster really did seem to be designed to hit the G5's sweetspot. For Mathematica, on the other hand, the rep really did make a very, very strong statement in favor of the Mac; that was something that I really took notice of.
One omission from the public presentation was any mention of BLAST; I believe Apple trumpets this on the website, but for one of their core markets, BLAST is the only app that matters.:-) Another omission was Matlab, which might be more understandable due to the relative immaturity of Matlab on newer Mac hardware. Given that, it would be really cool to know about Octave performance on the G5 with optimized BLAS; guess we'll have to wait to see that, though.
Safari is off to an excellent start
on
Safari 1.0 Released
·
· Score: 4, Insightful
It really is very impressive, but Jobs is being...cheeky when he claims it is the best browser on any platform. Camino and Mozilla are still competitive, and that's a great thing. Where Safari can do better is in filling in support for features where it is already strong.
So, for example, Safari provides very good support for @media style rules, but (oddly) doesn't support things like the "page-break-before" property or @page {size: landscape}. This is a bit surprising, and I'd like to see Safari reduce the number of surprises in general.
Mozilla can learn from Safari, as well. Safari's bookmark system is better. It's tabbed browsing implementation is nicer. I suspect these features will be adapted into other browsers, and as the competition heats up again (now that the IE giant is sleeping), everybody wins.
Tell me, have you actually been _following_ the Perl 6 development process?
Didn't think so.
Tell me: you don't have the decency to sign you name here, do you? Didn't think so.
But that was too easy. Here's a more complete answer to your question.
Have I been following the Perl6 development process closely? Not really, since it depends on high-volume mailing lists whose weekly summaries themselves are kinda inside. But what I do know is this:
The Perl RFCs were done by Fall 2000, and since then we've had 6 Apocalypses (3 in 2001, 2 in 2002, only 1 so far in 2003), one half-finished VM (whose FAQ still has an annoying HTML error that I reported weeks ago), some perl design documents, and now, to my utter amazement, a *book* by O'Reilly due anytime now. In other word, over 3 years of effort, and more that you can read than you can run. Again, I'm a casual observer, but I have no particular reason to believe that finishing this will take less than 3 years. I do remember how long Perl5 took, but unless I'm really going crazy, Perl6 progress has been much slower.
I know this is a frustrating situation for all concerned; the timing of the tech downturn could not have been worse for the development of this or most other similarly ambitious projects. But here we are, and I still have concerns about whether Perl6 will ever happen. Looking at what's new or fresh at dev.perl.org/perl6 doesn't make me feel more confident, for that matter.
It's posts like this one which name the names of everyone but the author of the post that remind me how apt the C is in AC.
I agree with you. I was not the author of the AC comments above, and I do not completely agree with them. However...
I was at the (excellent) YAPC::NA in St. Louis in 2002, and it was pretty clear to me then that the grant-funded work had not turned out as well as it might have. I am a long-time fan of Damian Conway's, and I was worshipfully grateful that Larry Wall signed my *old* 1st edition Camel Book. But it was at this conference that I first had the very uneasy feeling that Perl6 would never really "happen". Either it would not be released, or, if it were released, it might not be as relevant as expected. It's now a year later, and therefore even more likely that Perl6 won't "happen". I'm pretty sad about this.
I was very happy to donate money to the grant fund (not a lot of money, but more than millions of others), and I was hoping that it would become a viable model for free software development in general. Unfortunately, since the Perl grants were not viewed as a big success, I think people will be more cautious next time. That can't be the right result, however well or poorly you think the money was spent. And I do wish that the results of the grant program had done more to advance the cause of Perl6, since it might have been an interesting language. Sigh...
In the interests of Slashdot's non francophile readers and despite the fact that I might screw up some of this translation, here is what that item says:
Grek sends us evidence (lit. testimony) that not only confirms rumors of the release of PPC 970 machines, but but also this time for servers.
A friend of mine in in contact with a salesman from Apple
for the purchase of an XServe solution. After several exchanges, the salesman arranged a meeting with him immediately after June 23, a date of important announcements according to him, to talk about some new things in the server sphere. So there will not only be
PowerMacs with the 970 that we are expecting (lit. waiting for), but also Xserve units!
I would remind you that the WWDC last year was the the time when Steve announced the first generation of the XServe in addition to Jaguar. This year, Panther will be there, but it remains to be seen if there will aslo be a speedboat for the XServe at the WWDC. [sorry, I'm not sure how to translate "une vedette de la WWDC]
Now, I do not find the logic here completely compelling; this could be just a price drop, or an announcement about software improvements or what have you, but WWDC wouldn't be a silly place to announce changes in the server line by any means.
On the subject of what rumors have been pulled by Apple legal, squiggleslash writes:
Those stories were about the iChat videoconferencing thing though weren't they? (Think Secret didn't change the URL names or titles - one of the URLs was http://www.thinksecret.com/news/videoconf.html)
Now, whether Apple Legal had these pulled because they were accurate, or merely scurilous but potentially hurting hardware sales, is another question.
Most hardware stocks NOT depeleted at Apple Store
on
Massive WWDC Rumor Roundup
·
· Score: 4, Insightful
For what it's worth, I just checked online at the Apple Store, and pretty much *everything* was listed as shipping on the same day. It is possible that the Apple has emptied the distributors' channels and is holding the remaining inventory, but I would not be very sure about this.
Interestingly, what *wasn't* shipping the same day were two versions of the XServe (not the low-end model or the cluster unit, but the other two). Those were listed as 3-5 days. I haven't done this drill recently, so I don't know how unusual this is for the XServe.
In any case, it might be worthwhile "pinging" the Apple Store this week for the appearance of PowerMac shortages. right now, I don 't see any.
Hi, we're from the Slashdot Geekness Enforcement Group. We've determined that your inability to see the rationale for IP connectivity to headphones and remote controls violates our standards.
Specifically, if your TV remote does not have an IP address, how in the world can you:
Run Linux on it?
Use it as part of your household appliance Beowulf cluster?
Have the thing wirelessly H4x0r'ed by an attacker who has already defaced your neighbor's blender's web site?
The assumption that the 970 will have enough initial volume and be priced in such a way that Apple can REALLY incorporate it across their main product lines. Right now a basic dual 1.4GHz machine is ~$3k without adding in any extras such as RAM/etc. The towers aren't selling well for a lot of reasons, and while the same machine with dual 1.6GHz+ 970's might have double the performance and hold their own very well against competing X86 systems... if the machine costs $5-7k in a recession when 3.xGHz P4's cost half the price might not change much. No one seems to have an idea of what the thing will cost in volume, and that's a big gotcha.
You can always worry about things, but I think the best reason to assume that Apple (at least) will get decent or better pricing on the chip is just the fact that Apple's interests are aligned with IBM's while they surely aren't with Motorola's. Both Apple and IBM really need an affordable high performance follow-on to the current PPC architecture that doesn't involve Motorola, and IBM has a big interest in having everybody see how screamingly great their new chip is in consumer hardware, since that will (undeservedly) speak louder than all the whitepapers you can write about how well your new servers based on the chip will perform. So I'm less concerned with the price (since Apple will really have to make it more affordable than the current dead tower offerings) and more concerned with ramping up the volume.
That said, the fact that NOBODY is saying ANYTHING officially gives me hope that things are going really well. We shall, of course, see. Hopefully by August...
Let's do a little estimation, shall we? Let's call the "average" tornado as about 200 meters wide with a 10 kilometer path. That's actually a pretty big average, but let's take it for argument's sake. There are 1000 tornadoes in a year, on average. So, that's 2000 km^2 of damage per year. That translates roughly into a square patch of damage 9 miles on a side (80 mi^2).
Uh...looks like somebody dropped a zero there. 2000 km^2 is actually more like 800 mi^2.
Let's then further assume that all this damage happens in only Oklahoma proper. Again, a limiting and fanciful assumption, but one useful for these purposes. The area of Oklahoma is nearly 70,000 mi^2. So, the chances that your house will fall in tornado damage will be 80/70000, or 0.11% per year.
Substituting in corrected numbers, we get a probability that's about 10 times higher. But then when you redistribute tornado damage to all of the states, your number is probably not so bad.
Seriously, tornadoes can occur *anywhere* where a _thunderstorm_ can develop. That's pretty much most of Earth's surface between the Arctic and Antarctic circle latitudes.
What you say is true; there is very little absolute safety.
But, having said that, there are places on earth where tornadoes are 10x or 100x more likely to occur than a randomly selected place. Or, to put it another way, in the
almost 5 years I've lived in Columbia, MO, we have *averaged* one tornado warning (i.e., sirens go off, everybody scurries to the basement) per year. And there have been more severe thunderstorm warnings than I could count. In San Diego, where I lived for seven years, I think we had *one* severe thunderstorm, and I'm not sure if a tornado has ever been spotted in the county.
That said, the back of my envelope suggests that the chance of my house getting hit by a tornado in a given year are in the 100-300:1 range. Noticeable, and worth taking precautions over, but not by themselves sufficient reason to abandon the place.
This would be my choice, too. But note that there is still some hurricane danger, and a surprisingly high amount of seismic activity.
Re:Why are we always nitpicking?
on
Shuttle Politics
·
· Score: 5, Insightful
But how is it that we have had troops (US gov. employees) all over the world doing the most dangerous things for decades but 7 astronauts are unreasonable losses?
First, one could question how reasonable or unreasonable the size of the US military is. (Or one should be able to; these days even a hint that we should adjust the forcepool brings with it the accusation that you are a traitor.) Second, for me it's not the loss of the astronoauts' lives per se that makes the manned space program unreasonable. As you mention, the risks are concrete, obvious, and difficult to explain away, but people volunteer. The unreasonable loss is the loss of funding and opportunity to do better science, even space science, in the US. The expenditure of cash on the problem of how to keep a manned space program going when every launch makes you cringe with its "make-work" and PR mission content is just scandalous. People who think that *this* kind of thing will help us fight off near-earth asteroids or bring us closer to lunar colonization are really and truly just not thinking very critically. I would go so far as to argue that the people who are most interested in the eventual manned exploration of space should be the people who should be *least* interested in supporting the status quo.
Everyone is quick to claim this project is a success. I'm taking a wait and see attitude. Sure, it's nice to see people spending money on a legal on-line music distribution system, but I wouldn't call it a success until most of the people that use other systems illegally migrate to paid systems like Apple's.
I think you are missing one point, and worrying about another one excessively. The point you are missing is that there *were* other systems out there before, but the ones that were legal and had non-indie content on them were basically abject failures, for reasons that are obvious to us but apparently baffling to some in the music industry. Apple managed to convince some major labels that it was at least worth a shot to see if this other, more consumer-friendly way to sell music would work out better. Obviously, it has, at least for the people who are willing to pay for music at all.
Clearly, there are and will always be people, and maybe a lot of people, who will live life on the fringes of file-trading, dealing with what are generally fairly bad copies of music that you have to spend a surprising amount of effort to obtain in many situations. I am not really sure how many of those people there are out there, and how well they will be able to survive doing what they are doing given that everybody else pretty much hates them for the amount of bandwidth they suck up, to name their most obnoxious quality.
A million dollars a week is only 52 million dollars a year - that is CHUMP CHANGE. How much of that do you think went to the music industry? It's gonna take money, a whole lotta spendin' money, to make it worthwhile for the distribution oligopoly to embrace this.
You might think it's chump change, but given that we're talking about a service whose current availability is restricted to Mac OS X and iTunes4-using people, which is a 1% market if ever I've heard of one, I think it's fair to say that the potential growth rate is scary. Do I think the business will get 100 times bigger inside of 5 years? Maybe not, but 20 times bigger should be achievable, and that's $1 billion per year, and $350 million to Apple. In other words, still only a smallish share of the overall music industry, but a very high margin business and one that will definitely help sell the hardware.
From the Alternative list, it seems like they're definitely catering the people who have that one gotta-have hit song, but otherwise wouldn't buy much else by the group.
There's some of that, alright, but it's not quite as obvious as you'd think. So it's so...obvious that "Blister in the Sun" is way up there (why *not* buy it for $0.99?), but there are like ten other Violent Femmes tracks in the top 100 as well.
I also think it's interesting that TMBG is so popular, since anyone who was familiar with Napster back in the day would know that they had no shortage of TMBG songs, and Instanbul (Not Constantinople) has always been a favorite among file traders. It's not like there's any shortage of it floating around the internet. Looks like there are a lot of people willing to pay for a legal copy. Interesting.
There are only two TMBG albums up now, and you can buy them as albums or by the song. "Istanbul" does happen to be up, so... I have a hunch that this is one group whose fans might feel a bit bad about not buying. Heck, their label now is Rounder, and they do cheerfully give out some MP3s on their site. Maybe building up goodwill can be good for business? It's a nutty idea, but it might just work.:-)
Slashdot Mirror
I agree that it's not exactly a replacement for IM per se, but it's a much more practical person-to-person video-conferencing solution than I have seen, and with audio only and a broadband connection, it's free phonecalls at the margin. :-)
For the last time, if this really bugs you, just buy yourself a decent USB 3-button mouse!
That way, each copy of OS X will get to use 1.5 mouse buttons.
Why stop at just one?
Pff. Only if you just *look* at the *rendered* output. You might have noticed that a lot of content in an HTML document is not visible in the rendering. If I'm scraping a screen or spidering the web or doing any number of other things, I'll come up with situations where *most* PDF renderings of a document you'd see in practice are not going to be as useful. Really.
Wait, so you're claiming that page-break properties *aren't* the kind of thing that real world users want? Uhh...I've got news for you. :-) Moreover, yeah, I have read some of the standards (there are way more than one, as I know you know). No, they ain't page turners, but neither are most standards.
Uh...in most cases PDF is an ouput format that you end up with, not what you "use" to create content. (I say most cases because there are lots of things you can do with fillable forms in PDF, but that's a different story.) The great thing about HTML is that you don't need to care (and might not even need to know) where things are going to "end up". If I've got CSS support that includes even just a little control over page-breaks, and @media rules to render for the printed page or the screen, I've just handled probably 90% of most people's casual printing from the web needs without having to mess around with PDF at all.
In any case, I'm done responding to foul-mouthed Anonymous Cowards who really don't know as much as they think they do. It's way more boring than even the W3C specs.
I do agree with almost everything you say, but felt the usual geek urge to correct every micro-error. :-) In particular, I think you're falling into the "I can build a bottom of the barrel system for way cheap!" error.
The eMac does begin at $799, but that version of the eMac has only a CD-ROM (rather than a far superior combo drive), only an 800 MHz G4 (rather than a 1 GHz G4) and an undersized hard disk (40 GB; and I can't believe I just called that undersized but in this day and age...). Yeah, it exists, but the $999 version is clearly the cheapest "feature complete" one. You also really, truly want the swivel/tilt stand, and that's $59 if you have to pay full price. This is not to say that the eMac is a bad machine; I think it's a *superb* value for the price. But that price is more like $1000 than $800.
In a similar vein, the cheapest ($1999) G5 box actually has some subtle differences in specs from the wildly cool and high tech 1.8GHz and dual 2.0GHz box that might be worth mentioning. In particular, the low-end model lacks PCI-X slots, and has a slightly different memory arrangement (both slower, and less expandable). This might still be a good buy for many people, but it isn't just the 1.8GHz model with a slower processor, less memory and a smaller hard disk.
On the other hand, the pricing of these is exquisite marketing. Yes, you could buy the cheap eMac, but you really want the $1000 eMac...and then wouldn't it be nice to have a superdrive, too? Yes, you could buy the $2000 G5 system, but the $2400 model is really where the coolest high tech starts, and then for just $600 more, you can get a machine that is much more than 25% faster...
I think the marketing niches Apple is setting up now are pretty clear. On the desktop, for $1000, you get an eMac. For $2000, you'll get a maxed out iMac (I'm betting these will soon get faster processors). And for $3000, you'll get a G5. Way more rational than some of their older model line-ups.
OK, so the comments I was talking about were the ones on the G5 promotional video. There, Bruce Chizen makes the comparison explicitly to previous Macs.
But hey, it's still plenty fast enough. :-)
I think you're missing something here. HTML has some beautiful properties that PDF doesn't have, and I don't think that just because you sometimes want page breaks *not* to occur right after an H1 header when you print something is *not* sufficient reason to ditch perfectly reasonable web standards and go to pdf. I mean, don't you suppose the W3C committe had a clue when they introduced @media rules and the page-break-before/after properties? A lot of silly uses of PDF would go awy if people understood style sheets and if browsers could take a hint about where to break pages from time to time. Really. So we have a lot of nice parks and pools in town, but the only way to find their hours of operation on the web is to download a bloated PDF of the schedule. By your logic, this is reasonable, but I think it's just nuts.
Now, I'm not prejudiced against PDF; I think it is the best and most appropriate way to put out material with very heavy formatting or archival permanence, (like a journal article reprint). For my stuff, PDF is definitely a premature optimization. :-)
Uh...frankly *I* don't care about whether Safari implements all of the fiddly little bits of CSS2. Armenian style list items are a nice egalitarian thought, but I don't care if they ever get implemented.
But, believe me, if you had a bunch of undergraduates trying to print out your lecture notes (or a fresh copy of the syllabus, or...) then you would consider page-break-before and size: landscape to be jolly well a part of the user experience. Mozilla gets one of these right but not the other, as does IE. If I could rely on *both* of these working reliably, then life would be much nicer for me (just add a one-liner to my style section to get excellent print out) and for my students.
And, again, one *great* thing about Safari is that it *does* support @media rules really well. So I can make warning text bright red (say) on the screen but switch to bold italic on the print out...stuff that's useful and trivial to do if it's well supported. Doing page-breaks right would add on to a useful feature. Implementing XSLT style sheets in the browser...I'm not so sure.
The one big mega-feature I'd like to see supported, by the way, is SVG. Why? Because a lot of vector graphics look better as...vector graphics. And Illustrator will export stuff that way. But I didn't bring it up at first because SVG alone would be a feature that would justify a whole new major version-number. Page-breaks are a Software Update bug fix.
OK?
More to the point, IBM is almost certainly going to release low-price Linux stations using what Apple calls the G5 but whan IBM (it's maker and master) calls the 970.
What I think is actually more important here in the long-run is that *Wintel* will have some problems since the current windows OS is planned to be static until 2005, while Intel faces the challenges you mention. I think the very safest prediction you can make about desktop market shares and relative market shares is that Wintel's will be going down. How much, how fast, and how this affects Apple are additional questions.
As somebody else pointed out, *Veritest* turned off hyperthreading and SSE2 when doing the SPECmark tests. I have not found (but would love to know about) any similar tweaking in the application benchmarks.
Another reason that I am suspicious about this claim for the application benchmarks is that they had reps from Adobe and Wolfram Research on-hand saying very nice things about the Mac, even though I do not see any particular reason for them to with to favor it over their Windows versions of the same thing.
Actually, I think the rep comments were quite revealing about this. They were glowing recommendations...but not quite as glowing as Apple would have been alone. So, in the case of Adobe, all the rep would say in the promotional video was that the G5 was (so far) twice as fast as any previous Mac rather than twice as fast as any Dell. A big, meaningful difference, since the "Finding Nemo" poster really did seem to be designed to hit the G5's sweetspot. For Mathematica, on the other hand, the rep really did make a very, very strong statement in favor of the Mac; that was something that I really took notice of.
One omission from the public presentation was any mention of BLAST; I believe Apple trumpets this on the website, but for one of their core markets, BLAST is the only app that matters. :-) Another omission was Matlab, which might be more understandable due to the relative immaturity of Matlab on newer Mac hardware. Given that, it would be really cool to know about Octave performance on the G5 with optimized BLAS; guess we'll have to wait to see that, though.
So, for example, Safari provides very good support for @media style rules, but (oddly) doesn't support things like the "page-break-before" property or @page {size: landscape}. This is a bit surprising, and I'd like to see Safari reduce the number of surprises in general.
Mozilla can learn from Safari, as well. Safari's bookmark system is better. It's tabbed browsing implementation is nicer. I suspect these features will be adapted into other browsers, and as the competition heats up again (now that the IE giant is sleeping), everybody wins.
Tell me: you don't have the decency to sign you name here, do you? Didn't think so.
But that was too easy. Here's a more complete answer to your question.
Have I been following the Perl6 development process closely? Not really, since it depends on high-volume mailing lists whose weekly summaries themselves are kinda inside. But what I do know is this:
The Perl RFCs were done by Fall 2000, and since then we've had 6 Apocalypses (3 in 2001, 2 in 2002, only 1 so far in 2003), one half-finished VM (whose FAQ still has an annoying HTML error that I reported weeks ago), some perl design documents, and now, to my utter amazement, a *book* by O'Reilly due anytime now. In other word, over 3 years of effort, and more that you can read than you can run. Again, I'm a casual observer, but I have no particular reason to believe that finishing this will take less than 3 years. I do remember how long Perl5 took, but unless I'm really going crazy, Perl6 progress has been much slower.
I know this is a frustrating situation for all concerned; the timing of the tech downturn could not have been worse for the development of this or most other similarly ambitious projects. But here we are, and I still have concerns about whether Perl6 will ever happen. Looking at what's new or fresh at dev.perl.org/perl6 doesn't make me feel more confident, for that matter.
I agree with you. I was not the author of the AC comments above, and I do not completely agree with them. However...
I was at the (excellent) YAPC::NA in St. Louis in 2002, and it was pretty clear to me then that the grant-funded work had not turned out as well as it might have. I am a long-time fan of Damian Conway's, and I was worshipfully grateful that Larry Wall signed my *old* 1st edition Camel Book. But it was at this conference that I first had the very uneasy feeling that Perl6 would never really "happen". Either it would not be released, or, if it were released, it might not be as relevant as expected. It's now a year later, and therefore even more likely that Perl6 won't "happen". I'm pretty sad about this.
I was very happy to donate money to the grant fund (not a lot of money, but more than millions of others), and I was hoping that it would become a viable model for free software development in general. Unfortunately, since the Perl grants were not viewed as a big success, I think people will be more cautious next time. That can't be the right result, however well or poorly you think the money was spent. And I do wish that the results of the grant program had done more to advance the cause of Perl6, since it might have been an interesting language. Sigh...
In the interests of Slashdot's non francophile readers and despite the fact that I might screw up some of this translation, here is what that item says:
Grek sends us evidence (lit. testimony) that not only confirms rumors of the release of PPC 970 machines, but but also this time for servers.
I would remind you that the WWDC last year was the the time when Steve announced the first generation of the XServe in addition to Jaguar. This year, Panther will be there, but it remains to be seen if there will aslo be a speedboat for the XServe at the WWDC. [sorry, I'm not sure how to translate "une vedette de la WWDC]
Now, I do not find the logic here completely compelling; this could be just a price drop, or an announcement about software improvements or what have you, but WWDC wouldn't be a silly place to announce changes in the server line by any means.
On the subject of what rumors have been pulled by Apple legal, squiggleslash writes:
Actually, stories about G5 Macs have also been pulled from www.macbidouille.com, as has www.macrumors.com and www.osnews.com and tech-report.com. All of these were about 64-bit offerings being shown at WWDC.
Now, whether Apple Legal had these pulled because they were accurate, or merely scurilous but potentially hurting hardware sales, is another question.
Interestingly, what *wasn't* shipping the same day were two versions of the XServe (not the low-end model or the cluster unit, but the other two). Those were listed as 3-5 days. I haven't done this drill recently, so I don't know how unusual this is for the XServe.
In any case, it might be worthwhile "pinging" the Apple Store this week for the appearance of PowerMac shortages. right now, I don 't see any.
Specifically, if your TV remote does not have an IP address, how in the world can you:
(Sigh.) The kids these days...
You can always worry about things, but I think the best reason to assume that Apple (at least) will get decent or better pricing on the chip is just the fact that Apple's interests are aligned with IBM's while they surely aren't with Motorola's. Both Apple and IBM really need an affordable high performance follow-on to the current PPC architecture that doesn't involve Motorola, and IBM has a big interest in having everybody see how screamingly great their new chip is in consumer hardware, since that will (undeservedly) speak louder than all the whitepapers you can write about how well your new servers based on the chip will perform. So I'm less concerned with the price (since Apple will really have to make it more affordable than the current dead tower offerings) and more concerned with ramping up the volume.
That said, the fact that NOBODY is saying ANYTHING officially gives me hope that things are going really well. We shall, of course, see. Hopefully by August...
The bold font below is added by me:
Uh...looks like somebody dropped a zero there. 2000 km^2 is actually more like 800 mi^2.
Substituting in corrected numbers, we get a probability that's about 10 times higher. But then when you redistribute tornado damage to all of the states, your number is probably not so bad.
What you say is true; there is very little absolute safety. But, having said that, there are places on earth where tornadoes are 10x or 100x more likely to occur than a randomly selected place. Or, to put it another way, in the almost 5 years I've lived in Columbia, MO, we have *averaged* one tornado warning (i.e., sirens go off, everybody scurries to the basement) per year. And there have been more severe thunderstorm warnings than I could count. In San Diego, where I lived for seven years, I think we had *one* severe thunderstorm, and I'm not sure if a tornado has ever been spotted in the county.
That said, the back of my envelope suggests that the chance of my house getting hit by a tornado in a given year are in the 100-300:1 range. Noticeable, and worth taking precautions over, but not by themselves sufficient reason to abandon the place.
This would be my choice, too. But note that there is still some hurricane danger, and a surprisingly high amount of seismic activity.
First, one could question how reasonable or unreasonable the size of the US military is. (Or one should be able to; these days even a hint that we should adjust the forcepool brings with it the accusation that you are a traitor.) Second, for me it's not the loss of the astronoauts' lives per se that makes the manned space program unreasonable. As you mention, the risks are concrete, obvious, and difficult to explain away, but people volunteer. The unreasonable loss is the loss of funding and opportunity to do better science, even space science, in the US. The expenditure of cash on the problem of how to keep a manned space program going when every launch makes you cringe with its "make-work" and PR mission content is just scandalous. People who think that *this* kind of thing will help us fight off near-earth asteroids or bring us closer to lunar colonization are really and truly just not thinking very critically. I would go so far as to argue that the people who are most interested in the eventual manned exploration of space should be the people who should be *least* interested in supporting the status quo.
I think you are missing one point, and worrying about another one excessively. The point you are missing is that there *were* other systems out there before, but the ones that were legal and had non-indie content on them were basically abject failures, for reasons that are obvious to us but apparently baffling to some in the music industry. Apple managed to convince some major labels that it was at least worth a shot to see if this other, more consumer-friendly way to sell music would work out better. Obviously, it has, at least for the people who are willing to pay for music at all.
Clearly, there are and will always be people, and maybe a lot of people, who will live life on the fringes of file-trading, dealing with what are generally fairly bad copies of music that you have to spend a surprising amount of effort to obtain in many situations. I am not really sure how many of those people there are out there, and how well they will be able to survive doing what they are doing given that everybody else pretty much hates them for the amount of bandwidth they suck up, to name their most obnoxious quality.
You might think it's chump change, but given that we're talking about a service whose current availability is restricted to Mac OS X and iTunes4-using people, which is a 1% market if ever I've heard of one, I think it's fair to say that the potential growth rate is scary. Do I think the business will get 100 times bigger inside of 5 years? Maybe not, but 20 times bigger should be achievable, and that's $1 billion per year, and $350 million to Apple. In other words, still only a smallish share of the overall music industry, but a very high margin business and one that will definitely help sell the hardware.
No independents of any stripe right now. But that will change.
There's some of that, alright, but it's not quite as obvious as you'd think. So it's so...obvious that "Blister in the Sun" is way up there (why *not* buy it for $0.99?), but there are like ten other Violent Femmes tracks in the top 100 as well.
There are only two TMBG albums up now, and you can buy them as albums or by the song. "Istanbul" does happen to be up, so... I have a hunch that this is one group whose fans might feel a bit bad about not buying. Heck, their label now is Rounder, and they do cheerfully give out some MP3s on their site. Maybe building up goodwill can be good for business? It's a nutty idea, but it might just work. :-)