Why anyone would continue to use tabbed browsing given the existence of Expose is beyond me. Perhaps if Apple were to have the windows display their titles superimposed as soon as Expose is active, rather than only when they are moused over?
I would agree with you if I thought that tabbed browsing was only useful for having lots of windows without the clutter. As it turns out, I use tabs to *organize* my browsing experience. So there are many times when I have a set of tabs open for my teaching materials, a set open for BLAST analyses I'm running, a set for, well, bloggish stuff, and then one for google/deja/yahoo. That could easily be 20 windows, or just 4 sets of 5 tabs. (Note that I *never* quit the browser, and it starts up when I login.)
And as far as "mousing over" the windows in Expose goes...please. What do you think keyboard navigation is for?
I think the best compromise solution is for Expose to give you the option of exploding out your Safari tabs when you select one of its windows. Or, if you use F10, then maybe it auto-explodes them.
In any case, it really does turn out to be life changing. The command that just *really* became extinct, by the way, is "Hide..."
With any current X11 server of which I know, no. Apple made the mixed decision to implement a windowing system which is largely vector based, and which uses opengl textures for compositing of ui elements. Which is completely different from the behaviour of extant X11 servers.
I say "mixed" because Apple's approach has some fairly significant overhead for even the simplest of operations, but very complex operations (like Expose) are pretty much free beyond that.
Instead of "mixed" I would say "forward-looking". There is some cost, but Apple has (I think correctly) decided that the cost is only going to be noticeable on hardware of this generation (pre-G5) and before. The benefits of the vector model go on and on forever. So what Expose does for every window on the computer (or in the application) you can expect Safari to be able to do for every tab in your window. For that matter, "Grab" *could* be made to grap a PDF of your screen instead of a tiff (this is actually a bug, I think; also, if you print to PDF from Grab, you only get the visible part of the window; that's *definitely* a bug. Everything can generate useful PDF.
Basically, once you drink the Kool-aid on the vector-based model, I don't think you'd really want to go back.
I disagree. The find function is nice, but the most exciting thing about Preview is that it now knows about bookmarks in PDFs, making it very easy to navigate large PDFs quickly.
You're obviously some kind of pathetic geek.:-) More seriously, that *is* very nice, but I should point out that some of the TeX documents I tried playing with turned out to be unsearchable by Safari; is this possibly caused by the fact that they were created in an odd fashion (didn't use pdflatex or something?). Is there something special you should do with your LaTeX to make sure it "plays nice" with more advanced PDF readers?
As expected, the reviewer points out that Expose is really, really cool. More interestingly to me, he also points out that Preview is much improved as well. I used to think it sucked, but now I've made it my default PDF viewer.
The most exciting thing about Preview, though, is that it's easy to predict that its beautiful new "Find" function will make its way into things like Safari. Seriously, "Find" is so good that it has pretty much changed the way I rifle through the dozens of PDFs of journal articles I have on my disk. Like, I can actually find what I'm looking for.:-)
Our university gave out free Panther upgrades to all faculty and staff, which was very decent of them.
However, they only gave us *3* CDs, so I don't have the updated compilers or development environment. I'm assuming, therefore, that I can't do anything to upgrade Fink until I can the tech people here to give me a copy of CD 4, right?
I knew this would come back to bite me, but since I got what amounted to hundreds of dollars worth of free sofware...
Blue tooth is dead because Apple is the big PC maker pushing it, and:
Apple is beleagured.
Mac OSX is based on BSD and BSD is dying.
Some guy out there can't copy files from the hard drive of 1994 era computer running OSX, so OSX is slow, and its Bluetooth support won't work on a MacPlus.
The logo is stupid.
Fortunately, you have people like me around to point this all out to you.:-)
M: worm eater writes "The Register has posted a correction to the widely-reported story that a 3.7 terraflop Dell cluster cost the University of Texas $38 million. As it turns out, the computer cost $3 million, vs. $5.2 million for the 17.6 terraflop Mac G5 cluster at Virginia Tech."
Indeed. And now we can see that UT got a totally killer deal. Why, they paid only about $800,000 per teraflop while Virginia Tech, using horribly over-priced Apple hardware, paid an absurd $300,000 per teraflop of computing performance. As you can see, uh...wait a minute. Hmm...[spin, spin, spin]
Ah. So what do those Hokies think? That money grows on trees? How on earth are we going to see a recovery in the tech sector when people are only willing to pay for cut-rate supercomuter solutions that can't even run Windows natively?
This is exactly where PLOS comes in. How about they expand into Brain Research's territory? Of course, first you have to convince the authors it's a better idea to publish in PLOS than in Brain Research.
Actually, if you want "psychotically expensive institutional online access", the journals to pick on are anything by Cell Press. If PLOS could replace Neuron, the world would be a much, much less expensive place.:-)
A big problem with PLoS is that an author is charged $1500 (!!!) to publish in the journal. This is going to bar a lot of people who lack significant funding from publishing in the journal. I don't see how passing the ridiculous costs of journals from subscribers to authors is a very good fix!
You just aren't thinking very hard about this then. Teh first journal in the PLoS line-up is PLoS:Biology; the vast majority of articles published here, if they really do make it the equivalent of Science/Nature/whatever *will* be published by people and labs receiving some amount of external funding. Barring that, they could probably apply for intramural funding to defray publication cost in a prestigious journal. After they get off the ground, I have no doubt that one or more philanthropists or corporate sponsors will not start shelling out for this cost. But the point is: there is a real cost to publishing science, and somebody will have to pay it.
As far as the "who pays?" question goes, I think it should be crushingly obvious that this is a big win, especially for people who are underfunded or who come from institutions that do not have large journal budgets. So on-line institutional pricing for Nature is pretty high; I think I remember a figure like $8000 per year being bandied about. There are probably about 5000 institutions paying this fee right now, or about $40 million coming into Nature. There are tens of thousands of others who pay somethingly like $200 per year to get the same access (it can go higher, but you can always get some discount or other). Probably another $40 million or so comes in that way. That's $80 milllion spent for what I believer will work out to be 1000 or so articles. So the total subscription cost is on the order of $80,000 per article published in Nature If these had all been published in PLOS journals, the total subscriber cost would be $1500 per article. Even if I have my Nature numbers high by a factor of 10, there is still a sizable community savings by going the PLOS route. As far as how much PLOS will cost, if they get to the point where they publish as many articles as Nature (say 1000 per year), then I figure they will need an editorial staff of about 10, an office or offices for the same, and whatever their web access costs them (right now they're slashdotted; that shouldn't happen). I don't find it hard to believe that this would cost $1.5 million per year, and so I have to conclude the cost is reasonable. Funds can always be raised to cover reasonable costs.
Actually, it's the smaller disciplines (in science anyway) that have some of the highest costs. Brain Research, for example, runs $10,000 per year, last time I checked.
I'd like to point out that the field of Neuroscience now qualifies as Pretty Darn Big, and, moreover, it is moving Pretty Darn Fast. This is why Elsevier and others can charge hugely for their journals: the demand *is* there, and the cost of *not* having Brain Research or (to pick a non-random example) the Journal of Comparative Neurology can be huge.
Historically, journals like J. Comp. Neur were really expensive because their field (neuroanatomy) basically forced them to have lots of high resolution (and even full-color) plates. This is just not a very cheap thing to do. These days, however, when people are increasingly reading articles on-line, that cost has gone away. A similar issue existed with type-setting. Type-setting is exepnsive. Running a conversion filter over somebody's LaTeX or Word file is NOT. There do remain some real costs: the infrastructure for peer review is not that cheap.
But the unit costs of production have gone DOWN a lot, especially for on-line publication. The fact that journal prices have gone UP a lot just demonstrates that these prices have no connection with publication costs and every connection with being a monopoly supplier of a service in demand. It really is this simple. And, at some point, journals *like* PLOS:Biology (if not PLOS:B itself) really do have a good chance to break the cycle.
In any case, isn't really the size of the production run that does (or will) dominate but plain old market factors. Brain Research costs a ton because you can't substitute another journal to get the articles that they alone have rights to publish, and there is demand for the work that appears there.
OK, so one of the lamest aspects of the current Mac OS X experience for me revolves around using Preview to view PDF documents. (An issue I won't get into is why Preview and Acrobat fight over who gets to do this.)
So Acrobat (and Safari, and every other sane program) use the obvious cmd+ and cmd- to zoom. Preview psychotically uses the lame cmd-uparrow and cmd-downarrow. These choices are clearly lame because while + and - only have obvious meanings here for magnification, arrow keys should be for navigation. If I asked one hundred naive users what down-arrow should do for a document you're reading, I think it's clear they would all say "scroll down" and *not* "next page".
I find this infuriating because with arrow keys and the command key, you *could* get the nirvana of using arrow keys alone to scroll and pan, and then use cmd-arrrow keys to move by page. As it is now, I'm forced to touch the mouse to pan, and that's just lame. The beauty of Mac OS X for me is that it allows you to control 98% of what you ever need to do from the keyboard, despite having a very handsome GUI.
Other lamenesses of Preview include the lack of full-page mode, and the (current; this will be fixed) limitations to using it with pdf files rather than ps files. And these things are too bad, since *other* features of preview (e.g., "Export...") are quite nice. For a system that is so decisively built on top of PDF, it's very odd that Preview is now a weak link.
So does anybody out there with a preview copy of Panther know if Preview has been improved beyond what they brag about at apple.com?
Monday's prize honors pioneering work done in the 1970s that laid the groundwork for making MRI a useful method, the assembly said.
Heck, the first whole-body MRI scanner was finished in 1977 -- and the Nobel Prize is being awarded just now? What am I missing on how long it takes for the committee to conclude that something has been revolutionary?
There's almost always an appreciable delay, even for obviously pioneering stuff. My feeling, though, is that the delay can be longer for breakthroughs that fall under "instrumentation" or "technical" rubric.
The *classic* case of this is this was the fifty-three year gap between the invention of the electron microscope and the 1986 Physics prize being awarded to Ernst Ruska. The two other co-winners that year were IBM's Binnig and Rohrer who won for the "scanning tunneling microscope". What apparently happened here was that the panel wanted to award the prize for the scanning tunneling instrument (which had been invented less than 10 years previously) but then were reminded that Ruska had never won the prize for the invention of the EM. Oops...
In this case, I can conjecture that somebody on the panel was thinking ahead to the (inevitable and eventual) prizes that could be awarded for applications of functional MRI to neuroscience, and then somebody realized that nobody had ever won for the development of anatomical MRI techniques. Oops...
The real problem, alas, is that there are never enough Nobel Prizes available to hand out these days, especially in Physiology and Medicine. So everything that gets awarded comes across as long overdue given the huge backlog of potentially worthy recipients.
Re:Why do you think Bush gave them tax cuts?
on
Tech Rich Get Richer
·
· Score: 4, Insightful
This phrase is really starting to annoy me. If you define "the rich" as "everybody who pays taxes", then yes, the recent tax cuts were "tax cuts for the rich". But that definition is obviously ridiculous, so the phrase "tax cuts for the rich" really doesn't apply to the Bush tax cuts.
First of all, I don't know how everybody involved in these kinds of debates manages to ignore *payroll* taxes, which are just as surely taxes as any other kind of tax, and which fall disproportionately (meaning, a larger fraction of income) on people with the lowest earned incomes. Those taxes have not gone down, although they are in many cases the *majority* of the taxes paid by people with lower incomes. And that's really just the federal taxes. State income taxes, in the states that have them, often have a top bracket at some pathetically low amount; those taxes have not been going down, either.
And then there are sales taxes and gasoline taxes, which end up being a higher marginal rate on lower incomes for reasons that I'm sure should be pretty obvious.
You can agree or disagree with the reasoning behind the Bush tax cuts, but because they were cuts in income taxes primarily for the very highest brackets, there is very little way in which they could not have been tax cuts for the wealthiest.
In a similar vein, the plan to eliminate the estate tax by definition only affects states that are quite a bit larger than the vast majority of estates. By your reasoning, it would be unfair to say that this is a tax cut for the rich because it's a tax cut for the only people paying any estate tax. But those are the rich people. Hence, the point stands.
These depressing statistics, though, come with two caveats. First, poorer Americans are better off than they once were. The proportion of Americans in poverty now stands at 12%; in Mr Krugman's supposedly golden 1950s, it reached 22%.
And those caveats come with caveats...
First, the "proportion in poverty" number has some issues.
It is based on income and not wealth, and the poorest fifth, whether they live in poverty or not, are further in debt now than they were 20 years ago. It would be more interesting to look at "total real disposable income" or some measure that looks at what you've really got after taxes, mandatory debt service, and other things. Also, the poor, whether below the official poverty line or not, also tend to be the uninsured, and the importance of benefits like health insurance is far greater today than it was in the past.
A second issue with the poverty line is just that: it's just a standard threshold that was set with the hope that it was reflective of real poverty. Moreover, it's a standard threshold that you could, if you had the political power, mess around with a bit. In any case, looking at the raw data on income and wealth and household varianbles gives you much more information than looking at an under/over the threshold variable.
Second, America is a remarkably mobile society. As this year's Economic Report of the President points out, 50-80% of the unfortunates in America's bottom quintile push themselves into a higher quintile after 10 years.
The US *does* have rather more mobility than a lot of places, but you have to be really careful in how you look at it. So, if memory serves, the number quoted here has some oddities that may not be apparent. One of them is that for such a definitive sounding number, it has one heck of an error bar around it (50%-80%? sheesh!). Also, the number of people who inhabit an income quintile over the course of a decade is way more than one fifth of the people around. A lot of people are in low income quintiles for reasons that are basically trivial and have little to do with effects of poverty. When I was in grad school, I know that my wife and I were *always* in the lower 20-25% of incomes, and now a decade later I'm doing better than that. Of course, I had a low income because I was getting an advanced degree. Do you really think that mobility is that high for people with high school or less as their top educational attainment?
Beyond that, if X% of the bottom quintile leave it in ten years, Dogbert would point out than other people join it.
So imagine a time and place where there is no real progress in incomes, but a given person's income percentile is noisy, and bounced around +/- 10% over the course of a decade. In a situation like that, a goodly proportion of people in the lower quintile would leave it, and a goodly proportion of people in the second lowest quintile would join it, but progress here is (by definition) zero.
Now, real economists know all of these caveats and caveats to the caveats, which is why they never just say something as rosy sounding as "most people in the lowest quintile leave it in 10 years!". Politicians and journalists on the other hand...
1/3? I don't know what tax bracket you're in, but after Fed, SS, Medicare, State, Local, property tax, Sales tax, auto registration, and other various fees, most of us in the U.S. pay over 50% of our salaries to the gov.
In a word, no we don't. Not if by "most of us" you mean "most US taxpayers". For a decently readable account of this and other economic "facts", there's a piece in the NY Times (free registration blah blah).
I t___k y__r p__t is p___f t__t we d________y do n__d t_e m____e l_____s
Actually, I find this *very* interesting since the initial belief that this would work was based on an auditory study, and the fact that internal graphemes can be scrambled (at least many of them) is pretty stunning. I think the problem with using just underlines is not just that it hides the identity of the inner letters, but it makes it more difficult to know the true "length" of the word. If I'm right then,
Txxs sxxxxd be exxxxr to rxxd txxn
t__xt w___e y_u u_e u________s in p___e of t_e l_____s.
Or, to put it another way,
T__s s____d be h____r to r__d t__n
txxt wxxxe yxu uxe uxxxxxxxxs in pxxxe of txe lxxxxxs.
Or maybe not. They're both harder than
txet werhe you do not use udlerneins in pcale of the ltretes.
A commodity is only overpriced if one of two things is true:
1) It does not sell.
2) You have reliable evidence that a lower price point would have increased your earnings (note I did not say "unit sales").
Now, I know that's a point I've made before, too. But it works even better when you throw numbers out there. For the sake of argument, let us suppose that Apple has a robust 30% margin on the powerbook, so that the $1600 model nets them $370 in gross profit. Profit would then be $370*units_sold. Now let's pretend that they go for a more Dell-like 10% margin. Gross profit now goes to $123*units_sold. So now Apple has to sell three times as many of the suckers to make the same gross profit. The model with the 10% margin now costs only $1353, and there's no doubt that they would sell more of them at this price point. But 3 times as many, for an item for which you suspect the early buyers represent pent-up demand? Doesn't seem too likely to me. That's an extreme scenario, but I think the point is clear. If Apple can get away with their high margins, they'll do it every time. (Note: for simplicity I am pretending that every Mac gets sold at an Apple store so that the gross profit isn't shared; the situation is even more dramatic if you have to share the gross with the retailer who made the sale.)
you face the immediate problem of whether or not to provide DVI connectors, or what size monitor you want to support.
today even entry level vid cards (e.g. radeon 9200) have both DVI and analog connectors, and they can operate at resolutions high enough to 'support' all popular sizes of screens (and many unpopular ones as well)
True, but that's basically my point. The problem with an ULTRA-CHEAP headless machine, which is what we were talking about, is that you have to make sure both that people would buy it, *and* that the people who do buy it would not have bought one of your higher margin products. You and I both probably know that a decently designed G4 headless system with a nice monitor would be a pretty awesome deal. The problem, though, is that it would cannibalize other sales. And that's the problem: if you need to offer a DVI connection to sell it, I think you're very likely to cannibalize sales. If you don't offer the DVI connection, then why should anybody buy this rather than an eMac?
Most Mac buyers are not switchers
It may very well be because apple has no atractive low/mid-range solution - and said headless box might be it.
No, that's not it. If you need Windows, the lock-in factor is obvious. If you don't, there are other options (including cheap Linux boxes, by the way). The iBook is very attractive in the cheap notebook market, and the eMac is (now, after price drops) quite capable at $999. The problem is that Apple basically can't do a $500 box that would run Mac OS X acceptably and turn a decent profit, so you're then stuck in this $700+ region where being headless won't compete against ultra-cheap PCs given Windows lock-in, but then $700+ for the slab plus a $200 monitor puts you in the eMac price slot. I think a better option would be to grind the price of an eMac down to $850 or something.
A lot of monitors people have lying around to use with a headless box are pretty poor quality
other then the fact that they tend to break up pretty quickly (2-3 years, which is less of an issue because their so cheap) todays cheap monitors are of rather high quality (image quality, that is). I can honestly say i can hardly tell the difference in image quality between low end monitors and high end monitors these days...
You really aren't looking hard enough, then. Seriously, look at the "monitor line-up" at your local computer superstore and tell me that the ultra-cheap monitors look the same. They really don't, and both Mac OS X and XP make this really obvious. Quality has improved recently, but the real headless market is for people who already own the nice monitor already (then buy a power mac) or those who want to be ultra-cheap. Apple has an offering for the first group, and tells the second group to go for just "cheap" and skip the "purchase really cheap monitor" step. Maybe that's not ideal for growing lowest end market share, but Apple is actually trying to maximize profits, and I can't see that they're leaving much on the table.
the iPod might be great but it is terribly overpriced.
I disbelieve. There are two heuristics to tell if something is over-priced. The first is: nobody buys the thing, or you sell fewer than you expected. The second is: you have *reliable* marketing research tells you that a lower price would sell you more than enough additional units to make up for the reduced price. I don't think it's always easy to justify the second course of action. If we pretend that the gross profit on an iPod is about $100, than a $50 reduction in price would halve that number, and we would have to sell twice as many of them to make up the difference. I think you can see the risk here. The iPod is the best seller in its category, and while you can find cheaper hard-drive based players, nobody has come out with a clearly better one. I think the nominal prices will drift down over the next couple of years, but the real price will (and has) sunk like a stone. The original iPod was 5 MB for $500. The top-of-the-line model is now 40 MB (8 times the capacity) and comes with much better interface options. That's huge real improvement in 3 years...
Why can't Apple make a less expensive headless box? All thier lower-priced units have some sort of monitor attached (e/iMac).
I thought about this, and I came with at least 4 reasons.
The demand for headless boxes is probably less than you'd imagine. Most Mac buyers are not switchers, so I don't think they are as likely to have a dreamy monitor hanging around *unless* they were already buying the PowerMac line, and that's headless.
Headless has gotten more complicated to do right and do well. A dirt-cheap headless box would ahve to be pretty well integrated, but then you face the immediate problem of whether or not to provide DVI connectors, or what size monitor you want to support.
A lot of monitors people have lying around to use with a headless box are pretty poor quality, and the one thing Apple *doesn't* want is people running Mac OS X and having the result look crappy. Graphics people have nicer monitors, and would know better than to blame the platform if they used a cheap monitor and it looked bad. The extreme entry level crowd is different.
The notebook is the new headless Mac! OK, so it's not really headless, but the low end notebooks (iBook and 12" powerbook) can fill a part of the headless niche, and be notebooks, too. They perhaps aren't optimized for this usage yet, but they can do the job. Or, put it this way: your choice would be a $700 slab or a $999 iBook or a $999 iMac. The difference here is that one of these is *also* a notebook, one of thes comes with a *very nice* LCD monitor, and the other one is just a slab. I don't think it would be such a hot seller.
Okay, I'm not going to be lazy. I went to Dell.com and configured a 2.4ghz Dual Xeon with Windows XP, 512mb RAM, a DVD drive and a modem. This is roughly the same configuration offered by the G5.
Price is $2,801.
So the Mac is about $200 more than a system with about the same performance once the Mac is optimized.
OK, so I am even less lazy, and I disbelieve. I did what you said you did, and my dual-Xeon Precision 650 with a USB mouse, USB keyboard, dual-monitor DVI capable video card 120 GB IDE drive (largest IDE drive there; Serial ATA is not an option and the SCSI drive prices gave me the giggles) and DVD-RW/CD-ROM 48X (hope it's writable) drive and no Dell support (to equal the Apple G5) would cost me a total of $3054.10.
Now, I was too lazy to find out if the Dell had gigE ethernet like the Mac, but this is a closer comparison *IF* you are right about the fact that an optimized dual 2 GHz G5 is only equivalent to a 2.4 GHz dual Xeon. If it actually turns out to be anywhere close to a 3.0 GHz dual Xeon, we can stop having this conversation right now, since that box starts at over $4000. Myself, I'm guessing the truth is somewhere in between.
So that makes the G5 about the same performance as a Dell that might run anwhere between $0 and $500 more. "Priced very attractively" is how I would have to put this. Likely not attractively enough to get *many* Windows shops to switch (based on hardware cost and software legacy alone anyway), but attractive enough to keep market share and probably then some.
Consider, for example, what would happen if they went for Mandrake. Could Mandrake afford to fight them? I doubt it, they're living hand to mouth. They can either fight it and go into Chapter 11, pay extortionate terms and... go into Chapter 11, or... cut a deal in which they don't contest, in return for an agreement from SCO to ask for token damages.
Then SCO goes after SuSE and Red Hat, waving their precedent.
Uh...what precedent? Your scenario is that SCO settles with
Mandrake, and there is no ruling on the merits of the case.
Maybe I'm raving, but I thought "precedent" means that you won in court. Now, you could try to engineer that scenario, but I doubt that it would carry much weight to wave such a ruling, from a case where the defendant could not afford to defend and did not contest, in a new case where the defendent aggressively defends and asks for the case to be thrown out for a dozen different reasons before the first court hearing. Of course, I'm not a lawyer, so maybe I don't know what I'm talking about here.
There are two things (yeah, just two) I'm not 100% happy about on the G5 series:
The relative lack of expandability (there are only two drive bays, for example, and both are already in used by default - one HD, one optical media)
NINE FANS! Holy crap Apple, what the HELL were you thinking?!
OK, so others have pointed out that you get two hard drive bays, and the SuperDrive is standard, making that a bit less of an issue in my mind.
About the 9 fans, if they turn out to do a better job of managing heat than the 747-style fans in the dual-G4 I've got on my desk, they could put 15 fans in there and I'd still be thrilled. Seriously, the G4 case is splendid, but the cooling system on the latest dual processor models...somebody should have been shot for it to be this loud. (Okay, I'm fine now...)
Another point about the apparent smallish number of hard drive bays is that I think this makes sense because we have crossed an important threshold:
From now on, you'll want a disk or two at most on a desktop. If you need more storage, drink the kool-aid and get yourself a proper RAID.
Honestly, the two disks the G5 will hold can bring you up to half a terabyte. Anymore than this, and arguably even *this* much, and what you really want is either a big old server in a closet somewhere or a RAID appliance of some kind. If you want to use the G5 itself as a server, as luck would have it, Apple also announced a nifty fibre channel
PCI card that lets you attach the XRAID among other amusing devices.
*My* only objection to the G5 machines is that I don't have one. (sniff)
You really have no clue about OOP before you have tried one of the dynamic OOP languages: Python, Smalltalk, or Ruby. Smalltalk has fallen to a legacy role these days, while Ruby is much less mature and has a smaller community than Python. Additionally, Ruby is less "tasteful", in that it borrows more heavily from perl, but that is a matter of controversy;-).
I am not sure what you mean by Ruby being "less mature"; as a language in this niche, it appears (to me) to be among the most mature. It does have a smaller user community right now, and there are some library and documentation gaps, but nothing that could not be fixed.
Back on the topic of text-wrangling, I should point out that Ruby is also *very* well-suited for this. So well-suited I'm not sure you'd ever want or need a big book about the subject. Do check out the Ruby Language Home Page.
Slashdot Mirror
I would agree with you if I thought that tabbed browsing was only useful for having lots of windows without the clutter. As it turns out, I use tabs to *organize* my browsing experience. So there are many times when I have a set of tabs open for my teaching materials, a set open for BLAST analyses I'm running, a set for, well, bloggish stuff, and then one for google/deja/yahoo. That could easily be 20 windows, or just 4 sets of 5 tabs. (Note that I *never* quit the browser, and it starts up when I login.)
And as far as "mousing over" the windows in Expose goes...please. What do you think keyboard navigation is for?
I think the best compromise solution is for Expose to give you the option of exploding out your Safari tabs when you select one of its windows. Or, if you use F10, then maybe it auto-explodes them.
In any case, it really does turn out to be life changing. The command that just *really* became extinct, by the way, is "Hide..."
Instead of "mixed" I would say "forward-looking". There is some cost, but Apple has (I think correctly) decided that the cost is only going to be noticeable on hardware of this generation (pre-G5) and before. The benefits of the vector model go on and on forever. So what Expose does for every window on the computer (or in the application) you can expect Safari to be able to do for every tab in your window. For that matter, "Grab" *could* be made to grap a PDF of your screen instead of a tiff (this is actually a bug, I think; also, if you print to PDF from Grab, you only get the visible part of the window; that's *definitely* a bug. Everything can generate useful PDF.
Basically, once you drink the Kool-aid on the vector-based model, I don't think you'd really want to go back.
You're obviously some kind of pathetic geek. :-) More seriously, that *is* very nice, but I should point out that some of the TeX documents I tried playing with turned out to be unsearchable by Safari; is this possibly caused by the fact that they were created in an odd fashion (didn't use pdflatex or something?). Is there something special you should do with your LaTeX to make sure it "plays nice" with more advanced PDF readers?
As expected, the reviewer points out that Expose is really, really cool. More interestingly to me, he also points out that Preview is much improved as well. I used to think it sucked, but now I've made it my default PDF viewer.
The most exciting thing about Preview, though, is that it's easy to predict that its beautiful new "Find" function will make its way into things like Safari. Seriously, "Find" is so good that it has pretty much changed the way I rifle through the dozens of PDFs of journal articles I have on my disk. Like, I can actually find what I'm looking for. :-)
Our university gave out free Panther upgrades to all faculty and staff, which was very decent of them. However, they only gave us *3* CDs, so I don't have the updated compilers or development environment. I'm assuming, therefore, that I can't do anything to upgrade Fink until I can the tech people here to give me a copy of CD 4, right? I knew this would come back to bite me, but since I got what amounted to hundreds of dollars worth of free sofware...
Fortunately, you have people like me around to point this all out to you. :-)
Indeed. And now we can see that UT got a totally killer deal. Why, they paid only about $800,000 per teraflop while Virginia Tech, using horribly over-priced Apple hardware, paid an absurd $300,000 per teraflop of computing performance. As you can see, uh...wait a minute. Hmm...[spin, spin, spin]
Ah. So what do those Hokies think? That money grows on trees? How on earth are we going to see a recovery in the tech sector when people are only willing to pay for cut-rate supercomuter solutions that can't even run Windows natively?
Actually, if you want "psychotically expensive institutional online access", the journals to pick on are anything by Cell Press. If PLOS could replace Neuron, the world would be a much, much less expensive place. :-)
And, because this is Apple we're talking about, you make a movie about it and call it Kill Bill, Volume 3.
Sorry about that.
You just aren't thinking very hard about this then. Teh first journal in the PLoS line-up is PLoS:Biology; the vast majority of articles published here, if they really do make it the equivalent of Science/Nature/whatever *will* be published by people and labs receiving some amount of external funding. Barring that, they could probably apply for intramural funding to defray publication cost in a prestigious journal. After they get off the ground, I have no doubt that one or more philanthropists or corporate sponsors will not start shelling out for this cost. But the point is: there is a real cost to publishing science, and somebody will have to pay it.
As far as the "who pays?" question goes, I think it should be crushingly obvious that this is a big win, especially for people who are underfunded or who come from institutions that do not have large journal budgets. So on-line institutional pricing for Nature is pretty high; I think I remember a figure like $8000 per year being bandied about. There are probably about 5000 institutions paying this fee right now, or about $40 million coming into Nature. There are tens of thousands of others who pay somethingly like $200 per year to get the same access (it can go higher, but you can always get some discount or other). Probably another $40 million or so comes in that way. That's $80 milllion spent for what I believer will work out to be 1000 or so articles. So the total subscription cost is on the order of $80,000 per article published in Nature If these had all been published in PLOS journals, the total subscriber cost would be $1500 per article. Even if I have my Nature numbers high by a factor of 10, there is still a sizable community savings by going the PLOS route. As far as how much PLOS will cost, if they get to the point where they publish as many articles as Nature (say 1000 per year), then I figure they will need an editorial staff of about 10, an office or offices for the same, and whatever their web access costs them (right now they're slashdotted; that shouldn't happen). I don't find it hard to believe that this would cost $1.5 million per year, and so I have to conclude the cost is reasonable. Funds can always be raised to cover reasonable costs.
I'd like to point out that the field of Neuroscience now qualifies as Pretty Darn Big, and, moreover, it is moving Pretty Darn Fast. This is why Elsevier and others can charge hugely for their journals: the demand *is* there, and the cost of *not* having Brain Research or (to pick a non-random example) the Journal of Comparative Neurology can be huge.
Historically, journals like J. Comp. Neur were really expensive because their field (neuroanatomy) basically forced them to have lots of high resolution (and even full-color) plates. This is just not a very cheap thing to do. These days, however, when people are increasingly reading articles on-line, that cost has gone away. A similar issue existed with type-setting. Type-setting is exepnsive. Running a conversion filter over somebody's LaTeX or Word file is NOT. There do remain some real costs: the infrastructure for peer review is not that cheap.
But the unit costs of production have gone DOWN a lot, especially for on-line publication. The fact that journal prices have gone UP a lot just demonstrates that these prices have no connection with publication costs and every connection with being a monopoly supplier of a service in demand. It really is this simple. And, at some point, journals *like* PLOS:Biology (if not PLOS:B itself) really do have a good chance to break the cycle.
In any case, isn't really the size of the production run that does (or will) dominate but plain old market factors. Brain Research costs a ton because you can't substitute another journal to get the articles that they alone have rights to publish, and there is demand for the work that appears there.
So Acrobat (and Safari, and every other sane program) use the obvious cmd+ and cmd- to zoom. Preview psychotically uses the lame cmd-uparrow and cmd-downarrow. These choices are clearly lame because while + and - only have obvious meanings here for magnification, arrow keys should be for navigation. If I asked one hundred naive users what down-arrow should do for a document you're reading, I think it's clear they would all say "scroll down" and *not* "next page".
I find this infuriating because with arrow keys and the command key, you *could* get the nirvana of using arrow keys alone to scroll and pan, and then use cmd-arrrow keys to move by page. As it is now, I'm forced to touch the mouse to pan, and that's just lame. The beauty of Mac OS X for me is that it allows you to control 98% of what you ever need to do from the keyboard, despite having a very handsome GUI.
Other lamenesses of Preview include the lack of full-page mode, and the (current; this will be fixed) limitations to using it with pdf files rather than ps files. And these things are too bad, since *other* features of preview (e.g., "Export...") are quite nice. For a system that is so decisively built on top of PDF, it's very odd that Preview is now a weak link.
So does anybody out there with a preview copy of Panther know if Preview has been improved beyond what they brag about at apple.com?
There's almost always an appreciable delay, even for obviously pioneering stuff. My feeling, though, is that the delay can be longer for breakthroughs that fall under "instrumentation" or "technical" rubric.
The *classic* case of this is this was the fifty-three year gap between the invention of the electron microscope and the 1986 Physics prize being awarded to Ernst Ruska. The two other co-winners that year were IBM's Binnig and Rohrer who won for the "scanning tunneling microscope". What apparently happened here was that the panel wanted to award the prize for the scanning tunneling instrument (which had been invented less than 10 years previously) but then were reminded that Ruska had never won the prize for the invention of the EM. Oops...
In this case, I can conjecture that somebody on the panel was thinking ahead to the (inevitable and eventual) prizes that could be awarded for applications of functional MRI to neuroscience, and then somebody realized that nobody had ever won for the development of anatomical MRI techniques. Oops...
The real problem, alas, is that there are never enough Nobel Prizes available to hand out these days, especially in Physiology and Medicine. So everything that gets awarded comes across as long overdue given the huge backlog of potentially worthy recipients.
First of all, I don't know how everybody involved in these kinds of debates manages to ignore *payroll* taxes, which are just as surely taxes as any other kind of tax, and which fall disproportionately (meaning, a larger fraction of income) on people with the lowest earned incomes. Those taxes have not gone down, although they are in many cases the *majority* of the taxes paid by people with lower incomes. And that's really just the federal taxes. State income taxes, in the states that have them, often have a top bracket at some pathetically low amount; those taxes have not been going down, either.
And then there are sales taxes and gasoline taxes, which end up being a higher marginal rate on lower incomes for reasons that I'm sure should be pretty obvious.
You can agree or disagree with the reasoning behind the Bush tax cuts, but because they were cuts in income taxes primarily for the very highest brackets, there is very little way in which they could not have been tax cuts for the wealthiest.
In a similar vein, the plan to eliminate the estate tax by definition only affects states that are quite a bit larger than the vast majority of estates. By your reasoning, it would be unfair to say that this is a tax cut for the rich because it's a tax cut for the only people paying any estate tax. But those are the rich people. Hence, the point stands.
And those caveats come with caveats...
First, the "proportion in poverty" number has some issues.
It is based on income and not wealth, and the poorest fifth, whether they live in poverty or not, are further in debt now than they were 20 years ago. It would be more interesting to look at "total real disposable income" or some measure that looks at what you've really got after taxes, mandatory debt service, and other things. Also, the poor, whether below the official poverty line or not, also tend to be the uninsured, and the importance of benefits like health insurance is far greater today than it was in the past.
A second issue with the poverty line is just that: it's just a standard threshold that was set with the hope that it was reflective of real poverty. Moreover, it's a standard threshold that you could, if you had the political power, mess around with a bit. In any case, looking at the raw data on income and wealth and household varianbles gives you much more information than looking at an under/over the threshold variable.
The US *does* have rather more mobility than a lot of places, but you have to be really careful in how you look at it. So, if memory serves, the number quoted here has some oddities that may not be apparent. One of them is that for such a definitive sounding number, it has one heck of an error bar around it (50%-80%? sheesh!). Also, the number of people who inhabit an income quintile over the course of a decade is way more than one fifth of the people around. A lot of people are in low income quintiles for reasons that are basically trivial and have little to do with effects of poverty. When I was in grad school, I know that my wife and I were *always* in the lower 20-25% of incomes, and now a decade later I'm doing better than that. Of course, I had a low income because I was getting an advanced degree. Do you really think that mobility is that high for people with high school or less as their top educational attainment?
Beyond that, if X% of the bottom quintile leave it in ten years, Dogbert would point out than other people join it. So imagine a time and place where there is no real progress in incomes, but a given person's income percentile is noisy, and bounced around +/- 10% over the course of a decade. In a situation like that, a goodly proportion of people in the lower quintile would leave it, and a goodly proportion of people in the second lowest quintile would join it, but progress here is (by definition) zero.
Now, real economists know all of these caveats and caveats to the caveats, which is why they never just say something as rosy sounding as "most people in the lowest quintile leave it in 10 years!". Politicians and journalists on the other hand...
In a word, no we don't. Not if by "most of us" you mean "most US taxpayers". For a decently readable account of this and other economic "facts", there's a piece in the NY Times (free registration blah blah).
Actually, I find this *very* interesting since the initial belief that this would work was based on an auditory study, and the fact that internal graphemes can be scrambled (at least many of them) is pretty stunning. I think the problem with using just underlines is not just that it hides the identity of the inner letters, but it makes it more difficult to know the true "length" of the word. If I'm right then,
Txxs sxxxxd be exxxxr to rxxd txxn
t__xt w___e y_u u_e u________s in p___e of t_e l_____s.
Or, to put it another way,
T__s s____d be h____r to r__d t__n
txxt wxxxe yxu uxe uxxxxxxxxs in pxxxe of txe lxxxxxs.
Or maybe not. They're both harder than
txet werhe you do not use udlerneins in pcale of the ltretes.
Now, I know that's a point I've made before, too. But it works even better when you throw numbers out there. For the sake of argument, let us suppose that Apple has a robust 30% margin on the powerbook, so that the $1600 model nets them $370 in gross profit. Profit would then be $370*units_sold. Now let's pretend that they go for a more Dell-like 10% margin. Gross profit now goes to $123*units_sold. So now Apple has to sell three times as many of the suckers to make the same gross profit. The model with the 10% margin now costs only $1353, and there's no doubt that they would sell more of them at this price point. But 3 times as many, for an item for which you suspect the early buyers represent pent-up demand? Doesn't seem too likely to me. That's an extreme scenario, but I think the point is clear. If Apple can get away with their high margins, they'll do it every time. (Note: for simplicity I am pretending that every Mac gets sold at an Apple store so that the gross profit isn't shared; the situation is even more dramatic if you have to share the gross with the retailer who made the sale.)
True, but that's basically my point. The problem with an ULTRA-CHEAP headless machine, which is what we were talking about, is that you have to make sure both that people would buy it, *and* that the people who do buy it would not have bought one of your higher margin products. You and I both probably know that a decently designed G4 headless system with a nice monitor would be a pretty awesome deal. The problem, though, is that it would cannibalize other sales. And that's the problem: if you need to offer a DVI connection to sell it, I think you're very likely to cannibalize sales. If you don't offer the DVI connection, then why should anybody buy this rather than an eMac?
No, that's not it. If you need Windows, the lock-in factor is obvious. If you don't, there are other options (including cheap Linux boxes, by the way). The iBook is very attractive in the cheap notebook market, and the eMac is (now, after price drops) quite capable at $999. The problem is that Apple basically can't do a $500 box that would run Mac OS X acceptably and turn a decent profit, so you're then stuck in this $700+ region where being headless won't compete against ultra-cheap PCs given Windows lock-in, but then $700+ for the slab plus a $200 monitor puts you in the eMac price slot. I think a better option would be to grind the price of an eMac down to $850 or something.
You really aren't looking hard enough, then. Seriously, look at the "monitor line-up" at your local computer superstore and tell me that the ultra-cheap monitors look the same. They really don't, and both Mac OS X and XP make this really obvious. Quality has improved recently, but the real headless market is for people who already own the nice monitor already (then buy a power mac) or those who want to be ultra-cheap. Apple has an offering for the first group, and tells the second group to go for just "cheap" and skip the "purchase really cheap monitor" step. Maybe that's not ideal for growing lowest end market share, but Apple is actually trying to maximize profits, and I can't see that they're leaving much on the table.
I disbelieve. There are two heuristics to tell if something is over-priced. The first is: nobody buys the thing, or you sell fewer than you expected. The second is: you have *reliable* marketing research tells you that a lower price would sell you more than enough additional units to make up for the reduced price. I don't think it's always easy to justify the second course of action. If we pretend that the gross profit on an iPod is about $100, than a $50 reduction in price would halve that number, and we would have to sell twice as many of them to make up the difference. I think you can see the risk here. The iPod is the best seller in its category, and while you can find cheaper hard-drive based players, nobody has come out with a clearly better one. I think the nominal prices will drift down over the next couple of years, but the real price will (and has) sunk like a stone. The original iPod was 5 MB for $500. The top-of-the-line model is now 40 MB (8 times the capacity) and comes with much better interface options. That's huge real improvement in 3 years...
I thought about this, and I came with at least 4 reasons.
OK, so I am even less lazy, and I disbelieve. I did what you said you did, and my dual-Xeon Precision 650 with a USB mouse, USB keyboard, dual-monitor DVI capable video card 120 GB IDE drive (largest IDE drive there; Serial ATA is not an option and the SCSI drive prices gave me the giggles) and DVD-RW/CD-ROM 48X (hope it's writable) drive and no Dell support (to equal the Apple G5) would cost me a total of $3054.10.
Now, I was too lazy to find out if the Dell had gigE ethernet like the Mac, but this is a closer comparison *IF* you are right about the fact that an optimized dual 2 GHz G5 is only equivalent to a 2.4 GHz dual Xeon. If it actually turns out to be anywhere close to a 3.0 GHz dual Xeon, we can stop having this conversation right now, since that box starts at over $4000. Myself, I'm guessing the truth is somewhere in between.
So that makes the G5 about the same performance as a Dell that might run anwhere between $0 and $500 more. "Priced very attractively" is how I would have to put this. Likely not attractively enough to get *many* Windows shops to switch (based on hardware cost and software legacy alone anyway), but attractive enough to keep market share and probably then some.
Uh...what precedent? Your scenario is that SCO settles with Mandrake, and there is no ruling on the merits of the case. Maybe I'm raving, but I thought "precedent" means that you won in court. Now, you could try to engineer that scenario, but I doubt that it would carry much weight to wave such a ruling, from a case where the defendant could not afford to defend and did not contest, in a new case where the defendent aggressively defends and asks for the case to be thrown out for a dozen different reasons before the first court hearing. Of course, I'm not a lawyer, so maybe I don't know what I'm talking about here.
OK, so others have pointed out that you get two hard drive bays, and the SuperDrive is standard, making that a bit less of an issue in my mind.
About the 9 fans, if they turn out to do a better job of managing heat than the 747-style fans in the dual-G4 I've got on my desk, they could put 15 fans in there and I'd still be thrilled. Seriously, the G4 case is splendid, but the cooling system on the latest dual processor models...somebody should have been shot for it to be this loud. (Okay, I'm fine now...)
Another point about the apparent smallish number of hard drive bays is that I think this makes sense because we have crossed an important threshold:
From now on, you'll want a disk or two at most on a desktop. If you need more storage, drink the kool-aid and get yourself a proper RAID.
Honestly, the two disks the G5 will hold can bring you up to half a terabyte. Anymore than this, and arguably even *this* much, and what you really want is either a big old server in a closet somewhere or a RAID appliance of some kind. If you want to use the G5 itself as a server, as luck would have it, Apple also announced a nifty fibre channel PCI card that lets you attach the XRAID among other amusing devices.
*My* only objection to the G5 machines is that I don't have one. (sniff)
I am not sure what you mean by Ruby being "less mature"; as a language in this niche, it appears (to me) to be among the most mature. It does have a smaller user community right now, and there are some library and documentation gaps, but nothing that could not be fixed.
Back on the topic of text-wrangling, I should point out that Ruby is also *very* well-suited for this. So well-suited I'm not sure you'd ever want or need a big book about the subject. Do check out the Ruby Language Home Page.