With 64-bit you can address over 4 terabytes. Do you feel the need for more than that?
OK, folks, a 64-bit address goes up to 2^64, which
is 2^4 * 2^60. Crudely, that's about 16 * (2^10)^6, or 16 * (10^3)^6. Now let's review our
metric prefixes, shall we?
kilo = (10^3)^1
mega = (10^3)^2
giga = (10^3)^3
tera = (10^3)^4
peta = (10^3)^5
exa = (10^3)^6
So, yes, a 64 bit processor can address more than
4 terabytes. Roughly 4 million times as much as that, actually. That could be
of some importance.:-)
More seriously, I can foresee within 5 years the
certainty that addressing 4 terabytes would not
be enough. Indeed, you could predict somebody
would whine about gnu tar's 4 terabyte limit, and how they now can't back up their RAID
full of pr0n.:-)
I'm aware of the 3-5 year cycle in replacing hardware.
Well, I knew that, but I wasn't sure that everybody else out ther reading this would be.
Software does not follow suit.
What would be the price for swapping/maintaining all the software on a 3 year cycle? Assuming they stick to MS Office, is there a "cross-platform" upgrade? Will they have to buy it AGAIN?
Actually, I think this the key point. MS has
been frantically trying to get everybody aboard
their new licensing bandwagon to *make sure* that
there really is a software upgrade treadmill. I
think it is fair to say that response to that
initiative has been...mixed. There are more places out there "looking around" now than there
have been.
As far as the MS Office question, I'm guessing that has to be negotiable. So, around here, Office is Office is Office. You pay your money,
you get your PC or mac version and a lovely certificate (suitable for framing:-)) and you
are in business. Other places might not have
asked for or gotten the cross-platform deal, but
I suspect that they could.
The other problem is perception. Most corporate types don't see the iMac as an office computer. It is a cute, candy-colored home unit. G4 towers are office units. Yes, it is overkill and the iMac would be a wonderful unit but the perception is still there.
Well, the candy-colored iMac was indeed candy-colored. These days you can get any color you like as long as it's white.:-) More seriously, your average iMac on a desk looks just
like a really nice LCD monitor, and lord knows
that enough of those are popping up. Now, where
I think there may be a problem is:
Besides, the price of the Dell w/MONITOR is close to the iMac. Most companies don't rotate monitors out as fast as PCs -- they stay until they die.
Yup, that's a very real potential problem with
the iMac in big companies: you can't do the hardware rotate/monitor stays thing. I know nothing on the inside, but if Apple *isn't* thinking of making separate "stalk" LCD
displays and cheaper brick CPUs again...well they should. The cube was a failure because Apple was
at least a year ahead of the curve on LCD monitors, while the Cube itself was waaay too
exotic looking. I think Apple should start a new
RABID design line (Reduced Asthetics for Business,
Industry, and Defense) for this market.
Besides, the price of the Dell w/MONITOR is close to the iMac. Most companies don't rotate monitors out as fast as PCs -- they stay until they die.
Well, AspectJ [aspectj.org] is doing pretty well, and a new 1.1 version with badly-needed incremental compilation is in the works - it just so happens that, yes, the Xerox PARC research center was spun off as a separate company by Xerox to cut costs. But AspectJ is certainly still alive!
Thanks for the information on this. I haven't really been keeping up with the AOP side of things
much. I do remember looking at some of the original work and thinking "My God, this is great,
but how will they ever make it take off?" Guess
I know the answer now.:-)
The fact is, for a business to install OS X they must trash most of their existing hardware, and that just isn't going to happen.
Well, the funny thing here is that most large organizations end up trashing about 20% of their hardware in a given year anway. (Actually, many places run on a 3-year cycle.) This is why I think it was very important that that Jaguar (Mac OS X 10.2) works so much better
in Windows shops than previous versions. You now could conceive of buying some Macs and adding them to the mix. Most Windows users I know who have seen Jaguar are not unimpressed; the big gripe is that they might be missing something performance-wise on the hardware end. In reality,
they pretty much won't be for anything most users do, and those who do need more cycles will probably have much stronger choices in the Mac line within the next year.
Of course, to be perfectly honest, the best way to get anybody to switch to a Mac is to show them your 17" LCD, since Apple makes *very* nice LCD monitors...
Purchase prices means a LOT when a company is talking 50, 100 or 5,000 machines. When a PC will do the same thing for $500 - $1,000 less than the Mac, no accountant is going to approve $2.5 - $5.0 million *more* on the basis of OS X.
Most big companies get hardware from Dell, and the differential on Optiplex hardware versus an iMac
isn't as large as you might think (disclaimer: when last I checked). There's also the question
of costs down the line, and accountants are now
*very* aware of Microsoft licensing costs these
days, as well as how many tech positions it seems to take to keep N windows boxes in order.
Anecdotally, the number of Macs at my place of business is pretty small, and the official line is that they aren't supported. Which has not ended up mattering, since nobody who has a Mac has ever needed support...
The company can be found at http://intentionalsoftware.com/ [intentionalsoftware.com] with some vague-but-cool-sounding stuff about changing the world.
Now the interesting thing I found out there is that the *other* founder is Kiczales, a Xerox PARC person who was a prime mover in the Aspect-Oriented programming movement. So it looks
like we have here is a start-up featuring really
smart people whose efforts to do world-changing programming tool/language research did not get anywhere in the large companies they previously worked for. Or something like that.
The success rate for start-ups is not very high,
but this is at least an interesting sort of venture, unlike so many of the dot-coms of the
past few years.
Actually, with all of the new Macs the developer tools are on the harddrive in a package. It only takes a double click.
Well, I'll be a monkey's distant cousin. You're right.:-) (For my fellow clueless: check in
"/Applications/Installers/Developer Tools".) And here I was, wondering when I'd remember to haul my Developer's tools disk into work (or wait for all bazillion megs of software to download from Apple) to get those up and running on my new G4. Now, you'd think that would get more prominently mentioned somewhere...
Of course, the Developer Tools get even more interesting when you realize that you can use
Interface Builder with Perl (using
CamelBones and with Ruby using the RubyCocoa
stuff, and probably with Python using a tool
taht I haven't tried to look up yet.
I think the only thing that would make me happier is if there turned out to be an easy-to-use and universal method to navigate around buttons and fields in dialog boxes.
Does the "Full Keyboard Access" stuff found in System Preferences under keyboard settings count? If not how does it fall short? (I don't use it myself)
The "full keyboard access" stuff can sort of do
this, but it still seems to have some issues.
In particular, with a busy dialog it can take you forever to cyle around to the (non-default) "cancel" button you really did want to hit, since
the decided to optimize text entry/radio-button
entry items over just plain buttons. That said,
it's definitely possible I'm still missing some
magic incantation. Documentation on most of this
stuff, is, as you said, pretty flimsy.
Anyway, my latest revelation was that my iBook
did indeed have "page up" and "page down" buttons
(fn arrow keys), which means that I can now cycle through Mozilla tabs. Woo hoo! (So does anybody
see a pattern here?:-))
OK, so I have an interesting situation. I've used
Macs since 1984, but I'm a fast typist who does NOT
want to be mousing around all the time. Given that background, the following things intrigued me:
Prior to 10.2 the "hide all other apps" didn't have a consistant short cut, but it was always in the same place in the application menu (second manu on the bar, between the apple and the file menus).
OK, so the one thing that continues to annoy
me about even Mac OS X is that keyboard shortcuts are apparently not believed to be that important, despite the exhortation in this "Switch" document
that you shouldn't, um, "innovate" there. Frankly, something that should be in every app
should *certainly* have a keyboard shortcut. I
know that's personally why this (very useful!)
feature didn't bubble to the top of my "things to know" list.
If you wanted that app un-hidden, you can return to the last app by doing Command-Tab in 10.2, or prior to 10.2 the shareware HotApp program let you use Opt-tab for that.
You have made my week!:-) Seriously, this almost completely vanquishes my
desire to have a virtual window manager available.
One of the big pains of all time in the Mac interface has always been having to mouse around for other running apps. Not any more! Not only does Command-Tab give you back your last app, it actually cycles through all the apps you have running on the dock. At least,
that's what it appears to do. I'm speechless with
joy! I really am!
But, to paraphrase Dr. Strangelove, this amazing
feature only saves the world if Apple tells people about it! This wasn't on the top Anything list of groovy new Jaguar features people were touting, but it's worth *at least* $50 to me all by itself!
I think the only thing that would make me happier
is if there turned out to be an easy-to-use and
universal method to navigate around buttons and fields in dialog boxes. That's something I still
truly miss.
One point that the review missed that surprised me
was exactly how spookily good the "networking" under system preferences has now evolved to become. As the saying goes...it just works. The other day
I had to tweak some stuff with wireless networking on a ThinkPad. It wasn't that horrific, since I did sort of know what I was doing, but I couldn't help but think "this really sucks compared to Jaguar...".
And I'll spare you the account of my tears of joy
when USB printer sharing both a) worked again and
b) was way more pleasant than under MacOS 9. Wow.
You really have to see this to believe it.
"An API for a real honest-to-god VWM. Space works for now. I guess."
These are terrible ideas. The idea behind OS X was to take Unix and make it act like a Mac, not take a Mac and make it act like Unix.
What you say? Virtual window managers have NOTHING to do with Unix. Heck, I could argue that the first VWM I ever saw was...Switcher, which
worked on a Mac 512KE back in 1985 or 1986. (Yes, Switcher was more than that, but let's focus on
that aspect of the thing.) One of the most hugely
sucky things these days on an otherwise great OS X 10.2 is being forced to have overlapping windows.
With my VWM set-up on a Linux box in my lab, I
can (and do) have a 3 x 3 screen containing up
to 9 applications. Now it's true that with the
tabbed browsing in Mozilla, I don't *need* 3 browser windows like I used to, but I sure as heck
still want a screen of nothing but emacs and terminal windows that can safely use 18 point fonts. Really, it's a very simple concept here:
the only thing you see at any given time should
be things that need to be seen together. Overlapping windows are a hack; VWMs are not.
Seriously, I thought my question was good, but the
questions that got answered *very* thoughtfully
were all better. I'm sincerely impressed with
the way this particular slashdot interview went, and I have to say that this gives me a bit more faith that slashdot will continue to be a real forum that dishes news to nerds...stuff that matters.
As far as the Larry Wall answers go, I think the most important one had to do with testing one's own slogans. So while it's safe
to say that TIMTOWTDI is still strong, I was floored by Larry's admission that there was indeed something to think about in the "Perl makes easy things easy and hard things possible" line. Now, if there's time and energy to ponder another imponderable slogan, I'd suggest a second look be given to the "perl is a diagonal language" (rather than an orthogonal language) notion. That one has always worried me. If I had to get picturesque about it, I would observe that a Bishop in chess is the most Diagonal piece, while the rook is slightly more powerful for being Orthogonal. But, of course, the Queen combines both qualities, and is more powerful than the other two put together. Perl should be like that (and Ruby tries to be like that).
OK, so that one *is* pretty cheesy, but I think that it might be true.:-)
Re:I am not Larry, but easy answers
on
Ask Larry Wall
·
· Score: 2
The use of * for array flattening in Perl 6 is stolen straight from Ruby.
Well,yeah. I was also thinking of the move to "dot" syntax for "methods" (lots of languages do this, but Ruby was the closest in other ways), some aspects of the new "module" keyword (disclaimer: I have not gone over the exegesis on this as carefully as I might have), and other things that don't come to mind at this instant since I'm making sure Milo doesn't tear down the house while I type this...
Ruby's lack of strict.pm, and the optional semi-colons are both big drawbacks.
I don't consider the semi-colons optional.:-)
I haven't misses strict.pm yet; dunno what that says about me. A big plus for me with ruby over python is actual delimiters for "block-like" things. I like the Perl6 notion that {} always
creates a lexical closure, except that it has forced other syntax for things you would think are in braces...until you remember the closure bit.
A philosophical difference: Perl retains its "syntax and operators indicates typing, with autoconversions where needed" philosophy where Ruby is quite intentionally opposite with its "uniformity of operaters and syntax with different datatypes defining each operator as makes sense to it.
Well, I personally now consider a lot of the automagical conversions of Perl to be a failed
experiment. What looks nice in a short script here really, truly falls down in the large. The dorkiest thing about Ruby, though, is the "everything including an integer is an object" notion. Integers are values; they do not have state.
Thanks for the feed-back, non-Larry.:-)
Perl and Ruby
on
Ask Larry Wall
·
· Score: 5, Interesting
In the beginning, I programmed in awk. I lived
life one line at a time, but it was good. Then
somebody turned me onto perl, and life was much more than good.
At that time, there was no credible competition to Perl in any of the niches it basically created. These days, there is more competition than I can comfortably list. Indeed, if I were choosing a language like Perl today, I would be very, very tempted to choose Ruby instead, and I am not the only Perl programmer who feels this way. Interestingly, Perl6 is beginning to look and feel a lot more like Ruby. Are there indeed aspects of Ruby that you were deliberately trying to have in Perl6? Are there any aspects of Ruby you are especially wary of?
On a vaguely related note, my dual 1GHz PowerMac, which was ordered on 8/13 and quoted to ship in 3-5 business days took 8 business days to get to shipping (Jaguar will be pre-loaded).
Will be? Are you still waiting for yours? I haunted the local CompuUSA for a mere three days until they got their second shipment in (one day late). The 1gig I got had Jaguar loaded.
A careful reading of my post does suggest that I
am still waiting, since the box only just got *to* shipping (i.e., FedEx ground). I've now
got a compulsive tick where I check its shipping status within Fedex every hour or so. It departed the FedEx sort facility in Sacramento
on 08/24/2002 at 04:47. Everybody claims it will have Jaguar pre-loaded.
I wonder if the "update" is a regular boxed version. Then I could sell it on ebay. Heh. Except I don't want to bother. And it would probably be the "coupon version" anyhow.
I'm not really sure how they're doing this. I have to send in for the update for some other
recently arriving boxes. The local store here
doesn't seem to think that they will end up playing any role in this. (Which is part of the
thing that pisses them off, since they *know* that some people will just bypass them for the on-line Apple Store for their current stock that does not ship with Jaguar today. It's basically a college bookstore place, and they had to pre-order their dozens of iBooks and and PowerBooks to have stock on hand for the big Fall rush, and it's looking a lot like dead inventory
at the moment due to the OS update fiasco.)
Killing the floppy with the iMac was another (great that they killed it - absolutely stupid that they didn't *replace* it).
You mean with a CDRW standard in all Macs? Yeah. Sucks that they didn't add that feature....
What sucks is that it took them 3 years for writable CD media of any kind to be standard on all Macs. Seriously, the iMac dates from 1998, and I don't think CD-R(W) was stock on everything until last year. Actually, Jobs famously had to say "we screwed up on that"
when analysts asked him what took them so long in
getting writable optical media into their product lines. I've got an (otherwise quite nice) iMac DV SE that has a DVD-ROM drive...but no built-in CD burner. It really was a bizarre screw-up.
To be frank, they really should stop shipping anything less than a combo drive (CD-RW/DVD) any minute now, and stop shipping anything less than a super drive (CD-RW/DVD-R(W?)) by next year.
If you bought 10.1 (either by itself or on a new computer) within the past month or two, you can get an upgrade either for free or for $20 - I'm not quite sure - either way, they mail you a CD, you can't download it.
Updates are $20 for computers bought between July 17 and August 24; getting the upgrade is a bit tricky since the upgrade form is accessible only via a tab that is almost unreadable thanks to the !@#$@#$ Jaguar skin "theme" on the OS X pages at www.apple.com.
For computers bought on or after August 24, things could get sticky. At our local (college) computer store, you might not get Jaguar with hardware sold after August 24, and the store manager types are really pissed that the customer in this case has to do ther upgrade drill mentioned above, rather than having something like the 10.1 upgrade "sleeve" to hand out with the new hardware. If Apple has any brains, they will start doing something like that in all sales locations, otherwise they could get a ton of hardware ordered in early August returned.
On a vaguely related note, my dual 1GHz PowerMac,
which was ordered on 8/13 and quoted to ship in 3-5 business days took 8 business days to get to
shipping (Jaguar will be pre-loaded). This may or may not be an early indication that deliveries could be slow on this hardware, although the online Apple Store never updated the availability of this hardware on the web site.
I never tought that such a thing would be possible, but the court decision is actually a good read.
Not all decisions are very interesting, of course, but this one did seem to stand out as being particularly lucid. I have noticed a few people posting on this thread who claimed that the Internet did infringe and that they wanted the patent invalidated due to prior art. I am almost certain that these people did not read the decision carefully, because it really does completely demolish any idea that infringement *of this patent as granted* ocurred. The prior art about hyperlinks etc. was really besides the point because the BT patent did not end up
describing hyperlinks anyway. Really,
this ruling is probably better and fairer than declaring the patent invalid, because, as far as I can tell, there really was a novel invention that the patent does describe...and which nobody is especially likely to infringe on since we just don't go around providing button interfaces to access specific hard disk track/sector data on centralized computer systems. For a few months there, the patented idea may well have been very useful and all, but it is irrelevent to any
modern practice that I know of.
I don't think OS X should be weighed so heavily against OS 9. For every feature that we lost in the transition to X, we gained two improvements.
I agree with you in many ways, but...
Alas, the two things *I* lost going from OS9 to OS X were USB printer sharing and Airport basestation, and correcting those were (potentially) big ticket items. I did go out and buy a WAP to solve the Airport issue (Mac OS9 software base station was always a bit quirky), but the lack of USB printer sharing was pretty
annoying unconditionally.
To put it another way, if you want to increase adoption rates for Mac OS X among the SOHO group in particular, a really bad strategy is to break parts of printing and wireless networking. Yes, we survived, but I think the gripes here are legitimate. (Compare with: "I need more RAM to run OS X; wah!" and the like that we did see back in the day.)
... is a way of creating hypertext docs with embedded logic. Like a scripted web site, only less accessible.
Well, it could be used like that, for sure. Every great programming idea out there seems to have at least one controlling metaphor that is so striking that you have to pay attention to it. Now, most people *think* the controlling metaphor of Hypercard is simply "life is a stack of cards!",
but that is actually more of an implementational
detail. In one of the Wired articles referred to
in this thread, Bill Atkinson mentions explicitly what I always thought was true and obvious in retrospect: Hypercard "stacks" could have been
spread over the network, and something eerily like
the World Wide Web could have come on the scene
at least 5 years earlier.
Now what I think was the most important advance
made by Hypercard was the idea that "Everything
is an object or a message, and the way to handle
messages in a GUI is via delegation, and you really don't need inheritance at all." Seriously,
I think it is still hard to see how forward thinking Hypercard really was in this respect, until you browse some W3C documentation struggling to define how to handle user events in and around the DOM. In Hypercard, any object could handle
any appropriate message itself and/or pass it up
the chain to it's "enclosing" object, or pass it
to any object of its choosing. (You couldn't
always choose the first target of an event like "mouseup", but you could always delegate to the
right object in the long run, which is key.) When given this much power, you could do almost any fool thing and that's what people actually did. So I remember when I first implemented EMACS-style editing commands for the message box in Hypercard
(including a rudimentary kill-ring!); nobody had ever dreamed of doing anything so nutso, but there
it was! I also remember starting on a chess tutor-like interface that would allow a novice to
see all of the possible next moves for a piece;
that never went anywhere under my power, but the
proof of concept did impress a few people who had
no idea you could do anything like this with software that came for free with your Mac.
On the other hand, I can't write a complete love letter to Hypercard in this post since the memory of the thing immediately brings to mind the most
hideous misfeatures of Hyper*Talk*, the language.
Like the no real variables part, or the fact that
you could not escape to a less verbose syntax, or
the fact that since it lacked explicit "anchors"
like HTML that some kinds of links were truly painful to construct... But HyperCard was there,
and it was great, and now it's gone. And the real shame is that nobody in the Apple chain of command really knows why it was great: it was an astonishingly powerful programming environment that made some people absurdly happy and led to at least as many Apple hardware purchases in the late 80s as anything else. (Once somebody had a stack
that did X, you were better off buying more Macs than trying to re-do X some other way...) That
is just my opinion, of course, but any geek who
delights in the fact that Mac OS X is basically
just BSD under the Aqua seas might understand the
glee of users who suddenly found out that they could do powerful object-oriented programming with a piece of "address book" software added to the system almost as an afterthought.
Most kitchen shows these days are obviously supported by corporate sponsors - whose business apparently depends on pressuring people into buying needlessly specialized gadgets for the kitchen.
Actually, my take on the Food Network's commercials and sponsors is that they *aren't* makers of very specialized gadgets (okay, mixers and blenders) but rather of 3 other categories:
Big-ticket items like stoves
Wine, designer ice cream, and the like.
Stuff I have no idea why it's there, like Lean Cuisine frozen dinners...stuff that is basically
the *opposite* of cooking your own food well.
For that matter, we know that most people who buy Viking stoves do not actually cook very much, and my disturbing suspicion is that 90% of the people who watch shows on the Food Network really aren't as into cooking as you might assume. I guess the bright side of this is that 10% are really, truly
into it, and people like AB really matter to this bunch, but I have always wondered whether the profile of the average Food Network watcher bothers the on-screen talent that does work so hard to create great food.
Lots of carrying cases. Lots of fonts. Not very much software.
Ah yes; that takes me back.:-) One other notable piece of software was Megamax C, and the amazing debugger whose name now escapes me...
I got Stephen Chernicoff to sign my copy of "Macintosh Revealed."
Me too!:-)
Someone was demoing software that created a small amount of RAM cache for the floppy drive. If you had a whopping 512K of memory, that RAM cache actually could speed things up a bit.
OK, so was MacBottom (hard-drive that fit under
your Mac) at that first Expo?
What else was there? Overvue, from Provue Development, I think... Filevision from Telos, which was really mindblowing at the time.
I believe that RecordHolder may have had or shared a small booth.
I believe it was the 1986 MacWorld that had the huge inflated Macintosh outside promoting MacPublisher, a very early desktop publishing product.
I think 1986 was also the debut of Fontographer;
Aldus PageMaker was out in 1985, but I'm not sure if it was out in time for the Expo.
In the first few years, MacWorld was really great. You could belly up to a booth and really try out and learn about new software. The people exhibiting the software generally knew a lot about it and were often developers.
In either 1985 or 1986, a weird British bearded
guy was demo-ing this wild and nutty computer
algebra system for the (not so large) masses.
Guy's name was "Wolfram", like the element, and
his software was called "Mathemagica" or something like that.:-)
Also in 1986 I believe was the introduction
of this wild and crazy product called "HyperCard",
a product that was so important but so ahead of
its time that many younger people i talk to these days can't believe there was anyting like this in
the bad old days.
But my favorite early MacWorld was 1987, since
I basically had to sneak out from under my soon-to-be-inlaws in order to attend the thing...3 days before I got married in another state. I
randomly ran into a Boston Globe reporter who made me part of her story in the paper on how...hyper-dedicated macophiles were in those days.
Ah, well... MacWorld may come back to Boston, but it will never be like 1985.
No, but neither you or I will ever be the same,
either.
Why does it piss you off that your new iBook is less than 2 months old by the time 10.2 comes out? You bought that iBook for the functionality it had, and not the functionality it might have in the future--and if you didn't, that's your fault and not Apple's.
One might reasonably have expected that any 10.2 upgrade price would be less than the full price (even Microsoft does that). Or, perhaps one
might have assumed that a computer company that
makes most of its money on hardware sales would not do something that will tend to encourage people not to buy a new Mac when an OS upgrade is expected "soon" but not announced. Or, perhaps Apple might have
considered that people who already own multiple Apple systems will be less likely to upgrade or buy more new Apple hardware if the cost of upgrading system software (which used to be notably cheaper than Windows) gets too high. Or, Apple might have realized that owners of their computers tend to be idealistic zealots who will
turn on them very quickly if pissed off, even if
you could give them a brilliant economic argument for charging them more.
Frankly, Apple cannot afford to tweak its ownership base too much. They famously raised prices on one very popular model this spring and the scratched their heads when sales came in below plan after that. Heck, I doubt that Mac sales are too great even right now since you can save at least $20 by waiting until after August 24. Petty? Sure, but humans are like that. Apple really could have handled this better; they had
no reason to spring the "bad news" on pricing during one of their big positive press events of
the year. I mean, what were they thinking? People came out of the Jobs keynote
feeling a bit let down about new hardware, but
feeling good about OS X 10.2, and then the realization sinks in that they're going to fork
over hundreds of dollars (if they have more than
one Mac) to upgrade. Really, was that the best
plan?
Now, here's what bothers me to some extent, and by no means do you have to agree with me on this issue. But according to those numbers, this company is profitable. Granted, its not profitable enough to justify the high price it was bought back in April of 2000, but its in the positive, and appears to be staying there for awhile.
The problem with the argument here is in the blind acceptance of the statement that ZD is
truly profitable. As everybody should be quite
thoroughly aware of these days, GAAP can be used
to play with earnings to a disturbing extent even
before you get to outright Enron/Worldom style
fraud. If you read the whole NYT article, you can see that they are currently in a cash crunch,
which is what will force the filing. Yes it might seem weird that a company could "produce"
a profit up until the point it files for chapter
11, but it does happen (I think it used to be
normal for this to happen with banks that failed). For that matter, they *aren't* profitable now. You write:
The company's earnings before interest, taxes, depreciation and amortization are projected to be $6.5 million in 2002.
First off, that isn't earnings, but EBITDA, which
is wildly significant because their interest expense is absolutely huge. Further, the trends for future business here have to be strongly negative. I severely doubt, for example, that year-over-year ad revenue is up or will be up in the foreseeable future. I cannot believe that subscriptions to the flagship (PC Magazine) are going anywhere but down (unless they have expanded some of the "loss-leader" deals they used to offer).
And, for that matter, I'm very skeptical about
the breath-taking turnaround they suggest is going to happen after the pre-packaged chapter 11 filing. Note that they are pretty much *forced* to paint a rosy scenario here because they are trying to get their bondholders to accept equity in lieu of cash. For that
matter, chapter 11 does give them some leverage
here over the bondholders who already rejected
a similar deal (again, this is clear from the NYT
piece). When push comes to shove in bankruptcy
court, I wouldn't be surprised to see monthly operating reports deep in the red if they stay
there for long, and the bondholders (and any banks) are going to get heart pretty badly even
if they take the equity. OK, so that's my opinion based on the expected current value of any enterprise purchased in April 2000.
I have no current interest in the equity or debt of ZD or any of its partners or subsidiaries.
So you think you are old?....I'm not old (26) but HD didn't exists on the consumer market when I started.....my first real personal computer was an Apple II. and then the macintosh. Anyway, just remember how much application you could put on a single 800k floppy.
TI-99/4A [99er.net]!:) I had a 5.25" single-sided single-density floppy disk drive, with a whopping 90k per diskette.
Floppy drives (and both Apple and TI hardware) were for wusses. Real men used Scotch brand cassette tapes with the "data recorders" on the radically new and innovative VIC-20 system. We
ruled, not like those johnny-come-lately scum
with their Radio Shack Color Computers. (Sorry
to have to mention the members of that particular
psycho cult; if you want to revisit the horror,
see the web site of the kookoo CoCo High Priest. The best link on that page is about an innovative 4 MHz acceleration project they are putting together. I'm not kidding.)
Slashdot Mirror
OK, folks, a 64-bit address goes up to 2^64, which is 2^4 * 2^60. Crudely, that's about 16 * (2^10)^6, or 16 * (10^3)^6. Now let's review our metric prefixes, shall we?
So, yes, a 64 bit processor can address more than 4 terabytes. Roughly 4 million times as much as that, actually. That could be of some importance. :-)
More seriously, I can foresee within 5 years the certainty that addressing 4 terabytes would not be enough. Indeed, you could predict somebody would whine about gnu tar's 4 terabyte limit, and how they now can't back up their RAID full of pr0n. :-)
Well, I knew that, but I wasn't sure that everybody else out ther reading this would be.
Actually, I think this the key point. MS has been frantically trying to get everybody aboard their new licensing bandwagon to *make sure* that there really is a software upgrade treadmill. I think it is fair to say that response to that initiative has been...mixed. There are more places out there "looking around" now than there have been.
As far as the MS Office question, I'm guessing that has to be negotiable. So, around here, Office is Office is Office. You pay your money, you get your PC or mac version and a lovely certificate (suitable for framing :-)) and you
are in business. Other places might not have
asked for or gotten the cross-platform deal, but
I suspect that they could.
Well, the candy-colored iMac was indeed candy-colored. These days you can get any color you like as long as it's white.Yup, that's a very real potential problem with the iMac in big companies: you can't do the hardware rotate/monitor stays thing. I know nothing on the inside, but if Apple *isn't* thinking of making separate "stalk" LCD displays and cheaper brick CPUs again...well they should. The cube was a failure because Apple was at least a year ahead of the curve on LCD monitors, while the Cube itself was waaay too exotic looking. I think Apple should start a new RABID design line (Reduced Asthetics for Business, Industry, and Defense) for this market. Besides, the price of the Dell w/MONITOR is close to the iMac. Most companies don't rotate monitors out as fast as PCs -- they stay until they die.
Thanks for the information on this. I haven't really been keeping up with the AOP side of things much. I do remember looking at some of the original work and thinking "My God, this is great, but how will they ever make it take off?" Guess I know the answer now. :-)
Well, the funny thing here is that most large organizations end up trashing about 20% of their hardware in a given year anway. (Actually, many places run on a 3-year cycle.) This is why I think it was very important that that Jaguar (Mac OS X 10.2) works so much better in Windows shops than previous versions. You now could conceive of buying some Macs and adding them to the mix. Most Windows users I know who have seen Jaguar are not unimpressed; the big gripe is that they might be missing something performance-wise on the hardware end. In reality, they pretty much won't be for anything most users do, and those who do need more cycles will probably have much stronger choices in the Mac line within the next year.
Of course, to be perfectly honest, the best way to get anybody to switch to a Mac is to show them your 17" LCD, since Apple makes *very* nice LCD monitors...
Most big companies get hardware from Dell, and the differential on Optiplex hardware versus an iMac isn't as large as you might think (disclaimer: when last I checked). There's also the question of costs down the line, and accountants are now *very* aware of Microsoft licensing costs these days, as well as how many tech positions it seems to take to keep N windows boxes in order. Anecdotally, the number of Macs at my place of business is pretty small, and the official line is that they aren't supported. Which has not ended up mattering, since nobody who has a Mac has ever needed support...
Now the interesting thing I found out there is that the *other* founder is Kiczales, a Xerox PARC person who was a prime mover in the Aspect-Oriented programming movement. So it looks like we have here is a start-up featuring really smart people whose efforts to do world-changing programming tool/language research did not get anywhere in the large companies they previously worked for. Or something like that.
The success rate for start-ups is not very high, but this is at least an interesting sort of venture, unlike so many of the dot-coms of the past few years.
Well, I'll be a monkey's distant cousin. You're right. :-) (For my fellow clueless: check in
"/Applications/Installers/Developer Tools".) And here I was, wondering when I'd remember to haul my Developer's tools disk into work (or wait for all bazillion megs of software to download from Apple) to get those up and running on my new G4. Now, you'd think that would get more prominently mentioned somewhere...
Of course, the Developer Tools get even more interesting when you realize that you can use Interface Builder with Perl (using CamelBones and with Ruby using the RubyCocoa stuff, and probably with Python using a tool taht I haven't tried to look up yet.
Life is just grand.
The "full keyboard access" stuff can sort of do this, but it still seems to have some issues. In particular, with a busy dialog it can take you forever to cyle around to the (non-default) "cancel" button you really did want to hit, since the decided to optimize text entry/radio-button entry items over just plain buttons. That said, it's definitely possible I'm still missing some magic incantation. Documentation on most of this stuff, is, as you said, pretty flimsy.
Anyway, my latest revelation was that my iBook did indeed have "page up" and "page down" buttons (fn arrow keys), which means that I can now cycle through Mozilla tabs. Woo hoo! (So does anybody see a pattern here? :-))
Is this a job for Protozilla?
OK, so the one thing that continues to annoy me about even Mac OS X is that keyboard shortcuts are apparently not believed to be that important, despite the exhortation in this "Switch" document that you shouldn't, um, "innovate" there. Frankly, something that should be in every app should *certainly* have a keyboard shortcut. I know that's personally why this (very useful!) feature didn't bubble to the top of my "things to know" list.
You have made my week! :-) Seriously, this almost completely vanquishes my
desire to have a virtual window manager available.
One of the big pains of all time in the Mac interface has always been having to mouse around for other running apps. Not any more! Not only does Command-Tab give you back your last app, it actually cycles through all the apps you have running on the dock. At least,
that's what it appears to do. I'm speechless with
joy! I really am!
But, to paraphrase Dr. Strangelove, this amazing feature only saves the world if Apple tells people about it! This wasn't on the top Anything list of groovy new Jaguar features people were touting, but it's worth *at least* $50 to me all by itself!
I think the only thing that would make me happier is if there turned out to be an easy-to-use and universal method to navigate around buttons and fields in dialog boxes. That's something I still truly miss.
And I'll spare you the account of my tears of joy when USB printer sharing both a) worked again and b) was way more pleasant than under MacOS 9. Wow. You really have to see this to believe it.
What you say? Virtual window managers have NOTHING to do with Unix. Heck, I could argue that the first VWM I ever saw was...Switcher, which worked on a Mac 512KE back in 1985 or 1986. (Yes, Switcher was more than that, but let's focus on that aspect of the thing.) One of the most hugely sucky things these days on an otherwise great OS X 10.2 is being forced to have overlapping windows. With my VWM set-up on a Linux box in my lab, I can (and do) have a 3 x 3 screen containing up to 9 applications. Now it's true that with the tabbed browsing in Mozilla, I don't *need* 3 browser windows like I used to, but I sure as heck still want a screen of nothing but emacs and terminal windows that can safely use 18 point fonts. Really, it's a very simple concept here: the only thing you see at any given time should be things that need to be seen together. Overlapping windows are a hack; VWMs are not.
As far as the Larry Wall answers go, I think the most important one had to do with testing one's own slogans. So while it's safe to say that TIMTOWTDI is still strong, I was floored by Larry's admission that there was indeed something to think about in the "Perl makes easy things easy and hard things possible" line. Now, if there's time and energy to ponder another imponderable slogan, I'd suggest a second look be given to the "perl is a diagonal language" (rather than an orthogonal language) notion. That one has always worried me. If I had to get picturesque about it, I would observe that a Bishop in chess is the most Diagonal piece, while the rook is slightly more powerful for being Orthogonal. But, of course, the Queen combines both qualities, and is more powerful than the other two put together. Perl should be like that (and Ruby tries to be like that).
OK, so that one *is* pretty cheesy, but I think that it might be true. :-)
Well,yeah. I was also thinking of the move to "dot" syntax for "methods" (lots of languages do this, but Ruby was the closest in other ways), some aspects of the new "module" keyword (disclaimer: I have not gone over the exegesis on this as carefully as I might have), and other things that don't come to mind at this instant since I'm making sure Milo doesn't tear down the house while I type this...
I don't consider the semi-colons optional. :-)
I haven't misses strict.pm yet; dunno what that says about me. A big plus for me with ruby over python is actual delimiters for "block-like" things. I like the Perl6 notion that {} always
creates a lexical closure, except that it has forced other syntax for things you would think are in braces...until you remember the closure bit.
Well, I personally now consider a lot of the automagical conversions of Perl to be a failed experiment. What looks nice in a short script here really, truly falls down in the large. The dorkiest thing about Ruby, though, is the "everything including an integer is an object" notion. Integers are values; they do not have state.
Thanks for the feed-back, non-Larry. :-)
At that time, there was no credible competition to Perl in any of the niches it basically created. These days, there is more competition than I can comfortably list. Indeed, if I were choosing a language like Perl today, I would be very, very tempted to choose Ruby instead, and I am not the only Perl programmer who feels this way. Interestingly, Perl6 is beginning to look and feel a lot more like Ruby. Are there indeed aspects of Ruby that you were deliberately trying to have in Perl6? Are there any aspects of Ruby you are especially wary of?
A careful reading of my post does suggest that I am still waiting, since the box only just got *to* shipping (i.e., FedEx ground). I've now got a compulsive tick where I check its shipping status within Fedex every hour or so. It departed the FedEx sort facility in Sacramento on 08/24/2002 at 04:47. Everybody claims it will have Jaguar pre-loaded.
I'm not really sure how they're doing this. I have to send in for the update for some other recently arriving boxes. The local store here doesn't seem to think that they will end up playing any role in this. (Which is part of the thing that pisses them off, since they *know* that some people will just bypass them for the on-line Apple Store for their current stock that does not ship with Jaguar today. It's basically a college bookstore place, and they had to pre-order their dozens of iBooks and and PowerBooks to have stock on hand for the big Fall rush, and it's looking a lot like dead inventory at the moment due to the OS update fiasco.)
What sucks is that it took them 3 years for writable CD media of any kind to be standard on all Macs. Seriously, the iMac dates from 1998, and I don't think CD-R(W) was stock on everything until last year. Actually, Jobs famously had to say "we screwed up on that" when analysts asked him what took them so long in getting writable optical media into their product lines. I've got an (otherwise quite nice) iMac DV SE that has a DVD-ROM drive...but no built-in CD burner. It really was a bizarre screw-up.
To be frank, they really should stop shipping anything less than a combo drive (CD-RW/DVD) any minute now, and stop shipping anything less than a super drive (CD-RW/DVD-R(W?)) by next year.
Updates are $20 for computers bought between July 17 and August 24; getting the upgrade is a bit tricky since the upgrade form is accessible only via a tab that is almost unreadable thanks to the !@#$@#$ Jaguar skin "theme" on the OS X pages at www.apple.com.
For computers bought on or after August 24, things could get sticky. At our local (college) computer store, you might not get Jaguar with hardware sold after August 24, and the store manager types are really pissed that the customer in this case has to do ther upgrade drill mentioned above, rather than having something like the 10.1 upgrade "sleeve" to hand out with the new hardware. If Apple has any brains, they will start doing something like that in all sales locations, otherwise they could get a ton of hardware ordered in early August returned.
On a vaguely related note, my dual 1GHz PowerMac, which was ordered on 8/13 and quoted to ship in 3-5 business days took 8 business days to get to shipping (Jaguar will be pre-loaded). This may or may not be an early indication that deliveries could be slow on this hardware, although the online Apple Store never updated the availability of this hardware on the web site.
Not all decisions are very interesting, of course, but this one did seem to stand out as being particularly lucid. I have noticed a few people posting on this thread who claimed that the Internet did infringe and that they wanted the patent invalidated due to prior art. I am almost certain that these people did not read the decision carefully, because it really does completely demolish any idea that infringement *of this patent as granted* ocurred. The prior art about hyperlinks etc. was really besides the point because the BT patent did not end up describing hyperlinks anyway. Really, this ruling is probably better and fairer than declaring the patent invalid, because, as far as I can tell, there really was a novel invention that the patent does describe...and which nobody is especially likely to infringe on since we just don't go around providing button interfaces to access specific hard disk track/sector data on centralized computer systems. For a few months there, the patented idea may well have been very useful and all, but it is irrelevent to any modern practice that I know of.
But don't believe me; read it yourself.
I agree with you in many ways, but...
Alas, the two things *I* lost going from OS9 to OS X were USB printer sharing and Airport basestation, and correcting those were (potentially) big ticket items. I did go out and buy a WAP to solve the Airport issue (Mac OS9 software base station was always a bit quirky), but the lack of USB printer sharing was pretty annoying unconditionally.
To put it another way, if you want to increase adoption rates for Mac OS X among the SOHO group in particular, a really bad strategy is to break parts of printing and wireless networking. Yes, we survived, but I think the gripes here are legitimate. (Compare with: "I need more RAM to run OS X; wah!" and the like that we did see back in the day.)
Well, it could be used like that, for sure. Every great programming idea out there seems to have at least one controlling metaphor that is so striking that you have to pay attention to it. Now, most people *think* the controlling metaphor of Hypercard is simply "life is a stack of cards!", but that is actually more of an implementational detail. In one of the Wired articles referred to in this thread, Bill Atkinson mentions explicitly what I always thought was true and obvious in retrospect: Hypercard "stacks" could have been spread over the network, and something eerily like the World Wide Web could have come on the scene at least 5 years earlier.
Now what I think was the most important advance made by Hypercard was the idea that "Everything is an object or a message, and the way to handle messages in a GUI is via delegation, and you really don't need inheritance at all." Seriously, I think it is still hard to see how forward thinking Hypercard really was in this respect, until you browse some W3C documentation struggling to define how to handle user events in and around the DOM. In Hypercard, any object could handle any appropriate message itself and/or pass it up the chain to it's "enclosing" object, or pass it to any object of its choosing. (You couldn't always choose the first target of an event like "mouseup", but you could always delegate to the right object in the long run, which is key.) When given this much power, you could do almost any fool thing and that's what people actually did. So I remember when I first implemented EMACS-style editing commands for the message box in Hypercard (including a rudimentary kill-ring!); nobody had ever dreamed of doing anything so nutso, but there it was! I also remember starting on a chess tutor-like interface that would allow a novice to see all of the possible next moves for a piece; that never went anywhere under my power, but the proof of concept did impress a few people who had no idea you could do anything like this with software that came for free with your Mac.
On the other hand, I can't write a complete love letter to Hypercard in this post since the memory of the thing immediately brings to mind the most hideous misfeatures of Hyper*Talk*, the language. Like the no real variables part, or the fact that you could not escape to a less verbose syntax, or the fact that since it lacked explicit "anchors" like HTML that some kinds of links were truly painful to construct... But HyperCard was there, and it was great, and now it's gone. And the real shame is that nobody in the Apple chain of command really knows why it was great: it was an astonishingly powerful programming environment that made some people absurdly happy and led to at least as many Apple hardware purchases in the late 80s as anything else. (Once somebody had a stack that did X, you were better off buying more Macs than trying to re-do X some other way...) That is just my opinion, of course, but any geek who delights in the fact that Mac OS X is basically just BSD under the Aqua seas might understand the glee of users who suddenly found out that they could do powerful object-oriented programming with a piece of "address book" software added to the system almost as an afterthought.
Actually, my take on the Food Network's commercials and sponsors is that they *aren't* makers of very specialized gadgets (okay, mixers and blenders) but rather of 3 other categories:
For that matter, we know that most people who buy Viking stoves do not actually cook very much, and my disturbing suspicion is that 90% of the people who watch shows on the Food Network really aren't as into cooking as you might assume. I guess the bright side of this is that 10% are really, truly into it, and people like AB really matter to this bunch, but I have always wondered whether the profile of the average Food Network watcher bothers the on-screen talent that does work so hard to create great food.
Ah yes; that takes me back. :-) One other notable piece of software was Megamax C, and the amazing debugger whose name now escapes me...
I got Stephen Chernicoff to sign my copy of "Macintosh Revealed."
Me too! :-)
Someone was demoing software that created a small amount of RAM cache for the floppy drive. If you had a whopping 512K of memory, that RAM cache actually could speed things up a bit.
OK, so was MacBottom (hard-drive that fit under your Mac) at that first Expo?
What else was there? Overvue, from Provue Development, I think... Filevision from Telos, which was really mindblowing at the time.
I believe that RecordHolder may have had or shared a small booth.
I believe it was the 1986 MacWorld that had the huge inflated Macintosh outside promoting MacPublisher, a very early desktop publishing product.
I think 1986 was also the debut of Fontographer; Aldus PageMaker was out in 1985, but I'm not sure if it was out in time for the Expo.
In the first few years, MacWorld was really great. You could belly up to a booth and really try out and learn about new software. The people exhibiting the software generally knew a lot about it and were often developers.
In either 1985 or 1986, a weird British bearded guy was demo-ing this wild and nutty computer algebra system for the (not so large) masses. Guy's name was "Wolfram", like the element, and his software was called "Mathemagica" or something like that. :-)
Also in 1986 I believe was the introduction of this wild and crazy product called "HyperCard", a product that was so important but so ahead of its time that many younger people i talk to these days can't believe there was anyting like this in the bad old days.
But my favorite early MacWorld was 1987, since I basically had to sneak out from under my soon-to-be-inlaws in order to attend the thing...3 days before I got married in another state. I randomly ran into a Boston Globe reporter who made me part of her story in the paper on how...hyper-dedicated macophiles were in those days.
Ah, well... MacWorld may come back to Boston, but it will never be like 1985.
No, but neither you or I will ever be the same, either.
One might reasonably have expected that any 10.2 upgrade price would be less than the full price (even Microsoft does that). Or, perhaps one might have assumed that a computer company that makes most of its money on hardware sales would not do something that will tend to encourage people not to buy a new Mac when an OS upgrade is expected "soon" but not announced. Or, perhaps Apple might have considered that people who already own multiple Apple systems will be less likely to upgrade or buy more new Apple hardware if the cost of upgrading system software (which used to be notably cheaper than Windows) gets too high. Or, Apple might have realized that owners of their computers tend to be idealistic zealots who will turn on them very quickly if pissed off, even if you could give them a brilliant economic argument for charging them more.
Frankly, Apple cannot afford to tweak its ownership base too much. They famously raised prices on one very popular model this spring and the scratched their heads when sales came in below plan after that. Heck, I doubt that Mac sales are too great even right now since you can save at least $20 by waiting until after August 24. Petty? Sure, but humans are like that. Apple really could have handled this better; they had no reason to spring the "bad news" on pricing during one of their big positive press events of the year. I mean, what were they thinking? People came out of the Jobs keynote feeling a bit let down about new hardware, but feeling good about OS X 10.2, and then the realization sinks in that they're going to fork over hundreds of dollars (if they have more than one Mac) to upgrade. Really, was that the best plan?
The problem with the argument here is in the blind acceptance of the statement that ZD is truly profitable. As everybody should be quite thoroughly aware of these days, GAAP can be used to play with earnings to a disturbing extent even before you get to outright Enron/Worldom style fraud. If you read the whole NYT article, you can see that they are currently in a cash crunch, which is what will force the filing. Yes it might seem weird that a company could "produce" a profit up until the point it files for chapter 11, but it does happen (I think it used to be normal for this to happen with banks that failed). For that matter, they *aren't* profitable now. You write:
First off, that isn't earnings, but EBITDA, which is wildly significant because their interest expense is absolutely huge. Further, the trends for future business here have to be strongly negative. I severely doubt, for example, that year-over-year ad revenue is up or will be up in the foreseeable future. I cannot believe that subscriptions to the flagship (PC Magazine) are going anywhere but down (unless they have expanded some of the "loss-leader" deals they used to offer).
And, for that matter, I'm very skeptical about the breath-taking turnaround they suggest is going to happen after the pre-packaged chapter 11 filing. Note that they are pretty much *forced* to paint a rosy scenario here because they are trying to get their bondholders to accept equity in lieu of cash. For that matter, chapter 11 does give them some leverage here over the bondholders who already rejected a similar deal (again, this is clear from the NYT piece). When push comes to shove in bankruptcy court, I wouldn't be surprised to see monthly operating reports deep in the red if they stay there for long, and the bondholders (and any banks) are going to get heart pretty badly even if they take the equity. OK, so that's my opinion based on the expected current value of any enterprise purchased in April 2000.
I have no current interest in the equity or debt of ZD or any of its partners or subsidiaries.
Floppy drives (and both Apple and TI hardware) were for wusses. Real men used Scotch brand cassette tapes with the "data recorders" on the radically new and innovative VIC-20 system. We ruled, not like those johnny-come-lately scum with their Radio Shack Color Computers. (Sorry to have to mention the members of that particular psycho cult; if you want to revisit the horror, see the web site of the kookoo CoCo High Priest. The best link on that page is about an innovative 4 MHz acceleration project they are putting together. I'm not kidding.)