Safari is doing this (not BOMArchiver). And it does it for.dmg files too, so those could be exploited as easily as.zip or.sit.
This is unfortunately a rather Microsoft-like behavior -- building too much "intelligence" into the way the system handles an operation, rather than using a simpler, more transparent approach. Safari identifies archive files when you download them (.sit,.zip,.dmg,...?). Then, it tries to make.sit or.zip files act like Internet-enabled DMG files -- it extracts the contents and deletes the original archive file. Alternatively, it will simply mount any regular, non-Internet-enabled.dmg file.
Finally, if the archive (.sit,.zip or.dmg) contained only one file, and Safari deems that a "safe" file (subject to extension/creator spoofing) Safari will launch that file.
If you download any of these archive files in Camino, the behavior is different, and more like a regular social exploit (with file-type spoofing). The archive file will be dumped in your download folder. Then, if Camino is set to open "safe" files, or if you choose "open" in Camino's download window, or double-click on your download in the Finder, the archive will be handed off to the operating system. Then the operating system will treat it differently from Safari -- it will just mount the.dmg, or extract the contents of the.zip or.sit file. After that, it's up to you to find the enclosed file and open it.
The key difference between Safari and Camino, when both have "Automatically open safe files" turned on is this: When Camino downloads an archive, it opens it, leaving the contents in your download folder. Safari goes a step further: it removes the archive, leaving only the contents, and then (this is the key point), if the archive contained only one file and the file is deemed "safe," Safari will open that file.
This works well for some things, like recursively opening.tar.gz files, or maybe for opening some safe files. But Safari doesn't seem to be asking the operating system how the file will be opened -- it just assumes that if the extension is "safe," then the file must be safe. However, since these archives can hold metadata that will change the way the file is opened, this assumption can be incorrect (as shown in this exploit).
This couldn't happen with an unarchived file downloaded directly from the Internet, since the only metadata is the file extension. But when the file is stored in an archive (or e-mailed via Apple's Mail.app), extra metadata can go along that makes the file open with a different application (e.g., Terminal). This makes it possible for a web page to induce Safari to open any file it wants, without any user action, other than going to that web page.
If I were Apple, I'd redesign the code in Safari that checks what files are really "safe" to open. With their dual metadata system, they need to use the same routines everywhere to identify how a file will be opened, and avoid this kind of mix-up.
I would also change the file-opening code at the system level to give a warning anytime an executable-type file is opened, if it has a misleading extension. This would probably go in the same place as the code that currently prompts you when you open a document that will cause an application to open for the first time.
All the links you gave seem to support the original poster's point ("It's not an Intel motherboard, it's an Intel *chipset*."). As far as I could tell, all your links that discussed the motherboard design either cited or plagiarized an article at http://www.appleinsider.com/article.php?id=1431 . That article seems to say that Apple designed its own motherboards for the iMac, MacBook and forthcoming iBook and Mac Mini, and contracted with Intel to develop a custom motherboard for the Power Mac.
I heard once, long ago, that this is what happened:
IBM published all the hardware specs for the original PC, in order to encourage third parties to make lots of add-on cards for it. This would have been enough for anyone to assemble the hardware of an IBM PC. However, IBM retained control (through copyright?) of the BIOS, which actually made that hardware a real "IBM PC." Then, a company (I think it was Phoenix Technologies) got two groups of programmers together. They put one group in a room with an IBM PC and had them go through the BIOS and figure out what every system call did. Then they gave this information (not the actual code, just the detailed functional specifications) to the programmers in the other room. The second group, who had never seen the original code, then wrote their own code to match exactly those specs, but without copying any of the original code. Once this un-copied BIOS was readily available, anyone could then build a "PC-Compatible" computer.
IBM tried to regain control later, by introducing the new and improved Microchannel architecture, which had much stricter license agreements for accessories. But by then the horse was already out the door, and the industry stuck with ISA and its successors. Evenually, IBM caved, and went along with them.
"I want to copy files to that location and iTunes to notice them (=adding them to the library DB)."
There are two ways to do this easily, if not automatically.
(1) If you're using the "Keep iTunes Music Folder Organized" option: make a habit of dropping your new music files into a folder inside the main music folder (e.g., one called "New Music"). Then, periodically drag and drop this folder onto iTunes, or choose File->Add Folder to Library... and navigate to this folder. iTunes will add all the songs to your library database and move the files into appropriate folders. Advanced tips: (a) stick at least one non-music file in your "New Music" folder, so iTunes won't completely empty and delete the folder. (b) Create a folder action on the "New Music" folder, so it automatically feeds itself to iTunes whenever a file is added.
(2) If you're managing your library folders manually: Once in a while, drag the top level music library folder onto iTunes and drop it on the Library list. iTunes will go through the whole folder hierarchy and quickly add any music files that aren't already in the database. It's smart about this, so you won't get any doubles. You can also do this via File->Add Folder to Library...
Toshiba claims that this design sets "a new benchmark for areal density: 80-gigabits of data per square inch."
I don't know much about HD design, but I'm assuming that the reason you get faster transfers from drives with higher RPM is that the head passes over more bits per second, which it can read in and hand over to the CPU. So, couldn't you get the same effect from a lower RPM drive with the bits packed closer together?
e.g., If you double the areal bit-density, you should multiply the number of bits per track by approximately sqrt(2)=1.4, so the bits per revolution will be multiplied by 1.4, which makes a 4200 RPM drive equivalent to a 5900 RPM drive, in terms of the number of bits the head sees per second. (But also by this theory, physically small drives will always be slower than larger drives with the same RPM, since there are fewer bits per track, unless they can manage to acheive a higher bit-density. So maybe the Toshiba just comes out even with a 4200 RPM desktop drive.)
I think you missed the point that the "gravityplane of the sea" would have to "fly" underwater. The extra buoyancy comes from the fact that the surrounding medium would be more dense than air. Boats are already good at gliding on water, so some machine that dives underwater and back up wouldn't be much of an innovation in the rapid-water-transport area. It also wouldn't work the same way as the air gravityplane, because you can't compress water to store it and change your buoyancy in the same way you can with air.
There's not much point worrying about that anyway, because both versions of the machine are a complete sham. The promotional video plays a few sleights of hand with the energy transfers, so they're easy to lose track of. But in the end, you can never get around conservation of energy.
On the airplane, basically, the compressed air tanks are being used as batteries, and they're going to run out eventually. It won't take long either, because a tank of compressed air doesn't hold nearly as much energy as a tank of fuel.
Here's how the energy will flow:
Suppose you start on the ground with the (huge) buoyancy canisters full of helium, and your vehicle weighs less than the surrounding air. You have some lift. Up you go!
Eventually, you will reach the point where your vehicle weighs the same amount as the surrounding air, and you will stop rising. Now you can use some of the energy in your compressed air tank to turn a pump and bring more air on board (this is OK energetically, because you're venting high-pressure air and bringing on a larger amount of low-pressure air). Most likely you would do this by compressing your helium and letting ambient air come in to replace it. You might also vent some of your high-pressure air to start your vehicle moving forward.
Now here's the tricky part. If you do one joule of work in this process (forcing helium back into its tank and/or starting your vehicle moving horizontally), you will be able to collect (at best), slightly less than one joule of energy on the way down to ground level. If you try to collect energy faster than that on your way down (with your super whiz-bang turbines), the vehicle will stop moving forward. You can't get enough power from your downward/forward motion to put more pressure into the high-pressure air tank than you let out when you were changing your bouyancy or jetting forward.
Now, once you reach the ground, you will still be heavier than air. But you can let some helium back out into your bouyancy canister, displacing the ambient air and making your vehicle slightly bouyant again. If you run a little pump off this helium stream, you can pump a little more high-pressure air into your main air tank. And if you've lost no energy to friction or diabatic processes anywhere along the way, this extra air will be the last little increment you need to replace the one joule of energy you let out of your main air tank when you were at the top of the flight.
Basically, if it were a no-loss system, this thing could move up, down and side-to-side indefinitely, which is true for any no-loss system. But this won't be a no-loss system, and it won't be any better than dirigibles, boats or trains (which all have pretty low horizontal power losses already). Every one of the moving parts on this thing will have friction, which will reduce the compressed-air energy budget. There will also be losses from thermal transfer; e.g., when you compress the helium at the top, it will get hotter, and you will lose that heat to the surrounding environment. That will hurt your thermodynamic budget later. An electric or hybrid car is probably closer to a no-loss vehicle than the "gravityplane" would be.
I agree with you that it would be nice to have a state-saving option in Safari. I just don't want it as often as you. I quite happily open windows in Safari and let them sit until I have time to come back and read them (days later). State saving is only an issue when I want to reboot my computer for a software update or if it's starting to get cranky (every few weeks).
That said, I come to the main point of my post: As far as I know, your hard disk receives no additional protection by shutting the computer down, instead of putting it to sleep. Someone should correct me if I'm wrong, but my understanding is that the hard disk is completely powered down in either case, at which point it parks its heads and is safe to move. (Apple's instructions tell you to wait till the "sleep" light comes on before you move the computer, but after that it's fine.) I've carried my iBook to all sorts of places in sleep mode, and never had a hard disk problem (logic board problems are a different thing...). You might even be putting more "miles" on your hard disk by shutting down and restarting every day, instead of just letting the computer go to sleep. (Although I doubt those "miles" do any long-term harm either.)
"Terminal windows don't have borders, and changing the background color is extremely difficult, so the windows run together."
In the spirit of your own message... I had never thought of changing Terminal window colors to keep them distinct. But with a little experiment, it turns out to be pretty easy, at least in 10.3, if you don't mind a little screen clutter.
With a Terminal window open, choose Terminal->Window Settings... Then choose Color from the popup list. Then click on the color swatch next to "Use this background color." The system Color palette will come up, and it will be in "background color changing mode." You can now close the "Terminal Inspector" window and keep just the "Color" palette open. And every time you click on a color in the Color palette, the frontmost Terminal window will assume (and keep) that color.
>> "Preview does not support bookmarks, making navigating large PDFs a pain. Has this been added in Panther?"
If by "bookmarks," you mean "URLs," then you should be in luck. The Preview link of the "What's New" section of the online help says, "URL support: If a PDF file contains a URL that links to another part of the document, to another document on your computer, or to a website on the Internet, you can click the link in Preview to open the document or website." (I haven't actually tried any URLs myself, but it sounds good.)
Incidentally, the online help seems to load faster in 10.3 than in 10.2, and the actual help text may come from two different locations, depending on whether you have an active internet connection. e.g., for the fact noted above:
I'm impressed by the new ability to read any eps or PostScript file and convert it into a PDF file for viewing or saving, which may come in useful for graphic design work or for reading formatted documents distributed for the UNIX crowd.
Also, the graphic-mode select-and-copy feature in Preview copies the data to the clipboard in a vector format (probably PDF), so you can copy vector illustrations (and even text blocks) from a PDF file into another program that can handle vector graphics (e.g., Keynote or Adobe Illustrator), and get the images at their full resolution. This beats Acrobat, which only copies the graphics as bitmaps at your current viewing resolution. Unfortunately, if you paste the graphics into a program that cannot handle vector images (like MS Word), you get a bitmap at 72 dpi, even if you zoom in before copying it, which is a little worse than you can do with Acrobat.
Text-mode copy-and-paste in Preview seems to be roughly comparable to the same feature in Acrobat. Acrobat preserves some of the formatting (which Preview does not), but introduces more artifacts (like lots of line breaks). Both of them do a good but not perfect job of keeping text together when you copy from multiple columns.
Preview cannot edit text in PDF files, so I'll need to keep Acrobat around to do that.
So I was thinking more about how far you can go with a 64-bit architecture, and here's my roadmap, built from nothing more than my fuzzy recollection of the computers of my youth, and a little curve fitting. It seems like we're doubling demand for RAM every 6 months (i.e., we need two more address bits every year).
Apple usually uses the largest RAM modules currently available to make their "maximum memory" claims. Since 1 GB is the biggest you can get today, and the machine has 8 RAM slots, Apple reports that the maximum is 8 GB.
But if I recall correctly, Apple's machines generally work fine with larger modules when they come out later. So the real limit might be much more (so much more it makes my brain hurt).
Long ago, I transferred e-mail from a tired old Windows laptop to an OS 9 laptop via IMAP. This seemed to be a pretty universal solution, so it could work for you.
First get access to an IMAP e-mail account (if you don't have one already, you might be able to run an IMAP server on OS X). Then access the IMAP account from your Windows PC. Create appropriate folders within the IMAP account using your Windows mail client (I'm pretty sure Outlook can do this). Then drag messages into those folders. Outlook should copy the messages over to the IMAP server. Then connect to the IMAP server using Apple Mail, and transfer all the messages onto your eMac.
This worked for me using a Windows 3.1-era e-mail client from Microsoft (I think it was Mail and News or something like that), and Outlook Express on the Mac. I haven't tried it with Outlook and Apple Mail, but "in principle it should work." It's pretty bandwidth-hungry if you have a lot of messages, though.
Here's another method I think I used once: First, I dragged messages out of Outlook (or something similar) into folders on the Windows Desktop. In my case, the e-mail program created plain text versions of all the messages. Then I transferred them to the Mac, changed the creator type to match Outlook Express and dragged them into Outlook Express (which accepted them quite happily). Somewhere in there, I think I ran an Applescript to add a "Received" date header in the plain text files, since Outlook Express uses that instead of the "Sent" date. I got lucky in making this work, so it may not work for you, but it's worth a try. You may also be able to import plain-text messages straight into Mail.
"The waveforms are not only different but... the the music plays just slightly (imperceptibly) slower on the iTunes version."
Apple went back to the master tapes for a lot of their recordings. Jobs claimed in his announcement that some Music Store files would actually sound better than commercial CDs, which are not always made from such good sources.
If the track was converted to digital format from scratch, rather than using the same digital source as your CD, it would be reasonable to expect (minor) differences in playback speed and waveform.
(Sorry for the double-posting. This one is slightly better.)
Most people would connect the DSL modem to a router or NAT device, and then connect the router to their hub, so everyone on the network can share the connection. The router runs its own PPPoE software which connects through the modem to Yahoo, and then it automatically provides local IP addresses via DHCP to computers on your home network. This is what I do in my house, and it works great (with SBC/Yahoo DSL). SBC may not provide technical support for this, but at least they officially allow it, unlike Earthlink.
Your case is unusual, in that you don't want to use a router and don't want to share the connection with other computers. However, it should be easy to setup. Just plug the DSL modem into your hub (The one wrinkle is that you'll probably have to use an ethernet crossover cable between the modem and hub, because the modem expects to be connected directly to a computer instead of a hub). Then plug your computer into the hub. As far as your computer and the PPPoE software are concerned, your computer will look like it is connected directly to the modem, and everything will work fine (i.e., the hub is transparent to your computer). Also, you can use any PPPoE software you want. I would highly recommend using Mac OS X's built in PPPoE settings instead of EnterNet Classic, whatever that is.
Good luck!
P.S. I can't remember for sure what kind of cable the modem comes with. But the rule to follow is that you should use the opposite type of cable to connect the modem to a hub as you would use to connect the modem directly to a computer. So if the modem comes with a crossove cable, use a straight ethernet cable to connect it to the hub. If it comes with a straight cable (which I think is the case), then you'll need to use a crossover cable to connect the modem to the hub.
This is easy to setup.
Most people would connect the DSL modem to a router or NAT device, and then connect the router to their hub, so everyone on the network can share the connection. The router runs its own PPPoE software which connects through the modem to Yahoo, and then it automatically provides local IP addresses via DHCP to computers on your home network. This is what I do in my house, and it works great (with SBC/Yahoo DSL). SBC may not provide technical support for this, but at least they officially allow it, unlike Earthlink.
Your case is unusual, in that you don't want to use a router and don't want to share the connection with other computers. However, it should be easy to setup. Just plug the DSL modem into your hub (The one wrinkle is that you'll probably have to use an ethernet crossover cable between the modem and hub, because the modem expects to be connected directly to a computer instead of a hub). Then plug your computer into the hub. As far as your computer and the PPPoE software are concerned, your computer will look like it is connected directly to the modem, and everything will work fine (i.e., the hub is transparent to your computer). Also, you can use any PPPoE software you want. I would highly recommend using Mac OS X's built in PPPoE settings instead of EnterNet Classic, whatever that is.
Good luck!
P.S. I can't remember for sure what kind of cable the modem comes with. But the rule to follow is that you should use the opposite type of cable to connect the modem to a hub as you would use to connect the modem directly to a computer. So if the modem comes with a crossove cable, use a straight ethernet cable to connect it to the hub. If it comes with a straight cable (which I think is the case), then you'll need to use a crossover cable to connect the modem to the hub.
Cable modems are notorious for creating security openings. In many cases, you and all the other computers in your neighborhood are bridged onto a single network. So it's the same as if you were on one big LAN.
This issue affects your dad's computer whether or not your mom's computer is connected via it (the in-house network is just an extra wrinkle).
So you need to do a careful job of insulating your dad's computer from the outside network. Start by turning off all unnecessary services that could be carried on the Ethernet adapter. (i.e., make sure these services are not allowed to communicate over the Ethernet adapter. It's fine to let them run over the Airport adapter if your software base station is configured correctly, but you will have to discriminate between the two). OS X does a pretty good job of not loading too many services in the default configuration. But you can fine tune what's going on using OS X's internal firewall. You should also turn off any file or printer sharing on the Ethernet adapter (using the Sharing preference panel). I'm not sure whether you can turn off Rendezvous on one particular adapter, but if you can, that would be a good idea too.
Another way to restrict data from being sent over the Ethernet connection out to your neighbors, would be to install firewall or routing hardware between your Dad's computer and the cable modem. Then you won't really have to worry about reconfiguring your dad's computer at all. Anything that is labeled for "cable modem sharing" or "DSL connection sharing" should work fine for you. However, if you're going to get a connection sharing box, you might as well get one that can provide a connection directly to both your dad's computer and your mom's, so hers doesn't have to go through his to get to the Internet. There are plenty of cable modem routers out there that also include 802.11b support, and any of these should solve all your problems at once (i.e., they will hide your computers from your neighbors, and they will allow both of your computers to connect to the Internet independently via Airport or Ethernet). Apple's Airport base station is particularly nice, but there are other boxes in the $100 range that will work fine.
Additionally, haven't you noticed that if you pause it, after 5 minutes of inactivity it turns off by itself?
The advantage is that it will remember the last song played and continue from there. If you turn it off it will forget the last song and start playing at the beginning.
You can also get the auto-resume-at-the-same-song feature by pressing and holding the play button on the remote to turn off the iPod, instead of using the button on the iPod itself. I don't know why the two buttons work differently, but they do.
One other thing I noticed: keeping the "alerts" setting turned on makes the iPod battery run down much faster than otherwise (at least with older firmware). I think the iPod stays half awake waiting for an iCal alert to come up (even if you have none set for the future), and the battery ends up running down much faster. I turned off this feature, and my standby time doubled (or more).
. . . is not bandwidth throttling, but rather packet prioritization. It will be unnecessarily restrictive (and not ideal for you either) to give her, e.g., 300 kbps at all times. It would be much better to give yourself all the bandwidth you want at any given moment, and give her whatever's left over. That way she could use the full bandwidth most of the time, but not interfere with your work at all in the rare moments when you need a lot of bandwidth. Unfortunately, this kind of prioritization between traffic from different hosts would have to happen in the router, so it's probably not available on your home network.
Is Apple building an audio-video monopoly?
on
Apple Buys Emagic
·
· Score: 1
With all Apple's recent acquisitions of audio and video software producers
and their existing strengths, I wonder
whether Apple is trying to "pull a Microsoft" in the area of audio and
video authoring.
Microsoft beat out their competitors and made their office suite the de
facto standard for business users (who happen to be the biggest Windows
customers anyway). Maybe Apple has decided to let Microsoft keep the
business software monopoly, and is now trying to build their own monopoly
in audio and video production, so that if you want to work in this field,
you have to buy Apple software, and maybe even a Macintosh. Seems like a
pretty sensible strategy for Apple, capitalizing on their existing
strengths.
The shopping mall is really a straw man argument. I don't think there's any constituency that wants to block cell phone service for whole shopping malls, so that will never happen. (Most of the anti-cell phone people are probably anti-mall too, so they don't care. And the mall managers know where their bread is buttered.)
This would be used in small, enclosed places where people are already expected to be quiet -- movie theaters, libraries, etc. By definition, these are places that you could and should step outside of in order to make a phone call, either by pay phone or by cell phone. The pay phones would be outside of the "cone of silence," so they'd usually be outside of the EM blocking area, too.
I may be quibbling now, but I dont' think Caller ID would be enough to let you identify urgent calls, which was why I assumed you'd want to answer every call. I would guess that the hypothetical call about your girlfriend would come from one of the same people that would normally call you about other, random things. How will you know by Caller ID alone whether it's one of your friends/parents/in-laws calling for a chat and unaware that you're in the movie theater, versus one of the same people calling with an urgent message, unless you take the call? I suppose you could tell them to call repeatedly if it's really important, but other than that, I don't know any way to distinguish important from unimportant calls except by answering them all.
I'm also not sure how much trouble and turmoil you can avert by hearing about things like this an hour earlier than you might otherwise. That was the gist of my comment that we don't all need to be by a phone 24/7. It's easy to raise the once-in-a-few-lifetimes spectre of a critically injured relative whom you hear about just in time to rush to their bedside and have a few last words. But other than that, most situations (even the worst of them) can keep for an hour or two. We'll probably have to agree to disagree on this point.
There are lots of places you can go where cell phone service doesn't work. Movie theaters would just be one of those. I don't see how you'd have the right to sue anyone over that.
I'd rather throw popcorn at them then not be able to recieve an important call
So you really think you should be able to receive every phone call, even in the movie theater? I'm not sure how you propose to receive only the important ones and not the unimportant ones, after all.
I think that's pushing some sort of limit. If you absolutely must be "in touch" every moment of every day, maybe you should consider staying at home (like the old days), or anywhere your cell phone continues to work (lucky you). If you can't bear to be out of touch for more than a few minutes at a time, you could always step out and check for messages now and then. But I have to wonder whether anyone really needs to be accessible by phone every minute of every day. Most people weren't 5 or 10 years ago, and their lives were just fine, weren't they?
it gives the phone companies the ability to extort money out of people
This would not open up any special opportunities for pay phone companies. You can always step outside (where the pay phone probably is) and use your cell phone. On the other hand, cell phones do not act as price competitors with payphones, anyway. People seem to use their cellphones if they have them, whether payphones cost more or less. This may be what really drives up pay phone prices (and eventually makes pay phones extinct). As fewer people use payphones, the price per call has to go up in order to continue covering the cost of the service (capital expenses, maintenance, etc.).
Are you by chance using a CRT instead of an LCD? I'm wondering how much difference that makes.
I appreciate the "use whatever you like" notion, but as OS X comes into its own, I'm getting a distinct impression that everyone's going to end up with anti-aliased text, like it or not, without any real discussion of its merits and demerits. That's why I brought it up here.
As I noted before, I think anti-aliased text is the best choice in many cases: very small text, large text, or screen versions of print-oriented documents (e.g., PDFs, where the alternative is weirdly spaced aliased text).
But I still think the best thing for doing lots of on-screen reading may be well-hinted, aliased fonts, spaced and pixel-aligned with on-screen reading in mind. That's a pretty tall order, and somewhat kludgy, since it harkens back to the old screen-font/printer-font days -- it isn't "true" typography and doesn't fit that well with WYSIWYG workflow across different devices. So maybe I'll just have to get used to anti-aliasing. I do like the look of it, at least.
I think you may have it exactly right. I only have problems with antialiased text when the font size is small. If the strokes of the aliased text are solid black and one pixel wide, they usually become two pixels wide and middle-gray when they are antialiased, which makes them much harder to read. Also, once the "natural" stroke width drops below a pixel (i.e., very small text), antialiasing seems to help again -- I can recognize the patterns of tiny fuzzy text more easily than tiny distorted text.
I think this problem may be more pronounced on digital flat panels (e.g., my iBook). CRTs are much fuzzier than LCDs at the individual pixel level to begin with, so anti-aliasing may help (or at least not hurt) them, while it makes text less sharp on an LCD.
Slashdot Mirror
Safari is doing this (not BOMArchiver). And it does it for .dmg files too, so those could be exploited as easily as .zip or .sit.
.zip, .dmg, ...?). Then, it tries to make .sit or .zip files act like Internet-enabled DMG files -- it extracts the contents and deletes the original archive file. Alternatively, it will simply mount any regular, non-Internet-enabled .dmg file.
.zip or .dmg) contained only one file, and Safari deems that a "safe" file (subject to extension/creator spoofing) Safari will launch that file.
.dmg, or extract the contents of the .zip or .sit file. After that, it's up to you to find the enclosed file and open it.
This is unfortunately a rather Microsoft-like behavior -- building too much "intelligence" into the way the system handles an operation, rather than using a simpler, more transparent approach. Safari identifies archive files when you download them (.sit,
Finally, if the archive (.sit,
If you download any of these archive files in Camino, the behavior is different, and more like a regular social exploit (with file-type spoofing). The archive file will be dumped in your download folder. Then, if Camino is set to open "safe" files, or if you choose "open" in Camino's download window, or double-click on your download in the Finder, the archive will be handed off to the operating system. Then the operating system will treat it differently from Safari -- it will just mount the
The key difference between Safari and Camino, when both have "Automatically open safe files" turned on is this: When Camino downloads an archive, it opens it, leaving the contents in your download folder. Safari goes a step further: it removes the archive, leaving only the contents, and then (this is the key point), if the archive contained only one file and the file is deemed "safe," Safari will open that file.
.tar.gz files, or maybe for opening some safe files. But Safari doesn't seem to be asking the operating system how the file will be opened -- it just assumes that if the extension is "safe," then the file must be safe. However, since these archives can hold metadata that will change the way the file is opened, this assumption can be incorrect (as shown in this exploit).
This works well for some things, like recursively opening
This couldn't happen with an unarchived file downloaded directly from the Internet, since the only metadata is the file extension. But when the file is stored in an archive (or e-mailed via Apple's Mail.app), extra metadata can go along that makes the file open with a different application (e.g., Terminal). This makes it possible for a web page to induce Safari to open any file it wants, without any user action, other than going to that web page.
If I were Apple, I'd redesign the code in Safari that checks what files are really "safe" to open. With their dual metadata system, they need to use the same routines everywhere to identify how a file will be opened, and avoid this kind of mix-up.
I would also change the file-opening code at the system level to give a warning anytime an executable-type file is opened, if it has a misleading extension. This would probably go in the same place as the code that currently prompts you when you open a document that will cause an application to open for the first time.
All the links you gave seem to support the original poster's point ("It's not an Intel motherboard, it's an Intel *chipset*."). As far as I could tell, all your links that discussed the motherboard design either cited or plagiarized an article at http://www.appleinsider.com/article.php?id=1431 . That article seems to say that Apple designed its own motherboards for the iMac, MacBook and forthcoming iBook and Mac Mini, and contracted with Intel to develop a custom motherboard for the Power Mac.
I heard once, long ago, that this is what happened: IBM published all the hardware specs for the original PC, in order to encourage third parties to make lots of add-on cards for it. This would have been enough for anyone to assemble the hardware of an IBM PC. However, IBM retained control (through copyright?) of the BIOS, which actually made that hardware a real "IBM PC." Then, a company (I think it was Phoenix Technologies) got two groups of programmers together. They put one group in a room with an IBM PC and had them go through the BIOS and figure out what every system call did. Then they gave this information (not the actual code, just the detailed functional specifications) to the programmers in the other room. The second group, who had never seen the original code, then wrote their own code to match exactly those specs, but without copying any of the original code. Once this un-copied BIOS was readily available, anyone could then build a "PC-Compatible" computer. IBM tried to regain control later, by introducing the new and improved Microchannel architecture, which had much stricter license agreements for accessories. But by then the horse was already out the door, and the industry stuck with ISA and its successors. Evenually, IBM caved, and went along with them.
"I want to copy files to that location and iTunes to notice them (=adding them to the library DB)."
There are two ways to do this easily, if not automatically.
(1) If you're using the "Keep iTunes Music Folder Organized" option: make a habit of dropping your new music files into a folder inside the main music folder (e.g., one called "New Music"). Then, periodically drag and drop this folder onto iTunes, or choose File->Add Folder to Library... and navigate to this folder. iTunes will add all the songs to your library database and move the files into appropriate folders. Advanced tips: (a) stick at least one non-music file in your "New Music" folder, so iTunes won't completely empty and delete the folder. (b) Create a folder action on the "New Music" folder, so it automatically feeds itself to iTunes whenever a file is added.
(2) If you're managing your library folders manually: Once in a while, drag the top level music library folder onto iTunes and drop it on the Library list. iTunes will go through the whole folder hierarchy and quickly add any music files that aren't already in the database. It's smart about this, so you won't get any doubles. You can also do this via File->Add Folder to Library...
Toshiba claims that this design sets "a new benchmark for areal density: 80-gigabits of data per square inch."
I don't know much about HD design, but I'm assuming that the reason you get faster transfers from drives with higher RPM is that the head passes over more bits per second, which it can read in and hand over to the CPU. So, couldn't you get the same effect from a lower RPM drive with the bits packed closer together?
e.g., If you double the areal bit-density, you should multiply the number of bits per track by approximately sqrt(2)=1.4, so the bits per revolution will be multiplied by 1.4, which makes a 4200 RPM drive equivalent to a 5900 RPM drive, in terms of the number of bits the head sees per second. (But also by this theory, physically small drives will always be slower than larger drives with the same RPM, since there are fewer bits per track, unless they can manage to acheive a higher bit-density. So maybe the Toshiba just comes out even with a 4200 RPM desktop drive.)
I think you missed the point that the "gravityplane of the sea" would have to "fly" underwater. The extra buoyancy comes from the fact that the surrounding medium would be more dense than air. Boats are already good at gliding on water, so some machine that dives underwater and back up wouldn't be much of an innovation in the rapid-water-transport area. It also wouldn't work the same way as the air gravityplane, because you can't compress water to store it and change your buoyancy in the same way you can with air.
There's not much point worrying about that anyway, because both versions of the machine are a complete sham. The promotional video plays a few sleights of hand with the energy transfers, so they're easy to lose track of. But in the end, you can never get around conservation of energy.
On the airplane, basically, the compressed air tanks are being used as batteries, and they're going to run out eventually. It won't take long either, because a tank of compressed air doesn't hold nearly as much energy as a tank of fuel.
Here's how the energy will flow:
Suppose you start on the ground with the (huge) buoyancy canisters full of helium, and your vehicle weighs less than the surrounding air. You have some lift. Up you go!
Eventually, you will reach the point where your vehicle weighs the same amount as the surrounding air, and you will stop rising. Now you can use some of the energy in your compressed air tank to turn a pump and bring more air on board (this is OK energetically, because you're venting high-pressure air and bringing on a larger amount of low-pressure air). Most likely you would do this by compressing your helium and letting ambient air come in to replace it. You might also vent some of your high-pressure air to start your vehicle moving forward.
Now here's the tricky part. If you do one joule of work in this process (forcing helium back into its tank and/or starting your vehicle moving horizontally), you will be able to collect (at best), slightly less than one joule of energy on the way down to ground level. If you try to collect energy faster than that on your way down (with your super whiz-bang turbines), the vehicle will stop moving forward. You can't get enough power from your downward/forward motion to put more pressure into the high-pressure air tank than you let out when you were changing your bouyancy or jetting forward.
Now, once you reach the ground, you will still be heavier than air. But you can let some helium back out into your bouyancy canister, displacing the ambient air and making your vehicle slightly bouyant again. If you run a little pump off this helium stream, you can pump a little more high-pressure air into your main air tank. And if you've lost no energy to friction or diabatic processes anywhere along the way, this extra air will be the last little increment you need to replace the one joule of energy you let out of your main air tank when you were at the top of the flight.
Basically, if it were a no-loss system, this thing could move up, down and side-to-side indefinitely, which is true for any no-loss system. But this won't be a no-loss system, and it won't be any better than dirigibles, boats or trains (which all have pretty low horizontal power losses already). Every one of the moving parts on this thing will have friction, which will reduce the compressed-air energy budget. There will also be losses from thermal transfer; e.g., when you compress the helium at the top, it will get hotter, and you will lose that heat to the surrounding environment. That will hurt your thermodynamic budget later. An electric or hybrid car is probably closer to a no-loss vehicle than the "gravityplane" would be.
I agree with you that it would be nice to have a state-saving option in Safari. I just don't want it as often as you. I quite happily open windows in Safari and let them sit until I have time to come back and read them (days later). State saving is only an issue when I want to reboot my computer for a software update or if it's starting to get cranky (every few weeks).
That said, I come to the main point of my post: As far as I know, your hard disk receives no additional protection by shutting the computer down, instead of putting it to sleep. Someone should correct me if I'm wrong, but my understanding is that the hard disk is completely powered down in either case, at which point it parks its heads and is safe to move. (Apple's instructions tell you to wait till the "sleep" light comes on before you move the computer, but after that it's fine.) I've carried my iBook to all sorts of places in sleep mode, and never had a hard disk problem (logic board problems are a different thing...). You might even be putting more "miles" on your hard disk by shutting down and restarting every day, instead of just letting the computer go to sleep. (Although I doubt those "miles" do any long-term harm either.)
"Terminal windows don't have borders, and changing the background color is extremely difficult, so the windows run together."
... I had never thought of changing Terminal window colors to keep them distinct. But with a little experiment, it turns out to be pretty easy, at least in 10.3, if you don't mind a little screen clutter.
In the spirit of your own message
With a Terminal window open, choose Terminal->Window Settings... Then choose Color from the popup list. Then click on the color swatch next to "Use this background color." The system Color palette will come up, and it will be in "background color changing mode." You can now close the "Terminal Inspector" window and keep just the "Color" palette open. And every time you click on a color in the Color palette, the frontmost Terminal window will assume (and keep) that color.
>> "Preview does not support bookmarks, making navigating large PDFs a pain. Has this been added in Panther?"
n glish.lproj/pgs2/wn11.html or file:///Library/Documentation/Help/MacHelp.help/Co ntents/Resources/English.lproj/pgs2/wn11.html
If by "bookmarks," you mean "URLs," then you should be in luck. The Preview link of the "What's New" section of the online help says, "URL support: If a PDF file contains a URL that links to another part of the document, to another document on your computer, or to a website on the Internet, you can click the link in Preview to open the document or website." (I haven't actually tried any URLs myself, but it sounds good.)
Incidentally, the online help seems to load faster in 10.3 than in 10.2, and the actual help text may come from two different locations, depending on whether you have an active internet connection. e.g., for the fact noted above:
http://helposx.apple.com/MacHelpR4/MacHelp.help/E
There's also a little more (or less) info about Preview here: http://www.apple.com/macosx/features/preview/
I'm impressed by the new ability to read any eps or PostScript file and convert it into a PDF file for viewing or saving, which may come in useful for graphic design work or for reading formatted documents distributed for the UNIX crowd.
Also, the graphic-mode select-and-copy feature in Preview copies the data to the clipboard in a vector format (probably PDF), so you can copy vector illustrations (and even text blocks) from a PDF file into another program that can handle vector graphics (e.g., Keynote or Adobe Illustrator), and get the images at their full resolution. This beats Acrobat, which only copies the graphics as bitmaps at your current viewing resolution. Unfortunately, if you paste the graphics into a program that cannot handle vector images (like MS Word), you get a bitmap at 72 dpi, even if you zoom in before copying it, which is a little worse than you can do with Acrobat.
Text-mode copy-and-paste in Preview seems to be roughly comparable to the same feature in Acrobat. Acrobat preserves some of the formatting (which Preview does not), but introduces more artifacts (like lots of line breaks). Both of them do a good but not perfect job of keeping text together when you copy from multiple columns.
Preview cannot edit text in PDF files, so I'll need to keep Acrobat around to do that.
So I was thinking more about how far you can go with a 64-bit architecture, and here's my roadmap, built from nothing more than my fuzzy recollection of the computers of my youth, and a little curve fitting. It seems like we're doubling demand for RAM every 6 months (i.e., we need two more address bits every year).
"Why only 8-gig of RAM?"
Apple usually uses the largest RAM modules currently available to make their "maximum memory" claims. Since 1 GB is the biggest you can get today, and the machine has 8 RAM slots, Apple reports that the maximum is 8 GB.
But if I recall correctly, Apple's machines generally work fine with larger modules when they come out later. So the real limit might be much more (so much more it makes my brain hurt).
Long ago, I transferred e-mail from a tired old Windows laptop to an OS 9 laptop via IMAP. This seemed to be a pretty universal solution, so it could work for you.
First get access to an IMAP e-mail account (if you don't have one already, you might be able to run an IMAP server on OS X). Then access the IMAP account from your Windows PC. Create appropriate folders within the IMAP account using your Windows mail client (I'm pretty sure Outlook can do this). Then drag messages into those folders. Outlook should copy the messages over to the IMAP server. Then connect to the IMAP server using Apple Mail, and transfer all the messages onto your eMac.
This worked for me using a Windows 3.1-era e-mail client from Microsoft (I think it was Mail and News or something like that), and Outlook Express on the Mac. I haven't tried it with Outlook and Apple Mail, but "in principle it should work." It's pretty bandwidth-hungry if you have a lot of messages, though.
Here's another method I think I used once: First, I dragged messages out of Outlook (or something similar) into folders on the Windows Desktop. In my case, the e-mail program created plain text versions of all the messages. Then I transferred them to the Mac, changed the creator type to match Outlook Express and dragged them into Outlook Express (which accepted them quite happily). Somewhere in there, I think I ran an Applescript to add a "Received" date header in the plain text files, since Outlook Express uses that instead of the "Sent" date. I got lucky in making this work, so it may not work for you, but it's worth a try. You may also be able to import plain-text messages straight into Mail.
Good luck!
"The waveforms are not only different but ... the the music plays just slightly (imperceptibly) slower on the iTunes version."
Apple went back to the master tapes for a lot of their recordings. Jobs claimed in his announcement that some Music Store files would actually sound better than commercial CDs, which are not always made from such good sources.
If the track was converted to digital format from scratch, rather than using the same digital source as your CD, it would be reasonable to expect (minor) differences in playback speed and waveform.
(Sorry for the double-posting. This one is slightly better.)
Most people would connect the DSL modem to a router or NAT device, and then connect the router to their hub, so everyone on the network can share the connection. The router runs its own PPPoE software which connects through the modem to Yahoo, and then it automatically provides local IP addresses via DHCP to computers on your home network. This is what I do in my house, and it works great (with SBC/Yahoo DSL). SBC may not provide technical support for this, but at least they officially allow it, unlike Earthlink.
Your case is unusual, in that you don't want to use a router and don't want to share the connection with other computers. However, it should be easy to setup. Just plug the DSL modem into your hub (The one wrinkle is that you'll probably have to use an ethernet crossover cable between the modem and hub, because the modem expects to be connected directly to a computer instead of a hub). Then plug your computer into the hub. As far as your computer and the PPPoE software are concerned, your computer will look like it is connected directly to the modem, and everything will work fine (i.e., the hub is transparent to your computer). Also, you can use any PPPoE software you want. I would highly recommend using Mac OS X's built in PPPoE settings instead of EnterNet Classic, whatever that is.
Good luck!
P.S. I can't remember for sure what kind of cable the modem comes with. But the rule to follow is that you should use the opposite type of cable to connect the modem to a hub as you would use to connect the modem directly to a computer. So if the modem comes with a crossove cable, use a straight ethernet cable to connect it to the hub. If it comes with a straight cable (which I think is the case), then you'll need to use a crossover cable to connect the modem to the hub.
This is easy to setup. Most people would connect the DSL modem to a router or NAT device, and then connect the router to their hub, so everyone on the network can share the connection. The router runs its own PPPoE software which connects through the modem to Yahoo, and then it automatically provides local IP addresses via DHCP to computers on your home network. This is what I do in my house, and it works great (with SBC/Yahoo DSL). SBC may not provide technical support for this, but at least they officially allow it, unlike Earthlink. Your case is unusual, in that you don't want to use a router and don't want to share the connection with other computers. However, it should be easy to setup. Just plug the DSL modem into your hub (The one wrinkle is that you'll probably have to use an ethernet crossover cable between the modem and hub, because the modem expects to be connected directly to a computer instead of a hub). Then plug your computer into the hub. As far as your computer and the PPPoE software are concerned, your computer will look like it is connected directly to the modem, and everything will work fine (i.e., the hub is transparent to your computer). Also, you can use any PPPoE software you want. I would highly recommend using Mac OS X's built in PPPoE settings instead of EnterNet Classic, whatever that is. Good luck! P.S. I can't remember for sure what kind of cable the modem comes with. But the rule to follow is that you should use the opposite type of cable to connect the modem to a hub as you would use to connect the modem directly to a computer. So if the modem comes with a crossove cable, use a straight ethernet cable to connect it to the hub. If it comes with a straight cable (which I think is the case), then you'll need to use a crossover cable to connect the modem to the hub.
Cable modems are notorious for creating security openings. In many cases, you and all the other computers in your neighborhood are bridged onto a single network. So it's the same as if you were on one big LAN.
This issue affects your dad's computer whether or not your mom's computer is connected via it (the in-house network is just an extra wrinkle).
So you need to do a careful job of insulating your dad's computer from the outside network. Start by turning off all unnecessary services that could be carried on the Ethernet adapter. (i.e., make sure these services are not allowed to communicate over the Ethernet adapter. It's fine to let them run over the Airport adapter if your software base station is configured correctly, but you will have to discriminate between the two). OS X does a pretty good job of not loading too many services in the default configuration. But you can fine tune what's going on using OS X's internal firewall. You should also turn off any file or printer sharing on the Ethernet adapter (using the Sharing preference panel). I'm not sure whether you can turn off Rendezvous on one particular adapter, but if you can, that would be a good idea too.
Another way to restrict data from being sent over the Ethernet connection out to your neighbors, would be to install firewall or routing hardware between your Dad's computer and the cable modem. Then you won't really have to worry about reconfiguring your dad's computer at all. Anything that is labeled for "cable modem sharing" or "DSL connection sharing" should work fine for you. However, if you're going to get a connection sharing box, you might as well get one that can provide a connection directly to both your dad's computer and your mom's, so hers doesn't have to go through his to get to the Internet. There are plenty of cable modem routers out there that also include 802.11b support, and any of these should solve all your problems at once (i.e., they will hide your computers from your neighbors, and they will allow both of your computers to connect to the Internet independently via Airport or Ethernet). Apple's Airport base station is particularly nice, but there are other boxes in the $100 range that will work fine.
You can also get the auto-resume-at-the-same-song feature by pressing and holding the play button on the remote to turn off the iPod, instead of using the button on the iPod itself. I don't know why the two buttons work differently, but they do.
One other thing I noticed: keeping the "alerts" setting turned on makes the iPod battery run down much faster than otherwise (at least with older firmware). I think the iPod stays half awake waiting for an iCal alert to come up (even if you have none set for the future), and the battery ends up running down much faster. I turned off this feature, and my standby time doubled (or more).
. . . is not bandwidth throttling, but rather packet prioritization. It will be unnecessarily restrictive (and not ideal for you either) to give her, e.g., 300 kbps at all times. It would be much better to give yourself all the bandwidth you want at any given moment, and give her whatever's left over. That way she could use the full bandwidth most of the time, but not interfere with your work at all in the rare moments when you need a lot of bandwidth. Unfortunately, this kind of prioritization between traffic from different hosts would have to happen in the router, so it's probably not available on your home network.
With all Apple's recent acquisitions of audio and video software producers and their existing strengths, I wonder whether Apple is trying to "pull a Microsoft" in the area of audio and video authoring.
Microsoft beat out their competitors and made their office suite the de facto standard for business users (who happen to be the biggest Windows customers anyway). Maybe Apple has decided to let Microsoft keep the business software monopoly, and is now trying to build their own monopoly in audio and video production, so that if you want to work in this field, you have to buy Apple software, and maybe even a Macintosh. Seems like a pretty sensible strategy for Apple, capitalizing on their existing strengths.
I forgot to reply to the part about the mall...
The shopping mall is really a straw man argument. I don't think there's any constituency that wants to block cell phone service for whole shopping malls, so that will never happen. (Most of the anti-cell phone people are probably anti-mall too, so they don't care. And the mall managers know where their bread is buttered.)
This would be used in small, enclosed places where people are already expected to be quiet -- movie theaters, libraries, etc. By definition, these are places that you could and should step outside of in order to make a phone call, either by pay phone or by cell phone. The pay phones would be outside of the "cone of silence," so they'd usually be outside of the EM blocking area, too.
I may be quibbling now, but I dont' think Caller ID would be enough to let you identify urgent calls, which was why I assumed you'd want to answer every call. I would guess that the hypothetical call about your girlfriend would come from one of the same people that would normally call you about other, random things. How will you know by Caller ID alone whether it's one of your friends/parents/in-laws calling for a chat and unaware that you're in the movie theater, versus one of the same people calling with an urgent message, unless you take the call? I suppose you could tell them to call repeatedly if it's really important, but other than that, I don't know any way to distinguish important from unimportant calls except by answering them all.
I'm also not sure how much trouble and turmoil you can avert by hearing about things like this an hour earlier than you might otherwise. That was the gist of my comment that we don't all need to be by a phone 24/7. It's easy to raise the once-in-a-few-lifetimes spectre of a critically injured relative whom you hear about just in time to rush to their bedside and have a few last words. But other than that, most situations (even the worst of them) can keep for an hour or two. We'll probably have to agree to disagree on this point.
you can bet I'd sue instantly
There are lots of places you can go where cell phone service doesn't work. Movie theaters would just be one of those. I don't see how you'd have the right to sue anyone over that.
I'd rather throw popcorn at them then not be able to recieve an important call
So you really think you should be able to receive every phone call, even in the movie theater? I'm not sure how you propose to receive only the important ones and not the unimportant ones, after all.
I think that's pushing some sort of limit. If you absolutely must be "in touch" every moment of every day, maybe you should consider staying at home (like the old days), or anywhere your cell phone continues to work (lucky you). If you can't bear to be out of touch for more than a few minutes at a time, you could always step out and check for messages now and then. But I have to wonder whether anyone really needs to be accessible by phone every minute of every day. Most people weren't 5 or 10 years ago, and their lives were just fine, weren't they?
it gives the phone companies the ability to extort money out of people
This would not open up any special opportunities for pay phone companies. You can always step outside (where the pay phone probably is) and use your cell phone. On the other hand, cell phones do not act as price competitors with payphones, anyway. People seem to use their cellphones if they have them, whether payphones cost more or less. This may be what really drives up pay phone prices (and eventually makes pay phones extinct). As fewer people use payphones, the price per call has to go up in order to continue covering the cost of the service (capital expenses, maintenance, etc.).
Are you by chance using a CRT instead of an LCD? I'm wondering how much difference that makes.
I appreciate the "use whatever you like" notion, but as OS X comes into its own, I'm getting a distinct impression that everyone's going to end up with anti-aliased text, like it or not, without any real discussion of its merits and demerits. That's why I brought it up here.
As I noted before, I think anti-aliased text is the best choice in many cases: very small text, large text, or screen versions of print-oriented documents (e.g., PDFs, where the alternative is weirdly spaced aliased text).
But I still think the best thing for doing lots of on-screen reading may be well-hinted, aliased fonts, spaced and pixel-aligned with on-screen reading in mind. That's a pretty tall order, and somewhat kludgy, since it harkens back to the old screen-font/printer-font days -- it isn't "true" typography and doesn't fit that well with WYSIWYG workflow across different devices. So maybe I'll just have to get used to anti-aliasing. I do like the look of it, at least.
I think you may have it exactly right. I only have problems with antialiased text when the font size is small. If the strokes of the aliased text are solid black and one pixel wide, they usually become two pixels wide and middle-gray when they are antialiased, which makes them much harder to read. Also, once the "natural" stroke width drops below a pixel (i.e., very small text), antialiasing seems to help again -- I can recognize the patterns of tiny fuzzy text more easily than tiny distorted text.
I think this problem may be more pronounced on digital flat panels (e.g., my iBook). CRTs are much fuzzier than LCDs at the individual pixel level to begin with, so anti-aliasing may help (or at least not hurt) them, while it makes text less sharp on an LCD.
There's an interesting article on the topic here.