I've always wanted a radio-controlled helicopter with a wireless video transmitter, so I could launch the thing out my window and go and harass the construction workers in the shell of the building next door. And it would have to have enough range/air-time/maneuverability that when they noticed it and started swinging shovels at it, I could get away. Sadly, my research seems to indicate that RC copters don't work that way: the people who are into this stuff are into RC modelling, which means that they like building the things, not actually flying them, and so there aren't any that are actually stable or work well. They all seem to be very fragile and either have very short flight times, or be absolutely huge (like, 6' long) which kind of defeats the purpose, if you ask me. (``I've got a map of the world. It's actual size. Maybe you've seen it.'')
An AudioRequest would be cool, though a PC preconfigured to do the same thing would be even cooler.
Oh, I would also like someone to find me a missile silo home, but it has to be within ten miles of San Francisco (the Presidio or the Marin Headlands would be acceptable locations.)
It would be better (on a philosophical level, at least) to live in a world with NO software than in a world with proprietary software. [...] I don't care what anyone says, nobody could possibly want to be restricted when they have the choice to be free. AS long as non-free software exists, someone is living in a limited form of slavery.
A BSD license potentially helps non-free software.
This is true.
If the goal is to have lots of free software, than the GPL is necessary.
This does not follow. If the goal is to have only GPLed software and no non-GPLed software, then the GPL is necessary.
You assume that to have lots of free software, there must be no non-free software, which is obviously not the case. If 10% of the world's software is free and 90% is non-free, then increasing the amount of software in the world would increase both the amout of free and non-free software (assuming the ratio held.) This is called ``growing the market'' and it benefits all who participate in that market, even if they are direct competitors.
So there are cases where non-free software can help the goal of having lots of free software. In fact, given that very nearly every piece of free software is a clone of some piece of non-free software, one could argue that non-free software is a necessary catalyst for free software! Would GIMP be as good today if they didn't have Photoshop to mine for ideas? Much though I respect the GIMP developers, I most sincerely doubt it.
Also note that the GPL is not the only license that is considered a ``free software'' license, even by GPL advocates. Don't say ``free'' when you really mean ``GPLed.''
Solaris is the name of a particular bundle of software from Sun: it is the environment which includes SunOS,
Sorry. If Sun OS is included in Solaris, why doesn't software that worked with Sun OS 4.x need to have system calles changed to work with Solaris?
Uh, because SunOS 4.x and SunOS 5.x aren't totally compatible? You should have said "...to work with SunOS 5.x", not "...to work with Solaris." Because Solaris 1.x and Solaris 2.x are incompatible in the same way as SunOS 4.x and SunOS 5.x.
Saying ``SunOS software doesn't work on Solaris'' is obviously false, because Solaris 1.0 was SunOS 4.1.1. So your use of the word ``Solaris'' is wrong.
You should have specified version numbers.:-)
No, ``Solaris includes SunOS'' is true without qualification. You should have said SunOS 5 when you meant SunOS 5, not Solaris (which can mean either SunOS 4 or 5, despite what the ambiguous common usage is.)
Not that anyone really cares... ``Solaris'' means ``SYSV'' in most people's minds just like ``hacking'' means ``breaking into computers.''
...seems to have replaced ``Amiga Persecution Complex'' as their affliction of the day. Has there been a single example of an Amiga-related announcement actually having been followed up by something more substantial than, say, more announcements? Any time in the last eight years? Just wondering...
(Oh, and how's that Amiga Mozilla port coming along? No, not that one, the other one.)
Jamie, who still has a functional A1000, but doesn't expect it to be making a comeback any time soon.
You can bring up the menus with F10, but you can't select in them unless the thing you want to do has an accelerator key. That's annoying. Or am I just missing something?
Hit F10 to bring up the menu. Use the left and right arrows to select the menu you want. Hit Enter to pop down the menu. Note that all the menu items on the menu have one character underlined. If you type that character, it's the same as selecting that menu item.
The difference between these underlined characters (called "mnemonics") and the global keybindings (called "accelerators", and listed on the right side of each menu) is that accelerators have to be global to the application, and mnemonics only have to be unique within their menu. Also mnemonics are single characters and accelerators are usually chorded (being Control or Alt or Shift keys.)
I think you're supposed to be able to use the cursor keys to move around on the menu and then hit return to select one, but that doesn't seem to work for some reason. I'm not sure why.
By the way, if you're interested in playing with selections, XEmacs is a great way to do it, since it provides direct low-level access, which otherwise you'd have to write C code to get. For example:
In a Netscape window, select some text with the mouse.
Fire up XEmacs, and switch to the *scratch* buffer. (This is a lisp-interaction buffer, if you type a lisp expression and then type Ctrl-j after it, it will evaluate it and show you the result.)
(x-get-selection-internal 'PRIMARY 'TARGETS) [TARGETS TIMESTAMP TEXT STRING LENGTH FILE_NAME OWNER_OS HOST_NAME USER CLASS NAME CLIENT_WINDOW PROCESS COMPOUND_TEXT]
That just asked Netscape what data types it is capable of converting the PRIMARY selection to. The list was returned. Let's try type TEXT:
(x-get-selection-internal 'PRIMARY 'TEXT) "xterm"
There's the text I selected in Netscape. Pretty basic, let's try another:
That says that 5 characters are selected (for not very good reasons, it's returning the high and low 16 bits of the 32-bit number as a pair of shorts. This is an emacs quirk.)
And so on... By letting the owner announce the types to which the selection can be converted, and letting the requester of the selection look at that list and decide which one they would prefer, very complex and efficient interations are possible. It allows format negotiation, and good decisions about pass-by-reference versus pass-by-value (e.g., if an image is selected, do you want its name, or its bits?)
Wouldn't it be nice if when you selected some text in Netscape, and you pasted it into a text editor, it was pasted as plain-text, but when you pasted it into an HTML editor, it was pasted as raw HTML, with formatting intact? Sadly, Netscape doesn't do that, but that's the sort of thing the X selection mechanism is for.
Netscape's clipboard handling is... quirky. It seems to have its own internal clipboard (which corresponds to the Edit menu Cut/Copy/Paste), and that seems to be completely independent of the X clipboard functionality: you can select text in a browser window, then paste it to the editor using the middle button, but if you want to use Edit/Paste (or Alt+V) you have to do Edit/Copy (Alt+C) first.
This is not quirky, this is how X selections work.
Think of it this way: you drag out some text. That's the "primary selection." Now you do "File/Copy". That copies the primary selection to the clipboard selection.
Middle-click pastes the primary selection. File/Paste pastes the clipboard selection.
To get Mac behavior, eliminate the middle mouse button (that is, give access only to the clipboard selection, not the primary selection.)
People often confuse the primary selection and the clipboard for the simple reason that xterm doesn't have a menubar, and therfore doesn't have File/Paste, so people raised on Unix instead of real GUIs tend not to notice that the clipboard selection exists at all.
I don't know why Netscape insists on using Alt for so many keyboard accelerators, unless it's an attempt to avoid colliding with the Emacs key bindings it offers in some situations.
That's exactly why. Having basic Emacs keybindings in all text areas was a hard requirement from day one, so the keyboard accelerators for the menu items couldn't take over the Control key.
You are mistaken. The X selection mechanism is actually one of the few parts of X that I think was done exceptionally well. It's very powerful and uses bandwidth intelligently without getting in the way too much.
Now, it happens that there is a dramatic lack of applications out there that are capable of cutting and pasting types of data that are not text. But they do exist (Lucid's Energize did it, and XEmacs has APIs for asserting selection values of arbitrary types -- though I don't think any emacs-lisp applications take advantage of that, either.)
(By the way, the "primary selection" (the click-and-drag-with-left, paste-with-middle thing) and the "clipboard" (the File/Cut, File/Paste thing) are the same thing in different namespaces: they work exactly the same way.)
The point, surely, is to be able to drang-and-drop and cut-and-paste objects and components.
Drag-and-drop and cut-and-paste aren't exactly the same thing, but the X selection mechanism can (and should) be used to implement them both. That's what it's for...
I actually never learned how Motif drag-and-drop works, but I would expect that it uses selections inside.
It's just bad business practice and it hurts us all.
Well, it doesn't hurt RSA. Monopolies, de-facto and otherwise, are very profitable.
And important thing to note here is that this is not an example of the patent system breaking down: this is an example of the patent system working as intended. A patent is a grant of a time-limited monopoly, with the condition of full disclosure of the technique. The US government granted RSA a monopoly on these algorithms, and so it's not terribly surprising that they've been able to parlay that into market domination.
I point this out because it's important to understand the differences between bad things that happen when the patent system is working as intended, versus when it's not working as intended. Most of the time, the patent-related complaints we hear are about bogus patents, patents that were obvious or trivial or already in the prior art. Those are examples of the patent system breaking down.
But if you fixed the problems at the patent office that caused bogus patents to be issued, and caused the patent office to execute their mandate correctly, situations like the RSA one would still occur.
In other words, if you're against the recent Amazon patent, you're against stupid patents. If you're against the RSA patents, you're against software patents.
From: cary@afone.as.arizona.edu (Cary Kittrell) Newsgroups: alt.fan.cecil-adams Subject: Re: Consumer electronics (was: A few random questions) Date: 22 Jul 1999 22:52:23 GMT
In article pg@sff.net writes:
After quitting from Apple Computer, Steve Wozniak designed a universal remote that apparently was VERY programmable. I would love to get one, but unfortunately they were sold for a very short time (too powerful for the FCC, or something like that).
Ohmigod. Oh my god! That was the CORE, and I have two or three of them sitting in my drawer right now. Want one? They broke easily, and didn't repair too solidly.
I have long lost count of how many languages I've programmed, but this was easily the hardest -- and I'm including languages like Ladder Diagram Notation here. You could do anything, but damn! did you have to be tenacious far beyond the call of geekery. As I recall, basically you could set a timer to act as if you had pushed a buton at a certain time, and you programmed by putting instructions under that button -- mostly pushes of other buttons, under which you had put pushes of other buttons -- and the only way you really got anything useful done was to include under a given button instructions which said "next time you're pushed, edit your own current instructions, changing them to the following set:...". In other words, this was not an environment in which you employed self-modifying code as a space-saver or a geeky tour-de-force -- self-modifying code was the _only_ way to implement things like counters and possibly (if I recall correctly) even IF/ELSE.
I now write programs which generate and polish really really big telescope mirrors. I was much prouder of being able to write programs which would tell the CORE "turn on both VCRs, then after 5:30 change the channel to..."
I think I still have the manual, if you'd like me to copy it.
Re:For the Future (was Re: Cool)
on
Linux on Palm
·
· Score: 2
In fact, to date, the only PDA that looks reasonably useful to me is the Sony Viaio (or however you spell it;) The little pentium-powered handheld sublaptop. So you say, "but it's not a PDA". WHO CARES? It's what I want. A little system that can do some processing on the road, that I can uplink to my main computers.
The Vaio's a nice machine, but it hasn't crossed the size quantum. There are three sizes of computers that are interesting: ``it sits on my desk,'' ``it fits in a backpack,'' and ``it fits in a pocket.''
Now I've got big pockets, but the tinier Vaio still doesn't fit in them. And the keyboard is so small that my thumbs rub together when I'm on the home row. Also, it doesn't have a serial port, only USB (which means you can't use Ricochet with it.) (The bigger Vaio has a plug-on port-replicator thingy that gives you a serial port, but not the smaller version.)
The reason that CE doesn't make sense in many palm tops is because the UI was made for desktops. However, I highly doubt that anyone is going to try and run X on a palmtop
Agreed on both points. But today, you can write Palm programs with gcc, and all the usual Posix libc goodies. You can't use X, you can't use Gtk. So this begs the question, what tangible benefits would Linux provide that PalmOS would not?
So, linux on the palm pilot isn't bad. But what might make it actually good?
Exactly. As far as I can tell, using Linux rather than PalmOS wouldn't change either the programming experience or the end-user experience. In particular, there are basically no end-user Linux applications that will run on Palm Linux without first being redesigned to work with a completely different user interface.
It does have some areas that are better than PalmOS, like networking.
Isn't the primary problem with Palm networking the networking hardware, not software?
Plus, writing C code for PalmOS has some strange quirks.
Interesting, can you give some examples?
Personally, I would love a PalmPilot running linux. It might inspire me to write my own hand writing recognition software
What does Linux have to do with that? Do you think it would be easier to write handwriting recognition software under Linux than under PalmOS? Why?
This is only tangentially related, but how do I limit the maximum size of a process? You would think (as I did) that this is what the limit command in sh/csh is for, but it doesn't seem to work on Linux 2.2.12.
See, every now and then, xmms loses its mind and grows to hundreds of megabytes. I'd like to set things up so that as soon as it hits 20M or so, it just dumps core instead.
But I find that neither limit datasize 1000 nor limit memoryuse 1000 cause subsequently-launched processes to be unable to malloc enormous amounts of memory (I tried it with a simple malloc-bomb program; its mallocs never fail.)
Is there some kernel flag I have to set to make heap limits be obeyed or something?
I might buy one, though, if I can easily write applications for it under some OS I happen to actually use. If I can quickly and easily develop applications on my desktop and then simply recompile them for a Palm,
You can already develop Palm apps on Linux: see the Palm Development HOWTO. Then you can debug it in XCopilot and never actually try it on a handheld until it's finished...
Now, if you insist on everything being the same as on Linux, meaning you want the Palm to use all the same libraries you're already used to, like Xlib and Gtk, then I think you're going to be disappointed even if you get it. Try this: pick your favorite Gtk program and launch it with --geometry =160x160 and see how well that works... Then try to find the middle-button on your Pilot's mouse!
Apps for handhelds have to be written differently, because the input and output devices are so different from what's available on desktops and laptops.
Re:Get non-Wince version, or get a Learning remote
on
The Do-It-All Remote?
·
· Score: 2
The Marantz one IS the Pronto. Same company, same body, largely the same software.
I had heard, though, that they do run WinCE. I don't know how to verify this.
I would assume that if the Pronto was running Wince, there would be a Microsoft logo in the manual somewhere. There's not.
It doesn't feel Wincey. E.g., there's no Start button...
But then, I've seen a lot of ATMs and other kiosks crashed to NT lately. Oh for the good old days when one only saw them crashed to DOS!
AOL owns Netscape. Do you think they are going to just let that investment go to waste?
AOL bought Netscape for the Netcenter portal. They didn't want the server business, and they didn't want the client. There were legal and tax reasons why Netscape couldn't be easily broken up and sold off for parts, or I'm pretty sure that's what would have happened. This whole ``Netscape/Sun Alliance'' thing is AOL's way of selling-but-not-really-selling that side of the business to Sun.
Of course not! AOL is going to replace IE with Netscape in their client; it's just a matter of time.
If there is ever an end-user version of Mozilla, and it's highly compatible with MSIE, AOL may yet do this (as a target of opportunity.) That would be a very good thing for the web, and for Linux. But I seriously doubt it's keeping anyone in AOL management up at night; they're perfectly happy using MSIE, and they won't stop until something better comes along. And if nothing ever does, it won't matter to them.
When they built the World's Largest Tesla Coil for some sheep farmer in New Zealand last March, they tested it at an old Naval yard at Hunter's Point in San Francisco, and sent out an invitation to the public to watch. It was incredible! Seeing the lightning come out of the thing, shoot out forty feet, and then just... hang there, seething... The central shaft of each lightning bolt looked solid, it looked like a physical thing you could reach out and touch.
When the mad scientist running the show got inside the cage at the top, he first said, ``and now we're going to have a highly graphical demonstration of Faraday's law.'' He said that he didn't think anyone had ever been inside a running Tesla coil before, because nobody had ever built one large enough to get inside of before. Afterward, he said, ``I feel odd, but it's not entirely unpleasant.''
(This is why I like San Francisco. ``Hey, let's go check out the World's Largest Tesla Coil tonight'' just isn't the kind of thing you're going to hear in other parts of the world.)
Another interesting site is www.austinrichards.com. I think this was the guy at Burning Man this year who had a tesla coil mounted on top of a delivery van: he was standing on top of the truck wearing a Faraday suit and waving a metal rod around, letting the lightning hit him directly, the whole time shouting through a loudspeaker, ``I am Doctor Megavolt'' or something like that. It was sublime.
If you're like me, you have a DVD and VCR. I happen to have a decent VCR, meaning there is a jog shuttle. I use this to edit, and cannot do without it. There is no universal remote that has a jog shuttle of any kind, making the whole concept of universal remote moot if you have decent video gear, let alone professional.
It's true, it would be nice if the Pronto was an inch and a half taller and had a programmable jog-shuttle wheel on it. What I did when I laid out the pages for my devices with J/S wheels was add buttons for several speeds:
[] [>>] [>>>]
Then I had it memorize 3 positions in either direction on the wheel (well, 5 positions plus "play".) This works fairly well for normal viewing usage. (Though technically I guess I've only programmed "shuttle", not "jog".)
However, I can't imagine trying to do video editing with an IR remote at all! Do you actually do that? My VCRs (Panasonic AG-1950s, thirteen years old and still going strong!) have a hardwired edit controller that can control both decks in sync or independently, that has its own J/S wheel, as well as controls for edit-lists.
IR is nice, but never underestimate the power and flexibility of a 20' cable. I can reach the whole room with it, and I don't have to point it in any particular direction.
The Woz made the ultimate IR remote a few years back, called the "Cloud 9", it could learn any other remote and was fully programmable. Of course, it was 65C02 based.
A rare find indeed, the market was not ready for that much remote.
Woz's remote was called the CORE (Cloud 9 was the company name.) I believe it was the first memorizing programmable remote control on the market.
I owned one (still do, it's around here somewhere...) The thing was a complete piece of junk. Programming it was like using a combination of assembler and APL, it had lousy range, and it burned through its batteries in about two weeks (forgetting all its programming, of course.)
It did have buttons labelled 0 through F, though, which I thought was cute...
It was a very neat device for the time, and definitely groundbreaking. But as far as being a usable remote control -- well, sadly, no. It wasn't that the market wasn't ready for it, it was that the thing just wasn't any good.
Uhhh for $400 you could buy the nicest palm out there-- the Vx. Or two visors, providing you could actually get them shipped to you successfully. And had the patience to order and reorder (rinse and repeat) until they arrived...
First of all, it's not really $400, that's the list price; you can get them for half that.
Second, yes, you could get a nice Palm for that price. And I'd consider doing that, just as soon as someone demonstrates to me that a Palm will work as well as this remote does, in terms of ease-of-programmability, ability to memorize codes of other remote controls, and broadcast range.
(And don't tell me "just solder in a new resistor." Please.)
You're not paying for a Palm clone here, folks. You're paying for both hardware and software. You're paying for a solution to a particular problem in a particular niche. If someone has a software-only solution, I'd like to see it. But to my knowlege it doesn't exist.
I can't emphasize enough how well the Pronto actually works.
I've used many programmable remotes over the last decade, and they all suck! The Pronto is the first I've used that actually does what I want it to do, and so I consider it a bargain at just about any price...
I also have a Pronto, and it is without a doubt the best remote control I have ever seen in my life. I've had it for about six months (and it's still on the first set of batteries!)
The Pronto Edit software is very easy to use; I generally do all my programming of it with the computer instead of by hand on the remote.
I only wish I could figure out how to make VMware talk to my serial port, so that I wouldn't have to boot Windows to do this, sigh...
One of the interesting things you will discover the first time you try to use a touchscreen as a remote control is that it's slightly harder to operate in the dark: because you have to actually turn your head and look at the (backlit) screen to find the buttons, you can't just feel around for them. But that's another of the benefits of the Pronto over something more Pilot-like: it has seven physical buttons that are also per-page programmable (laid out for Mute, Ch +/-, Vol +/-, Left and Right) so I tend to use those for the most frequent actions, and don't actually have to look at it much except when changing pages.
Slashdot Mirror
An AudioRequest would be cool, though a PC preconfigured to do the same thing would be even cooler.
The ArcadePC MAME Cabinet is pretty sweet. So is the Arcade2000 cabinet.
Everybody loves BRAINS. And MORE BRAINS!!
For the audiophile in your life, perhaps you should consider an $80 steel brick or a $20 green pen.
Oh, I would also like someone to find me a missile silo home, but it has to be within ten miles of San Francisco (the Presidio or the Marin Headlands would be acceptable locations.)
94304-5533 is a zip code in Palo Alto, California.
Perhaps someone should modify XTraceRoute to show the physical location that corresponds to the current time_t.
It's so much fun to be a teenager.
This is true.
This does not follow. If the goal is to have only GPLed software and no non-GPLed software, then the GPL is necessary.
You assume that to have lots of free software, there must be no non-free software, which is obviously not the case. If 10% of the world's software is free and 90% is non-free, then increasing the amount of software in the world would increase both the amout of free and non-free software (assuming the ratio held.) This is called ``growing the market'' and it benefits all who participate in that market, even if they are direct competitors.
So there are cases where non-free software can help the goal of having lots of free software. In fact, given that very nearly every piece of free software is a clone of some piece of non-free software, one could argue that non-free software is a necessary catalyst for free software! Would GIMP be as good today if they didn't have Photoshop to mine for ideas? Much though I respect the GIMP developers, I most sincerely doubt it.
Also note that the GPL is not the only license that is considered a ``free software'' license, even by GPL advocates. Don't say ``free'' when you really mean ``GPLed.''
Uh, because SunOS 4.x and SunOS 5.x aren't totally compatible? You should have said "...to work with SunOS 5.x", not "...to work with Solaris." Because Solaris 1.x and Solaris 2.x are incompatible in the same way as SunOS 4.x and SunOS 5.x.
Saying ``SunOS software doesn't work on Solaris'' is obviously false, because Solaris 1.0 was SunOS 4.1.1. So your use of the word ``Solaris'' is wrong.
No, ``Solaris includes SunOS'' is true without qualification. You should have said SunOS 5 when you meant SunOS 5, not Solaris (which can mean either SunOS 4 or 5, despite what the ambiguous common usage is.)
Not that anyone really cares... ``Solaris'' means ``SYSV'' in most people's minds just like ``hacking'' means ``breaking into computers.''
Sorry, you're wrong. (If you're going to pick nits, get the facts right!)
(and so on, through 2.6 = 5.6.)
(released long after Solaris 2.0 due to customer backlash).
Oh, and
Hope that helps...
...seems to have replaced ``Amiga Persecution Complex'' as their affliction of the day. Has there been a single example of an Amiga-related announcement actually having been followed up by something more substantial than, say, more announcements? Any time in the last eight years? Just wondering...
(Oh, and how's that Amiga Mozilla port coming along? No, not that one, the other one.)
Hit F10 to bring up the menu. Use the left and right arrows to select the menu you want. Hit Enter to pop down the menu. Note that all the menu items on the menu have one character underlined. If you type that character, it's the same as selecting that menu item.
The difference between these underlined characters (called "mnemonics") and the global keybindings (called "accelerators", and listed on the right side of each menu) is that accelerators have to be global to the application, and mnemonics only have to be unique within their menu. Also mnemonics are single characters and accelerators are usually chorded (being Control or Alt or Shift keys.)
I think you're supposed to be able to use the cursor keys to move around on the menu and then hit return to select one, but that doesn't seem to work for some reason. I'm not sure why.
By the way, if you're interested in playing with selections, XEmacs is a great way to do it, since it provides direct low-level access, which otherwise you'd have to write C code to get. For example:
In a Netscape window, select some text with the mouse.
Fire up XEmacs, and switch to the *scratch* buffer. (This is a lisp-interaction buffer, if you type a lisp expression and then type Ctrl-j after it, it will evaluate it and show you the result.)
[TARGETS TIMESTAMP TEXT STRING LENGTH FILE_NAME OWNER_OS
HOST_NAME USER CLASS NAME CLIENT_WINDOW PROCESS COMPOUND_TEXT]
That just asked Netscape what data types it is capable of converting the PRIMARY selection to. The list was returned. Let's try type TEXT:
"xterm"
There's the text I selected in Netscape. Pretty basic, let's try another:
(0 . 5)
That says that 5 characters are selected (for not very good reasons, it's returning the high and low 16 bits of the 32-bit number as a pair of shorts. This is an emacs quirk.)
"file:/tmp/foo.html"
That's the URL of the document in which the selection exists.
(19110 . 3060)
That's a time_t of when the selection was made.
And so on... By letting the owner announce the types to which the selection can be converted, and letting the requester of the selection look at that list and decide which one they would prefer, very complex and efficient interations are possible. It allows format negotiation, and good decisions about pass-by-reference versus pass-by-value (e.g., if an image is selected, do you want its name, or its bits?)
Wouldn't it be nice if when you selected some text in Netscape, and you pasted it into a text editor, it was pasted as plain-text, but when you pasted it into an HTML editor, it was pasted as raw HTML, with formatting intact? Sadly, Netscape doesn't do that, but that's the sort of thing the X selection mechanism is for.
This is not quirky, this is how X selections work.
Think of it this way: you drag out some text. That's the "primary selection." Now you do "File/Copy". That copies the primary selection to the clipboard selection.
Middle-click pastes the primary selection. File/Paste pastes the clipboard selection.
To get Mac behavior, eliminate the middle mouse button (that is, give access only to the clipboard selection, not the primary selection.)
People often confuse the primary selection and the clipboard for the simple reason that xterm doesn't have a menubar, and therfore doesn't have File/Paste, so people raised on Unix instead of real GUIs tend not to notice that the clipboard selection exists at all.
That's exactly why. Having basic Emacs keybindings in all text areas was a hard requirement from day one, so the keyboard accelerators for the menu items couldn't take over the Control key.
You are mistaken. The X selection mechanism is actually one of the few parts of X that I think was done exceptionally well. It's very powerful and uses bandwidth intelligently without getting in the way too much.
Now, it happens that there is a dramatic lack of applications out there that are capable of cutting and pasting types of data that are not text. But they do exist (Lucid's Energize did it, and XEmacs has APIs for asserting selection values of arbitrary types -- though I don't think any emacs-lisp applications take advantage of that, either.)
(By the way, the "primary selection" (the click-and-drag-with-left, paste-with-middle thing) and the "clipboard" (the File/Cut, File/Paste thing) are the same thing in different namespaces: they work exactly the same way.)
Drag-and-drop and cut-and-paste aren't exactly the same thing, but the X selection mechanism can (and should) be used to implement them both. That's what it's for...
I actually never learned how Motif drag-and-drop works, but I would expect that it uses selections inside.
Well, it doesn't hurt RSA. Monopolies, de-facto and otherwise, are very profitable.
And important thing to note here is that this is not an example of the patent system breaking down: this is an example of the patent system working as intended. A patent is a grant of a time-limited monopoly, with the condition of full disclosure of the technique. The US government granted RSA a monopoly on these algorithms, and so it's not terribly surprising that they've been able to parlay that into market domination.
I point this out because it's important to understand the differences between bad things that happen when the patent system is working as intended, versus when it's not working as intended. Most of the time, the patent-related complaints we hear are about bogus patents, patents that were obvious or trivial or already in the prior art. Those are examples of the patent system breaking down.
But if you fixed the problems at the patent office that caused bogus patents to be issued, and caused the patent office to execute their mandate correctly, situations like the RSA one would still occur.
In other words, if you're against the recent Amazon patent, you're against stupid patents. If you're against the RSA patents, you're against software patents.
The Vaio's a nice machine, but it hasn't crossed the size quantum. There are three sizes of computers that are interesting: ``it sits on my desk,'' ``it fits in a backpack,'' and ``it fits in a pocket.''
Now I've got big pockets, but the tinier Vaio still doesn't fit in them. And the keyboard is so small that my thumbs rub together when I'm on the home row. Also, it doesn't have a serial port, only USB (which means you can't use Ricochet with it.) (The bigger Vaio has a plug-on port-replicator thingy that gives you a serial port, but not the smaller version.)
Agreed on both points. But today, you can write Palm programs with gcc, and all the usual Posix libc goodies. You can't use X, you can't use Gtk. So this begs the question, what tangible benefits would Linux provide that PalmOS would not?
Exactly. As far as I can tell, using Linux rather than PalmOS wouldn't change either the programming experience or the end-user experience. In particular, there are basically no end-user Linux applications that will run on Palm Linux without first being redesigned to work with a completely different user interface.
Isn't the primary problem with Palm networking the networking hardware, not software?
Interesting, can you give some examples?
What does Linux have to do with that? Do you think it would be easier to write handwriting recognition software under Linux than under PalmOS? Why?
See, every now and then, xmms loses its mind and grows to hundreds of megabytes. I'd like to set things up so that as soon as it hits 20M or so, it just dumps core instead.
But I find that neither limit datasize 1000 nor limit memoryuse 1000 cause subsequently-launched processes to be unable to malloc enormous amounts of memory (I tried it with a simple malloc-bomb program; its mallocs never fail.)
Is there some kernel flag I have to set to make heap limits be obeyed or something?
You can already develop Palm apps on Linux: see the Palm Development HOWTO. Then you can debug it in XCopilot and never actually try it on a handheld until it's finished...
Now, if you insist on everything being the same as on Linux, meaning you want the Palm to use all the same libraries you're already used to, like Xlib and Gtk, then I think you're going to be disappointed even if you get it. Try this: pick your favorite Gtk program and launch it with --geometry =160x160 and see how well that works... Then try to find the middle-button on your Pilot's mouse!
Apps for handhelds have to be written differently, because the input and output devices are so different from what's available on desktops and laptops.
I would assume that if the Pronto was running Wince, there would be a Microsoft logo in the manual somewhere. There's not.
It doesn't feel Wincey. E.g., there's no Start button...
But then, I've seen a lot of ATMs and other kiosks crashed to NT lately. Oh for the good old days when one only saw them crashed to DOS!
AOL bought Netscape for the Netcenter portal. They didn't want the server business, and they didn't want the client. There were legal and tax reasons why Netscape couldn't be easily broken up and sold off for parts, or I'm pretty sure that's what would have happened. This whole ``Netscape/Sun Alliance'' thing is AOL's way of selling-but-not-really-selling that side of the business to Sun.
If there is ever an end-user version of Mozilla, and it's highly compatible with MSIE, AOL may yet do this (as a target of opportunity.) That would be a very good thing for the web, and for Linux. But I seriously doubt it's keeping anyone in AOL management up at night; they're perfectly happy using MSIE, and they won't stop until something better comes along. And if nothing ever does, it won't matter to them.
Don't count on AOL to save the day.
When the mad scientist running the show got inside the cage at the top, he first said, ``and now we're going to have a highly graphical demonstration of Faraday's law.'' He said that he didn't think anyone had ever been inside a running Tesla coil before, because nobody had ever built one large enough to get inside of before. Afterward, he said, ``I feel odd, but it's not entirely unpleasant.''
(This is why I like San Francisco. ``Hey, let's go check out the World's Largest Tesla Coil tonight'' just isn't the kind of thing you're going to hear in other parts of the world.)
Another interesting site is www.austinrichards.com . I think this was the guy at Burning Man this year who had a tesla coil mounted on top of a delivery van: he was standing on top of the truck wearing a Faraday suit and waving a metal rod around, letting the lightning hit him directly, the whole time shouting through a loudspeaker, ``I am Doctor Megavolt'' or something like that. It was sublime.
It's true, it would be nice if the Pronto was an inch and a half taller and had a programmable jog-shuttle wheel on it. What I did when I laid out the pages for my devices with J/S wheels was add buttons for several speeds:
Then I had it memorize 3 positions in either direction on the wheel (well, 5 positions plus "play".) This works fairly well for normal viewing usage. (Though technically I guess I've only programmed "shuttle", not "jog".)
However, I can't imagine trying to do video editing with an IR remote at all! Do you actually do that? My VCRs (Panasonic AG-1950s, thirteen years old and still going strong!) have a hardwired edit controller that can control both decks in sync or independently, that has its own J/S wheel, as well as controls for edit-lists.
IR is nice, but never underestimate the power and flexibility of a 20' cable. I can reach the whole room with it, and I don't have to point it in any particular direction.
Woz's remote was called the CORE (Cloud 9 was the company name.) I believe it was the first memorizing programmable remote control on the market.
I owned one (still do, it's around here somewhere...) The thing was a complete piece of junk. Programming it was like using a combination of assembler and APL, it had lousy range, and it burned through its batteries in about two weeks (forgetting all its programming, of course.)
It did have buttons labelled 0 through F, though, which I thought was cute...
It was a very neat device for the time, and definitely groundbreaking. But as far as being a usable remote control -- well, sadly, no. It wasn't that the market wasn't ready for it, it was that the thing just wasn't any good.
First of all, it's not really $400, that's the list price; you can get them for half that.
Second, yes, you could get a nice Palm for that price. And I'd consider doing that, just as soon as someone demonstrates to me that a Palm will work as well as this remote does, in terms of ease-of-programmability, ability to memorize codes of other remote controls, and broadcast range.
(And don't tell me "just solder in a new resistor." Please.)
You're not paying for a Palm clone here, folks. You're paying for both hardware and software. You're paying for a solution to a particular problem in a particular niche. If someone has a software-only solution, I'd like to see it. But to my knowlege it doesn't exist.
I can't emphasize enough how well the Pronto actually works .
I've used many programmable remotes over the last decade, and they all suck! The Pronto is the first I've used that actually does what I want it to do, and so I consider it a bargain at just about any price...
The Pronto Edit software is very easy to use; I generally do all my programming of it with the computer instead of by hand on the remote.
I only wish I could figure out how to make VMware talk to my serial port, so that I wouldn't have to boot Windows to do this, sigh...
One of the interesting things you will discover the first time you try to use a touchscreen as a remote control is that it's slightly harder to operate in the dark: because you have to actually turn your head and look at the (backlit) screen to find the buttons, you can't just feel around for them. But that's another of the benefits of the Pronto over something more Pilot-like: it has seven physical buttons that are also per-page programmable (laid out for Mute, Ch +/-, Vol +/-, Left and Right) so I tend to use those for the most frequent actions, and don't actually have to look at it much except when changing pages.