> Italics means BabelFish blew it so thoroughly I couldn't make sense of it.
Well huh. I looked said stuff up in a German dictionary (just Ask Jeeves where you can find a German dictionary) and here's what I got:
Digitalausgang = Digital, so it's SPDIF Digital. I guess that means that if you have SPDIF outputs, this technology defeats them. Problem might be that the CD player uses SPDIF internally. My sound card (SB Live!) has SPDIF in and out. Thing is though: SPDIF streams can be flagged "copyrighted". I guess this is to make absolutely sure that you can't somehow defeat that flag.
"c't": c't is Heise.de's "Magazine of Computer Technology" (a rough translation of the magazine's subtitle by someone who doesn't know German, but it sounds right). It's not a German word!
The part of the sentence in question with "concomitantly" is:
und - and
damit - with it
auch - likewise
the
Digitalsignal - guess...
What a difference a human makes!
Kenneth
Disclaimer: This is an educated choice from the possible translations of a German dictionary (Foreign Language Master). I know no German besides that which sounds a lot like English (according to one of my friends who is taking German, this is about 50%).
Remember the Ask Slashdot about simulating walking in a VR environment? One of the topics raised many times was the problem of perception of motion. Well here is our solution.
IE5/Win lets you do zones, which work all right but it lumps all sorts of other security stuff with it, like Java and ActiveX security. Just because I want cookies to be accepted by all of some subset of sites doesn't mean that for each site I want it to treat my security the same way.
Missed one important option: selective cookie acceptance. Almost all browsers allow this.
Idea for the protocol for such an exchange:
The servers would be distributed, like the DNS system. One per ad server or whatever.
When a browser receives a cookie:
It checks the cookie against the "allow" list.
It checks the cookie against the "ask" list.
It checks the cookie against the "deny" list.
It sends the cookie to the cookie exchange server.
The packet to the exchange server would contain at least the following information:
Protocol version (for extensibility)
Cookie given
Site visiting when cookie received (you must be able to opt out of sending this information, for privacy purposes.)
The src= for the img tag
(possibly a user tag - like/not like)
Cookie sending would involve a similar exchange (since the mere presence of a cookie on a computer can be used for tracking without sending another cookie). If a cookie is sent more than once to a site that the user has specifically not said "allow" or "ask" to, it would also be exchanged. On the same subject, if the cookie was originally on the "ask" list when it was received, also ask before sending it back (option to exchange it also).
The entire system must be structured to minimize the bandwidth necessary and the time delay in loading the page. This is why exchanging would be done on the receiving side always, and on the sending side sometimes; you can display the page without knowning whether or not to store a cookie, but you may need a cookie to load a page properly.
The server would contain information by server as to whether or not scrambling the ID would work, and if it did work, the server would send back a scrambled cookie instead of a "recycled" cookie. The server would also keep track of how many times each cookie was sent so it could even out the averages.
It would be necessary to keep users from accidently exchanging a cookie with a username/password in it. Maybe the browser could check whether a form submission included the text from an item whose input type was "password", then either "ask" or "allow" all cookies from that site by user preference.
The worst-case scenario would be that the user was not tracked at all. The more likely case is that Internet cookie-based advertising dies. Be careful not to let up your guard if that happens, though.
I set up a firewall (simple ipchains... -d ad.doubleclick.net -j DENY) and it was a tremendous pain, especially with IE. In my experience, if IE5 has trouble loading an ad, it will cover the entire page with a location unavailable or something like that error. (I have seen IE only cover the image with the same, but that depends on some things that I don't want to spend my time investigating.) My bigger problem, though, was the problem of a heck of a lot of ad servers under doubleclick.net. I block one, the next day they have a different one and I block that, etc. It gets tiring, and I am still not guaranteed that somehow it isn't sneaking a cookie by (like a 1x1 image the same color as the background or something). I want the ability to block cookies from specific sites. This should be implemented in the browser, and is already sort of implemented in IE through the use of "zones". However, I have found this not to be flexible enough. I want to be able to specifically block cookies from specific sites, or even better, block cookies transmitted along with non-html (i.e. banner ads) content. Since we have the code for Mozilla, why not do it there? A simple list in the user preferences file would be enough to accomplish the desired result. Block cookies that come from *.doubleclick.net. Block all cookies that are attached to things not of type text/html. Or act like IE: only allow cookies from Slashdot, etc. Go ahead and tell me it's already been done and make me feel stupid... or make me happy by saying that you're working on it. If no one else wants to work on it, I'll try it. Can't be that hard.
Kenneth (who wishes to remain "anonymous" but is definately not).
PS - I don't like out-of-browser filters because (a) they are usually not cross-platform, while good browsers are, and (b) many rely on proxies, which I don't like, not to mention (c) they take up speed and resources on my computer that I don't really want to give away if I can help it.
Italicizing that was a question not of its meaning, but of its usage in the sentence, which I was pretty sure I got right, looking back at the original German. In any case this was just a fix-up of a computer translation.
In that case, we must ensure that Red Book==the logo, but in either case, there are better threads for discussing the legal implications of this "technology".
Win98 telnet is just as bad. It always cuts off the last line when you use 'less', for example, which really annoys me, not to mention that the arrow keys, page-up and page-down, and a bunch of other stuff simply does not work. I recommend getting one of the many freeware/shareware telnet clients available at download.com in the Windows:Internet section (this direct link may work).
Not complely clear about this (translation is kind of muddled) but do they have a "CD Digital Audio" logo on them? If they do, go ahead and sue. If they don't have such a logo, it doesn't have to be CD Digital Audio.
> BMG Entertainment, the music division of the Bertelsmann company, has started selling CDs with copy protection using the Cactus DATA Shield. This system was developed in co-operation with the Israeli software enterprise Midbar and Sonopress from Germany. So far the albums "Razorblade Romance" by Him and "My Private War" by former Independent Heroen Philip Boa & The Voodoo club(?). The copy protection prevents not only playing and copying of the audio tracks with the PC, but also playing on all D-CRcOcM-cDrives(?) as well as on older audio CD players.
The italics are an uncertain fix. I (or preferrably someone else) should look them up in a German dictionary to verify the exact meaning.
If you like RPN and are stuck with a TI (89, I could port it to the 83 if you want), check out my RPN program at ticalc.org (I just submitted it so it's probably not there yet). There is also a French RPN program that has a lot of nice features but requires the French language pack, which requires AMS 2.03, which is a royal pain if you are even thinking of ASM or complex Basic.
Up at my website in a few minutes (I have to boot to Linux and back to Windows) will be the last released version of a guy named James's code. He hacked the AOL IP tunneling code. I personally have not been able to get it to work, but I bet a bunch of Slashdotters with AOL as their only "ISP" could:-). It doesn't have any of the AOL front-end interface things, but the guy also hacked the AOL mail protocol, and there are a bunch of AIM clients for all sorts of OSes (I like gAIM, check Freshmeat for it).
A possible use of this code, besides AOL access under Linux, is a port to Windows as a networking driver. On my local AOL number, you dial it from a terminal and login as "aol", so you should be able to interface this with Windows Dial-Up Networking (username aol, password whatever). You'd again only get IP tunneling.
The reason I am posting it myself and not linking to his site is because, last I checked, his site is down. I hope I am not violating any licenses by redistributing this.
If you figure this out or have a better link, tell us.
...that little VB to GTK converter program that you wrote?
Hey, you write a chat program, what can you expect? Everyone wants their own chat program, and it makes things so much easier if they can find it for free. They ignore the GPL and copy it, making a few stupid changes if any. I got a couple of things that I'm planning to finish and release and I am worried about whether or not they will be stolen by someone else. I suppose that if I get an "official" Freshmeat announcement about it, no one can steal it and get away with it for that long. That's my suggestion: make sure that you announce it on some prominant site.
>If you wanted RAM for Quake or something, you could quadruple your PC's tally by borrowing from your appliances, or multiply your framerate tenfold by plugging your Voodoo card into your toaster and installing Linux on it!
So basically what you're saying is "Beowulf cluster" (well not exactly). We've heard that already. Yeah, maybe you'd get some framerate at all out of your toaster! A toaster with Linux on it isn't a toaster anymore. True, Linux has a heck of a lot of credibility as an embedded OS, but, "twit", A TOASTER IS NOT AN EMBEDDED SYSTEM, nor will it ever be, because if it does become one, then it immediately changes from a respectable status as a useful toaster into an impractical mess. Ever heard of making a car that turns into an airplane? Well at least the two both involve the similar concepts of propulsion and steering. Computing and toasters? Yeah, right. Burn up your expensive chips, that's what. No, a Crusoe wouldn't help.
I'm not spending the $100 or whatever it is for a calc just so I can feel justified when I get a superior model for not that many bucks more. Don't worry, dude. I struggled with a TI-83. I'm glad to get rid of that, and it sure was a blow to my intellegence to find out how little more symbolic manipulation, hi-res screen, fast processor, and lots more was. I was doing all sorts of stuff with trying to get lower-case characters in my programs, simplifying radicals, doing the quadratic formula, etc. The 89 was like a sigh of relief to me, a loosening of my shackles, a breath of fresh air. Okay, I'll shut up. Anyway, I did my fair share of struggling. I got the 83 right before Geometry and AlgII (same year), so I went through Algebra I with a handheld calculator. I got my 89 this recent Christmas. So I'm not cheating, friend. But unfourtunately this applies specifically to me. There are kids I know of who totally depend on the calc for everything, even simple things like 25^.5 . Calc in moderation is good.
A program is an algorithmic construct. By putting your mental processes into algorithmic form, you gain a better understanding of how they work. Your friend is (no bragging here) a bad programmer. Programs can be pieces of art, or pieces of "crap". I try to write art. I try to write an elegant program that gets the job done in the "nicest" way. That is what is very educational for me.
It doesn't matter what platform it's for. I so happen to carry around a TI-89, and when I'm bored I take it out and start making a program. If I was by my computer I might use that, but for computer programming I think that one concentrates too much on the interface, etc. details. With the graphing calc, you just write the program. All the math, string, list, etc. stuff is there and you use it often normally.
If you are not a good programmer, yeah a lot of your programs are going to come from friends. I prefer to write my own, though, because I usually can do in my head what they write programs to help themselves with. They write bad programs anyway.
Take CTY's Theoretical Foundations of Computer Science (and understand it) and I think you'll figure it out.
Linux works fine on the platforms that other OSes already run fine on, not absurd gadgets that it would have no point running on. In fact, it would be an insult to the Linux community if Linux was ported to, say, a wrist-watch MP3 player. Linux would get the press for being the "most pointless OS in existence".
I once wrote a post-fix interpreter in Scheme. My problem, though, was that it was kind of difficult to understand for a while. I suppose that I could eventually get used to it.
That is the problem with RPN. It is more compliccated to the beginner, who is used to seeing equations written in infix form. A novice user who sees x^2+2x+2=0 in a book will naturally want to enter it that way, instead of x 2 ^ 2 x * + 2 + (then however you enter the =0). Admittedly, there are some really nice things about post-fix, namely the exclusion of nasty parentheses, some of which I have found out through this thread and others in this discussion. That is not to say that there isn't a lot to like about in-fix, too. In the end, it is one's personal preferrence, either way. Your original reply should have been moderated Redundant if it were not for the ensuing thread because I had said in my original post that it is a matter of personal preferrence. If someone likes infix, fine. If you like postfix, fine. However, whomever keeps a closed mind to the other type is missing out on something.
I think that this is enough for the in/post-fix debate. I'll probably get an HP for my next calc (when I get the $ and can give up a CPU upgrade) and try out post-fix. In the mean time I might write a postfix parser for my TI-89. Might be fun:-). But for beginners and many intermediate students, the obvious choice is the TI. There is the issue of "what everybody else has got", which includes what the teacher has (and thus what the teacher can help you with), games (the best games are those that everybody else has:-), and, of course, in-fix.
We have learned these days to use the graphing calculator as a tool. I myself am fully capable of solving equations, graphing, finding zeros, etc., but those things are secondary to an understanding of how a problem works. The key thing in my discussion is that I use the calculator as a tool to aid me in solving the nitty-gritty details of a problem while I concentrate on the big picture. What is this problem trying to say? What can I learn from it? I never put anything into the calculator that I couldn't do myself in enough time. The calculator allows me to concentrate on what I really need to do instead of what takes up all of the time in problem-solving. I sometimes enter a complex equation into the calculator, and then tell it to do a certain set of steps to solve it (divide by a, then add b*c/2, etc.), and use the symbolic manipulation to check my work. I will also use the symbolic manipulation when the calculation is trivial and I have done it myself ad nauseum.
My age has nothing to do with my skill level. I think that if every student was introduced to mathematics in the same way as I was and had the same attitude towards it as I did, that every student would be taking precalc in middle school. I didn't get the calculator until Geometry/AlgII, and I didn't use it much, if at all, in Geometry. For Algebra II, the curriculum was carefully designed around using the calculator as a tool. And don't you mean, "you can't be less than 17 years old."?
It will usually take more time to develop a program to automate the solution, then you would learning do it by yourself.
Aha! I get to respond to this rant! You see, in order to write a program to do something you must have an intimate knowledge of it yourself. The fact is that computers are inherently stupid, and to teach a computer to do something is like teaching a child to do something; thus it is like the job of a teacher. Although a teacher can teach a class something without knowing it him/herself, the students gain a much better understanding of it if the teacher knows the material very well. When I write programs, I am really just teaching the calculator how to do something. Unless it's a guess-and-check kind of program, I must be fluent in whatever I want the program to do. On the 89 you can kind of "cheat" by using its internal symbolic manipulation capabilities, but I find it more fun to do it myself. I write programs to help myself, not so much as to have an aid on the calculator (although this does help), but to check how much I really know whatever it is I am working with. If I find that, while writing a program, something doesn't work out right and I fix it, or some algorithm is cumbersome and I improve it, I have learned something about the topic in question on my own. That really helps me.
Linux won't be ported to the TI calculator series because there's not enough memory and TI won't release enough specs for someone with enough interest to develop the standard C libraries on the calculator. So scratch that idea.
First thought: dan! Then: (1) there are devices available to expand the calculator's memory, and some work quite nicely (we're talking up to 4 MB in some cases, and I don't see a reason why (at least for the 68k calcs) that they can't go higher), and (2) many libraries have already been developed for the 89/92(+) shells that do similar things. You wouldn't be able to do much, but it would be kind of neat. And what are you talking about, "specs". Just about everything anyone needs to know about these calculators has been figured out in a short time by the assembly programmers.
The TI calculators use hardware which used to be closed. But again, now that the ASM programmers attacked it, the hardware is open. And for most of the TI calcs, the processor is the Z80, which is thoroughly documented, as is its accompanying linkport hardware. And although the TI GraphLink grey cable is a mysterious grey box at the end of a cord, again people have hacked it (it has been hacked for quite some time now). I have reason to believe that the protocol for the grey link cable is no more than a pass-through, possibly with some ioctl-like commands. I may (if I am not lazy) look at the GtkTiLink code to verify this. There are also other link cables available, both serial and parallel. Of couse since users made them, users know how they work. As for the protocol, that was somewhat documented by TI and has been expanded to an almost full knowledge by GtkTiLink. So linkport is No Problem (tm).
The TIs can be very durable, depending on what you call durable. However, it is quite likely that the HPs have gained a better track record due to increased care. Example: where do you store your calculator when you carry it around? I shove it in the back of my backpack, push it around, occasionally drop it on the floor (by accident), spill things on it (LCD display), etc., and that only the anti-slip pad came loose was a marvel to me. I could get a new one if it mattered, but it doesn't. I don't need it.Or I could just take off the other one also and then at least it would lay flat (though it doesn't really matter to me). The only problem that I really do have with the TIs' durability is that the covers that usually click shut and hold the calculator gradually get to the point where they won't latch on to the calculator anymore. This is a stupid flaw in the TI case design, and if I had made it it wouldn't do that (yeah right). If you are talking about maintainence costs to keep it near-new, I could replace my cover, solving both problems, for not that much (don't remember off-hand).
Anyway, I have my calculator and I'm not spending any more money to get another one until I have to. My TI-89 contains enough mathematical features to last me through my graduate studies (I a freshman in high school).
Okay, I don't own an HP, and have only seen them a few times. My question is: how does the HP48GX or whatever similar calc display what it has figured out? And is this compatible with the way your mind works? Though I do find it difficult sometimes to enter the equation/expression I want into the calculator, with the Pretty Print capability I am able to see what I entered as it should look. Another nice thing that you mention about the RPN calcs also applies to my TI-89 -- if I find a complex equation/expression, I will look for the main part/operation(/connective), and enter this into the calculator. I will then find the next operation and enter ans(1) (op) whatever. The TI-89 automatically simplifies this so that it kind of looks like on the screen what I see in the book/board/whatever, and it's easy for me to tell what I missed. With this method, I rarely bother about parentheses except in situations such as (x+1)/(x+2), and I can't conceive how you could enter this without using something at least similar to parentheses.
The TI-82 definately has limitations, so I am not suprised that you noticed a substantial improvement when moving up to a real calculator. The TI-82 is basically a hand-held calculator with the ability to make a few simple decisions. Suprisingly it is Turing-complete, but it offers not that much more over that. The TI-89 (and the HP) extend the calculator's functions to much more closely relate to the human brain. When you get to be that complex (brain == the absolute most complex computing device ever constructed), personal preferences have a big impact on how easy it is for the brain to parse and do good things with what it is given, I sometimes have a fine line between what I "get" and what flies over my head. (to continue the metaphor, I guess I just "duck" more or less each time) Thus, if what the calculator throws at me is a bit too complex, often I will just gawk at it and do it on my own. Back to my initial question: how does the HP48 display what it figures out?
Again, the critical question is of the interface between your brain and the calculator. Your brain has built a set of routines for converting its internal representation of a mathematical form into RPN notation (redundant), and conversely finding a brain-internal representation for what the calc throws at you. My set of interface routines work with infix-like stuff, and it works nicely for me because on the input side, I only have to maintain one set of routines to convert both what my calc shows me and what the book shows me (and what is written on the board) into "internal representation". Also, what I throw at the calculator is often similar to wat I throw at other humans. It just works that way for me.
Cross-platform: Check out GtkTiLink. And just think: the 89, 92, and 92+ are all based on the Motorola 68000 processor, and we have a version of Linux for that processor, don't we?
Durability: My TI-83 still works just fine after a year and a half; the only problem I have had is one of the anti-slip pads coming loose, but I used that thing a heck of a lot. If I hadn't have gotten a Ti-89 for Christmas, I would still be using that 83 for years to come.
Slashdot Mirror
> Italics means BabelFish blew it so thoroughly I couldn't make sense of it.
Well huh. I looked said stuff up in a German dictionary (just Ask Jeeves where you can find a German dictionary) and here's what I got:
What a difference a human makes!
Kenneth
Disclaimer: This is an educated choice from the possible translations of a German dictionary (Foreign Language Master). I know no German besides that which sounds a lot like English (according to one of my friends who is taking German, this is about 50%).
Remember the Ask Slashdot about simulating walking in a VR environment? One of the topics raised many times was the problem of perception of motion. Well here is our solution.
IE5/Win lets you do zones, which work all right but it lumps all sorts of other security stuff with it, like Java and ActiveX security. Just because I want cookies to be accepted by all of some subset of sites doesn't mean that for each site I want it to treat my security the same way.
Missed one important option: selective cookie acceptance. Almost all browsers allow this.
Idea for the protocol for such an exchange:
When a browser receives a cookie:
The packet to the exchange server would contain at least the following information:
Cookie sending would involve a similar exchange (since the mere presence of a cookie on a computer can be used for tracking without sending another cookie). If a cookie is sent more than once to a site that the user has specifically not said "allow" or "ask" to, it would also be exchanged. On the same subject, if the cookie was originally on the "ask" list when it was received, also ask before sending it back (option to exchange it also).
The entire system must be structured to minimize the bandwidth necessary and the time delay in loading the page. This is why exchanging would be done on the receiving side always, and on the sending side sometimes; you can display the page without knowning whether or not to store a cookie, but you may need a cookie to load a page properly.
The server would contain information by server as to whether or not scrambling the ID would work, and if it did work, the server would send back a scrambled cookie instead of a "recycled" cookie. The server would also keep track of how many times each cookie was sent so it could even out the averages.
It would be necessary to keep users from accidently exchanging a cookie with a username/password in it. Maybe the browser could check whether a form submission included the text from an item whose input type was "password", then either "ask" or "allow" all cookies from that site by user preference.
The worst-case scenario would be that the user was not tracked at all. The more likely case is that Internet cookie-based advertising dies. Be careful not to let up your guard if that happens, though.
Suggestions for improvements?
Ken, eating his cookies
I set up a firewall (simple ipchains ... -d ad.doubleclick.net -j DENY) and it was a tremendous pain, especially with IE. In my experience, if IE5 has trouble loading an ad, it will cover the entire page with a location unavailable or something like that error. (I have seen IE only cover the image with the same, but that depends on some things that I don't want to spend my time investigating.) My bigger problem, though, was the problem of a heck of a lot of ad servers under doubleclick.net. I block one, the next day they have a different one and I block that, etc. It gets tiring, and I am still not guaranteed that somehow it isn't sneaking a cookie by (like a 1x1 image the same color as the background or something). I want the ability to block cookies from specific sites. This should be implemented in the browser, and is already sort of implemented in IE through the use of "zones". However, I have found this not to be flexible enough. I want to be able to specifically block cookies from specific sites, or even better, block cookies transmitted along with non-html (i.e. banner ads) content. Since we have the code for Mozilla, why not do it there? A simple list in the user preferences file would be enough to accomplish the desired result. Block cookies that come from *.doubleclick.net. Block all cookies that are attached to things not of type text/html. Or act like IE: only allow cookies from Slashdot, etc. Go ahead and tell me it's already been done and make me feel stupid... or make me happy by saying that you're working on it. If no one else wants to work on it, I'll try it. Can't be that hard.
Kenneth (who wishes to remain "anonymous" but is definately not).
PS - I don't like out-of-browser filters because (a) they are usually not cross-platform, while good browsers are, and (b) many rely on proxies, which I don't like, not to mention (c) they take up speed and resources on my computer that I don't really want to give away if I can help it.
Italicizing that was a question not of its meaning, but of its usage in the sentence, which I was pretty sure I got right, looking back at the original German. In any case this was just a fix-up of a computer translation.
In that case, we must ensure that Red Book==the logo, but in either case, there are better threads for discussing the legal implications of this "technology".
Win98 telnet is just as bad. It always cuts off the last line when you use 'less', for example, which really annoys me, not to mention that the arrow keys, page-up and page-down, and a bunch of other stuff simply does not work. I recommend getting one of the many freeware/shareware telnet clients available at download.com in the Windows:Internet section (this direct link may work).
Not complely clear about this (translation is kind of muddled) but do they have a "CD Digital Audio" logo on them? If they do, go ahead and sue. If they don't have such a logo, it doesn't have to be CD Digital Audio.
Looking back at it, my "fix" of DS to CD is obviously incorrect. Is this supposed to be "audio DVDs"?
> BMG Entertainment, the music division of the Bertelsmann company, has started selling CDs with copy protection using the Cactus DATA Shield. This system was developed in co-operation with the Israeli software enterprise Midbar and Sonopress from Germany. So far the albums "Razorblade Romance" by Him and "My Private War" by former Independent Heroen Philip Boa & The Voodoo club(?). The copy protection prevents not only playing and copying of the audio tracks with the PC, but also playing on all D-CRcOcM-cDrives(?) as well as on older audio CD players.
The italics are an uncertain fix. I (or preferrably someone else) should look them up in a German dictionary to verify the exact meaning.
If you like RPN and are stuck with a TI (89, I could port it to the 83 if you want), check out my RPN program at ticalc.org (I just submitted it so it's probably not there yet). There is also a French RPN program that has a lot of nice features but requires the French language pack, which requires AMS 2.03, which is a royal pain if you are even thinking of ASM or complex Basic.
Kenneth
The proper link is aol-0_5.tgz.
Kenneth
Up at my website in a few minutes (I have to boot to Linux and back to Windows) will be the last released version of a guy named James's code. He hacked the AOL IP tunneling code. I personally have not been able to get it to work, but I bet a bunch of Slashdotters with AOL as their only "ISP" could :-). It doesn't have any of the AOL front-end interface things, but the guy also hacked the AOL mail protocol, and there are a bunch of AIM clients for all sorts of OSes (I like gAIM, check Freshmeat for it).
A possible use of this code, besides AOL access under Linux, is a port to Windows as a networking driver. On my local AOL number, you dial it from a terminal and login as "aol", so you should be able to interface this with Windows Dial-Up Networking (username aol, password whatever). You'd again only get IP tunneling.
The reason I am posting it myself and not linking to his site is because, last I checked, his site is down. I hope I am not violating any licenses by redistributing this.
If you figure this out or have a better link, tell us.
Kenneth
...that little VB to GTK converter program that you wrote?
Hey, you write a chat program, what can you expect? Everyone wants their own chat program, and it makes things so much easier if they can find it for free. They ignore the GPL and copy it, making a few stupid changes if any. I got a couple of things that I'm planning to finish and release and I am worried about whether or not they will be stolen by someone else. I suppose that if I get an "official" Freshmeat announcement about it, no one can steal it and get away with it for that long. That's my suggestion: make sure that you announce it on some prominant site.
Ken
>If you wanted RAM for Quake or something, you could quadruple your PC's tally by borrowing from your appliances, or multiply your framerate tenfold by plugging your Voodoo card into your toaster and installing Linux on it!
So basically what you're saying is "Beowulf cluster" (well not exactly). We've heard that already. Yeah, maybe you'd get some framerate at all out of your toaster! A toaster with Linux on it isn't a toaster anymore. True, Linux has a heck of a lot of credibility as an embedded OS, but, "twit", A TOASTER IS NOT AN EMBEDDED SYSTEM, nor will it ever be, because if it does become one, then it immediately changes from a respectable status as a useful toaster into an impractical mess. Ever heard of making a car that turns into an airplane? Well at least the two both involve the similar concepts of propulsion and steering. Computing and toasters? Yeah, right. Burn up your expensive chips, that's what. No, a Crusoe wouldn't help.
Kenneth
What did you get the 85 to do?
I'm not spending the $100 or whatever it is for a calc just so I can feel justified when I get a superior model for not that many bucks more. Don't worry, dude. I struggled with a TI-83. I'm glad to get rid of that, and it sure was a blow to my intellegence to find out how little more symbolic manipulation, hi-res screen, fast processor, and lots more was. I was doing all sorts of stuff with trying to get lower-case characters in my programs, simplifying radicals, doing the quadratic formula, etc. The 89 was like a sigh of relief to me, a loosening of my shackles, a breath of fresh air. Okay, I'll shut up. Anyway, I did my fair share of struggling. I got the 83 right before Geometry and AlgII (same year), so I went through Algebra I with a handheld calculator. I got my 89 this recent Christmas. So I'm not cheating, friend. But unfourtunately this applies specifically to me. There are kids I know of who totally depend on the calc for everything, even simple things like 25^.5 . Calc in moderation is good.
nuf said. Ken
A program is an algorithmic construct. By putting your mental processes into algorithmic form, you gain a better understanding of how they work. Your friend is (no bragging here) a bad programmer. Programs can be pieces of art, or pieces of "crap". I try to write art. I try to write an elegant program that gets the job done in the "nicest" way. That is what is very educational for me.
It doesn't matter what platform it's for. I so happen to carry around a TI-89, and when I'm bored I take it out and start making a program. If I was by my computer I might use that, but for computer programming I think that one concentrates too much on the interface, etc. details. With the graphing calc, you just write the program. All the math, string, list, etc. stuff is there and you use it often normally.
If you are not a good programmer, yeah a lot of your programs are going to come from friends. I prefer to write my own, though, because I usually can do in my head what they write programs to help themselves with. They write bad programs anyway.
Take CTY's Theoretical Foundations of Computer Science (and understand it) and I think you'll figure it out.
Kenneth
waffles
Okay, so that wasn't a common one.
Linux works fine on the platforms that other OSes already run fine on, not absurd gadgets that it would have no point running on. In fact, it would be an insult to the Linux community if Linux was ported to, say, a wrist-watch MP3 player. Linux would get the press for being the "most pointless OS in existence".
-ka
I once wrote a post-fix interpreter in Scheme. My problem, though, was that it was kind of difficult to understand for a while. I suppose that I could eventually get used to it.
That is the problem with RPN. It is more compliccated to the beginner, who is used to seeing equations written in infix form. A novice user who sees x^2+2x+2=0 in a book will naturally want to enter it that way, instead of x 2 ^ 2 x * + 2 + (then however you enter the =0). Admittedly, there are some really nice things about post-fix, namely the exclusion of nasty parentheses, some of which I have found out through this thread and others in this discussion. That is not to say that there isn't a lot to like about in-fix, too. In the end, it is one's personal preferrence, either way. Your original reply should have been moderated Redundant if it were not for the ensuing thread because I had said in my original post that it is a matter of personal preferrence. If someone likes infix, fine. If you like postfix, fine. However, whomever keeps a closed mind to the other type is missing out on something.
I think that this is enough for the in/post-fix debate. I'll probably get an HP for my next calc (when I get the $ and can give up a CPU upgrade) and try out post-fix. In the mean time I might write a postfix parser for my TI-89. Might be fun :-). But for beginners and many intermediate students, the obvious choice is the TI. There is the issue of "what everybody else has got", which includes what the teacher has (and thus what the teacher can help you with), games (the best games are those that everybody else has :-), and, of course, in-fix.
cya, Kenneth
TI-89:
(integral) (e^(-x^2),x) (ENTER)
(integral) (e^(-x^2),x)So your answer is no. You can, however, write a program to do this by using the part() function.
Ken
We have learned these days to use the graphing calculator as a tool. I myself am fully capable of solving equations, graphing, finding zeros, etc., but those things are secondary to an understanding of how a problem works. The key thing in my discussion is that I use the calculator as a tool to aid me in solving the nitty-gritty details of a problem while I concentrate on the big picture. What is this problem trying to say? What can I learn from it? I never put anything into the calculator that I couldn't do myself in enough time. The calculator allows me to concentrate on what I really need to do instead of what takes up all of the time in problem-solving. I sometimes enter a complex equation into the calculator, and then tell it to do a certain set of steps to solve it (divide by a, then add b*c/2, etc.), and use the symbolic manipulation to check my work. I will also use the symbolic manipulation when the calculation is trivial and I have done it myself ad nauseum.
My age has nothing to do with my skill level. I think that if every student was introduced to mathematics in the same way as I was and had the same attitude towards it as I did, that every student would be taking precalc in middle school. I didn't get the calculator until Geometry/AlgII, and I didn't use it much, if at all, in Geometry. For Algebra II, the curriculum was carefully designed around using the calculator as a tool. And don't you mean, "you can't be less than 17 years old."?
Kenneth
It will usually take more time to develop a program to automate the solution, then you would learning do it by yourself.
Aha! I get to respond to this rant! You see, in order to write a program to do something you must have an intimate knowledge of it yourself. The fact is that computers are inherently stupid, and to teach a computer to do something is like teaching a child to do something; thus it is like the job of a teacher. Although a teacher can teach a class something without knowing it him/herself, the students gain a much better understanding of it if the teacher knows the material very well. When I write programs, I am really just teaching the calculator how to do something. Unless it's a guess-and-check kind of program, I must be fluent in whatever I want the program to do. On the 89 you can kind of "cheat" by using its internal symbolic manipulation capabilities, but I find it more fun to do it myself. I write programs to help myself, not so much as to have an aid on the calculator (although this does help), but to check how much I really know whatever it is I am working with. If I find that, while writing a program, something doesn't work out right and I fix it, or some algorithm is cumbersome and I improve it, I have learned something about the topic in question on my own. That really helps me.
Linux won't be ported to the TI calculator series because there's not enough memory and TI won't release enough specs for someone with enough interest to develop the standard C libraries on the calculator. So scratch that idea.
First thought: dan! Then: (1) there are devices available to expand the calculator's memory, and some work quite nicely (we're talking up to 4 MB in some cases, and I don't see a reason why (at least for the 68k calcs) that they can't go higher), and (2) many libraries have already been developed for the 89/92(+) shells that do similar things. You wouldn't be able to do much, but it would be kind of neat. And what are you talking about, "specs". Just about everything anyone needs to know about these calculators has been figured out in a short time by the assembly programmers.
The TI calculators use hardware which used to be closed. But again, now that the ASM programmers attacked it, the hardware is open. And for most of the TI calcs, the processor is the Z80, which is thoroughly documented, as is its accompanying linkport hardware. And although the TI GraphLink grey cable is a mysterious grey box at the end of a cord, again people have hacked it (it has been hacked for quite some time now). I have reason to believe that the protocol for the grey link cable is no more than a pass-through, possibly with some ioctl-like commands. I may (if I am not lazy) look at the GtkTiLink code to verify this. There are also other link cables available, both serial and parallel. Of couse since users made them, users know how they work. As for the protocol, that was somewhat documented by TI and has been expanded to an almost full knowledge by GtkTiLink. So linkport is No Problem (tm).
The TIs can be very durable, depending on what you call durable. However, it is quite likely that the HPs have gained a better track record due to increased care. Example: where do you store your calculator when you carry it around? I shove it in the back of my backpack, push it around, occasionally drop it on the floor (by accident), spill things on it (LCD display), etc., and that only the anti-slip pad came loose was a marvel to me. I could get a new one if it mattered, but it doesn't. I don't need it.Or I could just take off the other one also and then at least it would lay flat (though it doesn't really matter to me). The only problem that I really do have with the TIs' durability is that the covers that usually click shut and hold the calculator gradually get to the point where they won't latch on to the calculator anymore. This is a stupid flaw in the TI case design, and if I had made it it wouldn't do that (yeah right). If you are talking about maintainence costs to keep it near-new, I could replace my cover, solving both problems, for not that much (don't remember off-hand).
Anyway, I have my calculator and I'm not spending any more money to get another one until I have to. My TI-89 contains enough mathematical features to last me through my graduate studies (I a freshman in high school).
Kenneth
Okay, I don't own an HP, and have only seen them a few times. My question is: how does the HP48GX or whatever similar calc display what it has figured out? And is this compatible with the way your mind works? Though I do find it difficult sometimes to enter the equation/expression I want into the calculator, with the Pretty Print capability I am able to see what I entered as it should look. Another nice thing that you mention about the RPN calcs also applies to my TI-89 -- if I find a complex equation/expression, I will look for the main part/operation(/connective), and enter this into the calculator. I will then find the next operation and enter ans(1) (op) whatever. The TI-89 automatically simplifies this so that it kind of looks like on the screen what I see in the book/board/whatever, and it's easy for me to tell what I missed. With this method, I rarely bother about parentheses except in situations such as (x+1)/(x+2), and I can't conceive how you could enter this without using something at least similar to parentheses.
The TI-82 definately has limitations, so I am not suprised that you noticed a substantial improvement when moving up to a real calculator. The TI-82 is basically a hand-held calculator with the ability to make a few simple decisions. Suprisingly it is Turing-complete, but it offers not that much more over that. The TI-89 (and the HP) extend the calculator's functions to much more closely relate to the human brain. When you get to be that complex (brain == the absolute most complex computing device ever constructed), personal preferences have a big impact on how easy it is for the brain to parse and do good things with what it is given, I sometimes have a fine line between what I "get" and what flies over my head. (to continue the metaphor, I guess I just "duck" more or less each time) Thus, if what the calculator throws at me is a bit too complex, often I will just gawk at it and do it on my own. Back to my initial question: how does the HP48 display what it figures out?
Again, the critical question is of the interface between your brain and the calculator. Your brain has built a set of routines for converting its internal representation of a mathematical form into RPN notation (redundant), and conversely finding a brain-internal representation for what the calc throws at you. My set of interface routines work with infix-like stuff, and it works nicely for me because on the input side, I only have to maintain one set of routines to convert both what my calc shows me and what the book shows me (and what is written on the board) into "internal representation". Also, what I throw at the calculator is often similar to wat I throw at other humans. It just works that way for me.
Enough,
Ken
Cross-platform: Check out GtkTiLink. And just think: the 89, 92, and 92+ are all based on the Motorola 68000 processor, and we have a version of Linux for that processor, don't we?
Durability: My TI-83 still works just fine after a year and a half; the only problem I have had is one of the anti-slip pads coming loose, but I used that thing a heck of a lot. If I hadn't have gotten a Ti-89 for Christmas, I would still be using that 83 for years to come.
Availability of programs: Here's the link
For ease of use, see my top-level post.