I want a bigger Newton. I had two different MessagePad versions, and I loved the interface. It was *designed* for a pen and that showed. But it was too big to carry with me everywhere (which is why I have a Palm to keep my schedule and contacts), and yet too small to be able to see very much info at once.
I'd spend a fair amount for a steno-pad-size or 8.5x11" tablet running NewtonOS. Not that that's likely to ever happen, but I can dream...
I would like to be able to laugh at this, but two weeks ago I mentioned Firefox to a coworker in passing and he said "What?" And he's a technologist, a software designer, architect, and developer.
...people in society should work for the profits of the small percentage of people who "own" companies.
At least in the United States, we're increasingly working for each other. That is, more people in the US own stock either directly or indirectly (e.g., through money market funds or retirement accounts) than ever before.
I'm not saying that IBM is right or wrong here, I just wanted to refute the misconception that there's a tiny group of owners and a huge class of non-owners. The group of part-owners is really big these days.
I've used NB to build some Java applications that were mostly GUI. Prior to 4.0, the project support in NB was pretty broken, but now I think it's great. I really like being able to customize how the IDE does builds, javadoc creation, etc., by customizing the ant files.
I also really like the GUI builder in NB. I haven't tried the GUI builder in Eclipse yet, but from what I see of it on the net, it doesn't seem as polished. Also, I happen to like Swing, and have no need or interest in learning SWT.
My current project doesn't have a GUI, and my teammates have convinced me to try Eclipse.
The big win for me with Eclipse is the "quick fix" (ctrl-1 by default). It brings up a pop-up menu of common options for fixing the problem where the cursor is. This can include adding an import (it nearly always figures out the right one), adding a parameter to a method call, changing the invoked method to include the parameter you've used (with the correct type), creating a new local variable or class member variable, and wrapping the current line in a try/catch block (with all requisite catches). Once you get used to it, you can just type code and later use quick fix to define variables for you, so you don't have to manually type declarations as much.
As someone else pointed out, the refactoring is also a lot better. I'm really happy to see NB getting better refactoring, both built-in and through commercial add-ons, but the built-in refactoring is much better in Eclipse. I agonize a lot less about naming of classes and methods now, because it's really easy to change it later. More sophisticated refactoring, like pushing methods up or down in the class hierarchy, is also easy.
One thing I still use NB for is javadocs. The NB javadoc problems browser is a really nice way to find problems with javadocs and easiliy fix them. Eclipse has some support for writing javadocs, but no centralized "fixer" like that.
Another area where I prefer Eclipse is version control. The "Team synchronization" perspective is nice because it lets you easily view conflicts, changes you've made, and changes in the repository separately or all together. Also, the visual diff is nice.
I will probably continue to use both of them at times (and emacs now and then, too), and wish them both great success.
I don't think you should be so quick to dismiss people who are concerned with licenses. I guess you missed the Bitkeeper saga.
I don't want to (re)start a debate about whether Linus made the right choice to use BK at all. My point is that as that whole episode showed, it's not at all obvious that you should pick the software with the most features regardless of the licensing terms.
For some people, the terms of the license are themselves an important "feature". For others, not.
I used a Newton for years (first the original MessagePad and then the MP100). For me, the fatal flaw was that it was both too big and too small:
* Too big to fit in my pocket so I could carry it with me everywhere * Too small to be able to see very much data at once
On the plus side, the interface was amazing. It was actually designed to be used with a pen, not just a modified desktop UI.
So now I use a Palm, because it lets me have my calendar and contact info with me all the time (as well as other stuff, of course, but the main thing I use it for is calendar and contacts). And its interface is ok.
But I still miss my Newton. I'd love to have a Newtonesque tablet. Even one with a display the size of a steno pad would be excellent.
This reminds me of a quote by Steve Jobs from Triumph of the Nerds: "The only problem with Microsoft is they just have no taste, they have absolutely no taste...I don't mean that in a small way I mean that in a big way."
There are different styles of programming. To name a few styles (with a representative language): * Imperative: C * Object-oriented: Smalltalk * Functional: LISP * Declarative: PROLOG
When you're programing well in a certain style you think differently than when you programing well in another style.
The pragmatic reason to learn languages in other styles is it'll make you a better programmer, in all languages you know.
LISP is a good one to learn because it's the canonical functional language.
I appreciate seeing Kevin Smith's comments. I think he's a great writer and director.
However, his review didn't encourage me much about the quality of RotS because it didn't address the big flaw the other two had: the writing. Sure, he said RotS was dark^n for a really big n, but dark doesn't automatically = good.
To me the big dramatic element overarching eps. 1-3 is this little innocent kid becoming *Darth Vader*, one of the biggest, baddest villians ever. And based on how it was handled in eps 1 and 2, I have little confidence that it will be handled well in ep 3.
And the fact that ep 3 has really cool lightsaber duels, special effects, etc., is definitely a plus, but basically irrelevant to whether I'll see the movie. Nowadays, there's good special effects lots of places.
I've seen people comment before about the amazing characterization, acting, and drama in Firefly. I saw all the eps that aired and I just don't see it.
I want to like Firefly. I loved Buffy and liked Angel very much. But in terms of characterization, I think Firefly is nowhere near Farscape (and even second place to Stargate SG-1, but maybe that's only because of the difference in length of series). And as far as drama, Firefly doesn't come close to Battlestar Galactica (the new one, of course:) ) or Babylon 5.
It's weird to me because people who to like so many things I like say Firefly is/was great. But I just don't see it.
I agree that Python is a great language. It's my favorite for small things. (Well, for really small things I use the shell or awk.)
If you're using Java already, I strongly recommend checking out jython, an implementation of Python in Java. I love it for small GUIs because it gives me access to Swing with much simpler syntax than Java.
Ok, I see what you mean about the different models of object. So far that hasn't been a problem for me.
As to the benefits of Jython vs. CPython: Sometimes I like to use some of the Java libraries in a Python program, such as Swing. (Although most of the time, if I'm writing the whole thing in Python, its large library base is sufficient.) Often I use Jython because I have a program that's partly in Java and I'm using Jython to debug it, or write a test wrapper, or do something else using my Java code from Jython.
I'm not familiar with Groovy, but I find that Python and Java together in Jython is a good combo. I'm not sure what you mean by saying that Python is disconnected from the concepts of the JVM. At least at the language level, what I like about Jython is that Python and Java *are* different languages.
From a theoretical security perspective brute force does not count as a "break", and so DES is essentially unbroken although insecure due to the small keyspace.
For all us non-theoreticians out here, it seems to me that there's a big difference between two systems if one is "insecure due to the small key space" and the other is not, even if the cryptographic algorithm behind each is equally good (or just plain equal, as in this case). If I'm depending on a system that is in fact insecure, whether it's a practical or theoretical insecurity is less important than the fact that it is insecure.
And until or unless DES is broken, it is highly unlikely that 3DES will be broken. And 3DES has a sufficiently large key size that it will be secure for many years to come, barring new weaknesses in DES.
Agreed.
I don't know what banks use, but standard DES uses 40-bit keys, which can be broken by brute force in a matter of hours at most. Maybe you're thinking of triple-DES, which uses longer keys (120-bit, I think) and is quite strong by today's (2005) standards.
Cannons are really good in that respect. We recently bought a color inkjet photoprinter, and probably would have bought a Cannon except that it didn't have slots for the storage cards from digital cameras. The Epson did, and was comparable to the Cannon, so we got it. Thus far, we've been really happy with the results.
A quick look at price/page, and it was easy to decide that HP was right out.
A lot of RPGs are real-time, which I think would probably make them pretty tough, but not all are.
My all-time second favorite RPG, Fallout, by default goes into turns when you enter combat. Other times it's real-time, but I don't think that should be a problem. It's also heavily mouse-based (and maybe can be played using only the mouse--it's been a while so I don't remember). Its successor, Fallout 2, is also a very worthy game and uses the same interface.
My #1 all-time favorite RPG is Planescape: Torment. The NPC interaction and storyline are better than anything else I've seen before or since. I think it's always real-time, but pausable at will with the space bar.
Turn-based strategy is one of my favorite game genres. I've logged a lot of hours on Civs 1, 2, and 3 and on SMAC (Sid Myer's Alpha Centauri). I can't really decide whether Civ3 or SMAC is better; both have definite strengths and both are really, really fun. I still enjoy firing them both up every now and then.
As another poster said in this thread, the Heroes of Might and Magic series (HOMM) is also quite fine.
A lesser known series is Age of Wonders. It's sort of a cross between HOMM and Warlords. There's 2 in the series so far, and maybe an expansion pack. I really enjoyed the first one, and the part of the second that I had time to play. (Warlords is ok, but not as much to my taste as the others I've mentioned.)
I agree that undo trees would be cool to have in editors, but they have all the problems and issues of version control systems that allow branches (i.e., pretty much all VCSs).
For example, if you want to take some edits from one branch and some from another, you have to worry about conflicting changes to the same object, modifications to an object in one branch that was deleted in another branch, etc. It's certainly possible to handle these problems, but most computer users aren't used to these concepts, and presenting them in a simple, easy-to-understand way is hard.
There were some research projects in the 90s on visualizations and interactions for managing undo trees. I don't think I've ever seen any in a released piece of software (proprietary or OSS). I don't have any links handy, but I know some of the techniques were presented at the ACM UIST symposium.
...since C, Java, Perl and Python are but different languages where you are trying to speak to a machine, not a human.
Early in my formal CS education, a professor told us "Think of programming not as telling the computer what to do, but telling other people what you want the computer to do." Unless you are and will always be the sole developer of a piece of software, it's not just about writing code that can get the computer to do the desired thing, it's also about making it clear to other humans what's going on.
I know this isn't the main point of the parent (or possibly even one of the points), but I've read enough bad code that I wanted to argue for legible code.
That's about as far as I got. I guess he didn't really express his ideas in the same way that I wanted to think about them.
I don't understand your point. Why did you stop reading? Because you didn't understand what the authors were trying to say? Or because you disagreed with the authors? Or some other reason?
You're right that there is a semantic gap. But the authors suggest that we might be able to bridge it by moving the language closer to what humans naturally use than it currently is. I don't think they expect to have an English (or pick your favorite NL) compiler any time soon (if ever).
Slashdot Mirror
I agree. I think there are two benefits apart from the direct result of (potentially) cool ideas to turn into lucrative products.
One you state explicitly: happier employees.
The other you imply, but I want to make explicit: you get more, smarter job candidates to choose from.
I want a bigger Newton. I had two different MessagePad versions, and I loved the interface. It was *designed* for a pen and that showed. But it was too big to carry with me everywhere (which is why I have a Palm to keep my schedule and contacts), and yet too small to be able to see very much info at once.
I'd spend a fair amount for a steno-pad-size or 8.5x11" tablet running NewtonOS. Not that that's likely to ever happen, but I can dream...
If you're in Pittsburgh, you can use http://wheeldeliver.net/. I've used it many times and it works quite well.
I would like to be able to laugh at this, but two weeks ago I mentioned Firefox to a coworker in passing and he said "What?" And he's a technologist, a software designer, architect, and developer.
If you like that style, you may also like Dave Patterson's "How to have a Bad Career in Research/Academia".
At least in the United States, we're increasingly working for each other. That is, more people in the US own stock either directly or indirectly (e.g., through money market funds or retirement accounts) than ever before.
I'm not saying that IBM is right or wrong here, I just wanted to refute the misconception that there's a tiny group of owners and a huge class of non-owners. The group of part-owners is really big these days.
As someone who's used both, I'll chime in.
I've used NB to build some Java applications that were mostly GUI. Prior to 4.0, the project support in NB was pretty broken, but now I think it's great. I really like being able to customize how the IDE does builds, javadoc creation, etc., by customizing the ant files.
I also really like the GUI builder in NB. I haven't tried the GUI builder in Eclipse yet, but from what I see of it on the net, it doesn't seem as polished. Also, I happen to like Swing, and have no need or interest in learning SWT.
My current project doesn't have a GUI, and my teammates have convinced me to try Eclipse.
The big win for me with Eclipse is the "quick fix" (ctrl-1 by default). It brings up a pop-up menu of common options for fixing the problem where the cursor is. This can include adding an import (it nearly always figures out the right one), adding a parameter to a method call, changing the invoked method to include the parameter you've used (with the correct type), creating a new local variable or class member variable, and wrapping the current line in a try/catch block (with all requisite catches). Once you get used to it, you can just type code and later use quick fix to define variables for you, so you don't have to manually type declarations as much.
As someone else pointed out, the refactoring is also a lot better. I'm really happy to see NB getting better refactoring, both built-in and through commercial add-ons, but the built-in refactoring is much better in Eclipse. I agonize a lot less about naming of classes and methods now, because it's really easy to change it later. More sophisticated refactoring, like pushing methods up or down in the class hierarchy, is also easy.
One thing I still use NB for is javadocs. The NB javadoc problems browser is a really nice way to find problems with javadocs and easiliy fix them. Eclipse has some support for writing javadocs, but no centralized "fixer" like that.
Another area where I prefer Eclipse is version control. The "Team synchronization" perspective is nice because it lets you easily view conflicts, changes you've made, and changes in the repository separately or all together. Also, the visual diff is nice.
I will probably continue to use both of them at times (and emacs now and then, too), and wish them both great success.
I don't think you should be so quick to dismiss people who are concerned with licenses. I guess you missed the Bitkeeper saga.
I don't want to (re)start a debate about whether Linus made the right choice to use BK at all. My point is that as that whole episode showed, it's not at all obvious that you should pick the software with the most features regardless of the licensing terms.
For some people, the terms of the license are themselves an important "feature". For others, not.
I used a Newton for years (first the original MessagePad and then the MP100). For me, the fatal flaw was that it was both too big and too small:
* Too big to fit in my pocket so I could carry it with me everywhere
* Too small to be able to see very much data at once
On the plus side, the interface was amazing. It was actually designed to be used with a pen, not just a modified desktop UI.
So now I use a Palm, because it lets me have my calendar and contact info with me all the time (as well as other stuff, of course, but the main thing I use it for is calendar and contacts). And its interface is ok.
But I still miss my Newton. I'd love to have a Newtonesque tablet. Even one with a display the size of a steno pad would be excellent.
This reminds me of a quote by Steve Jobs from Triumph of the Nerds: "The only problem with Microsoft is they just have no taste, they have absolutely no taste...I don't mean that in a small way I mean that in a big way."
I'll try to restate his point more pragmatically.
There are different styles of programming. To name a few styles (with a representative language):
* Imperative: C
* Object-oriented: Smalltalk
* Functional: LISP
* Declarative: PROLOG
When you're programing well in a certain style you think differently than when you programing well in another style.
The pragmatic reason to learn languages in other styles is it'll make you a better programmer, in all languages you know.
LISP is a good one to learn because it's the canonical functional language.
I appreciate seeing Kevin Smith's comments. I think he's a great writer and director.
However, his review didn't encourage me much about the quality of RotS because it didn't address the big flaw the other two had: the writing. Sure, he said RotS was dark^n for a really big n, but dark doesn't automatically = good.
To me the big dramatic element overarching eps. 1-3 is this little innocent kid becoming *Darth Vader*, one of the biggest, baddest villians ever. And based on how it was handled in eps 1 and 2, I have little confidence that it will be handled well in ep 3.
And the fact that ep 3 has really cool lightsaber duels, special effects, etc., is definitely a plus, but basically irrelevant to whether I'll see the movie. Nowadays, there's good special effects lots of places.
I've seen people comment before about the amazing characterization, acting, and drama in Firefly. I saw all the eps that aired and I just don't see it.
:) ) or Babylon 5.
I want to like Firefly. I loved Buffy and liked Angel very much. But in terms of characterization, I think Firefly is nowhere near Farscape (and even second place to Stargate SG-1, but maybe that's only because of the difference in length of series). And as far as drama, Firefly doesn't come close to Battlestar Galactica (the new one, of course
It's weird to me because people who to like so many things I like say Firefly is/was great. But I just don't see it.
I guess there's no accounting for taste.
If you're using Java already, I strongly recommend checking out jython, an implementation of Python in Java. I love it for small GUIs because it gives me access to Swing with much simpler syntax than Java.
Ok, I see what you mean about the different models of object. So far that hasn't been a problem for me.
As to the benefits of Jython vs. CPython: Sometimes I like to use some of the Java libraries in a Python program, such as Swing. (Although most of the time, if I'm writing the whole thing in Python, its large library base is sufficient.) Often I use Jython because I have a program that's partly in Java and I'm using Jython to debug it, or write a test wrapper, or do something else using my Java code from Jython.
I'm not familiar with Groovy, but I find that Python and Java together in Jython is a good combo. I'm not sure what you mean by saying that Python is disconnected from the concepts of the JVM. At least at the language level, what I like about Jython is that Python and Java *are* different languages.
Two great tastes...
YMMV. Void where prohibited.
From a theoretical security perspective brute force does not count as a "break", and so DES is essentially unbroken although insecure due to the small keyspace.
For all us non-theoreticians out here, it seems to me that there's a big difference between two systems if one is "insecure due to the small key space" and the other is not, even if the cryptographic algorithm behind each is equally good (or just plain equal, as in this case). If I'm depending on a system that is in fact insecure, whether it's a practical or theoretical insecurity is less important than the fact that it is insecure.
And until or unless DES is broken, it is highly unlikely that 3DES will be broken. And 3DES has a sufficiently large key size that it will be secure for many years to come, barring new weaknesses in DES. Agreed.
I don't know what banks use, but standard DES uses 40-bit keys, which can be broken by brute force in a matter of hours at most. Maybe you're thinking of triple-DES, which uses longer keys (120-bit, I think) and is quite strong by today's (2005) standards.
Cannons are really good in that respect. We recently bought a color inkjet photoprinter, and probably would have bought a Cannon except that it didn't have slots for the storage cards from digital cameras. The Epson did, and was comparable to the Cannon, so we got it. Thus far, we've been really happy with the results.
A quick look at price/page, and it was easy to decide that HP was right out.
A lot of RPGs are real-time, which I think would probably make them pretty tough, but not all are.
My all-time second favorite RPG, Fallout, by default goes into turns when you enter combat. Other times it's real-time, but I don't think that should be a problem. It's also heavily mouse-based (and maybe can be played using only the mouse--it's been a while so I don't remember). Its successor, Fallout 2, is also a very worthy game and uses the same interface.
My #1 all-time favorite RPG is Planescape: Torment. The NPC interaction and storyline are better than anything else I've seen before or since. I think it's always real-time, but pausable at will with the space bar.
Turn-based strategy is one of my favorite game genres. I've logged a lot of hours on Civs 1, 2, and 3 and on SMAC (Sid Myer's Alpha Centauri). I can't really decide whether Civ3 or SMAC is better; both have definite strengths and both are really, really fun. I still enjoy firing them both up every now and then.
As another poster said in this thread, the Heroes of Might and Magic series (HOMM) is also quite fine.
A lesser known series is Age of Wonders. It's sort of a cross between HOMM and Warlords. There's 2 in the series so far, and maybe an expansion pack. I really enjoyed the first one, and the part of the second that I had time to play. (Warlords is ok, but not as much to my taste as the others I've mentioned.)
I also really liked the more flexible government/society/economic models than Civ had. It would be great to see something like that in Civ 4.
I agree that undo trees would be cool to have in editors, but they have all the problems and issues of version control systems that allow branches (i.e., pretty much all VCSs).
For example, if you want to take some edits from one branch and some from another, you have to worry about conflicting changes to the same object, modifications to an object in one branch that was deleted in another branch, etc. It's certainly possible to handle these problems, but most computer users aren't used to these concepts, and presenting them in a simple, easy-to-understand way is hard.
There were some research projects in the 90s on visualizations and interactions for managing undo trees. I don't think I've ever seen any in a released piece of software (proprietary or OSS). I don't have any links handy, but I know some of the techniques were presented at the ACM UIST symposium.
Early in my formal CS education, a professor told us "Think of programming not as telling the computer what to do, but telling other people what you want the computer to do." Unless you are and will always be the sole developer of a piece of software, it's not just about writing code that can get the computer to do the desired thing, it's also about making it clear to other humans what's going on.
I know this isn't the main point of the parent (or possibly even one of the points), but I've read enough bad code that I wanted to argue for legible code.
I don't understand your point. Why did you stop reading? Because you didn't understand what the authors were trying to say? Or because you disagreed with the authors? Or some other reason?
You're right that there is a semantic gap. But the authors suggest that we might be able to bridge it by moving the language closer to what humans naturally use than it currently is. I don't think they expect to have an English (or pick your favorite NL) compiler any time soon (if ever).