The solution to spyware problems is either technological(although I have no idea how, using an non-Widnows OS isn't really "the answer") or social(teach people how to 1. Avoid spyware and 2. Avoid giving any kind of financial incentive to any company associated with spyware).
I think some kind of system, like a "web of trust", would be useful in distinguishing between good and bad software. You could use a hash of the installer EXE or source tarball as a unique identifier for each piece of software.
Any serious flaws to this idea, besides the fact that it requires widespread adoption before it's really useful?
It's actually a known phenomenon that "real nice and clean and well-documented codebase" can in fact be _evil_! Because everyone except really lousy coders are afraid to touch it. "It's so beautiful it's practically dead" one could say.
Interesting thought. On the other hand, if it's gotten that far, and it's been worked on by more than one or two people, it probably already has a documented procedure for modifying the code: naming conventions, unit test requirements, etc. Maybe it seems tedious or intimidating, but sloppy code slows development in the long run, and usually leads to more bugs.
Undergrads who get up off their bums, seek employment, and still continue college are generally...not taking a very demanding course load? Honestly, if you have 20+ hours per week to do intellectually-challenging work on top of class work, you're either a total genius or you're wasting your tuition by not taking many/challenging classes.
A lot of people say 2 nics would be good for a firewall/router. I don't see that, either. Why not just up an external and an internal IP on one NIC and set all the other computers to the internal IP address scheme, with a subnet of [not enough ips to get to the internet on their own] and set the gateway to the firewall?
Okay, and how do you propose to plug two Ethernet cables (one to your LAN, one to the outside world) into one NIC?
For OO to succeed it needs to have a marketing campaign similar to FireFox. It needs to be a product that people get recommended to them from non-geeks.
For OpenOffice.org to succeed, they need to improve the product to the point where it can actually compete with MS Office. It's good, and adequate for most people who just need to do simple word processing and spreadsheets, but it's also ugly, slow, and lacking in features (compare Excel's graphing abilities to OO Calc's). It may seem petty, but they really need to drop the Win95-esque look. It's ugly on Windows, and it's even uglier on KDE.
In the web app world in particular it is often very hard to write real unit tests without getting into a whole variety of special rigging.
True enough, but if you've written the code in good abstracted OO with the "special rigging" (ie, mock objects) in mind, you'll be much better off. I'd say there's very little complex code that doesn't require mock objects or the equivalent to test, so it's something well worth learning. As you say, once you've got all the groundwork set up, it's much easier to extend.
See if you can write some tests to ensure basic functionality. Then tear it apart and start refactoring. Writing properly testable code is no simple task.
Yes, Solaris does kinda suck for a desktop. It's more appropriate to compare it to the BSDs rather than Linux. It's primarily a secure, stable, mature server OS, and it does a fine job at that. Personally, I don't know of any reasons to use it instead of OpenBSD or FreeBSD.
While I strongly disagree with the journal Science's policy of rejecting articles which have been made public, it is a consistent and well known policy
I'm not sure that's the case here. To quote the summary: "the points he make had been 'widely dispersed on the internet.'".
If your article doesn't contain any original thought or research, it probably isn't going to be published, especially not in Nature.
Re:When is the Hack Apache contest?
on
Hack IIS6 Contest
·
· Score: 1
He doesn't even use the Java naming convention of putting an underscore in front of your instance variables. I spent a few seconds wondering where optionsVector came from, because I thought pretty much all Java coders followed the naming conventions.
And the rest of the comments are just as bad. If you're just going to restate the name of the method, please don't bother. What the fuck is "Return the list of values" supposed to mean?
I don't know about elsewhere, but the fake-tree cell phone towers in New York look awful. Yes, we do have real trees, and no, this doesn't look like a tree. It looks like a fucking cell phone tower with a few tiny fake branches at the top. Is it so hard to make it look a little more realistic?
By refusing to purchase CDs, I vote against the RIAA with my wallet.
Sorry, but this simply isn't an option for people who actually enjoy music from artists on RIAA labels. Boycotts can be effective, but it's not the greatest idea to make the artists suffer because of the actions of an organization that their record label belongs to.
The source code for a work means the preferred form of the work for making modifications to it. For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable.
This is to prevent people from releasing obfuscated or otherwise unusable code. If Apple really is releasing nothing but big tarballs most of the time, they're probably not in violation of the GPL, but it is contrary to the spirit of the GPL.
I really agree. Daggerfall, as I remember it, was very difficult.
Morrowind was just too easy after you gained a few levels.
I mean, where did they get the brilliant idea to give the player the second-best suit of light armor at the very beginning of the game (Tribunal, assassin)?
True enough, but OpenOffice really sucks for graphs. It can't, for example, draw two plots on the same graph. Maybe this is vastly improved in 2.0; I haven't had the time to play around with it.
Back on topic, I have to agree that doing any kind of serious development with a spreadsheet app is typically a very bad idea. Your "code" is going to be ugly and unmaintainable.
If Bob writes a program, releases it under the GPL, and incorporates contributed code into the project, that's another can of worms. I would think if he wanted to "go private" with the code base at that point he would need to get the permission of everyone who contributed any code, much like Mozilla did. If he couldn't get their permission he would have to rewrite those chunks of code.
This isn't necessary if all contributors assign copyright to one entity. That's the FSF's policy, for example.
Slashdot Mirror
Pay us. Otherwise, I'll keep working on the distro I like best, thanks.
I think some kind of system, like a "web of trust", would be useful in distinguishing between good and bad software. You could use a hash of the installer EXE or source tarball as a unique identifier for each piece of software.
Any serious flaws to this idea, besides the fact that it requires widespread adoption before it's really useful?
Interesting thought. On the other hand, if it's gotten that far, and it's been worked on by more than one or two people, it probably already has a documented procedure for modifying the code: naming conventions, unit test requirements, etc. Maybe it seems tedious or intimidating, but sloppy code slows development in the long run, and usually leads to more bugs.
Undergrads who get up off their bums, seek employment, and still continue college are generally...not taking a very demanding course load? Honestly, if you have 20+ hours per week to do intellectually-challenging work on top of class work, you're either a total genius or you're wasting your tuition by not taking many/challenging classes.
Okay, and how do you propose to plug two Ethernet cables (one to your LAN, one to the outside world) into one NIC?
Well, sort of. Most programmers use make to automate all that stuff.
Correction: not just vi, vim and LaTeX.
For OpenOffice.org to succeed, they need to improve the product to the point where it can actually compete with MS Office. It's good, and adequate for most people who just need to do simple word processing and spreadsheets, but it's also ugly, slow, and lacking in features (compare Excel's graphing abilities to OO Calc's). It may seem petty, but they really need to drop the Win95-esque look. It's ugly on Windows, and it's even uglier on KDE.
Every mutation of a virus that grants it immunity to another antibiotic.
True enough, but if you've written the code in good abstracted OO with the "special rigging" (ie, mock objects) in mind, you'll be much better off. I'd say there's very little complex code that doesn't require mock objects or the equivalent to test, so it's something well worth learning. As you say, once you've got all the groundwork set up, it's much easier to extend.
See if you can write some tests to ensure basic functionality. Then tear it apart and start refactoring. Writing properly testable code is no simple task.
Yes, Solaris does kinda suck for a desktop. It's more appropriate to compare it to the BSDs rather than Linux. It's primarily a secure, stable, mature server OS, and it does a fine job at that. Personally, I don't know of any reasons to use it instead of OpenBSD or FreeBSD.
I'm not sure that's the case here. To quote the summary: "the points he make had been 'widely dispersed on the internet.'".
If your article doesn't contain any original thought or research, it probably isn't going to be published, especially not in Nature.
Try just about 70%.
And the rest of the comments are just as bad. If you're just going to restate the name of the method, please don't bother. What the fuck is "Return the list of values" supposed to mean?
I don't know about elsewhere, but the fake-tree cell phone towers in New York look awful. Yes, we do have real trees, and no, this doesn't look like a tree. It looks like a fucking cell phone tower with a few tiny fake branches at the top. Is it so hard to make it look a little more realistic?
Sorry, but this simply isn't an option for people who actually enjoy music from artists on RIAA labels. Boycotts can be effective, but it's not the greatest idea to make the artists suffer because of the actions of an organization that their record label belongs to.
Of course, such a filter is nearly worthless. It's trivial for the RIAA to get a consumer ISP account and gather data with that.
Check out MusicBrainz. It's spiffy.
Well, kinda. IIS has always been one of Microsoft's worst products. Apache market-share continues to grow at the expense of everyone else.
Wrong. Nearly all packages get hosted on and downloaded from our distfiles mirrors. The address in SRC_URI is only used as a last resort.
The source code for a work means the preferred form of the work for making modifications to it. For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable.
This is to prevent people from releasing obfuscated or otherwise unusable code. If Apple really is releasing nothing but big tarballs most of the time, they're probably not in violation of the GPL, but it is contrary to the spirit of the GPL.
I really agree. Daggerfall, as I remember it, was very difficult. Morrowind was just too easy after you gained a few levels. I mean, where did they get the brilliant idea to give the player the second-best suit of light armor at the very beginning of the game (Tribunal, assassin)?
Back on topic, I have to agree that doing any kind of serious development with a spreadsheet app is typically a very bad idea. Your "code" is going to be ugly and unmaintainable.
This isn't necessary if all contributors assign copyright to one entity. That's the FSF's policy, for example.