Twillerror (536681) writes "Over the last few weeks a flood of SOPA related articles have been on slashdot. One common theme in comments is that piracy is not a problem at all. Using piracy as way to improve sales is modded up way over any suggestion that people are stealing regardless of lost sales or not. Is part of the problem here that techies are in denial that there is an issue that needs solving in the first place?" top
Twillerror (536681) writes "As a web developer I'm a little put off by the success of browser plugins like AdBlock that modify the content of pages and remove ads.
I like site like Slashdot and I'm perfectly fine looking at an ad or two since the content is free. I think I learned of a few products as well. One rationalization is to remove flash ads that slow your computer, but if the site is free the appropriate option in my opinion is to not visit the offending site.If the site is funded via ads by removing them you might as well be stealing it. What does Slashdot think?" top
Twillerror (536681) writes "RedmondMag reports on problems with the new new Volume licensing site that MS has recently launched. As someone who has used this system it has to be in the top 5 worst major sites I've ever used. I'm not as big of anit-MS guy as the normal Slashdot reader, but licensing in MS I think is clearly "just bad". Cost is debatable, but CALs are just confusing and hard to implement. At what point will MS just drop CALs and start charging reasonable fees for Exchange Server, Active Directory access, File server access and the like. I think people will pay for Windows Server 2008, but when you have to upgrade all your existing 2003 CALs it makes the adoption to hard. Why should I have to pay extra for a copy of windows to hit a file server...when will MS make enough off SMB? Will the site problems finally nudge MS to revamp their licensing approach?" Link to Original Source top
Twillerror writes "The other day I heard a story/myth about nuclear plants running VMWare to emulate Windows 95 systems. I guess the software would only run on Win95 and getting new hardware to run Win95 can be hard.
Got me thinking about Linux and older versions of it. Is anyone running much older versions of the kernel in production like situations or maybe just for fun?" top
Twillerror (536681) writes "Recently I got an orginally Windows developed web based application up and running on Linux by using JFS without case sentivity. Code referenced files in a case insentive fashion(not ideal I know). Which got me thinking about the merits of both approaches.
This question also seems to have a lot of emotion behind it from the *nix community which confuses me. I can't for the life of me understand why you would want a directory with both text.txt and Test.txt in it. I like the fact that most OSs keep the fact that it is Test.txt, and don't store test.txt, but that is where it stops. Many of the conversations I see out there bring programming languages into the mix which I think is seperate. Also, I've seen misguided posts that say Linux can do "ls -a" and "ls -A" which has nothing to do with this as well.
Can we have an honest discussion about exactly what would break on Linux and why are worlds would stop functioning? I get the consistency argument, but can't that be implemented at the programming languaging or script processing level(determined by that community) and not at the OS level? If you dont' want bash to execute "LS" and only execute "ls" is that for bash to decide and not ext3? I'm also concerned about interoperability." top
Twillerror (536681) writes "I've been a backend web developer for many years. Recently though I've been doing more HTML, CSS, DOM writing...more of the GUI end of things. I've come to the conclusion that HTML sucks, but its the path of least resitance and so gets the most attention. People are willing to work for hours just to get things to align correctly and work in the all browsers. I can't count how many times I've discovered some weird DOM parsing bug in both IE and/or Mozzila causing some JS toolkit to fail.
Working with Flex (Flash based applications) I was mostly pleased with how much easier it was to get a good looking application and how much and more reliable it was. It was actually compiled!. It was more like your traditional GUI API where you have containers and components. You have real input forms, trees, tables, etc. You want an input that only accepts numbers...a few characters. In general a better toolkit for laying out a desktop like application. The three biggest problems I see are that A) it is not open/standard, B) it tends to get abused for things like ads and other annoying things, and C) the embed or object tags are annoying and trying to get things to play nice inside of HTML is just bad.
I'm not pushing Flash, but I'm pushing for a more Flash like future. Suppose we had an open source Flash...there would be a mute button in the options so that no ad would ever be able to play sound without your approval. We might have a standard format and compiler..etc...etc. Web apps could be deployed in a mostly binary format without the need for a slow and complex parser.
What I'd like to propose is that W3C rethink the problem and make the browser a "JVM like" application. Expose an API which allows an engine to draw graphics, save files to disk, send an email, open a socket, etc. Then have plugins that could do whatever we need. Then HTML 4 could be a plugin and HTML 5 could be a plugin. We won't have some ultra complex browser that has to parse HTML 1-5 and all the various quirks mode and crap. Then Flex\Flash can run in there as well as some open source alternative. These plugins can then register with the main browsers through some sort of approval process. So Firefox would just auto download whatever engine they needed and we would have a trusted list of applications.
Twillerror (536681) writes "It seems like almost every day some site is hacked or another exploit is found in an application. The theme is usually the same. A buffer overflow, XSS attack or SQL Injection. The standard solutions then follow. Either check the array length, validate the input, or strip bad tags or attributes. All of which are reactive.
The NX bit(Wikipedia) is a step in the right direction. Should we not be focusing more on similar solutions that prevent or eliminate the problem.
For instance on the SQL front imagine having your SQL statements start with a security line. An example might be...
allow select : users; expected rowcount <= 1;
select username, password from users where login = '%form.login%'
The developer could specify what his/her statement is intended to do. The SQL engine could easily detect if it tries something else and refuse to run it, or better yet silently reject the invalid part of the statement and alert the application.
What else could we do to be more proactive about security in the programming languages themselves?" top
Twillerror writes "After reading the article on slashdot concering bands refusing iTunes I was reminded about what I consider to be digital music's biggest flaw. This to me is the lack of a standardized album file. Imagine an MP3 or Ogg that had a bit of XML stuck onto it's header of footer. This had the normal ID3 Tag information, but also had a list of tracks and start\offset times.
If say WinAmp and other players knew of this format, they could make the list of songs collapsable, and make them part of a randomizer. Imagine that you could say, play two songs off each random album I choose.
This to me is why iTunes has done better as a player, because of it's organizational skills. On the other hand, I think that ultimately if all the tracks are not in one binary file that they will get broken up. Also, backup would be far faster for people who only scanned in albums.
Yes sometimes I'd rather just get a track off an album, but most of the time I'd rather get the whole thing. Mostly because most of the bands I listen to might actually do care about their albums...mostly. This is a personal opinion, but I that almost everyone out there has listened to a great album in it's entirity and would like it stuck together."