Quite frankly, it seems like a bit of a knee-jerk reaction to me. Do you think it will hurt SCO if GCC removes support? Not in the slightest. Will it hurt the users? Probably not in the short run, since they will be able to continue using older versions. The message should be "let's show the users that we're on their side", and not "fsck SCO" -- granted that the users aren't immediately being hurt, but the perception from the user's perspective is the same.
Sorry, but filters are not the final answer. Even when the filters can "learn", the user still has to expend a certain amount of effort to "teach" the software. And quite frankly, spammers (or the people who write automated spamming software) just need to study the filters and learn to get around them. And worse, you can never be sure that the filter is not deleting email that you actually want, unless you set it to never delete suspect mail, allowing you to examine and delete it manually. But at this point, you've gained absolutely nothing -- simply setting your email client to put all email that's from addresses not in your address book, or that doesn't contain your exact address in the "To:" line will achieve exactly the same effect.
The only thing that can truly save email is to switch to a service that requires authentication of senders.
It's easier and more cost-effective to sue in the US, where they're almost certain to win, and recoup court costs as well. At the moment I happen to be living in the Czech Republic, and there's a huge problem with pirated CDs at most of the vietnamese open-air markets, esp. near the German and Austrian borders (this is not a racist remark, it is a comment on the state of affairs...) The Czech police are almost helpless to stop it -- most of the time, as soon as the police show up to raid the markets, the owners simply walk away from their stands, and the police confiscate what's on display, but arrest noone. Worse, court cases in the CR are notoriously prone to dragging out for years, so it's no wonder that RIAA wants to go after US-based downloaders.
"autonomyously"? Your groovitude is certainly showing...
I'm assuming that you meant "autonomously", but unfortunately even that doesn't make sense here. Autonomously implies that it works and functions as an agent independently -- mail arrives and Outlook automatically opens itself and chooses to play the file attached to the email, and just because it chose to do so. Now, if you'd said "automatically", that might have made more sense.
Agreed that Visual Age blows (makes a PIII 700 run like a P166), and Eclipse is by and large quite usable. That being said, it's missing a few things that after a while really start to bug you.
#1 - it allows you to set a code style (tab vs. spaces, tab size, brackets, etc...), but when you're coding, it doesn't automatically apply the style as you go. You must explicitly apply the selected format.
#2 - Its debugger is not as intuitive as it could be. When debugging a method, it automatically shows the local variables, but for non-static methods, it should be able to automatically display the current object's properties. Instead, you must set a watch to be able to see them.
#3 - When removing a file from a project, it's not completely obvious whether the file is being deleted from the project, or from the FILE SYSTEM. It's not well documented, and it certainly doesn't warn you appropriately.
.txt is simply not sufficient to replace a book. A book contains graphics, charts, indexes, etc. and there's no reason these should not be exploited in any eBook format. Also, there's no reason not to extend the paradigm to allow for readers to make annotations in the book, just like you can in real paper-based books.
But then again, why not? Former prosecutors actually tend to make the best defense lawyers... and if anybody is going to know anything about how privacy abuses work, it's someone who worked for a noted abuser.
I know more Java programmers than C programmers.... and all C programmers I know program Java now.
All industry software projects I'm involved in are in... ah forgett it you wont belive it anyway...
Absolutely... I love C/C++, but haven't done anything work-related in it for at least two years now.
Not really clear on the speed, but transferring 20 GB sounds like it might take a while...
Why does this really matter? So it might take a while to transfer the music, but it'll take significantly longer to listen to it. Your average user will probably transer music as necessary -- no need to shoot the entire collection over in one shot.
... combine the failed trip by 'N-Sync "singer" Lance Bass to the ISS with this trip... they could leave him on the face of the moon holding up a *BIG* sign, and anybody with a decent telescope could verify the veracity of the Apollo missions themselves.
At least he'd finally be doing something useful and (only slightly) more entertaining than usual.
It just handled MP3 files (worth mentioning twice)
The courts didn't kill their technology... they did, by not trying to develop it further themselves. I, for one, will not shed a single tear for the demise of Napster.
After reading the article, I decided to send an email to the Dallas Morning News' general manager. It really infuriates me when companies insist on using lawyers to solve a purely technological problem:
Dear Mr. Barker,
Up until today, I have always been proud to be a native Texan from the Dallas area. After reading an article recently in Wired (http://www.wired.com/news/politics/0,1283,52213,0 0.html) about your parent company's stance on deep linking, I must say that I as ashamed to be from Dallas. Why does the newspaper insist on using lawyers to solve a purely technological problem?
It is a quite simple matter to set a web server to only serve pages that have a valid "Http-Referrer" header -- every request sent by a browser sends this header, which contains that address of the page where the link was located. If it wasn't from your site, simply refuse the request. Technologically easy. No lawyers.
But if your intention is to simply harrass someone, which it must be since you opted for the more expensive option of litigation rather than the cheaper and more effective technological route.
Funny how MidbarTech keeps showing up in these types of articles. Their premier copy protection scheme is actually pretty easy to circumvent. Since Cactus uses a "bogus" multi-session TOC to cause PC's to fail to read the CD, all you have to do is hide the second session:
Cut up a laser printer label into small pieces. Place four of these pieces along the outer rim of the CD in question. Multi-session TOC's are stored on the outside edge of CD's. This will cause the CD-ROM unit to treat the CD as an unclosed multi-session CD.
After the CD's TOC has been read by the drive, wait for the unit to spin down. Then use an ordinary paperclip to "emergency" open the drive and remove the CD. Do not hit the eject button, as this will cause the TOC to be reread!
Carefully remove the laser printer labels from the CD.
Replace the CD and close the drive. Rip audio to your heart's content.
But don't do this in the USA, unless you want to see laser printer label stock and paperclips declared illegal under the DMCA.
Text editor + Jakarta Ant
on
Java IDEs?
·
· Score: 1
Actually, the best way to improve developer productivity is to throw away IDE's completely -- use a good text editor such as SciTE (www.scintilla.org) and for a build tool I'd recommend Jakarta Ant. IDE's tend to make programmers lazy, and thus don't actually think things through before they start coding.
I suspect that it does conform. Audio CD players only read the first session of a CD, and don't allow for multi-session disks. If the first session is valid, then the CD is valid. I haven't seen a 'mediacloq' disk yet, but I've got a Cactus-protected disk at home. It thwarts playing in PC's by using a bogus multi-session TOC... by setting the lead-out in the final track at the beginning of the first track, most PC hardware will refuse to play, seek, or read sectors after the start of the lead-out. Which leaves us with an effectively unreadable CD in a PC which plays perfectly in a normal audio player.
If it allows people to download the songs, then the TOC must be readable. ISRC is only readable from the subchannel information, which implies that the sectors themselves are readable (and rippable). So, to identify the CD for download, the TOC must be readable.
Thanks for the link... I'm the author of an open-source ripper, and so I'm more than interested in this. Oh, since I just happen to be in Prague at the moment, I think I'll stop at BontonLand over lunch and have a look.
How did the prosecution prove that Hensen sent the messages in question? I didn't notice any messages signed by Hensen in that thread. If he was dumb enough to testify in court that he wrote 'em, even if he considered the entire case against him to be a sham... well, hope he enjoys his new life in Canada.
There's a better reason to drop the original cddb protocol on freedb... it was designed for personal use and just kind of grew into a monster being used by everyone. After it hit critical mass, it was too difficult to force the thousands of applications that used it to change.
The major problems:
Genre is treated as part of the primary key for each lookup. Genre should be an attribute of a CD, and not a key item. (Not to mention the limited list of accepted genres).
Only one CD allowed for each discid/genre. Since the discid is not guaranteed to be unique, this is obviously a bogus restriction. This is fine for small collections, but as the number of CDs indexed grows, the likelyhood of a collision increases.
Screwed-up submit/update/delete semantics. The same process is used to submit and update CDs to the index -- except for submit, you have to increment the entry revision. The cleaner solution is to use seperate mechanisms for submit and update -- there's a reason that UPDATE and INSERT in SQL are seperate.
The database itself is file-based -- *very* non-portable (ever try to unpack 150000+ files into a VFAT filesystem?). Gracenote itself apparently switched to a relational database a long time ago, because they have been "poisoning" the entries sent back to clients with invalid track offsets.
I could go on, but there's not to much point in beating a dead horse.
I'd urge all of you to immediately switch all of your applications to use freedb instead. And for the developer types among you, volunteer to help.
-Andy Key
When Reuters was first starting out, there were certain areas where the telegraph hadn't been run through yet. So until the lines were completed, they had to send messages (packets) via carrier pigeons. It's by no means a new concept.
Slashdot Mirror
Quite frankly, it seems like a bit of a knee-jerk reaction to me. Do you think it will hurt SCO if GCC removes support? Not in the slightest. Will it hurt the users? Probably not in the short run, since they will be able to continue using older versions. The message should be "let's show the users that we're on their side", and not "fsck SCO" -- granted that the users aren't immediately being hurt, but the perception from the user's perspective is the same.
Sorry, but filters are not the final answer. Even when the filters can "learn", the user still has to expend a certain amount of effort to "teach" the software. And quite frankly, spammers (or the people who write automated spamming software) just need to study the filters and learn to get around them. And worse, you can never be sure that the filter is not deleting email that you actually want, unless you set it to never delete suspect mail, allowing you to examine and delete it manually. But at this point, you've gained absolutely nothing -- simply setting your email client to put all email that's from addresses not in your address book, or that doesn't contain your exact address in the "To:" line will achieve exactly the same effect.
The only thing that can truly save email is to switch to a service that requires authentication of senders.
Can someone please explain why we would want to boot a 5MB version of an operating system that came out over 8 years ago?
Why not? If it's only 5MB, then then next step is to port Bochs to PalmOS, and then I could at least finally play solitaire on my m515.
Or someone else, much more mentally ill than I, could port Bochs to Windows CE...
It's easier and more cost-effective to sue in the US, where they're almost certain to win, and recoup court costs as well. At the moment I happen to be living in the Czech Republic, and there's a huge problem with pirated CDs at most of the vietnamese open-air markets, esp. near the German and Austrian borders (this is not a racist remark, it is a comment on the state of affairs...) The Czech police are almost helpless to stop it -- most of the time, as soon as the police show up to raid the markets, the owners simply walk away from their stands, and the police confiscate what's on display, but arrest noone. Worse, court cases in the CR are notoriously prone to dragging out for years, so it's no wonder that RIAA wants to go after US-based downloaders.
"... now we are spared the exercise of raising our hand to signal the waiting staff for a refill."
But at least the wait staff will stay thin...
"autonomyously"? Your groovitude is certainly showing...
I'm assuming that you meant "autonomously", but unfortunately even that doesn't make sense here. Autonomously implies that it works and functions as an agent independently -- mail arrives and Outlook automatically opens itself and chooses to play the file attached to the email, and just because it chose to do so. Now, if you'd said "automatically", that might have made more sense.
I prefer IntelliJ IDEA for Java development.
Agreed that Visual Age blows (makes a PIII 700 run like a P166), and Eclipse is by and large quite usable. That being said, it's missing a few things that after a while really start to bug you.
#1 - it allows you to set a code style (tab vs. spaces, tab size, brackets, etc...), but when you're coding, it doesn't automatically apply the style as you go. You must explicitly apply the selected format.
#2 - Its debugger is not as intuitive as it could be. When debugging a method, it automatically shows the local variables, but for non-static methods, it should be able to automatically display the current object's properties. Instead, you must set a watch to be able to see them.
#3 - When removing a file from a project, it's not completely obvious whether the file is being deleted from the project, or from the FILE SYSTEM. It's not well documented, and it certainly doesn't warn you appropriately.
.txt is simply not sufficient to replace a book. A book contains graphics, charts, indexes, etc. and there's no reason these should not be exploited in any eBook format. Also, there's no reason not to extend the paradigm to allow for readers to make annotations in the book, just like you can in real paper-based books.
But then again, why not? Former prosecutors actually tend to make the best defense lawyers... and if anybody is going to know anything about how privacy abuses work, it's someone who worked for a noted abuser.
I know more Java programmers than C programmers .... and all C programmers I know program Java now. ... ah forgett it you wont belive it anyway ...
All industry software projects I'm involved in are in
Absolutely... I love C/C++, but haven't done anything work-related in it for at least two years now.
Not really clear on the speed, but transferring 20 GB sounds like it might take a while...
Why does this really matter? So it might take a while to transfer the music, but it'll take significantly longer to listen to it. Your average user will probably transer music as necessary -- no need to shoot the entire collection over in one shot.
... combine the failed trip by 'N-Sync "singer" Lance Bass to the ISS with this trip... they could leave him on the face of the moon holding up a *BIG* sign, and anybody with a decent telescope could verify the veracity of the Apollo missions themselves.
At least he'd finally be doing something useful and (only slightly) more entertaining than usual.
- It just handled MP3 files
- It requires a centralized server
- Most people are already used to KaZaa
- It just handled MP3 files (worth mentioning twice)
The courts didn't kill their technology... they did, by not trying to develop it further themselves. I, for one, will not shed a single tear for the demise of Napster.After reading the article, I decided to send an email to the Dallas Morning News' general manager. It really infuriates me when companies insist on using lawyers to solve a purely technological problem:
0 0.html) about your parent company's stance on deep linking, I must say that I as ashamed to be from Dallas. Why does the newspaper insist on using lawyers to solve a purely technological problem?
Dear Mr. Barker,
Up until today, I have always been proud to be a native Texan from the Dallas area. After reading an article recently in Wired (http://www.wired.com/news/politics/0,1283,52213,
It is a quite simple matter to set a web server to only serve pages that have a valid "Http-Referrer" header -- every request sent by a browser sends this header, which contains that address of the page where the link was located. If it wasn't from your site, simply refuse the request. Technologically easy. No lawyers.
But if your intention is to simply harrass someone, which it must be since you opted for the more expensive option of litigation rather than the cheaper and more effective technological route.
sincerely,
Jay A. Key
But don't do this in the USA, unless you want to see laser printer label stock and paperclips declared illegal under the DMCA.
Actually, the best way to improve developer productivity is to throw away IDE's completely -- use a good text editor such as SciTE (www.scintilla.org) and for a build tool I'd recommend Jakarta Ant. IDE's tend to make programmers lazy, and thus don't actually think things through before they start coding.
I suspect that it does conform. Audio CD players only read the first session of a CD, and don't allow for multi-session disks. If the first session is valid, then the CD is valid. I haven't seen a 'mediacloq' disk yet, but I've got a Cactus-protected disk at home. It thwarts playing in PC's by using a bogus multi-session TOC... by setting the lead-out in the final track at the beginning of the first track, most PC hardware will refuse to play, seek, or read sectors after the start of the lead-out. Which leaves us with an effectively unreadable CD in a PC which plays perfectly in a normal audio player.
If it allows people to download the songs, then the TOC must be readable. ISRC is only readable from the subchannel information, which implies that the sectors themselves are readable (and rippable). So, to identify the CD for download, the TOC must be readable.
I thought the SPARC bios uses interpreted Forth...
Thanks for the link... I'm the author of an open-source ripper, and so I'm more than interested in this. Oh, since I just happen to be in Prague at the moment, I think I'll stop at BontonLand over lunch and have a look.
It's really exciting to see several different groups independantly and together...
See, SmartTags(tm) aren't so bad after all.
How did the prosecution prove that Hensen sent the messages in question? I didn't notice any messages signed by Hensen in that thread. If he was dumb enough to testify in court that he wrote 'em, even if he considered the entire case against him to be a sham... well, hope he enjoys his new life in Canada.
The major problems:
- Genre is treated as part of the primary key for each lookup. Genre should be an attribute of a CD, and not a key item. (Not to mention the limited list of accepted genres).
- Only one CD allowed for each discid/genre. Since the discid is not guaranteed to be unique, this is obviously a bogus restriction. This is fine for small collections, but as the number of CDs indexed grows, the likelyhood of a collision increases.
- Screwed-up submit/update/delete semantics. The same process is used to submit and update CDs to the index -- except for submit, you have to increment the entry revision. The cleaner solution is to use seperate mechanisms for submit and update -- there's a reason that UPDATE and INSERT in SQL are seperate.
- The database itself is file-based -- *very* non-portable (ever try to unpack 150000+ files into a VFAT filesystem?). Gracenote itself apparently switched to a relational database a long time ago, because they have been "poisoning" the entries sent back to clients with invalid track offsets.
I could go on, but there's not to much point in beating a dead horse.I'd urge all of you to immediately switch all of your applications to use freedb instead. And for the developer types among you, volunteer to help. -Andy Key
When Reuters was first starting out, there were certain areas where the telegraph hadn't been run through yet. So until the lines were completed, they had to send messages (packets) via carrier pigeons. It's by no means a new concept.