What features did you want to get into Perl 6 that won't be there, and why were they left out? Do you think they'll be in Perl 7, or is Perl 6 the end of the line?
"Other kings said I was daft to build a castle on a swamp, but I built it all the same, just to show 'em. It sank into the swamp. So, I built a second one. That sank into the swamp. So, I built a third one. That burned down, fell over, then sank into the swamp, but the fourth one... stayed up!"
Perl 6 naming
on
Ask Larry Wall
·
· Score: 2, Interesting
Did you consider using a name other than Perl 6 for your new language? Many current Perl users recoil at the extreme changes made to their favorite language, even though they can still write Perl 5 for the most part. (And I really like the new regexp design.)
In retrospect do you think a new name or version nomenclature would have warmed the reception and/or reduced confusion?
Wow, I'm looking forward to seeing that movie when it makes its way into the public domain in 2173. Assuming they don't change the law retroactively--again.
Yes, maybe 10 years from now we'll all look back on this and laugh at how Valenti thought he had us on the rocks. But until the DMCA has been struck down it acts as a great big wet blanket to free speech and fair use. I wish we could just get a test case into court and get this over with, but Valenti et al would prefer to use threats since the court case may not go their way.
In the article, Kerr is quoted as saying "If the public believes that the DMCA is stopping Professor Felten and other researchers from conducting legitimate research, then that is a major victory for opponents of the law." Well, the public believes that because it is true. Felten wanted to present a paper and was told by his employer not to do it. No matter what that article says about lawsuits being unlikely to succeed, the mere threat of a DMCA lawsuit seems to be enough when you're dealing with deep pockets.
Right. Customers love sending a message to some support email address and then turning off their computer for the night, only to find a lame auto-email when they return and check mail the next evening after they get home from work. It's a day later and they're no closer to having their problem solved.
Wow. It's described down to a level of detail that would make you think they've already written the Outlook add-in for it. I wonder why we haven't seen it yet?
Why *won't* it work on an HTML mail that's only an image? First, as I understood it, the whole message including the header are examined. Second, if IMG, SRC, freeserve (in the domain) are among the most interesting words and high on the probability list, then it's spam. I would expect that to be the case since the only HTML email I seem to get is spam.
>> based on the assumption that it is written in English
There's no reason to think that Spanish, German, or even Chinese spam doesn't follow the same statistical word frequency rules.
>> following the simple steps outlined in the URL above
What if you are subscribed to mailing lists, or have mail bots that send you useful messages (like "your server is down")? The usual answer is "just configure those in advance" but that's a pain and not very robust. My hosting company was bought out and their automated server status messages just started to come from a new domain. If I had this kind of filter I would have missed them.
Sounds like you didn't know that developers can get every business and OS product that Microsoft makes for every international language in the MSDN Subscription on DVD for $2,500. Most US developers would only need the Professional subscription which is $1,200. That includes MS Office, Visual Studio and all the compilers, Project, SQL Server, SDKs, DDKs, every version of Windows since 95, and a year of updates. The MSDN versions of most products allow 10 licenses, which is plenty for most developers. The price of the Windows licenses alone far exceeds the cost of the subscription.
>> Later, when and if they got enough sales, they'd reconsider XP.
I don't know their application so I can't say for sure, but in most cases that's ass-backwards. You usually want to build your product for the biggest market first.
You miss the point about the goal of this. It isn't to harass the patent reviewers, it is to help them find the information they don't seem to be finding. To say "the reviewers are strapped for time as it is" is like saying "Linus is already busy, the last thing he needs is people to be sending him patched code to look at."
To prevent offtopic posts and flames it could be restricted, for example only to registered users. By registered I mean that the USPTO knows where they live, not that they set up some lame account with a fake email address. Abusers would be banned.
Isn't there some way that the Patent Office could open up this process so that the prior art could be waved in front of them before the patent is granted and expensive lawyers have to be called in to resolve the issue?
I'm thinking the USPTO could create a database of pending patents on their web site that have passed initial muster with the investigator and are likely to be approved. Interested parties could go and post links about prior art (or earlier filed but still pending patents) for the patent investigator to review.
If the MS contract just says they can't ship the system bare, why not install Linux?
Also, does this mean that Microsoft's contract clause will go unchallenged? Who the heck is Microsoft to say whether Dell ships their hardware with or without an OS on it?
Duh! The exploit is *intended* to be malicious code that causes a buffer overrun in an application and could be used to break into a system. That's the point!
This Win32 attack can't be removed without breaking literally thousands of programs.
Please provide your list of thousands of Win32 apps that run as a privileged service and interact with the desktop. My guess is that the number is more like 10 than 1000.
I agree that Microsoft should emphasize the security risk of services that interact with the desktop. A good place to do that would be on the page that documents the use of interactive services. There's a short discussion near the bottom, but it focuses on getting privileges, not restricting them.
As far as docs go, Linux isn't setting a shining example. Here's the understated comment about sprintf tacked on at the end of the Linux man page: Because sprintf and vsprintf assume an infinitely long string, callers must be careful not to overflow the actual space; this is often impossible to assure.
Something around 10MB/sec sounds about right for application-level I/O with a good ATA100 drive using DMA, at least from my experience with Windows. If you bypass all the application layers (file system with potential for fragmentation, disk cache with possibly non-optimal prefetch) and go directly to the device you can get better numbers, but few apps can do that.
Okay, let's compare these comments with the ones about the end-of-life plans for Windows 2000 and Windows 95. Executive summary: It's the right business and technical decision for Apple, but it's heavy-handed tactics by Microsoft. Most OS 9 apps run under OS X using classic mode, and this is why Apple is justified in killing OS 9. Most Windows 95 apps run under Windows XP, either out of the box or using compatibilty mode, but it's part of Microsoft's plan to make us use the license-enforced XP scourge. This is not contradictory logic because Apple is an underdog and Microsoft is a monopoly.
Consumer's Checkbook had a good survey of all the cell providers serving the region in their latest Washington DC area pub, and it looks like they did the same for San Francisco as well. I have Sprint and I can tell you they mentioned most of the dead spots Sprint has near my house.
To read the survey you'll either need to buy the dead-tree version at a local bookstore or subscribe to the web site.
Doesn't it seem odd to you that TV makers oppose a regulation that basically would require all consumers to buy new equipment over the next five years? If they don't buy a new TV, they will at least need to buy a converter box for their existing one. So why would the TV makers object?
The content providers, broadcasters, and equipment makers are still in a big tug of war about evil consumers recording Lion King in digital quality when it's broadcast on ABC/Disney. Equipment makers don't want to implement the copy protection that the content providers want because consumers won't want those restrictions.
I think they also suspect the govt may still back out of this mandate due to consumer revolt, and if that happens they'll be stuck with product nobody wants because it's more expensive than the old stuff.
For better or worse, the govt is the manager of the airwaves. When Congress gave away the valuable HDTV frequencies to existing broadcasters for free, it was supposedly with the understanding that they would give back their existing frequencies in 2006. That bandwidth will be auctioned off and the govt wants to make that money, so delay in switching to HDTV costs the govt. That's why they're trying to accelerate the process by laws and regulations now.
Slashdot Mirror
What features did you want to get into Perl 6 that won't be there, and why were they left out? Do you think they'll be in Perl 7, or is Perl 6 the end of the line?
"Other kings said I was daft to build a castle on a swamp, but I built it all the same, just to show 'em. It sank into the swamp. So, I built a second one. That sank into the swamp. So, I built a third one. That burned down, fell over, then sank into the swamp, but the fourth one... stayed up!"
Did you consider using a name other than Perl 6 for your new language? Many current Perl users recoil at the extreme changes made to their favorite language, even though they can still write Perl 5 for the most part. (And I really like the new regexp design.)
In retrospect do you think a new name or version nomenclature would have warmed the reception and/or reduced confusion?
There are extremists in every group, so don't go there. Earth First, Greenpeace, John Ashcroft, slavery reparations, ...
Wow, I'm looking forward to seeing that movie when it makes its way into the public domain in 2173. Assuming they don't change the law retroactively--again.
Yes, maybe 10 years from now we'll all look back on this and laugh at how Valenti thought he had us on the rocks. But until the DMCA has been struck down it acts as a great big wet blanket to free speech and fair use. I wish we could just get a test case into court and get this over with, but Valenti et al would prefer to use threats since the court case may not go their way.
In the article, Kerr is quoted as saying "If the public believes that the DMCA is stopping Professor Felten and other researchers from conducting legitimate research, then that is a major victory for opponents of the law." Well, the public believes that because it is true. Felten wanted to present a paper and was told by his employer not to do it. No matter what that article says about lawsuits being unlikely to succeed, the mere threat of a DMCA lawsuit seems to be enough when you're dealing with deep pockets.
Right. Customers love sending a message to some support email address and then turning off their computer for the night, only to find a lame auto-email when they return and check mail the next evening after they get home from work. It's a day later and they're no closer to having their problem solved.
Wow. It's described down to a level of detail that would make you think they've already written the Outlook add-in for it. I wonder why we haven't seen it yet?
Why *won't* it work on an HTML mail that's only an image? First, as I understood it, the whole message including the header are examined. Second, if IMG, SRC, freeserve (in the domain) are among the most interesting words and high on the probability list, then it's spam. I would expect that to be the case since the only HTML email I seem to get is spam.
>> based on the assumption that it is written in English
There's no reason to think that Spanish, German, or even Chinese spam doesn't follow the same statistical word frequency rules.
>> following the simple steps outlined in the URL above
What if you are subscribed to mailing lists, or have mail bots that send you useful messages (like "your server is down")? The usual answer is "just configure those in advance" but that's a pain and not very robust. My hosting company was bought out and their automated server status messages just started to come from a new domain. If I had this kind of filter I would have missed them.
>> Microsoft developer licenses can be pricey
Sounds like you didn't know that developers can get every business and OS product that Microsoft makes for every international language in the MSDN Subscription on DVD for $2,500. Most US developers would only need the Professional subscription which is $1,200. That includes MS Office, Visual Studio and all the compilers, Project, SQL Server, SDKs, DDKs, every version of Windows since 95, and a year of updates. The MSDN versions of most products allow 10 licenses, which is plenty for most developers. The price of the Windows licenses alone far exceeds the cost of the subscription.
>> Later, when and if they got enough sales, they'd reconsider XP.
I don't know their application so I can't say for sure, but in most cases that's ass-backwards. You usually want to build your product for the biggest market first.
Geez, that's great. So why didn't anyone see this one?
You miss the point about the goal of this. It isn't to harass the patent reviewers, it is to help them find the information they don't seem to be finding. To say "the reviewers are strapped for time as it is" is like saying "Linus is already busy, the last thing he needs is people to be sending him patched code to look at."
To prevent offtopic posts and flames it could be restricted, for example only to registered users. By registered I mean that the USPTO knows where they live, not that they set up some lame account with a fake email address. Abusers would be banned.
Isn't there some way that the Patent Office could open up this process so that the prior art could be waved in front of them before the patent is granted and expensive lawyers have to be called in to resolve the issue?
I'm thinking the USPTO could create a database of pending patents on their web site that have passed initial muster with the investigator and are likely to be approved. Interested parties could go and post links about prior art (or earlier filed but still pending patents) for the patent investigator to review.
If the MS contract just says they can't ship the system bare, why not install Linux?
Also, does this mean that Microsoft's contract clause will go unchallenged? Who the heck is Microsoft to say whether Dell ships their hardware with or without an OS on it?
That program was called fuzz. A pretty cool idea, maybe someone should send the link to Mcafee.
Huh? He even sends a message to the app that increases the buffer limit to ensure he can overrun the buffer.
Duh! The exploit is *intended* to be malicious code that causes a buffer overrun in an application and could be used to break into a system. That's the point!
Hint to moderators: Try "Funny" next time.
This Win32 attack can't be removed without breaking literally thousands of programs.
Please provide your list of thousands of Win32 apps that run as a privileged service and interact with the desktop. My guess is that the number is more like 10 than 1000.
I agree that Microsoft should emphasize the security risk of services that interact with the desktop. A good place to do that would be on the page that documents the use of interactive services. There's a short discussion near the bottom, but it focuses on getting privileges, not restricting them.
As far as docs go, Linux isn't setting a shining example. Here's the understated comment about sprintf tacked on at the end of the Linux man page:
Because sprintf and vsprintf assume an infinitely long string, callers must be careful not to overflow the actual space; this is often impossible to assure.
its a fact that someone, many someones actually, will write poor code. It is a flaw in the OS that lets this happen.
The OS can protect against dumb coding by someone who writes apps with elevated privilege? You underestimate the cleverness of those dumb coders.
Something around 10MB/sec sounds about right for application-level I/O with a good ATA100 drive using DMA, at least from my experience with Windows. If you bypass all the application layers (file system with potential for fragmentation, disk cache with possibly non-optimal prefetch) and go directly to the device you can get better numbers, but few apps can do that.
Okay, let's compare these comments with the ones about the end-of-life plans for Windows 2000 and Windows 95. Executive summary: It's the right business and technical decision for Apple, but it's heavy-handed tactics by Microsoft. Most OS 9 apps run under OS X using classic mode, and this is why Apple is justified in killing OS 9. Most Windows 95 apps run under Windows XP, either out of the box or using compatibilty mode, but it's part of Microsoft's plan to make us use the license-enforced XP scourge. This is not contradictory logic because Apple is an underdog and Microsoft is a monopoly.
Consumer's Checkbook had a good survey of all the cell providers serving the region in their latest Washington DC area pub, and it looks like they did the same for San Francisco as well. I have Sprint and I can tell you they mentioned most of the dead spots Sprint has near my house.
To read the survey you'll either need to buy the dead-tree version at a local bookstore or subscribe to the web site.
Gan you near be how? Could.
Comcast has been offering a premium service for a few months now:
/ Ad ditionalProducts/serviceupgrades.asp
http://comcast.comcastonline.com/memberservices
They don't seem to promote it though.
Doesn't it seem odd to you that TV makers oppose a regulation that basically would require all consumers to buy new equipment over the next five years? If they don't buy a new TV, they will at least need to buy a converter box for their existing one. So why would the TV makers object?
The content providers, broadcasters, and equipment makers are still in a big tug of war about evil consumers recording Lion King in digital quality when it's broadcast on ABC/Disney. Equipment makers don't want to implement the copy protection that the content providers want because consumers won't want those restrictions.
I think they also suspect the govt may still back out of this mandate due to consumer revolt, and if that happens they'll be stuck with product nobody wants because it's more expensive than the old stuff.
For better or worse, the govt is the manager of the airwaves. When Congress gave away the valuable HDTV frequencies to existing broadcasters for free, it was supposedly with the understanding that they would give back their existing frequencies in 2006. That bandwidth will be auctioned off and the govt wants to make that money, so delay in switching to HDTV costs the govt. That's why they're trying to accelerate the process by laws and regulations now.