Programming is inherently a solo act. Pair programming is not only less efficient, it is 4 times less efficient. Every time you interrupt a programmer's train of thought you are losing time more certainly than when he picks up his Coke for a drink. XP pair programming is an incredibly stupid idea, which is why no one NOT ASSOCIATED with XP ever gives it a positive third party review. If you try it, well, sorry to hear that, but if you try it, you will NEVER want to pair program again.
Also, refactoring can be good, but it takes a lot of time and effort to make code cleanly work, and that effort is usually duplicated when you re-factor. The cost in time is HUGE, and it should not be taken lightly.
The hard core, nitty-gritty, down and dirty facts are that the use of XP is four times as slow, irritating as can possibly be, and would not EVER be allowed by the vast majority of seasoned programmers.
Next time you are programming think of what it would be like to have someone looking over your shoulder arguing about class definitions or variable names. Projects take two programmers twice the time as the better of the two would use to do it alone. The best is when you get stuck halfway through a re-factor and just waste time backing the changes out so the code works again.
XP is a solution looking for a problem, it probably needs refactoring itself. Next time someone recommends it, ask for a successful neutral third party evaluation (ie: one not associated with the XP program). You won't find such a positive eval because they do not exist.
Fair use depends on the impact the copying will have. This is based on the fraction of the original copied, its impact on the market for the original, its impact on the reputation of the copyright holder, the intended market for the copied lines (commercial, or educational)...
There is no fixed amount that is OK in every case. A judge would weigh the above factors and decide if it constitutes fair use or not.
As an academic, I can copy entire journal papers verbatim to teach class, but I cannot copy entire book chapters. That is because the journal papers were never intended for classroom consumption (no impact on market), whereas the book was intended exactly for classroom consumption.
SGI is actually saving themselves and others a LOT of money. If SCO goes through with lawsuits against SGI, it would cost both parties a lot of money to litigate. Part of SGI's defense will be their code audit and removal of anything remotely arguably infringing, which then limits potential damages.
After that, SGI can argue about the copyright value associated with various code fragments. These are just insurance steps. However, as SGI has the System V code, they can surely determine what code snippets may be infringing at least as well as SCO can...
Linux is built to satisfy many different design criteria. Mainly, it seems that whatever pisses off developers the most gets worked on.
Lately, that has been latency issues. Several patch-forks developed for pre-emptible kernels and low latency issues, and kernel developers didn't like latencies for X and xmms and video playing. So, they worked on it, and found a small uniprocessor performance hit accompanied a vast improvement in scheduling and latency.
Most linux servers are not CPU limited anyway. Mail servers and DNS servers and web servers are not so often CPU limited, and when they are, a faster machine and/or SMP is a better solution than a kernel with the prior latency issues.
Basically, the kernel is not optimized to get the most out of a single CPU situation. If that is REALLY important to you, it may be worth evaluating different OSs (personally, I'd find it a lot easier to buy the latest speed-demon machine and still use linux, but that is just me).
Notice that while the new kernel 'kicks ass' on SMP systems, on uniprocessor systems the 2.4 kernel is the one kicking ass. Anyone benchmarked 2.4 against some of the pre-SMP kernels on a uniprocessor machine?
Yeah, they missed an important test - latency for interactive processes. A lot of scheduler work went into improving this, and it makes a huge difference when you have large memory processes working hard.
This aspect is improved across the board in 2.6, as well as the SMP issues. Sure, the uniprocessor machine may be a little slower, but response latencies in X are a lot better, and this makes more of a difference to users.
Sounds like RedHat is trying to achieve some of the advantages of Debian. I'll welcome this, although I won't switch any machines over right away.
That's sorta what I saw too. Debian uses almost all volunteers, and things work amazingly well. If RedHat could pull their distro up to Debian standards, and off-load the work to the community, it would be a huge plus. They would re-orient their business to consulting, and specialized packaging (like to OEM computer sales).
There is not enough money in it for a corporation to maintain a distro as their principal revenue stream. Especially not while Debian is around and so comprehensive.
You are wrong, qmail is not in the same class as proprietary software.
Qmail comes free, with source, with the ability to modify the source and re-distribute the original package as source, and any patches you might have as separate patches.
Debian packaging distributes qmail with a patch and a build script.
Very nearly all the Free Software guidelines are met by the distribution of qmail. The one lacking is not a true freedom, but something that makes life easier, the ability to package binaries any way you like and re-distribute them.
Besides, doncha just like to install something that works and not need to worry about it, like djbdns or qmail?
Actually, I think you will find there are a large number of blind people who are very interested in how their perceptual capabilities are different, or especially enhanced, relative to sighted persons.
This all stems from a very long psychological literature on the capabilities of the blind, almost all of which find there to be only very small enhancements in their perception. The studies on the "obstacle sense" show the largest differences I am aware of.
Of course you did. Blind people use this all the time, it is referred to properly as the "obstacle sense".
At one time this amazed scientists - blind people could walk through a room without hitting objects. So, they covered their bodies in thick felt, and the blind still had their obstacle sense. Then, they filled their ears with wax, and the blind bumped into things.
Sighted people lack the obstacle sense, but can learn it in a few hours. No clicking or other extra noise generation is necessary.
I'd much rather see all ports 0-1023 blocked by default, and the ISP customers being allowed, through a secure web interface, to re-configure the firewall rules for their server.
Let's face it, 99.9% of home users don't need input on a secure port ever. But you don't want to choke off an experienced user. I've run mailservers and nameservers on my home machines without issue. And so have lots of other. This is a privilege of the knowledgable, and I'd hate to see it become a privilege of those who spend more for web access.
Linus doesn't ship an operating system - he provides a kernel.
A kernel, by itself, doesn't open any ports on the outside world.
Of course Microsoft is to blame for this. They know a) users rarely change default settings b) rpc ports are open by default
If Microsoft took the very tiny but reasonable step of making the RPC port closed until sharing is enabled, then Blaster wouldn't have done much.
Likewise, Microsoft knows that users are horrible at patching systems, and should have a better system in place for autoupdating the system. It should, in a sense, appear as a higher priority to the user. Instead, Microsoft enables the MS Messenger by default, so the user thinks every message is a spam.
At Microsoft, a lot of the defects in security are defective by basic design, and the fact that an exploitable bug appeared was inevitable.
And you know what - there are still millions of machines with the RPC exploit that are on the net. Blaster only took down about 150,000. The other 20 million are still exploitable.
Excuse me, but I think this is nothing more than a land grab by some greedy "Intellectual Property" company.
It isn't a company - it is a person and about 4 lawyers. Some dude wrote this patent 9 years ago, and now he is worth $500 million (at least).
The patent is ridiculous and I hardly see the patent holder writing their own browser or selling their own system.
That is pretty much exactly what he wants to do. OF course, he will start with $520 million of Microsoft's money...and work on making an OS through the browser. No one else will be allowed to use bi-directional data flow through a browser until 2015!
Microsoft will, of course, do their best with a work-around.
There are probably 100 patents filed for every patent that evolves into an implementation. Maybe 1000. The US PTO is morally bankrupt, and the future of innovation through small inventors is essentially dead. But this case is at least kinda interesting in a soap-opera kinda way.
Yeah, the REALLY important aspect of all this is the contractual agreement about adding derivative code to the AT&T code. It said, pretty bluntly, that all code that was added belonged to the AT&T copyright holders.
This is the critical clause. SCO views this as an enormously viral clause that allows them to claim EVERYTHING as owned by SCO.
OF course, in court, they will go through it. Anything BSD'd or public domain'd gets exempted. Anything reverse engineered from specs, or inherited from a common source gets exempted.
Really, SCO needs to show definitively that valid code was contributed to linux by someone with access to the valid code. The proverbial smoking gun. This is why their case will fail. The smoking gun test can account for, at most, a few lines of code, and this code will be rewritten in days. We can already see that most of the things SCO talks about were either reverse engineered from specs, or inherited from a common "public domain" ancestor.
Then there will arise the GPL question - if SCO knowlingly distributes such code under the GPL, isn't it then fair game?
I'll never forgive him for signing those absolutely stupid power contracts. My power bill was FOUR HUNDRED DOLLARS last month.
California got into power trouble with deregulation. FERC laid off for far too long even as it was obvious out of state energy companies were bleeding California dry. I blame BUSH for that - he easily could have saved everyone (except the energy company officers) a lot of cash by advising FERC to step in earlier and cap sales.
Davis acted aggressively to keep the power on. It was a really bad situation - it is a little unclear what he could have done better, knowing what he knew at the time. I certainly don't blame him for what he did, nor do I think it is at all obvious that California would be better off if someone else were in the hot seat at that time.
If it's so easy to get those signatures, why is this the first time in 90 years that it's happened?
Someone thought Davis' approval was so low he could lose a recall. So, he was willing to pay for it.
The recall election was BOUGHT, plain and simple. The event says nothing about the position of California voters. You could get a million California voters to oppose the most popular governor.
Microsoft's social engineers have been analyzing Usenet for a long time, and they are providing their clients with benefits from this analysis. Things like better search engines, better ordering of newsgroups, better ordering of threads (by activity, for example).
Anyway, nothing in this mentions any changing of standards by Microsoft, or changing of Usenet protocol.
Of course I cannot begin to tell you how much fun the old begin bug is.
I do not find prior art or demonstrating infringing to be an issue.
The biggest beef I have is that patents are used as tollbooths on non-existent tollroads. That is, the principal way the system is abused is that people think up something they think could potentially be useful, try to patent it, and hope someone comes up with a working model so that the patent holder can use his monopoly to financial advantage. This occurs because most patents are granted to people who never intend to create a working model.
A very easy work-around would be setting up a "proving" period after the patent is granted. Say, 5 years. If the patent grantee doesn't demonstrate a working model in that time, the patent becomes public domain.
This simple addition would get rid of about 3/4 of all patents.
A public commentary period will not work. Most people ignore patents now, until they try to create a working model. You know X doesn't exist, so you make one. Then you find out although you've made the world's first working model, your model is covered by 15 patents by people who had ideas and never intended to implement them.
Fair use allows for the end user to make a copy for PERSONAL use. Not corporate use, not public use, not any other use. Personal, baby.
He is not proposing to exploit any loophole in fair use. He is proposing to exploit a loophole in co-operative ownership. If you and I pool our money and buy a CD together, can we each listen to a copy of it at the same time?
Now expand that concept to everyone in the "company" buying the single CD together, and listening to it whenever we feel like downloading it. That is the proposal, except there is a download fee also.
And somebody please fill me in, but is the SCO hotshot lawyer who lost the Gore case against Florida and worked against Microsoft such a hotshot? It seems to me that he's more of a loser.
His rep stretches back further. He headed the antitrust case defense for IBM, stretching the proceedings out for so long the gov't finally dropped the case. A clear, big, win for IBM at the time.
This time around, Boies is up against the law firm he worked for when he defended IBM.
It's a new Windows API designed to turn Windows into a virtual machine like Java so it can be architecture independent.
That statement is a laughable sham, and I am sure M$ is glad you brought it up. Windows controls the hardware, and not the other way around. It has been this way for a long time - Windows killed Alpha, for example..NET is all about providing a web programming interface that fits better with Windows than Java, to force lock-in on the operating system AND the network interface. It is like Java without platform independence, so that Microsoft can make even more money. Predictably, the developer tools are so simple even a Visual Basic monkey can make a web application. Predictably, the bytecode interpreter is buggy and insecure - this is not what will win the battle. Microsoft will make life REALLY easy for developers, they will make development costs low for web companies, and.NET will attempt to throw Java out the window.
Yeah. I can't stand capacitative membrane keyboards. They are OK when new, but start to suck long and hard after a little use.
I've been using mechanical switch keyboards (actually a Darwin Keyboards Smart Board), but I don't like the ergo layout, so I've been thinking about changing to a different mechanical switch keyboard - probably the lil big board.
By the way, it can't be emphasized enough that "unstable" is still really darn stable... stable enough, in my opinion, for any reasonable home user. While some packages occasionally have small problems, it's very rare for anything big to go wrong
A few years ago Perl was broken in unstable for a few days. As apt-get and dpkg depend on Perl, well, a little "under the hood" tweaking was necessary. That was the last time I remember, though.:wq
Slashdot Mirror
Programming is inherently a solo act. Pair programming is not only less efficient, it is 4 times less efficient. Every time you interrupt a programmer's train of thought you are losing time more certainly than when he picks up his Coke for a drink. XP pair programming is an incredibly stupid idea, which is why no one NOT ASSOCIATED with XP ever gives it a positive third party review. If you try it, well, sorry to hear that, but if you try it, you will NEVER want to pair program again.
Also, refactoring can be good, but it takes a lot of time and effort to make code cleanly work, and that effort is usually duplicated when you re-factor. The cost in time is HUGE, and it should not be taken lightly.
The hard core, nitty-gritty, down and dirty facts are that the use of XP is four times as slow, irritating as can possibly be, and would not EVER be allowed by the vast majority of seasoned programmers.
Next time you are programming think of what it would be like to have someone looking over your shoulder arguing about class definitions or variable names. Projects take two programmers twice the time as the better of the two would use to do it alone. The best is when you get stuck halfway through a re-factor and just waste time backing the changes out so the code works again.
XP is a solution looking for a problem, it probably needs refactoring itself. Next time someone recommends it, ask for a successful neutral third party evaluation (ie: one not associated with the XP program). You won't find such a positive eval because they do not exist.
XP is kinda like an informercial gone bad.
Fair use depends on the impact the copying will have. This is based on the fraction of the original copied, its impact on the market for the original, its impact on the reputation of the copyright holder, the intended market for the copied lines (commercial, or educational)...
There is no fixed amount that is OK in every case. A judge would weigh the above factors and decide if it constitutes fair use or not.
As an academic, I can copy entire journal papers verbatim to teach class, but I cannot copy entire book chapters. That is because the journal papers were never intended for classroom consumption (no impact on market), whereas the book was intended exactly for classroom consumption.
SGI is actually saving themselves and others a LOT of money. If SCO goes through with lawsuits against SGI, it would cost both parties a lot of money to litigate. Part of SGI's defense will be their code audit and removal of anything remotely arguably infringing, which then limits potential damages.
After that, SGI can argue about the copyright value associated with various code fragments. These are just insurance steps. However, as SGI has the System V code, they can surely determine what code snippets may be infringing at least as well as SCO can...
Linux is built to satisfy many different design criteria. Mainly, it seems that whatever pisses off developers the most gets worked on.
Lately, that has been latency issues. Several patch-forks developed for pre-emptible kernels and low latency issues, and kernel developers didn't like latencies for X and xmms and video playing. So, they worked on it, and found a small uniprocessor performance hit accompanied a vast improvement in scheduling and latency.
Most linux servers are not CPU limited anyway. Mail servers and DNS servers and web servers are not so often CPU limited, and when they are, a faster machine and/or SMP is a better solution than a kernel with the prior latency issues.
Basically, the kernel is not optimized to get the most out of a single CPU situation. If that is REALLY important to you, it may be worth evaluating different OSs (personally, I'd find it a lot easier to buy the latest speed-demon machine and still use linux, but that is just me).
Notice that while the new kernel 'kicks ass' on SMP systems, on uniprocessor systems the 2.4 kernel is the one kicking ass. Anyone benchmarked 2.4 against some of the pre-SMP kernels on a uniprocessor machine?
Yeah, they missed an important test - latency for interactive processes. A lot of scheduler work went into improving this, and it makes a huge difference when you have large memory processes working hard.
This aspect is improved across the board in 2.6, as well as the SMP issues. Sure, the uniprocessor machine may be a little slower, but response latencies in X are a lot better, and this makes more of a difference to users.
Sounds like RedHat is trying to achieve some of the advantages of Debian. I'll welcome this, although I won't switch any machines over right away.
That's sorta what I saw too. Debian uses almost all volunteers, and things work amazingly well. If RedHat could pull their distro up to Debian standards, and off-load the work to the community, it would be a huge plus. They would re-orient their business to consulting, and specialized packaging (like to OEM computer sales).
There is not enough money in it for a corporation to maintain a distro as their principal revenue stream. Especially not while Debian is around and so comprehensive.
You are wrong, qmail is not in the same class as proprietary software.
Qmail comes free, with source, with the ability to modify the source and re-distribute the original package as source, and any patches you might have as separate patches.
Debian packaging distributes qmail with a patch and a build script.
Very nearly all the Free Software guidelines are met by the distribution of qmail. The one lacking is not a true freedom, but something that makes life easier, the ability to package binaries any way you like and re-distribute them.
Besides, doncha just like to install something that works and not need to worry about it, like djbdns or qmail?
Actually, I think you will find there are a large number of blind people who are very interested in how their perceptual capabilities are different, or especially enhanced, relative to sighted persons.
This all stems from a very long psychological literature on the capabilities of the blind, almost all of which find there to be only very small enhancements in their perception. The studies on the "obstacle sense" show the largest differences I am aware of.
Of course you did. Blind people use this all the time, it is referred to properly as the "obstacle sense".
At one time this amazed scientists - blind people could walk through a room without hitting objects. So, they covered their bodies in thick felt, and the blind still had their obstacle sense. Then, they filled their ears with wax, and the blind bumped into things.
Sighted people lack the obstacle sense, but can learn it in a few hours. No clicking or other extra noise generation is necessary.
MS already lost the lawsuit. A judgment of $521 million was made, and an injunction granted.
However, this is all stayed upon the appeal. Victory in the appeal is unlikely.
I think you hit the nail on the head when you say
maybe they will just abandon standards altogether with the next version of IE and blame it on the lawsuit.
This is far too Draconian.
I'd much rather see all ports 0-1023 blocked by default, and the ISP customers being allowed, through a secure web interface, to re-configure the firewall rules for their server.
Let's face it, 99.9% of home users don't need input on a secure port ever. But you don't want to choke off an experienced user. I've run mailservers and nameservers on my home machines without issue. And so have lots of other. This is a privilege of the knowledgable, and I'd hate to see it become a privilege of those who spend more for web access.
What I'm very curious about is will MS make Word be able to open sxw files by default?
About the same time they release MS Office for Linux.
However, expect flying pigs, frozen hell, and a Republican governor of California first.
Linus doesn't ship an operating system - he provides a kernel.
A kernel, by itself, doesn't open any ports on the outside world.
Of course Microsoft is to blame for this. They know
a) users rarely change default settings
b) rpc ports are open by default
If Microsoft took the very tiny but reasonable step of making the RPC port closed until sharing is enabled, then Blaster wouldn't have done much.
Likewise, Microsoft knows that users are horrible at patching systems, and should have a better system in place for autoupdating the system. It should, in a sense, appear as a higher priority to the user. Instead, Microsoft enables the MS Messenger by default, so the user thinks every message is a spam.
At Microsoft, a lot of the defects in security are defective by basic design, and the fact that an exploitable bug appeared was inevitable.
And you know what - there are still millions of machines with the RPC exploit that are on the net. Blaster only took down about 150,000. The other 20 million are still exploitable.
It is gonna get worse.
Excuse me, but I think this is nothing more than a land grab by some greedy "Intellectual Property" company.
It isn't a company - it is a person and about 4 lawyers. Some dude wrote this patent 9 years ago, and now he is worth $500 million (at least).
The patent is ridiculous and I hardly see the patent holder writing their own browser or selling their own system.
That is pretty much exactly what he wants to do. OF course, he will start with $520 million of Microsoft's money...and work on making an OS through the browser. No one else will be allowed to use bi-directional data flow through a browser until 2015!
Microsoft will, of course, do their best with a work-around.
There are probably 100 patents filed for every patent that evolves into an implementation. Maybe 1000. The US PTO is morally bankrupt, and the future of innovation through small inventors is essentially dead. But this case is at least kinda interesting in a soap-opera kinda way.
Yeah, the REALLY important aspect of all this is the contractual agreement about adding derivative code to the AT&T code. It said, pretty bluntly, that all code that was added belonged to the AT&T copyright holders.
This is the critical clause. SCO views this as an enormously viral clause that allows them to claim EVERYTHING as owned by SCO.
OF course, in court, they will go through it. Anything BSD'd or public domain'd gets exempted. Anything reverse engineered from specs, or inherited from a common source gets exempted.
Really, SCO needs to show definitively that valid code was contributed to linux by someone with access to the valid code. The proverbial smoking gun. This is why their case will fail. The smoking gun test can account for, at most, a few lines of code, and this code will be rewritten in days. We can already see that most of the things SCO talks about were either reverse engineered from specs, or inherited from a common "public domain" ancestor.
Then there will arise the GPL question - if SCO knowlingly distributes such code under the GPL, isn't it then fair game?
Let's see them argue their way out of that one...
I'll never forgive him for signing those absolutely stupid power contracts. My power bill was FOUR HUNDRED DOLLARS last month.
;)
California got into power trouble with deregulation. FERC laid off for far too long even as it was obvious out of state energy companies were bleeding California dry. I blame BUSH for that - he easily could have saved everyone (except the energy company officers) a lot of cash by advising FERC to step in earlier and cap sales.
Davis acted aggressively to keep the power on. It was a really bad situation - it is a little unclear what he could have done better, knowing what he knew at the time. I certainly don't blame him for what he did, nor do I think it is at all obvious that California would be better off if someone else were in the hot seat at that time.
As for your energy bill, turn off a few lights
If it's so easy to get those signatures, why is this the first time in 90 years that it's happened?
Someone thought Davis' approval was so low he could lose a recall. So, he was willing to pay for it.
The recall election was BOUGHT, plain and simple. The event says nothing about the position of California voters. You could get a million California voters to oppose the most popular governor.
RTFA.
Microsoft's social engineers have been analyzing Usenet for a long time, and they are providing their clients with benefits from this analysis. Things like better search engines, better ordering of newsgroups, better ordering of threads (by activity, for example).
Anyway, nothing in this mentions any changing of standards by Microsoft, or changing of Usenet protocol.
Of course I cannot
begin to tell you how much fun the old begin
bug is.
I do not find prior art or demonstrating infringing to be an issue.
The biggest beef I have is that patents are used as tollbooths on non-existent tollroads. That is, the principal way the system is abused is that people think up something they think could potentially be useful, try to patent it, and hope someone comes up with a working model so that the patent holder can use his monopoly to financial advantage. This occurs because most patents are granted to people who never intend to create a working model.
A very easy work-around would be setting up a "proving" period after the patent is granted. Say, 5 years. If the patent grantee doesn't demonstrate a working model in that time, the patent becomes public domain.
This simple addition would get rid of about 3/4 of all patents.
A public commentary period will not work. Most people ignore patents now, until they try to create a working model. You know X doesn't exist, so you make one. Then you find out although you've made the world's first working model, your model is covered by 15 patents by people who had ideas and never intended to implement them.
Fair use allows for the end user to make a copy for PERSONAL use. Not corporate use, not public use, not any other use. Personal, baby.
He is not proposing to exploit any loophole in fair use. He is proposing to exploit a loophole in co-operative ownership. If you and I pool our money and buy a CD together, can we each listen to a copy of it at the same time?
Now expand that concept to everyone in the "company" buying the single CD together, and listening to it whenever we feel like downloading it. That is the proposal, except there is a download fee also.
And somebody please fill me in, but is the SCO hotshot lawyer who lost the Gore case against Florida and worked against Microsoft such a hotshot? It seems to me that he's more of a loser.
His rep stretches back further. He headed the antitrust case defense for IBM, stretching the proceedings out for so long the gov't finally dropped the case. A clear, big, win for IBM at the time.
This time around, Boies is up against the law firm he worked for when he defended IBM.
It's a new Windows API designed to turn Windows into a virtual machine like Java so it can be architecture independent.
.NET is all about providing a web programming interface that fits better with Windows than Java, to force lock-in on the operating system AND the network interface. It is like Java without platform independence, so that Microsoft can make even more money. Predictably, the developer tools are so simple even a Visual Basic monkey can make a web application. Predictably, the bytecode interpreter is buggy and insecure - this is not what will win the battle. Microsoft will make life REALLY easy for developers, they will make development costs low for web companies, and .NET will attempt to throw Java out the window.
That statement is a laughable sham, and I am sure M$ is glad you brought it up. Windows controls the hardware, and not the other way around. It has been this way for a long time - Windows killed Alpha, for example.
I wonder if Redhat and Sun's attempt to open source java will have any impact on this emerging battlefield??
Yeah. I can't stand capacitative membrane keyboards. They are OK when new, but start to suck long and hard after a little use.
I've been using mechanical switch keyboards (actually a Darwin Keyboards Smart Board), but I don't like the ergo layout, so I've been thinking about changing to a different mechanical switch keyboard - probably the lil big board.
By the way, it can't be emphasized enough that "unstable" is still really darn stable... stable enough, in my opinion, for any reasonable home user. While some packages occasionally have small problems, it's very rare for anything big to go wrong
:wq
A few years ago Perl was broken in unstable for a few days. As apt-get and dpkg depend on Perl, well, a little "under the hood" tweaking was necessary. That was the last time I remember, though.