Yes, that is definitely idiotic big-company-think. I think a manager should be responsible for informally overseeing or okaying use of random new tools. Using a tool is very different from incorporating source code, copy-and-pasting material, and doing something that creates a potential legal issue. As a manager, if somebody says they need Winzip, emacs, bash or whatever to feel comfortable and get work done, then more power to them. It should be made completely clear to them that they can't download source code or software from any source on the internet for use as a part of a product, runtime component or anything like that without approval from a manager and legal. Beyond that, there's nothing you can do but trust your employees, make the potential consequences really clear, and conduct regular code reviews to spot anything potentially suspect (primarily just to spot shitty, lazy code, but if somebody really cut and pasted a bunch of code, it would probably be obvious if you knew their coding style, your company's coding standards, and so on).
NOT like the Linux kernel. You can "use" the Linux kernel in any kind of project you want - rarely indeed do you need to create derived works of the kernel, which is a piece of OS-level infrastructure, just to build custom apps. The GPLness of the kernel pretty much only comes into play with hardware support and driver issues because they must link directly against kernel code and thus create derived works. You can generally call Unix standard functions without creating a derived work of the Linux kernel.
On the other hand, if you GUI toolkit library is under the GPL, than simply "using" it, which ALWAYS involved creating derived works to call into the API is possible only for GPL-compatible software. So this is not just like the linux kernel at all. I like Qt too, but developing commercial software with it is costly relative to other options. For a wide-use packaged software app for *nix, this cost is insignificant relative to other costs. But in a situation with a hobbyist developer wanting to write a cross-platform app, or make it so a primarily Windows-targetted app is also accessible to a *nix audience, that cost, on the margin, to gain access to an additional 1-2% desktop audience is very hard to justify.
I'm glad we have choice of desktop environment too. I just feel, as do many others, that the choice of API should not lock you into one widget set with its own drawing mechanisms, theme engine, and so on such that it becomes aesthetically impossible to run apps written for one desktop environment and apps written for the other desktop environment at the same time, when certain apps really only have a top notch representative in one widget set or the other.
I think hacking the widget set to make it use the drawing routines of the other widget set as much as possible is probably the best solution anybody has come up with yet to make them fit together without having to go through some manual configuration nightmare and cherry pick themes with matching engines on both the Gtk and Qt side. So kudos to this fellow and may his work continue.
I'll throw in my vote for allofmp3.com as well. The selection is different from iTunes Music Store - they have some things I couldn't find on iTunes, and they lack some things that iTunes does have, so YMMV. But in general the online encoding feature is fabulous, and I love being able to do a low-res preview of an entire song, not just the 30 second iTunes previews. Also the much lower cost makes it sensible to just buy entire albums that I never would have otherwise. This has encouraged me to recently explore a lot of music I never would have bothered to before.
I definitely wouldn't have done so much exploring of other genres and lesser known artists with Kazaa or other P2P services - they are all polluted with shitty quality rips and encodes, and of course 80% of the mainstream stuff is polluted with hundreds of fake MP3s these days thanks to the RIAA (though WinMX and other slightly less mainstream P2P services are better, the quality and ability to find non-mainstream stuff is still mixed).
Also - thanks to the other replier for pointing out Magnatunes - some interesting bands there, worth exploring too.
Read the article. They say that they regularly block 75-80% of incoming mail as spam. I am sure they aren't blocking all the spam, I've seen spam getting through to AOL accounts recently too - their spam filtering definitely isn't great. But it's not bad either. Also because they are so big, spammers will take extra effort to test, figure out and "game" their spam filters, I'm sure, which they would never bother with for a smaller ISP.
Fortunately, you are not a judge. There is a substantial body of case law in most states on what constitutes doing business in those states, and how much active involvement is required to establish personal jurisdiction over an issue.
There are what are called "long arm" statutes in most states that let states take jurisdiction under certain circumstances - like a foreign (i.e. out-of-state) business entity doing transactions in a state. However, just sending email, even spam doesn't count as that. Some states include a clause in their long-arm statute like "causing tortious injury" in the state if you regularly solicit business in the state. So presumably, the burden would be on the prosecution to show that they did more than just send spam (and on a regular basis), they also caused damage that was tort-worthy by their sending of spam (I'm guessing here since I don't know Virginia's specific laws).
Just calling people up or sending email to them, or sending snail mail, no matter how offensive any of it may be to you, is itself justification for that state asserting personal jurisdiction. Just imagine the mess that would occur if your (legitimate) business could be taken to court pretty much anywhere where the laws disfavored you enough that some litigious fuckhead thought he could win some arbitrary case against you. It's hard enough for most legitimate businesses to keep in the letter of all their state and local laws all the time - if you had to worry about the random legislation in all 50 states, it would be crazy. Just because these people are spammers and scumbags doesn't mean we can change the rules of jurisdiction because we want to fuck them over. Oh yeah, IANAL so don't take this as literal description of the law.
Oh because ethnic violence has never been portrayed in a movie? You think that nobody makes movies where a character in the movie says something like "Kill the [insert ethnic group here]"? In fact, there are plenty of movies where characters say much more offensive things than that - these movies of course are generally rated R, but that's the point - we know they are for adults, not children.
No, let's be honest about it. People are holding games up to an entirely different standard than they hold movies to. It's based on the broken assumption that games are for kids and movies can be for adults only.
I am very confused, as there are also plenty of movies in which characters say things plenty more offensive than "Kill The Hatians". When we see it in a movie we say "oh, that character sure is nasty". Generally, nobody says "the director of this movie is an awful racist and the studio that made it is terrible". Sure, some would argue that these movies glorify violence, contain bad language, and so on - and those people will generally choose not to watch movies that offend them, and they certainly wouldn't show such movies to their children.
The problem is that people seem to be stuck in this mindset that games are for children, whereas movies can be for adults. I even see posts here in this thread on Slashdot making these same basic assumptions. Why the hell should the way parents police the games their children play be any different from them policing the movies or TV their children watch? You have to do it for younger kids, and you have to accept that you won't really be able to do it for older kids, and that you'd damned well better make sure they know the difference between right and wrong, fantasy and reality before they get old enough to play games and watch movies you don't want them to behind your back.
Yes, I agree and you agree - we'd prefer that pretty graphics stay on the web and that email be used for text. The problem is that Joe Average seems absolutely enamored of using all sorts of atrocious fonts, background colors, and blinkenlights in his emails. I receive this things in the course of business all the time, and they make me want to gouge my eyes out.
The counterpoint to this is that Joe Average seems to respond to HTML emails with large images and huge blinkenlights much better than he responds to plain text. That is actually the primary reason spammers use it - the people saavy enough to prefer text email are exactly the same people who never buy something they receive an unsolicited email about.
The sad truth is that people are dumb, and people like shiny flashy things (my preciousssss...). Just deal with it. Don't expect them to change just because it would create a positive externality for all of us who use the Internet.
As do Packet8, Vonage, and quite a few others in the US - this is what most people mean when they say VoIP these days, not weird geek-only talk-through-your-computer systems. Vonage is in fact a bonafide CLEC, meaning there's number portability and all the other stuff you would expect with real phone service (I don't think the rest of them in the US are CLECs at this point). The only difference is the last mile to the phone - dedicated old-school copper, or broadband internet. Of course, you can also argue VoIP is just another internet service that runs over TCP/IP networks. Both arguments seem legitimate to me - frankly, I doubt that services that look, act, talk, and quack like ducks will be able to avoid taxation by claiming "hey, we're not really ducks".
The thing is XFree86 ALSO has a Board of Directors. The Core Team was like a Board of Directors, only they didn't do anything but add bureaucracy and private list discussion of issues that would then be cited as authority for decisions made. These are the fuckers that attacked Keith Packard for being "low class" because he set off to work on X outside of the XFree86 organization because they simply couldn't adopt their bureaucracy to accept innovative new patches and extensions to X.
Keith for those of you who don't know, wrote the Xft/XRender extensions that do anti-aliased font rendering and is generally the leading force pushing X (kicking and screaming, I might add) into the 21st century. The Core Team was one of the leading forces doing the kicking and screaming, next to the Board of Directors. I would be happy to see them go to, replaced by a more forward thinking, less bureaucracy-minded group of leaders.
It is a crazy emotional response to call for death. So perhaps you're right - maybe they should spend an amount of time in prison commensurate with the amount of other people's time they've wasted? If somebody like Ralsky sends spam messages to a million people per day (he has boasted of such figures), and wastes 2 seconds of each person's time dealing with and deleting the spam, that would be 2 million seconds of wasted time - or about 23 days. Basically this guy would end up spending the rest of his life in jail, probably not that far from fair for the kind of negative externality he's inflicted on all of us.
Dude, you are SO busted. I particularly like how you replied to your own original AC post a couple minutes after the post for some nice karma whoring. Hah.
I'm trying to find information on building/using Arch for Win32. I see some references to using it under Cygwin - less than ideal, IMHO. Sure, it depends on patch, tar and diff, but I have MinGW builds of these and they are readily available for Win32. Maybe I'll futz around and see if it's possible.
Ohhh, that one always gets a laugh out of me, even though it's posted on every Lord of the Rings/Star Trek Slashdot thread. You ever wonder what goes through somebody's head when they agree to record something like that? I guess back then they didn't realize it would end up immortalized on the Internet, passed around for generations as a monument to human stupidity.
That rant seems to focus on the wrong things. Subversion should probably be more stable than it is for a 3 year old project, and I think they do some stupid things - for example, binaries should be available, more emphasis on stable releases, not just "oh, use your own subversion to check out the latest subversion sources..." please... we don't all want to be fucking SCM system developers, we just want to use it.
But knocking Subversion because of the version numbering they use is dumb. Knocking Subversion because the system can integrate with Apache2 is dumb. Claiming that the features Subversion adds over CVS are trivial is crazy, and claiming every other SCM/version control system has them is just factually wrong. I agree that a good, modern SCM system has atomic commits, but if it was so damned simple, show me all the other systems that have implemented it that don't cost many hundreds to many thousands of dollars per seat license.
And there are graphical tools for Subversion, though I haven't really played with them much - rapidSVN looks like the rough equivalent of a cross-platform version of WinCVS, and tortoiseSVN is a rather interesting concept, integrating in as a Windows Explorer extension.
CVS is good. But not great. Subversion has the potential to be great - atomic commits, versioning of directories, moving files easily, cheap branching. All those things that ever made you want to smack your computer upside the head when you were using CVS because they were so obviously the WRONG way to do version control.
Unfortunately, subversion seems to be always _almost_ stable enough for real use. Maybe this has changed recently (I've just played with it, I still use CVS for real work). I haven't really checked out GNU Arch - it seems to claim to support changesets (groups of changes), and thus I presume atomic commits, better/faster branching and merging and so on - the other good stuff that CVS is lacking. My guess is that Arch is even less mature than Subversion though, since it appears to have not been around as long.
Anyone else know of any other good alternatives that are more mature?
It's the best solution to the general problem. I am not really aware of any way to take an app that uses non-native widgets and make it use native widgets - it would probably not be possible to write a general API wrapper to go from Gtk->Windows widgets. Another response suggested Gtk-Wimp, a nice Windows Gtk theme to make things blend in - in my experience, these solutions are always "almost perfect" - they never perfectly match the look AND feel of your native platform widgets.
In other words, there IS NO solution without rewriting a tool to use a real cross-platform native widget set, or using a different native tool on each platform, or dealing with the godawful clash of different toolkits and themes everywhere you go (which personally causes me so much aesthetic pain, I can't deal with it).
You might want to check out WxWindows for a toolkit that does exactly what you propose, and does it quite well. It uses Gtk widgets on X, and native Win32 widgets in Windows (and I assume Carbon or something similar on OS X and MacOS 9).
The API isn't quite as nice, clean consistent and well-documented as Qt, but it's definitely not bad, and I've written some fairly complex GUIs with it before. They look and feel truly native, though it may take some minor tweaking to get everything perfect on all the platforms you are targeting.
Yes, as you said 99% of the work is done by Mandrakesoft. This has been the case with every one of their recent releases. I have been a Mandrake fan for some time (I remember installing the first or second release of Mandrake when it was really just a patched, bugfixed RedHat). I still use Mandrake regularly, but what ticks me off about the recent releases is that they all seem to be about 99% of the way there, or maybe more like 95% of the way there.
Texstar did an absolutely admirable job of packaging fabulous RPMs to fix some of the atrociousness that came with out-of-the-box Mandrake back in 9.0/9.1 (and 8.2 if I remember properly). Check out the default font configuration on 9.1 to see an example of what I'm talking about - I couldn't look at the desktop, it pained my eyes. Between Texstar's RPMs and the PLF RPMs, you can actually make Mandrake 9.1 into a usable system.
If Texstar is going to build on Mandrake and take a 95% distro and make it into a 100% distro out of the box rather than distribute piecemeal patches fixing the things Mandrakesoft screwed up, then by all means, more power to him. That's fully in the spirit of the GPL and of Linux in general. And I should again point out that Mandrake got its start as basically a bugfixed/patched up version of RedHat - anybody else remember their first releases when it looked like they had just done a Find/Replace on "RedHat" and typed in "Linux Mandrake"?
I also might recommend trying MEPIS Linux out as a great bootable CD as well as general use distro. I just discovered it recently, and I have to give it immense credit for working out of the box with all the NVidia hardware (evil-tainted driver detection and all). MEPIS gets the fact that people want easy to use and easy to install, not ideological purity. Mind you, I still use Mandrake when I'm using Linux, but if you don't have the time or patience to make Mandrake not look and feel sucky, or to make it work with your hardware, MEPIS is a great alternative (and can let you experiment with a Debian-based alternative that's very easy to test out).
I am sure PCLinuxOS probably does as good a job, knowing the quality of all the old Texstar RPMs. I predict we'll all be hearing a lot more from these upstarts, and see them presenting a serious challenge to the most popular distros, especially with the major PR fuckup that RedHat has brought upon itself with Fedora (sorry, it had to be said) and with the middling quality of the Mandrake 9.2 release (as with the last several Mandrake releases, unfortunately - always _almost_ great).
Update - I checked the letter again today, and it seems they've already, in the last 24 hours, raised more than the $20k target, by several grand. Looks like the Slashdot posting helped after all. Hope they use it for good, unlike K5, where the site's owner raised $70k and took off with it.
Also, more people die of cancer because they're not dying of other diseases.
Right, that's what I said too, which is why it is a difficult question to answer with the data we have and statistics techniques available. But it's an important one to look at before you definitively pronounce that long-term, low-level exposure to agents known to be toxic or carcinogenic in high dose exposure is NOT a substantial factor in cancer rates.
Are they carcinogens? Isn't cancer (in its many forms) the second largest cause of death in the country? I don't know if we know enough to say that changes in environmental factors have or haven't had effects. As others have pointed out, people are less likely these days to die of other factors at a young age, and thus, more people reach old age and are susceptible to death from heart disease and cancer. Do we know that controlling for other factors, cancer rates have not increased with industrialization?
I'm not one of those tree-hugger types, but I do think it's reasonable to ask these questions.
Slashdot Mirror
Yes, that is definitely idiotic big-company-think. I think a manager should be responsible for informally overseeing or okaying use of random new tools. Using a tool is very different from incorporating source code, copy-and-pasting material, and doing something that creates a potential legal issue. As a manager, if somebody says they need Winzip, emacs, bash or whatever to feel comfortable and get work done, then more power to them. It should be made completely clear to them that they can't download source code or software from any source on the internet for use as a part of a product, runtime component or anything like that without approval from a manager and legal. Beyond that, there's nothing you can do but trust your employees, make the potential consequences really clear, and conduct regular code reviews to spot anything potentially suspect (primarily just to spot shitty, lazy code, but if somebody really cut and pasted a bunch of code, it would probably be obvious if you knew their coding style, your company's coding standards, and so on).
On the other hand, if you GUI toolkit library is under the GPL, than simply "using" it, which ALWAYS involved creating derived works to call into the API is possible only for GPL-compatible software. So this is not just like the linux kernel at all. I like Qt too, but developing commercial software with it is costly relative to other options. For a wide-use packaged software app for *nix, this cost is insignificant relative to other costs. But in a situation with a hobbyist developer wanting to write a cross-platform app, or make it so a primarily Windows-targetted app is also accessible to a *nix audience, that cost, on the margin, to gain access to an additional 1-2% desktop audience is very hard to justify.
I think hacking the widget set to make it use the drawing routines of the other widget set as much as possible is probably the best solution anybody has come up with yet to make them fit together without having to go through some manual configuration nightmare and cherry pick themes with matching engines on both the Gtk and Qt side. So kudos to this fellow and may his work continue.
I definitely wouldn't have done so much exploring of other genres and lesser known artists with Kazaa or other P2P services - they are all polluted with shitty quality rips and encodes, and of course 80% of the mainstream stuff is polluted with hundreds of fake MP3s these days thanks to the RIAA (though WinMX and other slightly less mainstream P2P services are better, the quality and ability to find non-mainstream stuff is still mixed).
Also - thanks to the other replier for pointing out Magnatunes - some interesting bands there, worth exploring too.
Read the article. They say that they regularly block 75-80% of incoming mail as spam. I am sure they aren't blocking all the spam, I've seen spam getting through to AOL accounts recently too - their spam filtering definitely isn't great. But it's not bad either. Also because they are so big, spammers will take extra effort to test, figure out and "game" their spam filters, I'm sure, which they would never bother with for a smaller ISP.
There are what are called "long arm" statutes in most states that let states take jurisdiction under certain circumstances - like a foreign (i.e. out-of-state) business entity doing transactions in a state. However, just sending email, even spam doesn't count as that. Some states include a clause in their long-arm statute like "causing tortious injury" in the state if you regularly solicit business in the state. So presumably, the burden would be on the prosecution to show that they did more than just send spam (and on a regular basis), they also caused damage that was tort-worthy by their sending of spam (I'm guessing here since I don't know Virginia's specific laws).
Just calling people up or sending email to them, or sending snail mail, no matter how offensive any of it may be to you, is itself justification for that state asserting personal jurisdiction. Just imagine the mess that would occur if your (legitimate) business could be taken to court pretty much anywhere where the laws disfavored you enough that some litigious fuckhead thought he could win some arbitrary case against you. It's hard enough for most legitimate businesses to keep in the letter of all their state and local laws all the time - if you had to worry about the random legislation in all 50 states, it would be crazy. Just because these people are spammers and scumbags doesn't mean we can change the rules of jurisdiction because we want to fuck them over. Oh yeah, IANAL so don't take this as literal description of the law.
No, let's be honest about it. People are holding games up to an entirely different standard than they hold movies to. It's based on the broken assumption that games are for kids and movies can be for adults only.
The problem is that people seem to be stuck in this mindset that games are for children, whereas movies can be for adults. I even see posts here in this thread on Slashdot making these same basic assumptions. Why the hell should the way parents police the games their children play be any different from them policing the movies or TV their children watch? You have to do it for younger kids, and you have to accept that you won't really be able to do it for older kids, and that you'd damned well better make sure they know the difference between right and wrong, fantasy and reality before they get old enough to play games and watch movies you don't want them to behind your back.
Face it. If you had breasts that looked like that, you'd be proud of them too.
The counterpoint to this is that Joe Average seems to respond to HTML emails with large images and huge blinkenlights much better than he responds to plain text. That is actually the primary reason spammers use it - the people saavy enough to prefer text email are exactly the same people who never buy something they receive an unsolicited email about.
The sad truth is that people are dumb, and people like shiny flashy things (my preciousssss...). Just deal with it. Don't expect them to change just because it would create a positive externality for all of us who use the Internet.
As do Packet8, Vonage, and quite a few others in the US - this is what most people mean when they say VoIP these days, not weird geek-only talk-through-your-computer systems. Vonage is in fact a bonafide CLEC, meaning there's number portability and all the other stuff you would expect with real phone service (I don't think the rest of them in the US are CLECs at this point). The only difference is the last mile to the phone - dedicated old-school copper, or broadband internet. Of course, you can also argue VoIP is just another internet service that runs over TCP/IP networks. Both arguments seem legitimate to me - frankly, I doubt that services that look, act, talk, and quack like ducks will be able to avoid taxation by claiming "hey, we're not really ducks".
Keith for those of you who don't know, wrote the Xft/XRender extensions that do anti-aliased font rendering and is generally the leading force pushing X (kicking and screaming, I might add) into the 21st century. The Core Team was one of the leading forces doing the kicking and screaming, next to the Board of Directors. I would be happy to see them go to, replaced by a more forward thinking, less bureaucracy-minded group of leaders.
It is a crazy emotional response to call for death. So perhaps you're right - maybe they should spend an amount of time in prison commensurate with the amount of other people's time they've wasted? If somebody like Ralsky sends spam messages to a million people per day (he has boasted of such figures), and wastes 2 seconds of each person's time dealing with and deleting the spam, that would be 2 million seconds of wasted time - or about 23 days. Basically this guy would end up spending the rest of his life in jail, probably not that far from fair for the kind of negative externality he's inflicted on all of us.
Dude, you are SO busted. I particularly like how you replied to your own original AC post a couple minutes after the post for some nice karma whoring. Hah.
I'm trying to find information on building/using Arch for Win32. I see some references to using it under Cygwin - less than ideal, IMHO. Sure, it depends on patch, tar and diff, but I have MinGW builds of these and they are readily available for Win32. Maybe I'll futz around and see if it's possible.
Ohhh, that one always gets a laugh out of me, even though it's posted on every Lord of the Rings/Star Trek Slashdot thread. You ever wonder what goes through somebody's head when they agree to record something like that? I guess back then they didn't realize it would end up immortalized on the Internet, passed around for generations as a monument to human stupidity.
But knocking Subversion because of the version numbering they use is dumb. Knocking Subversion because the system can integrate with Apache2 is dumb. Claiming that the features Subversion adds over CVS are trivial is crazy, and claiming every other SCM/version control system has them is just factually wrong. I agree that a good, modern SCM system has atomic commits, but if it was so damned simple, show me all the other systems that have implemented it that don't cost many hundreds to many thousands of dollars per seat license.
And there are graphical tools for Subversion, though I haven't really played with them much - rapidSVN looks like the rough equivalent of a cross-platform version of WinCVS, and tortoiseSVN is a rather interesting concept, integrating in as a Windows Explorer extension.
Unfortunately, subversion seems to be always _almost_ stable enough for real use. Maybe this has changed recently (I've just played with it, I still use CVS for real work). I haven't really checked out GNU Arch - it seems to claim to support changesets (groups of changes), and thus I presume atomic commits, better/faster branching and merging and so on - the other good stuff that CVS is lacking. My guess is that Arch is even less mature than Subversion though, since it appears to have not been around as long.
Anyone else know of any other good alternatives that are more mature?
In other words, there IS NO solution without rewriting a tool to use a real cross-platform native widget set, or using a different native tool on each platform, or dealing with the godawful clash of different toolkits and themes everywhere you go (which personally causes me so much aesthetic pain, I can't deal with it).
The API isn't quite as nice, clean consistent and well-documented as Qt, but it's definitely not bad, and I've written some fairly complex GUIs with it before. They look and feel truly native, though it may take some minor tweaking to get everything perfect on all the platforms you are targeting.
Texstar did an absolutely admirable job of packaging fabulous RPMs to fix some of the atrociousness that came with out-of-the-box Mandrake back in 9.0/9.1 (and 8.2 if I remember properly). Check out the default font configuration on 9.1 to see an example of what I'm talking about - I couldn't look at the desktop, it pained my eyes. Between Texstar's RPMs and the PLF RPMs, you can actually make Mandrake 9.1 into a usable system.
If Texstar is going to build on Mandrake and take a 95% distro and make it into a 100% distro out of the box rather than distribute piecemeal patches fixing the things Mandrakesoft screwed up, then by all means, more power to him. That's fully in the spirit of the GPL and of Linux in general. And I should again point out that Mandrake got its start as basically a bugfixed/patched up version of RedHat - anybody else remember their first releases when it looked like they had just done a Find/Replace on "RedHat" and typed in "Linux Mandrake"?
I am sure PCLinuxOS probably does as good a job, knowing the quality of all the old Texstar RPMs. I predict we'll all be hearing a lot more from these upstarts, and see them presenting a serious challenge to the most popular distros, especially with the major PR fuckup that RedHat has brought upon itself with Fedora (sorry, it had to be said) and with the middling quality of the Mandrake 9.2 release (as with the last several Mandrake releases, unfortunately - always _almost_ great).
Update - I checked the letter again today, and it seems they've already, in the last 24 hours, raised more than the $20k target, by several grand. Looks like the Slashdot posting helped after all. Hope they use it for good, unlike K5, where the site's owner raised $70k and took off with it.
Right, that's what I said too, which is why it is a difficult question to answer with the data we have and statistics techniques available. But it's an important one to look at before you definitively pronounce that long-term, low-level exposure to agents known to be toxic or carcinogenic in high dose exposure is NOT a substantial factor in cancer rates.
I'm not one of those tree-hugger types, but I do think it's reasonable to ask these questions.