Most "foreigners" -- whatever that means given the global nature of a message board like slashdot -- probably write better English than most Americans. That's not to say their speech isn't heavily accented. It's very difficult to learn to speak a second language except from a very young age. Interestingly enough, people with vastly different accents can still understand each other with little difficulty. Language can be weird like that.
Yep. I as thinking specifically of those cases where the rename was a result of a slow double click.
If you double click the text of an icon within the alloted time for a double click, the default action will be applied. If you wait too long for the second click, and click the text of an icon, the rename feature is invoked. Most people don't make a distinction between double clicking the icon and the text. That's what I was referring to. I should have been clearer on that. The rename is only invoked if the click is on the text, not the picture.
As you say, if you highlight an icon and then click the text, rename is invoked. You're right. There should be a time out feature so that if more than, say, 1 minute has passed since the item was highlighted, then the next click doesn't invoke the rename.
In either case, if the second click is on the icon, not the text, the rename is not invoked.
The fact that the two of us spent so much time discussing this and figuring out exactly when the rename feature is invoked is perhaps an indication that:
1) This feature needs some usability study to be improved.
D'oh. Forgot about EXIF. Does EXIF or IPTC have specific fields for things like subject matter, the way an MP3 or Ogg has a specific field for artist or album?
Just off the top of my head, here are the fields I can think of that would be useful in addition to the stuff like exposure, aperature, resolution, etc., that are in EXIF:
artist title subject [who or what is in the photo] composition type [still life, action, etc., kind of like genre for music, and allow multiple types] long description [free form here, like a lyrics tag] date
It does get triggered too often in Windows, that's true. But the solution is to set the doubleclick speed for the mouse, something you can do in Win, Gnome and KDE. Don't know about the Mac, but I would assume so. So if a person is a little slower on the double click, just increase the time between clicks for what counts as a double click. That way if someone has a muscle problem that prevents quickly repeated actions, a double click will still work. The "rename" feature only gets called if the time between the two clicks is greater than the time set for the upper limit of a double click. Increast the time interval and you eliminate accidental renames.
The nice thing about increasing the time for a double click is that you can still DC quickly and it excepts it as a double click. So by setting the interval higher, you take care of people with muscle problems and those people who accidentally rename files instead of launching them.
It would be possible, but not advisible for the interface to make a distinction between DCing an icon's label and DCing the icon. Their too close together and people can't aim that well.
ACDSee does this. The user still has to go in and identify Suzie by hand of course (at least until they get that pesky telepathic interface worked out) but once done, it doesn't matter where the photo is as long as it is in ACDSee's database you can search by wedding, niece or Suzie.
What we need is a standard for metadata for photos and videos like we have for music in oggs and mp3s. If there were a standard set of data for the filetype itself it would solve a lot of these issues more simply.
Yeah, there's a couple of different ways to do it in a dos batch file. I think you can even do a foreach to loop through all the files in a directory. And then there's windows script, another overly complicated solution for a simple problem. And of course, if I use a computer for more than half an hour, I put perl and php on it for command line scripting.
But for the average user with a digital camera, the software that comes with the camera normally has a batch rename function. I know my Nikon did and the others I looked at when shopping had similar software.
And even better, many photo programs allow batch renames. So while you're putting them in the wedding folder, rename them all to wedding####.jpg and let the program automatically append numbers.
Reminds me of Scotty's line, "The more they overthink the plumbing, the easier it is to stop up the drains." They've developed a complex solution for a simple problem that already had a simple solution.
While a database driven file system with the ability to let users define their own metadata fields in the database sounds really, really cool, I won't be using Microsoft's first or second version for anything I value.
So what's the status of the *nix version of a database file system?
And video phones have been around (or at least technology capable of producing them) for decades. It's rare that someone wants to actually see who their talking to when they use the phone. No more answering the phone right out of the shower, for a start. Well, unless you're in to that. And certainly no phones while driving.
And this doesn't even begin to address the bandwidth issues for the many, many people who are still on dialup.
I realize you are being funny, but Chaucer actually wrote in Middle English, not OE. OE was much more Germanic in nature; ME clearly shows the French influences.
The big question is why a 600 year old treatise is suddenly news. Did someone just now discover the Guttenberg Project?
The problem is that IE (and Firebird, and Mozilla) all display the URL as typed, including user name and password information. So if you type http://www.slashdot.org:foo@www.whitehouse.gov/ you get directed to a nasty site, even though the URL appears to say www.slashdot.org.
No. I can *not* reproduce the bug you describe in Firebird running under XP, and I can't imagine the bug would be present in the Linux version and not the windows. I tried both typing the url you entered and copying and pasting and in both cases Firebird displays the whole url and does not appear to say just www.slashdot.org.
Your example didn't work in the latest patched version of IE 6, so I went to http://www.zapthedingbat.com/security/ex01/vun1.ht m and tested from there. Even from that site, Firebird correctly display the entire url. It was not vulnerable to the bug. My copy of IE 6, with all latest patches, was vulnerable.
Your quote doesn't mention at all the number of appointments that were blocked, merely the ones that were voted on. So 20 nominees received a hearing within 2 weeks. That doesn't tell us how many did *not* receive a hearing. It says they confirmed 370+ nominees. But it doesn't tell us at all how many nominees Clinton actually had. It could be 372, it could be 500. This is called telling the truth selectively. Makes me think of that Monty Python news sketch. No parrots were harmed today in any plane crashes. No wombats were harmed today in any plane crashes. And on through several animals, never telling how many people were killed in the plane crash.
Nearly 60 Clinton nominees were not given hearings and/or votes, and others were filibustered or waited years to get their hearings.
At the time Democrats took over leadership of the Judiciary Committee in the summer of 2001, Democrats inherited 110 vacancies, and 40 additional vacancies occurred while Democrats were in charge. The Democratic Senate confirmed 100 Bush judicial nominees - 17 circuit and 83 district - in only 17 months. Presumably, nearly all 100 confirmed by the Democratic-led Senate were pro-life, conservative Republican nominees. The Democratic pace was faster and fairer than Republicans' pace since their slowdown began in 1996. Last year (2002) was the best single year (in terms of numbers of judicial nominees confirmed) since 1994.
Fair enough. I was specifically thinking of the following true story:
I come home from work about 4:30, which is earlier than I normally get home. I turn on the tv, but alas no cable. I turn on the computer. No internet. I call the cable company and tell them I can't get any channels or an ip address.
Luckily they have someone in the area and he comes out and fixes the problem (the guy in the next apartment complex got a little too aggressive mowing and severed a cable).
In my mind, this is not a computer problem, even if I had turned the computer on first and diagnosed it that way. For one reason or another, my IP was down. But I first noticed it because the cable tv was out.
You are probably right -- most people do assume it is a problem with their computer if their ISP is having problems. I work with computers, and I know how mine is set up. If I can't get an ip address, my first thought is that they've had an outage or something, not that my computer has had a problem.
This is computer related and does therefore count. No, it's not Microsoft's fault either. We're still at 11:20.
Just out of curiosity, if the power did go out -- say he plugged in a hair dryer, a blender and a space heater on the same circuit, blew the circuit breaker to the computer room and then spent an hour tracking down fuses (pretend it is an old house) and replacing them -- would that still count as a computer problem?
Ok, looking through the log there are a minimum of three computers. His computer, Mom's computer and the kids' computer.
Of the 11 hours and 20 minutes, he includes 30 minutes of a cable blackout. Now unless Bill Gates went over to the cable company and snipped some wires, that's hardly a Microsoft problem. It isn't a tv or light fixture problem if the electricity goes out, so I wouldn't call it a computer problem if the internet provider is down.
We're now down to 10:50:00.
He spent 5 minutes helping a friend with Word problems, 10:45:00.
That takes the time spent *per computer* down to 3:36 and a few seconds per month.
Of the remaining problems, a very small amount of user education can take care of a large chunk of the time. Let's start with Windows updates. 6 clicks automates this whole process and you get all the critical updates you need to stop the next worm. He devoted 1:20 minutes that he never should have to that problem. That takes the total time down to 9:25, or approx. 3:08 per computer. While this is a problem, the fix takes literally 10 seconds and from then on you spend no time keeping your computer up to date. It just happens automatically. I think this may even be turned on by default in XP now, but I could be wrong.
I've seen a lot of posts that didn't read the article and just started bashing MS and Windows. Of the remaining problems, a couple were from third party software. In fact, he even counts the time spent consoling his child when a game doesn't work as part of the computer problem time. While there are few things as sad as a Christmas toy that won't work on Christmas day, it simply isn't fair to allocate that time to the computer.
So we're down to about 3:05 per computer over the course of a month. This includes what is probably a one time event -- the 4 hours spent determining that motherboard drivers were needed and installing them. If this is a one time event, then the per computer time drops to 1:45 per month. Because this is such a limited time frame, it is unknown whether the average time spent per month is closer to 3 hours or 1 3/4 hours.
Yes, there are plenty of things wrong with Windows, Linux, OSX, and computer hardware and software in general. But this is not the article to use to get an accurate picture of how much time is wasted on poor design, bad programming, and out right errors.
The problem is that the peers reviewing it are often not experts in the field either.
And worse yet, because the articles are written by individuals in accordance with their preferences, some subjects are short of some basic information that a more well rounded article would include.
Note that the same rough edges often exist in free software projects.
Actually, I was not thinking of sources like the news media. That's not the sort of thing a general encyclopedia is for.
Knowing what is a reliable source and why is one of the biggest problems with research. General Encyclopedias, like World Book or Britannica are written by people who know something about the field when that field is especially specific. The entry on how a car engine works is probably written by a generic person with an engineering background, but something on quantum mechanics is probably written by someone with a bit of expertise
And that's the whole problem. 1000000 monkeys do not end up writing Shakespeare. Freedom of speech is great, and I think everyone should exercise it. But I'll get my information from a more reliable source, thanks all the same.
This is a textbook example of a poster not knowing the full story.
McDonald's was repeatedly warned that it was serving its coffee hotter than it should -- close to boiling. They were told to serve their coffee at a more normal temperature. When the woman received third degree burns to her thighs because McDonald's repeatedly refused to. All she did was remove the lid to add her cream and sugar and received third degree burns. Think about that.
"She was hospitalized for eight days, during which time she underwent skin grafting. Liebeck, who also underwent debridement treatments, sought to settle her claim for $20,000, but McDonalds refused.
During discovery, McDonalds produced documents showing more than 700 claims by people burned by its coffee between 1982 and 1992. Some claims involved third-degree burns substantially similar to Liebecks. This history documented McDonalds' knowledge about the extent and nature of this hazard."
So, McDonald's knew they were serving their coffee too hot, they'd been sued before, and now a woman who removed the lid to add cream and sugar had to had skin grafts as a result.
Umm, if every slashdotter bought a couple hundred dollars worth of stock, would that make us (as a group, disparate though it would be) majority stockholders, in which case we would have more pull to make them drop the suit?
I haven't done the math, but there can't be that many outstanding shares of stock compared to the number of registered slashdotters.
Impractical of course, but as long as you're talking about a coordinated effort, why not just solve the problem entirely.
I believe the response was because the writer felt that the reasons given for not choosing KDE boiled down to FUD. Whether or not that is true is a different issue. The writer seemed to feel that in writing on why he did not choose KDE, Perens was not giving a reasonable picture of the KDE project. And also perhaps that if you choose one product over another, it is good to list the positive reasons for the chosen product instead of the reasons you didn't choose the competitor.
"telling we gentle readers"
on
PC Annoyances
·
· Score: 0, Troll
Grammar Nazi alert. This should be "telling us gentle readers". Think about how the sentence would sound if you ended on the pronoun:
They did it without telling us. They did it withoug telling we.
OTOH, Thunderbird's spam filter is *so* much better than the one Eudora 6. I don't know what Eudora uses, but it constantly marks messages from mailing lists as junk, even though I have filters specifically set up to move list messages into folders for each list. WIth Eudora, I have to correct it every time I get mail. With Thunderbird and similar filters, a mail list message gets marked as junk maybe once a week.
One interface problem with Thunderbird though is the way it treats messages it decides as junk. When it puts them in the junk folder, they are not marked as junk. So it is unclear what happens if I mark a message as not junk and then read and delete without moving it to a non-junk folder. What I have to end up doing is marking all mail in the junk folder as junk and then explicitly unmarking false positives as not junk. Eudora at least solves this because once a message is in the junk folder, the only option is to mark it not junk and then eudora immediately moves the message to the right mailbox(excluding filters).
So while its algorithm for defining junk stinks, Eudora's human interface for dealing with the messages is better. With Thunderbird, it's the reverse.
Slashdot Mirror
Most "foreigners" -- whatever that means given the global nature of a message board like slashdot -- probably write better English than most Americans. That's not to say their speech isn't heavily accented. It's very difficult to learn to speak a second language except from a very young age. Interestingly enough, people with vastly different accents can still understand each other with little difficulty. Language can be weird like that.
Yep. I as thinking specifically of those cases where the rename was a result of a slow double click.
:
;)
If you double click the text of an icon within the alloted time for a double click, the default action will be applied. If you wait too long for the second click, and click the text of an icon, the rename feature is invoked. Most people don't make a distinction between double clicking the icon and the text. That's what I was referring to. I should have been clearer on that. The rename is only invoked if the click is on the text, not the picture.
As you say, if you highlight an icon and then click the text, rename is invoked. You're right. There should be a time out feature so that if more than, say, 1 minute has passed since the item was highlighted, then the next click doesn't invoke the rename.
In either case, if the second click is on the icon, not the text, the rename is not invoked.
The fact that the two of us spent so much time discussing this and figuring out exactly when the rename feature is invoked is perhaps an indication that
1) This feature needs some usability study to be improved.
2) We're so pathetic.
D'oh. Forgot about EXIF. Does EXIF or IPTC have specific fields for things like subject matter, the way an MP3 or Ogg has a specific field for artist or album?
Just off the top of my head, here are the fields I can think of that would be useful in addition to the stuff like exposure, aperature, resolution, etc., that are in EXIF:
artist
title
subject [who or what is in the photo]
composition type [still life, action, etc., kind of like genre for music, and allow multiple types]
long description [free form here, like a lyrics tag]
date
It does get triggered too often in Windows, that's true. But the solution is to set the doubleclick speed for the mouse, something you can do in Win, Gnome and KDE. Don't know about the Mac, but I would assume so. So if a person is a little slower on the double click, just increase the time between clicks for what counts as a double click. That way if someone has a muscle problem that prevents quickly repeated actions, a double click will still work. The "rename" feature only gets called if the time between the two clicks is greater than the time set for the upper limit of a double click. Increast the time interval and you eliminate accidental renames.
The nice thing about increasing the time for a double click is that you can still DC quickly and it excepts it as a double click. So by setting the interval higher, you take care of people with muscle problems and those people who accidentally rename files instead of launching them.
It would be possible, but not advisible for the interface to make a distinction between DCing an icon's label and DCing the icon. Their too close together and people can't aim that well.
ACDSee does this. The user still has to go in and identify Suzie by hand of course (at least until they get that pesky telepathic interface worked out) but once done, it doesn't matter where the photo is as long as it is in ACDSee's database you can search by wedding, niece or Suzie.
What we need is a standard for metadata for photos and videos like we have for music in oggs and mp3s. If there were a standard set of data for the filetype itself it would solve a lot of these issues more simply.
Yeah, there's a couple of different ways to do it in a dos batch file. I think you can even do a foreach to loop through all the files in a directory. And then there's windows script, another overly complicated solution for a simple problem. And of course, if I use a computer for more than half an hour, I put perl and php on it for command line scripting.
But for the average user with a digital camera, the software that comes with the camera normally has a batch rename function. I know my Nikon did and the others I looked at when shopping had similar software.
And even better, many photo programs allow batch renames. So while you're putting them in the wedding folder, rename them all to wedding####.jpg and let the program automatically append numbers.
Reminds me of Scotty's line, "The more they overthink the plumbing, the easier it is to stop up the drains." They've developed a complex solution for a simple problem that already had a simple solution.
While a database driven file system with the ability to let users define their own metadata fields in the database sounds really, really cool, I won't be using Microsoft's first or second version for anything I value.
So what's the status of the *nix version of a database file system?
And video phones have been around (or at least technology capable of producing them) for decades. It's rare that someone wants to actually see who their talking to when they use the phone. No more answering the phone right out of the shower, for a start. Well, unless you're in to that. And certainly no phones while driving.
And this doesn't even begin to address the bandwidth issues for the many, many people who are still on dialup.
I realize you are being funny, but Chaucer actually wrote in Middle English, not OE. OE was much more Germanic in nature; ME clearly shows the French influences.
The big question is why a 600 year old treatise is suddenly news. Did someone just now discover the Guttenberg Project?
The problem is that IE (and Firebird, and Mozilla) all display the URL as typed, including user name and password information. So if you type http://www.slashdot.org:foo@www.whitehouse.gov/ you get directed to a nasty site, even though the URL appears to say www.slashdot.org.
t m and tested from there. Even from that site, Firebird correctly display the entire url. It was not vulnerable to the bug. My copy of IE 6, with all latest patches, was vulnerable.
No. I can *not* reproduce the bug you describe in Firebird running under XP, and I can't imagine the bug would be present in the Linux version and not the windows. I tried both typing the url you entered and copying and pasting and in both cases Firebird displays the whole url and does not appear to say just www.slashdot.org.
Your example didn't work in the latest patched version of IE 6, so I went to http://www.zapthedingbat.com/security/ex01/vun1.h
You might want to check out this quote from http://leahy.senate.gov/press/200305/050803.html a source as unbiased as Orrin Hatch, the source of your quote.
Nearly 60 Clinton nominees were not given hearings and/or votes, and others were filibustered or waited years to get their hearings.
At the time Democrats took over leadership of the Judiciary Committee in the summer of 2001, Democrats inherited 110 vacancies, and 40 additional vacancies occurred while Democrats were in charge. The Democratic Senate confirmed 100 Bush judicial nominees - 17 circuit and 83 district - in only 17 months. Presumably, nearly all 100 confirmed by the Democratic-led Senate were pro-life, conservative Republican nominees. The Democratic pace was faster and fairer than Republicans' pace since their slowdown began in 1996. Last year (2002) was the best single year (in terms of numbers of judicial nominees confirmed) since 1994.
And considering the Repubs did the same thing to Clinton's nominees, it should hardly come as a surprise.
Fair enough. I was specifically thinking of the following true story:
I come home from work about 4:30, which is earlier than I normally get home. I turn on the tv, but alas no cable. I turn on the computer. No internet. I call the cable company and tell them I can't get any channels or an ip address.
Luckily they have someone in the area and he comes out and fixes the problem (the guy in the next apartment complex got a little too aggressive mowing and severed a cable).
In my mind, this is not a computer problem, even if I had turned the computer on first and diagnosed it that way. For one reason or another, my IP was down. But I first noticed it because the cable tv was out.
You are probably right -- most people do assume it is a problem with their computer if their ISP is having problems. I work with computers, and I know how mine is set up. If I can't get an ip address, my first thought is that they've had an outage or something, not that my computer has had a problem.
This is computer related and does therefore count. No, it's not Microsoft's fault either. We're still at 11:20.
Just out of curiosity, if the power did go out -- say he plugged in a hair dryer, a blender and a space heater on the same circuit, blew the circuit breaker to the computer room and then spent an hour tracking down fuses (pretend it is an old house) and replacing them -- would that still count as a computer problem?
Ok, looking through the log there are a minimum of three computers. His computer, Mom's computer and the kids' computer.
Of the 11 hours and 20 minutes, he includes 30 minutes of a cable blackout. Now unless Bill Gates went over to the cable company and snipped some wires, that's hardly a Microsoft problem. It isn't a tv or light fixture problem if the electricity goes out, so I wouldn't call it a computer problem if the internet provider is down.
We're now down to 10:50:00.
He spent 5 minutes helping a friend with Word problems, 10:45:00.
That takes the time spent *per computer* down to 3:36 and a few seconds per month.
Of the remaining problems, a very small amount of user education can take care of a large chunk of the time. Let's start with Windows updates. 6 clicks automates this whole process and you get all the critical updates you need to stop the next worm. He devoted 1:20 minutes that he never should have to that problem. That takes the total time down to 9:25, or approx. 3:08 per computer. While this is a problem, the fix takes literally 10 seconds and from then on you spend no time keeping your computer up to date. It just happens automatically. I think this may even be turned on by default in XP now, but I could be wrong.
I've seen a lot of posts that didn't read the article and just started bashing MS and Windows. Of the remaining problems, a couple were from third party software. In fact, he even counts the time spent consoling his child when a game doesn't work as part of the computer problem time. While there are few things as sad as a Christmas toy that won't work on Christmas day, it simply isn't fair to allocate that time to the computer.
So we're down to about 3:05 per computer over the course of a month. This includes what is probably a one time event -- the 4 hours spent determining that motherboard drivers were needed and installing them. If this is a one time event, then the per computer time drops to 1:45 per month. Because this is such a limited time frame, it is unknown whether the average time spent per month is closer to 3 hours or 1 3/4 hours.
Yes, there are plenty of things wrong with Windows, Linux, OSX, and computer hardware and software in general. But this is not the article to use to get an accurate picture of how much time is wasted on poor design, bad programming, and out right errors.
The problem is that the peers reviewing it are often not experts in the field either.
And worse yet, because the articles are written by individuals in accordance with their preferences, some subjects are short of some basic information that a more well rounded article would include.
Note that the same rough edges often exist in free software projects.
Actually, I was not thinking of sources like the news media. That's not the sort of thing a general encyclopedia is for.
Knowing what is a reliable source and why is one of the biggest problems with research. General Encyclopedias, like World Book or Britannica are written by people who know something about the field when that field is especially specific. The entry on how a car engine works is probably written by a generic person with an engineering background, but something on quantum mechanics is probably written by someone with a bit of expertise
It all depends on the users using it.
And that's the whole problem. 1000000 monkeys do not end up writing Shakespeare. Freedom of speech is great, and I think everyone should exercise it. But I'll get my information from a more reliable source, thanks all the same.
No keyboards?
;)
Nope. Just a numberpad and a send button. Or at least that's all the old one I used years ago had.
185 is many degrees away from boiling
Hmm. 27 degrees to be precise. So if I stuck your hand in 185 degree water, you'd be ok with that?
27 degrees is not that much of a difference at that temperature -- appproximately 15% difference. Close.
The rest is just trolling.
This is a textbook example of a poster not knowing the full story.
McDonald's was repeatedly warned that it was serving its coffee hotter than it should -- close to boiling. They were told to serve their coffee at a more normal temperature. When the woman received third degree burns to her thighs because McDonald's repeatedly refused to. All she did was remove the lid to add her cream and sugar and received third degree burns. Think about that.
"She was hospitalized for eight days, during which time she underwent skin grafting. Liebeck, who also underwent debridement treatments, sought to settle her claim for $20,000, but McDonalds refused.
During discovery, McDonalds produced documents showing more than 700 claims by people burned by its coffee between 1982 and 1992. Some claims involved third-degree burns substantially similar to Liebecks. This history documented McDonalds' knowledge about the extent and nature of this hazard."
So, McDonald's knew they were serving their coffee too hot, they'd been sued before, and now a woman who removed the lid to add cream and sugar had to had skin grafts as a result.
Umm, if every slashdotter bought a couple hundred dollars worth of stock, would that make us (as a group, disparate though it would be) majority stockholders, in which case we would have more pull to make them drop the suit?
I haven't done the math, but there can't be that many outstanding shares of stock compared to the number of registered slashdotters.
Impractical of course, but as long as you're talking about a coordinated effort, why not just solve the problem entirely.
I believe the response was because the writer felt that the reasons given for not choosing KDE boiled down to FUD. Whether or not that is true is a different issue. The writer seemed to feel that in writing on why he did not choose KDE, Perens was not giving a reasonable picture of the KDE project. And also perhaps that if you choose one product over another, it is good to list the positive reasons for the chosen product instead of the reasons you didn't choose the competitor.
Grammar Nazi alert. This should be "telling us gentle readers". Think about how the sentence would sound if you ended on the pronoun:
They did it without telling us.
They did it withoug telling we.
OTOH, Thunderbird's spam filter is *so* much better than the one Eudora 6. I don't know what Eudora uses, but it constantly marks messages from mailing lists as junk, even though I have filters specifically set up to move list messages into folders for each list. WIth Eudora, I have to correct it every time I get mail. With Thunderbird and similar filters, a mail list message gets marked as junk maybe once a week.
One interface problem with Thunderbird though is the way it treats messages it decides as junk. When it puts them in the junk folder, they are not marked as junk. So it is unclear what happens if I mark a message as not junk and then read and delete without moving it to a non-junk folder. What I have to end up doing is marking all mail in the junk folder as junk and then explicitly unmarking false positives as not junk. Eudora at least solves this because once a message is in the junk folder, the only option is to mark it not junk and then eudora immediately moves the message to the right mailbox(excluding filters).
So while its algorithm for defining junk stinks, Eudora's human interface for dealing with the messages is better. With Thunderbird, it's the reverse.