Make sure that you periodically fold the things up. I shoudl've made my workbench foldable. As it is, since the bench is a flat surface that's rarely disturbed, it accumulates junk rapidly. If it folded up to the wall when not in use, I might actually have a flat surface to work on once in a while.:(
I suppose I should add soem detail to that. Mailboxes are either single files, or directories full of files, depending on whether you're using maildir, mbox, or something else. I'm partial to Maildirs, where a mailbox is acutally a directory with three folders (cur, new, tmp) and those three folders contain one file per message. It's good for lock-free mailbox editing, and moderate volume, but the absolutely huge amount of files can tax some filesystems. Using mbox gets you one file containing all of the messages, which requires actual locking, and increases the odds that some bad process could mess the file up (I specifically worry about message deletions).
Most Anmerican people live in cities. Most American people don't pay attention in History class, and most don't have any idea what most of those things on the farm are for. Granted, I do, but I grew up on a farm. Most of the people who live in the cities with 75K+ people and who I've spoken to about farm implements of some type, those people generally don't have any idea what a plow is for (or a plough, if you wanna use eaxtrae lettears and have me read it to myself as "plawf"). I went to the Museum of Science and Industry in Chicago recently, and most of the people around the farm exhibit had no idea what the Combine was. I've actualy spoken to people who have no idea what a cow or a pig is, or where meat comes from.
So yeah, referring to "the plow in the sky" (never mind that it's probably a moldboard v/s a chisel plow) probably is gonna confuse as many Americans as it helps.:p
Yeah, they're all frontends to package management systems. And firefox's update could easily be a frontend for dpkg, apt, rpm, emerge, portage, pkgtool, etc. It takes what, one like to update a package? It'd be pretty trivial to add package support and auto-built packages to the distribution scheme, and it'd make more sense to do that than to worry about keeping track of the install path for future upgrades. It's not like installing on Linux is just like installing on Win32/OS X anyway - which reminds me, do they use the OS X built-in installer system yet, or is it still "drag this application to the Applications folder"?
Yeah, and when those flash cards are combined with ide-flash adaptors and a 3Ware 12-port raid card, they'll allow for a 384GB flash RAID-0 with no almost no seek time. However, all Samsung has right now are some chips that "could" be used for something.
Besides, the real news is the 50nm process, not the capacity.:)
Umm, if you get 540 images on a 4GB microdrive, you'll be able to get approximately 540 images on a 32Gb flash drive, since 32 * 2^30 bits is 4GBytes. Well, maybe a little more, since the flash uses 2^30 while the drive proably uses 10^3. Moving to solid state, though, that'd be a huge gain in reliability.
Not to rain on your parade, but there's a factor of 8 difference between Gb and GB. The article says Gb - gigabit... I'm not sure why Samsung feels compelled to discuss the things in terms of bits anyway, since no one else typically uses bits as a measure of capacity. Next thing you know, Samsung will be using the power-of-ten units to measure their flash memory capacity - like hard drives. If they are, their "16Gb" (16,000,000,000 bits) device sounds way more impressive than calling it a 1.8GB device, even though they could well be the same thing.
"Unresolvable server" isn't a server-generated error - it's an application-generated error. I read "descriptive error pages" as being for server-generated errors, though that doesn't agree at all with the screenshot, does it?:)
It's impossible for a human to make a valid sendmail config - but the head bashing will certainly make something that a human can't distinguish from a conf that's valid.
Bah, glasses are never looked down upon. Why, I've *never* heard anyone referred to as a nerd just because they were wearing glasses. I've also never seen one of those commercials where the difference between prudish librarian and super-hot party girl is just tied-up hair and glasses.
Yep, glasses are nothing like hearing aids - since you can't really see hearing aids most of the time.
I'm not saying that error checking in the backend is not needed - I'm saying that, with properly designed applications, it should never be needed. I'm always as restrictive as possible with the data being inserted, use well-tested stored procedures when possible so the programmer has fewer places to screw up, etc. But it's my feeling that all of that should be largely wasted time, because it *should* never be needed. Then again, I think insurance in general shoudl be wasted money - but that sure a heck doesn't mean I don't carry health, auto, personal property, etc insurance...:)
The point is, anyone coming across the discrepencies in MySQL's error conditions was doing something wrong. Ergo, the programmer shares the blame - it shouldn't all be "well, MySQL is teh suck", 'cuz the programmer is also teh suck.:)
Yeah, 'cuz "friendly error mesages" is my favorite feature of IE, too. I certainly don't turn it off as one of the first things I do after a clean install or anything. I mean, why in the world would I want to know what the server is telling me *actually* went wrong, when I can see a generic interpretation of what the error code might mean? The generic interpretation is *prettier*!
Yeah, things like aptitude, porthole (or it is portal? I forget) and red carpet - they're the devil. I hate when third party apps mess with my packages.:)
What if your programmer checks for the wrong number of characters? If you have 10 applications inserting information into your database, how sure are you that all this information is checked properly?
If the programmer doesn't know the size of the fields he's inserting in to, then he's a bad programmer. Furthermore, if the programmer depends on the DBMS's behavior in edge cases, he's a bad programmer.
When I have 10 apps inserting data into my database, I'm sure they're doing it right because those applications have been tested, and ere written by competent programmers.
The rest of your reply is a straw man - I have quite a bit of database experience, both as a DBA on multiple systems and as a developer targeting database backends, not to mention formal education.
Anyway, I agree that MySQL should behave according to the ways that the others do, largely on principle alone. The examples you cited, however, are areas where the behavior shouldn't really matter. No competent user will see those situations, and no developer should ever depend on the oucome of those things. When I develop for databases, I consider that kind of thing (inserting an out-of-range number, etc) to be an undefined behavior. Not coincidentally, I have never had a problem with MySQL or any other DBMS - relational or not - in those edge cases.
I still get hits from Code Red / etc infected IIS machines, years after the patch was very widely announced. Yet, possibly infectable machines make up less than 1/5 of the Internet...
You take raw data from a user and insert it into your database without doing any range checking? You execute SQL statements that involve boolean operations on strings? No wonder you're dissapointed in the behavior of various RDBMS backends - you're a bad programmer.
Boo hoo. I'll bet the local electric company sells electricity to the SCO offices, so therefore electricity is bad! The SCO offices probably have some printers and copiers - the people selling to SCO are bad! I'll bet there are cars from all major American auto manufacturers in the SCO parking lots, which means the auto manufacturers are selling cars to these evil people - the American automotive industry is bad!
You probably also blame human murders on gun and knife manufacturers too, right?
Oh, well, that rules me out as prior art. I allowed for unlimited depth of organization, under the assumption that the user would be intelligent enough to make reasoanble decisions. My implementation was artists/albums/songs, though who else would *possibly* consider organizing music in that way?:)
My car mp3 player (warning - half-done web site from many years ago) used that interface in 1999, if I'm understanding the patent correctly. If only I'd gotten the software up there for download sooner... Seriously - how *else* would you allow effectively navigating a lot of music in a small display space?
Slashdot Mirror
I'm inclined to believe that This one is more likely.
Make sure that you periodically fold the things up. I shoudl've made my workbench foldable. As it is, since the bench is a flat surface that's rarely disturbed, it accumulates junk rapidly. If it folded up to the wall when not in use, I might actually have a flat surface to work on once in a while. :(
What could you possibly know, bumpkin? Why, unless you spend hours trying to get three blocks from your residence, you know nothing. ;)
I suppose I should add soem detail to that. Mailboxes are either single files, or directories full of files, depending on whether you're using maildir, mbox, or something else. I'm partial to Maildirs, where a mailbox is acutally a directory with three folders (cur, new, tmp) and those three folders contain one file per message. It's good for lock-free mailbox editing, and moderate volume, but the absolutely huge amount of files can tax some filesystems. Using mbox gets you one file containing all of the messages, which requires actual locking, and increases the odds that some bad process could mess the file up (I specifically worry about message deletions).
Most Anmerican people live in cities. Most American people don't pay attention in History class, and most don't have any idea what most of those things on the farm are for. Granted, I do, but I grew up on a farm. Most of the people who live in the cities with 75K+ people and who I've spoken to about farm implements of some type, those people generally don't have any idea what a plow is for (or a plough, if you wanna use eaxtrae lettears and have me read it to myself as "plawf"). I went to the Museum of Science and Industry in Chicago recently, and most of the people around the farm exhibit had no idea what the Combine was. I've actualy spoken to people who have no idea what a cow or a pig is, or where meat comes from.
:p
So yeah, referring to "the plow in the sky" (never mind that it's probably a moldboard v/s a chisel plow) probably is gonna confuse as many Americans as it helps.
Yeah, they're all frontends to package management systems. And firefox's update could easily be a frontend for dpkg, apt, rpm, emerge, portage, pkgtool, etc. It takes what, one like to update a package? It'd be pretty trivial to add package support and auto-built packages to the distribution scheme, and it'd make more sense to do that than to worry about keeping track of the install path for future upgrades. It's not like installing on Linux is just like installing on Win32/OS X anyway - which reminds me, do they use the OS X built-in installer system yet, or is it still "drag this application to the Applications folder"?
Yeah, and when those flash cards are combined with ide-flash adaptors and a 3Ware 12-port raid card, they'll allow for a 384GB flash RAID-0 with no almost no seek time. However, all Samsung has right now are some chips that "could" be used for something.
:)
Besides, the real news is the 50nm process, not the capacity.
Umm, if you get 540 images on a 4GB microdrive, you'll be able to get approximately 540 images on a 32Gb flash drive, since 32 * 2^30 bits is 4GB ytes. Well, maybe a little more, since the flash uses 2^30 while the drive proably uses 10^3. Moving to solid state, though, that'd be a huge gain in reliability.
Not to rain on your parade, but there's a factor of 8 difference between Gb and GB. The article says Gb - gigabit... I'm not sure why Samsung feels compelled to discuss the things in terms of bits anyway, since no one else typically uses bits as a measure of capacity. Next thing you know, Samsung will be using the power-of-ten units to measure their flash memory capacity - like hard drives. If they are, their "16Gb" (16,000,000,000 bits) device sounds way more impressive than calling it a 1.8GB device, even though they could well be the same thing.
"Unresolvable server" isn't a server-generated error - it's an application-generated error. I read "descriptive error pages" as being for server-generated errors, though that doesn't agree at all with the screenshot, does it? :)
It's impossible for a human to make a valid sendmail config - but the head bashing will certainly make something that a human can't distinguish from a conf that's valid.
A filesystem guarantees file integrity. A mail system guarantees mail integrity. Do one thing and do it well. Or write a plugin for Reiser4. :)
Mailboxes *are* files. Using qmail or any other sane mail system, mail backups are just like any other file backup.
Bah, glasses are never looked down upon. Why, I've *never* heard anyone referred to as a nerd just because they were wearing glasses. I've also never seen one of those commercials where the difference between prudish librarian and super-hot party girl is just tied-up hair and glasses.
Yep, glasses are nothing like hearing aids - since you can't really see hearing aids most of the time.
I'm not saying that error checking in the backend is not needed - I'm saying that, with properly designed applications, it should never be needed. I'm always as restrictive as possible with the data being inserted, use well-tested stored procedures when possible so the programmer has fewer places to screw up, etc. But it's my feeling that all of that should be largely wasted time, because it *should* never be needed. Then again, I think insurance in general shoudl be wasted money - but that sure a heck doesn't mean I don't carry health, auto, personal property, etc insurance... :)
:)
The point is, anyone coming across the discrepencies in MySQL's error conditions was doing something wrong. Ergo, the programmer shares the blame - it shouldn't all be "well, MySQL is teh suck", 'cuz the programmer is also teh suck.
Don't test on live data?
Yeah, 'cuz "friendly error mesages" is my favorite feature of IE, too. I certainly don't turn it off as one of the first things I do after a clean install or anything. I mean, why in the world would I want to know what the server is telling me *actually* went wrong, when I can see a generic interpretation of what the error code might mean? The generic interpretation is *prettier*!
Yeah, things like aptitude, porthole (or it is portal? I forget) and red carpet - they're the devil. I hate when third party apps mess with my packages. :)
What if your programmer checks for the wrong number of characters? If you have 10 applications inserting information into your database, how sure are you that all this information is checked properly?
If the programmer doesn't know the size of the fields he's inserting in to, then he's a bad programmer. Furthermore, if the programmer depends on the DBMS's behavior in edge cases, he's a bad programmer.
When I have 10 apps inserting data into my database, I'm sure they're doing it right because those applications have been tested, and ere written by competent programmers.
The rest of your reply is a straw man - I have quite a bit of database experience, both as a DBA on multiple systems and as a developer targeting database backends, not to mention formal education.
Anyway, I agree that MySQL should behave according to the ways that the others do, largely on principle alone. The examples you cited, however, are areas where the behavior shouldn't really matter. No competent user will see those situations, and no developer should ever depend on the oucome of those things. When I develop for databases, I consider that kind of thing (inserting an out-of-range number, etc) to be an undefined behavior. Not coincidentally, I have never had a problem with MySQL or any other DBMS - relational or not - in those edge cases.
I still get hits from Code Red / etc infected IIS machines, years after the patch was very widely announced. Yet, possibly infectable machines make up less than 1/5 of the Internet...
You take raw data from a user and insert it into your database without doing any range checking? You execute SQL statements that involve boolean operations on strings? No wonder you're dissapointed in the behavior of various RDBMS backends - you're a bad programmer.
Boo hoo. I'll bet the local electric company sells electricity to the SCO offices, so therefore electricity is bad! The SCO offices probably have some printers and copiers - the people selling to SCO are bad! I'll bet there are cars from all major American auto manufacturers in the SCO parking lots, which means the auto manufacturers are selling cars to these evil people - the American automotive industry is bad!
You probably also blame human murders on gun and knife manufacturers too, right?
Oh, well, that rules me out as prior art. I allowed for unlimited depth of organization, under the assumption that the user would be intelligent enough to make reasoanble decisions. My implementation was artists/albums/songs, though who else would *possibly* consider organizing music in that way? :)
Yeah, you have subscriptions to "information" sites. Sure. :)
My car mp3 player (warning - half-done web site from many years ago) used that interface in 1999, if I'm understanding the patent correctly. If only I'd gotten the software up there for download sooner... Seriously - how *else* would you allow effectively navigating a lot of music in a small display space?