Well, yeah. There's no point talking about problems of older version... DB2, MySQL, Postgres, etc were total jokes (for other reasons) back when SQL Server 2000 came out, too. The only RDBMS that had legs to stand on back then was Oracle, if you didn't mind the buttrape price:) I mean, SQL Server 2008 is about to come out o.O
Sure you could create an extra table for the shipping info, along with another table for the billing info, but now you have 3 tables instead of 1
And having it all in one table will bite you in the ass sooner or later... especially with that example. I worked on a system where it was done that way originally... Ouf did it hurt when requirements changed to have N adress per customers, have multiple customers with the same adresses linked, etc... major refactoring there. Database design rules aren't there for show =P
Plus: Company Name, department name (pushing it), city, state and country should be lookup tables. Multiple adress lines in their own tables, and Telephone/Fax should be in their own with a "Phone Number Type" column. Bang, you're nowhere close to 8k anymore, and thats just basic normalisation (things you should do regardless of RDBMS limitations.
Of course, you seem quite knowledgeable, so I'm not going to pretend that I know better how to solve your database problems, or that I'm smarter than you. I'm most likely not:) But unless you're making a "quick and dirty" database for a web site or something (in which case, SQL Server isn't even the best tool for the job), that should be how such a database be designed.... and it actually become difficult to hit the 8k limit.
Well, unless you're like one of my old boss...
Inventory
----------
Product ID
Quantity Production Color 1
Quantity Production Color 2
Quantity Production Color 3 ...
Quantity Production Color 16
Quantity Shipping Color 1
Quantity Shipping Color 2...
you get the idea:) I think 450 columns that table had. And it was in Btrieve. Man that was painful.
You can. You just can't have more than 8k of numeric, datetimes, etc data. Types like varchar(max) and varbinary(max) don't count toward the 8k limit (and varchar(max) isn't like text...it is can be fully processed like normal varchar columns). So in any sensible database schema, you'll never break the 8k limit. Not saying they shouldn't lift that limitation: they should. But if it gets to be a problem, there is, 99% of the time, a problem with your database design.
Well, your college doesn't get new students every year? And they never get a new computer during their degree? Must be one tiny college, considering the amount of lap-tops Ive seen being shipped to campus back in the days, and that doesn't count people who bought it before.
And most people I know get a new computer right before college (I mean, their family still need theirs, they cant run off with it...you want a lap-top or something). Add all of the pirated upgrades...and you should have a decent amount of people...especially if its one of those 20k+ people public colleges.
We're in 2008. Even non-software products now get recalled, blow up, fall apart, are defective by design, are made in china (lol), all over the place. Go to Bestbuy and buy a headset at random (close your eyes and pick one), go up, and try it. 9 to 1 that thing will break within 2 weeks, sound will be crap, and it will be barely usuable.
All but the fanciest grocery stores will have expired stuff on the shelves if you look well enough. You have to be selective in what food you pick, make sure to read the expiration date, cook your meat to 160-170 degrees, etc.
Nothing works out of the box anymore. The only difference is that software doesnt always have to be recalled, it can be patched. But if you don't say informed, the ground beef you have in your fridge that got recalled...you'll never know it was. Thats "common sense" in this day and age.
Well, considering universities have a decent amount of people, that the update will be installed on a large number of users at exactly the same time....that you didnt get 200 people at the same time bitching about it means that its a rather uncommon issue. Not uncommon ENOUGH. But uncommon.
Indeed:) Though an heaily unpatched Linux machine is probably just as easy to take control of. Just less of em out there, haha. (I spent my spare time in college years rooting random Linux servers and changing the text that says "Welcome to blah blah Linux Redhat blah blah" to "Welcome to blah blah Windows ME blah blah....". That was fairly amusing, and harmless to boot.
Oh of course. Since the article was about photoshop, I was mostly refering to "anyone who uses their PC for something else than browsing, email and light word processing".
Linux will always be a top notch choice, regardless of apps, if you don't -specifically- need Windows (or Mac). I personally used Linux exclusively for years, but then switched direction in many things... now, the amount of Windows exclusive things I need is so great, I don't expect to be able to switch back to Linux for at least 5-8 years if that (thats the other extreme end of the story).
In the end, one just need to use what suits them best. Things like this article shows that people want to fit a round peg in a square hole.
The hackers collaborated online to attack and take control of as many as one million computers around the world that were not up to date with patches and didn't have users with common sense.
The thing is: there's a million things missing. When you look at discussions like these, what you see is:
"Linux only needs ONE thing to crush Windows: Exchange!!!"
Then later you see "Linux is getting Photoshop, its the one thing missing, Windows is doomed!!"
"The only reason to use Windows is games!!"
"the only reason to use Linux is autocad!!"
(add a bunch of Mac examples here too if you will).
And it goes on forever. And the way I end up seeing it is: there are a TON of things missing. Its just that the devil's in the detail...to you, its CAD and video editing, for others its something else...there are so many things missing, we'd never list them all. And its not just Linux thats like that: getting people to move away from an environment that they've been using for a while is always tricky, cuz whatever you bring to the table, its always missing -something-.
I live in north america. The only stopping most people of doing anything is the fear of getting caught. If there isn't a fear of getting caught, things automatically become "right", or even MORE than right, because they are "sticking it to the man".
The only question I have for you is the following: You said you lived in an underdeveloped country. Is it illegal in your country to pirate US-made stuff? I Don't live in the US myself, so I sure as hell don't follow US laws. I follow my country's laws (which however, support copyrights), and thats it. If tomorrow the law says that copyright is null and void, I'll only buy videogames (because I'm a collector), but movies will all get pirated, and software will be free (and as a software developer, I'll change job!).
Personally though, when I was young, my family could barely afford food (if that). If we couldn't afford something, we didn't buy it. Simple as that. Didn't go and get a bootleg. Sure, it meant I spent most of my life with no computer, no videogame, no nothing, but Im still alive.
(I'd make exception to that on softwares and tools that would help someone get OUT of poverty. Entertainment is one thing.... education and such, I feel is a RIGHT, not a priviledge, like games are.). Besides, if pirating tools and software get you rich, then you'll have money to buy them later:)
Not really. If you talk with a lot of "pirates" honestly (that is, not when they're in panic mode trying to justify themselves), what you'll hear (sometimes in nicer words) is "Anyone would be a retard to buy something they can get for free". The main difference, is when you see someone with 250 PS2 games. Well, obviously they wouldn't have bought that many...but its fairly safe to expect they would have been somewhere around the usual attach rate of the console if they hadn't. Like I said. Someone who pirated 20 games wouldn't have bought 20 games. You need to compute the ratio and make estimations. "Hey, which DS games do you have?" "Oh, too many to count, I use an R4. Used to buy douzans of games, but then I discovered R4s!". Thats just an example of a recent conversation I had with someone...and the guy is single, no car, with an income in the 6 digits. And thats fairly common (though an extreme case).
I mean, someone who -doesn't pirate- is much more likely to "not care" about game and be among the casuals who buy ZERO game beyond Wii Sport...so it averages out with the pirates who would have bought 0 game if not for piracy...
My point really was: there are ways to estimate that. Thinking that someone who pirated something would never have bought it (even at full price), is lunacy. They wouldn't have bought ALL of them, not all pirates would have bought any...but there is still a ratio that can be estimated.
For the chinese market, again, they lost something (as in potential purchases). Now, I'm not as knowledgeable there, but my entire family in law is chinese, most living actually in China, and I've still seen quite a bit how these people think (or don't think sometimes). A certain percentage of the copies would have been real sales and so on. But its probably insignificant, yes.
What it probably did hurt is the negotiations with the chinese phone providers, where perception is everything. "Why would we do business with you if everyone is going to buy unlocked fakes?"
Note: I hate Apple so I'm giggling at this event, and I hate locked phones even more, so I'm laughing my ass off and I'm quite delighted at this turn of event. So its not like the above arguments come from a "OH NOES POOR APPLE IS GETTING SCREWED!" side of me.
say they LOST a billion dollars because people who were never going to buy the product ended up not buying the product
While they ARE usually very skewed, you know how these reports work, yes? Hint: they don't count the amount of pirated products, multiply it by the cost, and come up with a figure. A lot of people WOULD have bought the thing if they couldn't pirate it, and there ARE ways to -estimate- how many.
Again, not saying its not skewed...but saying "people involved in piracy would have never bought it anyway!!" is even MORE skewed.
Really, XML does solve the problem. The only issue with it, is that its designed to solve ALL problems, instead of using the usual 80/20 rule... instead of being optimized for most problems, it uses the lowest common denominator to try and catch the 20 other %...and everything that makes XML suck comes from that extra 20%.
If it was easier to handle dates in JSON without schemas, we'd have one heck of a winner there though.
Low level CS and algorythms are just maths. My business workflow integrations have virtually no math in them aside the occasional check on money amounts.
My user interfaces are more design than math, thats for sure.
There's not a whole lot of math in Amazon's One click crap, either.
20 years ago, software was just math. Now, what makes the software run is just math and electronics, but the software itself? Depending on the field, not so much.
Well, for the installation, it actually gives detailed, app specific instructions. So yes, thats easier. For the start menu, if a user want to open notepad, they type "N" "O" "T" and its there. No menu to open at all.
I bow before you, as that was one of the best expression of the situation I've read in a long time. Finally someone else on Slashdot who gets it. Bravo!
I know its what you meant. And guiding people through command line was much harder. Actually, the easiest platform to support has always been MacOSX IMO (GUI only, even though obviously you could use the command line there too), and I've never even sat down in front of a Mac more than 15 minutes:)
The main issue with services I've always found, is that then people don't have a reason to make something that doesn't require services...
Hypothetical, geek-wise example: I make a free RDBMS to compete with Postgres, MySQL, etc. Now, I make money out of services (well, not unlike how the people behind MySQL do). Now, I need to choose which features to add (as a priority, of course all are good)... I could make the RDBMS run faster, have more programmatic features, blah blah, or I could have a good ecosystem of management tools, auto-tuning services, etc.
Well, of course, if I make money out of services, unless it would be a showstopper for adoption, I'd be RETARDED to do the later: it will cut in my cash flow, and thats all it will do (since all my extra "customers" will be jumping on the wagon because they -don't- need me).
Actually, that is fairly commonly seen in the world of software, and other products, that live mostly off of services. The thing works well, ONCE its all properly configured, setup, tuned....but good luck doing that yourself (Oracle, I'm looking at you).
Then you have things that simply don't require any support or services, and are mighty boring to make. That accounts for, like, 99% of the lacks in open source softwares: for example, most UI things. If its well done, it doesn't need to be supported (much), in general (most) people hate doing UI, its long and tedious. So the better UIs out there are made from projects that are directly or indirectly linked to commercial projects (KDE...).
In the world of music for example, you have a bit less of a problem: the better musicians, singers, etc out there, do it because its fun first, for the money later (thus why there's so much great indy music). In videogames though? Ouch. You'll virtually never get the 100+ people team to work for years on a PS3 game (especially since aside he occasional collector edition, you can't sell much lateral products). It is a pain, documentation is scarce, you have a 5 years artificial deadline to get anything done (the lifespan of the console). Well, unless everyone starts making MMORPGs (ugh....)
So by going fully services and related products, one side of things will continue to purr along just fine...the other side will probably suffer in product diversity a lot. Not everything can be supported. And there aren't enough people who do it "for fun" to completly fill the remaining gap.
Im actually not lying. I had to walk through my mom through a manual spyware/virus infestation for which there was no easy anti-virus fix (it was a virus that had just gotten in the wild, as my stepdad literally click on everything, including entering the admin password when he gets a prompt for it in Vista running as a normal user), so I had to get her to search the registery for specific patterns of keys and get her to delete them...keep in mind she can barely boot her computer.
I mean, of course if you say "Expand HKLM" they wont get it....but if you say "You see there's a list on the left...you see the first one, yes? the second? the third? Ok, click on the little plus on the left on that one...yeah".
Now I dont even need the daemon thing, I just use Gotomeeting with her...dont even need to install anything.
I've also worked in smaller companies for years, in both environments (a few years exclusively Linux, a few years exclusively Windows) where I had to support stuff like that... plus a few months (I quit that fast enough) not long after college where I worked as a support monkey for my ISP...and level 2 support actually has the agents deal with registery keys and stuff.... I've guided someone who was -DRUNK- on NEW YEARS at -2 AM- through it.... its easy enough:) You just have to word it right (same with Linux to be honest).
The registery thing can be done completly with the mouse aside for the initial regedit, so its pretty simple to explain to the less fortunates.
Actually, they've been "Helping" with it (whatever that means), and blogging about it and publishing books that refer to moonlight all over the place. Its just that moonlight is silverlight 2.0, while only silverlight 1.0 is fully done (Moonlight just follows the specs as they come, as far as I can tell)
Slashdot Mirror
Whoops. -1 point for me for forgetting to close a quote tag.
You can. You just can't have more than 8k of numeric, datetimes, etc data. Types like varchar(max) and varbinary(max) don't count toward the 8k limit (and varchar(max) isn't like text...it is can be fully processed like normal varchar columns). So in any sensible database schema, you'll never break the 8k limit. Not saying they shouldn't lift that limitation: they should. But if it gets to be a problem, there is, 99% of the time, a problem with your database design.
Well, your college doesn't get new students every year? And they never get a new computer during their degree? Must be one tiny college, considering the amount of lap-tops Ive seen being shipped to campus back in the days, and that doesn't count people who bought it before.
And most people I know get a new computer right before college (I mean, their family still need theirs, they cant run off with it...you want a lap-top or something). Add all of the pirated upgrades...and you should have a decent amount of people...especially if its one of those 20k+ people public colleges.
Hey, WinXP was trash until SP2. Win2k had registery corruption-caused bluescreens until SP3. WIndows ME....well...nevermind that one.
:)
So its always been that way
We're in 2008. Even non-software products now get recalled, blow up, fall apart, are defective by design, are made in china (lol), all over the place. Go to Bestbuy and buy a headset at random (close your eyes and pick one), go up, and try it. 9 to 1 that thing will break within 2 weeks, sound will be crap, and it will be barely usuable.
All but the fanciest grocery stores will have expired stuff on the shelves if you look well enough. You have to be selective in what food you pick, make sure to read the expiration date, cook your meat to 160-170 degrees, etc.
Nothing works out of the box anymore. The only difference is that software doesnt always have to be recalled, it can be patched. But if you don't say informed, the ground beef you have in your fridge that got recalled...you'll never know it was. Thats "common sense" in this day and age.
Well, considering universities have a decent amount of people, that the update will be installed on a large number of users at exactly the same time....that you didnt get 200 people at the same time bitching about it means that its a rather uncommon issue. Not uncommon ENOUGH. But uncommon.
Indeed :) Though an heaily unpatched Linux machine is probably just as easy to take control of. Just less of em out there, haha. (I spent my spare time in college years rooting random Linux servers and changing the text that says "Welcome to blah blah Linux Redhat blah blah" to "Welcome to blah blah Windows ME blah blah....". That was fairly amusing, and harmless to boot.
Oh of course. Since the article was about photoshop, I was mostly refering to "anyone who uses their PC for something else than browsing, email and light word processing". Linux will always be a top notch choice, regardless of apps, if you don't -specifically- need Windows (or Mac). I personally used Linux exclusively for years, but then switched direction in many things... now, the amount of Windows exclusive things I need is so great, I don't expect to be able to switch back to Linux for at least 5-8 years if that (thats the other extreme end of the story). In the end, one just need to use what suits them best. Things like this article shows that people want to fit a round peg in a square hole.
The thing is: there's a million things missing. When you look at discussions like these, what you see is:
"Linux only needs ONE thing to crush Windows: Exchange!!!"
Then later you see "Linux is getting Photoshop, its the one thing missing, Windows is doomed!!"
"The only reason to use Windows is games!!"
"the only reason to use Linux is autocad!!"
(add a bunch of Mac examples here too if you will).
And it goes on forever. And the way I end up seeing it is: there are a TON of things missing. Its just that the devil's in the detail...to you, its CAD and video editing, for others its something else...there are so many things missing, we'd never list them all. And its not just Linux thats like that: getting people to move away from an environment that they've been using for a while is always tricky, cuz whatever you bring to the table, its always missing -something-.
I live in north america. The only stopping most people of doing anything is the fear of getting caught. If there isn't a fear of getting caught, things automatically become "right", or even MORE than right, because they are "sticking it to the man".
:)
The only question I have for you is the following: You said you lived in an underdeveloped country. Is it illegal in your country to pirate US-made stuff? I Don't live in the US myself, so I sure as hell don't follow US laws. I follow my country's laws (which however, support copyrights), and thats it. If tomorrow the law says that copyright is null and void, I'll only buy videogames (because I'm a collector), but movies will all get pirated, and software will be free (and as a software developer, I'll change job!).
Personally though, when I was young, my family could barely afford food (if that). If we couldn't afford something, we didn't buy it. Simple as that. Didn't go and get a bootleg. Sure, it meant I spent most of my life with no computer, no videogame, no nothing, but Im still alive.
(I'd make exception to that on softwares and tools that would help someone get OUT of poverty. Entertainment is one thing.... education and such, I feel is a RIGHT, not a priviledge, like games are.). Besides, if pirating tools and software get you rich, then you'll have money to buy them later
I mean, someone who -doesn't pirate- is much more likely to "not care" about game and be among the casuals who buy ZERO game beyond Wii Sport...so it averages out with the pirates who would have bought 0 game if not for piracy...
My point really was: there are ways to estimate that. Thinking that someone who pirated something would never have bought it (even at full price), is lunacy. They wouldn't have bought ALL of them, not all pirates would have bought any...but there is still a ratio that can be estimated.
For the chinese market, again, they lost something (as in potential purchases). Now, I'm not as knowledgeable there, but my entire family in law is chinese, most living actually in China, and I've still seen quite a bit how these people think (or don't think sometimes). A certain percentage of the copies would have been real sales and so on. But its probably insignificant, yes.
What it probably did hurt is the negotiations with the chinese phone providers, where perception is everything. "Why would we do business with you if everyone is going to buy unlocked fakes?"
Note: I hate Apple so I'm giggling at this event, and I hate locked phones even more, so I'm laughing my ass off and I'm quite delighted at this turn of event. So its not like the above arguments come from a "OH NOES POOR APPLE IS GETTING SCREWED!" side of me.
Again, not saying its not skewed...but saying "people involved in piracy would have never bought it anyway!!" is even MORE skewed.
Really, XML does solve the problem. The only issue with it, is that its designed to solve ALL problems, instead of using the usual 80/20 rule... instead of being optimized for most problems, it uses the lowest common denominator to try and catch the 20 other %...and everything that makes XML suck comes from that extra 20%.
If it was easier to handle dates in JSON without schemas, we'd have one heck of a winner there though.
Low level CS and algorythms are just maths. My business workflow integrations have virtually no math in them aside the occasional check on money amounts.
My user interfaces are more design than math, thats for sure.
There's not a whole lot of math in Amazon's One click crap, either.
20 years ago, software was just math. Now, what makes the software run is just math and electronics, but the software itself? Depending on the field, not so much.
Well, for the installation, it actually gives detailed, app specific instructions. So yes, thats easier. For the start menu, if a user want to open notepad, they type "N" "O" "T" and its there. No menu to open at all.
Oh wait, you meant XP?
I bow before you, as that was one of the best expression of the situation I've read in a long time. Finally someone else on Slashdot who gets it. Bravo!
I know its what you meant. And guiding people through command line was much harder. Actually, the easiest platform to support has always been MacOSX IMO (GUI only, even though obviously you could use the command line there too), and I've never even sat down in front of a Mac more than 15 minutes :)
If you wanted to keep horny nerds away, you shouldn't have added that last one.
The main issue with services I've always found, is that then people don't have a reason to make something that doesn't require services...
Hypothetical, geek-wise example: I make a free RDBMS to compete with Postgres, MySQL, etc. Now, I make money out of services (well, not unlike how the people behind MySQL do). Now, I need to choose which features to add (as a priority, of course all are good)... I could make the RDBMS run faster, have more programmatic features, blah blah, or I could have a good ecosystem of management tools, auto-tuning services, etc.
Well, of course, if I make money out of services, unless it would be a showstopper for adoption, I'd be RETARDED to do the later: it will cut in my cash flow, and thats all it will do (since all my extra "customers" will be jumping on the wagon because they -don't- need me).
Actually, that is fairly commonly seen in the world of software, and other products, that live mostly off of services. The thing works well, ONCE its all properly configured, setup, tuned....but good luck doing that yourself (Oracle, I'm looking at you).
Then you have things that simply don't require any support or services, and are mighty boring to make. That accounts for, like, 99% of the lacks in open source softwares: for example, most UI things. If its well done, it doesn't need to be supported (much), in general (most) people hate doing UI, its long and tedious. So the better UIs out there are made from projects that are directly or indirectly linked to commercial projects (KDE...).
In the world of music for example, you have a bit less of a problem: the better musicians, singers, etc out there, do it because its fun first, for the money later (thus why there's so much great indy music). In videogames though? Ouch. You'll virtually never get the 100+ people team to work for years on a PS3 game (especially since aside he occasional collector edition, you can't sell much lateral products). It is a pain, documentation is scarce, you have a 5 years artificial deadline to get anything done (the lifespan of the console). Well, unless everyone starts making MMORPGs (ugh....)
So by going fully services and related products, one side of things will continue to purr along just fine...the other side will probably suffer in product diversity a lot. Not everything can be supported. And there aren't enough people who do it "for fun" to completly fill the remaining gap.
Im actually not lying. I had to walk through my mom through a manual spyware/virus infestation for which there was no easy anti-virus fix (it was a virus that had just gotten in the wild, as my stepdad literally click on everything, including entering the admin password when he gets a prompt for it in Vista running as a normal user), so I had to get her to search the registery for specific patterns of keys and get her to delete them...keep in mind she can barely boot her computer.
:) You just have to word it right (same with Linux to be honest).
I mean, of course if you say "Expand HKLM" they wont get it....but if you say "You see there's a list on the left...you see the first one, yes? the second? the third? Ok, click on the little plus on the left on that one...yeah".
Now I dont even need the daemon thing, I just use Gotomeeting with her...dont even need to install anything.
I've also worked in smaller companies for years, in both environments (a few years exclusively Linux, a few years exclusively Windows) where I had to support stuff like that... plus a few months (I quit that fast enough) not long after college where I worked as a support monkey for my ISP...and level 2 support actually has the agents deal with registery keys and stuff.... I've guided someone who was -DRUNK- on NEW YEARS at -2 AM- through it.... its easy enough
The registery thing can be done completly with the mouse aside for the initial regedit, so its pretty simple to explain to the less fortunates.
Well, right now Moonlight is actually -ahead- of Silverlight 2.0, funny enough.
That said, I've never heard of the IE thing...do you have any doc about that? I'm genuinly curious. That must be quite the funny read.
Actually, they've been "Helping" with it (whatever that means), and blogging about it and publishing books that refer to moonlight all over the place. Its just that moonlight is silverlight 2.0, while only silverlight 1.0 is fully done (Moonlight just follows the specs as they come, as far as I can tell)