You're right... that WAS a new product, not an upgrade.
Unfortunately for most Slashdotters, I gotta say that MS handles backwards compatibility about a million times better than Apple does. I remember when I had to do Apple tech support, and we had a giant chart of which OS worked on which piece of hardware. Uuugh. That was a nightmare. These days, for 80-90% of all Windows apps, you just have to be running some version of Windows that's at least as new as Windows 95. Personally, I was amazed when I went from Windows 3.1 to NT 3.51 (never used a DOS based OS again). That was a completely different product, but the transition was pretty damn smooth.
They're a hell of a lot smarter than most Slashdotters. I got into a big flame-fest the other day when I said I was still using VB6 and will for the forseeable future because it works great for us. I got a ton of "BUT IT'S NOT SUPPORTED" responses, with no real explination.
Windows 98 to XP wasn't really an upgrade... Going from DOS to the NT kernel really was like getting a whole new product that happened to look the same and was compatible with the old stuff. That's opposed to say, Mac OS.whatever that really are just ARE upgrades (a few new features here and there, but the same basic product) that they sell every few months.
You strike me as the kind of person who thinks doing a 'select' and then formatting the output via PHP makes good software, in my mind, it's barely even software at all.
Actually, from all of my experience working with HEAVY DUTY web apps (10+ TB databases 5 years ago), I've learned that for most web apps, the BEST performance you can get comes from putting as much work at the database as is possible. There's not a "*" in any of my select's, and most of my business logic is in stored procs.
If you're software is any good it should be able to run on more then one DB, at least Postgres.
Actually, I'd say that if your software is any good, it won't be able to run on multiple database. Why? If you can run the same code on multiple databases, then you're not taking advantage of any of the database-specific performance features. Heck, how do you get stored procedures to run across multiple databases? If you're calling "SELECT * FROM TABLENAME" good software because it can run on any database, then you've got some learnin' to do.
Why are you moving cross-country for a job? Is the job THAT great that you're going to destroy any local ties that you have jsut to keep them happy? Call me nuts, but work comes secondary to the rest of my life, and something as critical like where I live isn't decided by work.
That is one of the big problems. the fact that some of my queries will not go cross platform because of stupidities thrown in by Microsoft, MYSQL, and Oracle that cause pain and suffering like this.
I'm just curious... what's the situation where you need the same SQL to talk to multiple kinds of databases? Does your company just have one of everything, or do they gut their infrastructure on a regular basis for fun? When I make a committment to a database, I expect it to be at *least* a 5+ year committment. I don't see the reason for replacing the most important part of most IT infrastuctures unless there's an extemely serious reason to do so.
But for OSS vendors, I think the most important aspect is that the client gets a horse in the race, so to speak. As a paying client, they get a seat at the table, even if by proxy, and have a voice in what will happen with the product. They become Somebody. True, paying a developer to participate is another way - and even more influential - but if your business isn't software in the first place that is just not feasible. Paying a company to, in effect, pay developers by proxy is the next best thing.
That only works with small products. If I'm paying genericJoeSupportGuy for Apache support, that doesn't mean that I get to influence Apache in any way. But you'd also get this with small products with closed source, as well. If you're one of only a few customers, they'll help tailor it to your wants and needs.
Not really.... if the software is good, then little to no service is needed. The "service" model is bogus. It already assumes that broken or difficult to implement software is being sold, and that you're goping to have to pony up to fix it/figure out how to use it later. That doesn't instill very myuch confidence in me. I try to buy software that requires as little support as possible.
Oh, right. Hackers putting together beige-box PC's running Linux and this simple software are going to get to install voting machines. Riiiight. And monkeys might fly out of my butt. Nice idea, but pretty damn silly, nonetheless.
You go ahead and use cheap Linux boxes for routing. I prefer that important devices like routers don't contain parts that fail relatively quickly... like hard drives.
On top of that, you've got all of the complexity of securing every Linux box to deal with. I'd rather buy a box that works when I plug it into the wall, knowing that it was deisgned specifically for that purpose, and is less likely to have security problems, need updating, blah, blah, blah.
There are many compelling reasons to change from VB. "Cost to Market" is the biggest factor. "Time to Market" is what I call the "Bonus Paid Factor". VB? I shake my head an wonder why.
Why? Because if you want to slap a GUI onto a database, along with some business logic, there's nothing faster or easier. What's so difficult to grasp about that? I can make a program, that conforms to all Windows windowing standards (something that most Linux programs still can't do, even with whole teams of developers), and still *does* something, and have it compiled, and functioning in under 60 seconds.
They work now. Nothing guarantees they'll continue working in Windows 2007 or whatever. MS can easily break say, something in ADO, and all the programs using it will be screwed.
Why does everybody feel the need to upgrade every piece of software that comes out? I've never understood that. We use W2K and VB6. It all works. It does it's job. When (IF) we need more functionality out of our software, then I'll look at upgrading as a possiblity. I've been in computers in one way another my whole life, and I always thought that people who jumped on every upgrade were the same kind of suckers that lease their vehicles, and switch vehicles every year, just so they have the newest.
I'm certainly not spending hard earned time and money out of fear, that one day, for no good reason, my software may just stop working. That simply doesn't happen. That's why we have hard drives.
Probably because you'd like to get another job sometime.
Good point. The hamster-wheel career of programming that requires a new skill set, and another $1000 worth of books every 6 months was a real PITA (that was part of the reason I left that career). If I were still a professional developer, I would have learned.NET already.
Thanks for the good info. That's some of the first real feedback I've gotten about it. I definitely use all kinds of 3rd party COM objects, and if they're not going to plug in seamlessly, then I'm already losing time/money. And if.NET objects are *easier*, then there's really no need. I've never really run into any problems with the plain ol' vanilla OLE DB stuff except for vendor-supplied drivers (I found one from Oracle where they switched the order of parameters in some important functions that contradicted the documentation).
Thanks for the info. It sounds like there's no real reason to make the jump.
I think that the book should start off by trying to explain to me WHY I should move from VB6 to VB2005. From everything I've read, there's little reason to make the jump. I've got probably a dozen, working, mission-critical VB6 apps right now, and I just don't see the point.
And of course, there are people like me who downloaded it about 10 times at work, but ended up not using any of the copies because the lack of ActiveX makes it pretty much useless for us. The 150M downloaded means just that... it was downloaded 150 million times. There's NO POSSIBLE WAY to get even close to estimating the number of users with that statistic. So there. You're both trolls.
Your input is quite useful, but if you spent $x on a machine and weren't happy with it, why would you spend MORE money buying something else from the same company?
I couldn't agree more. The XBox 360 pales in comparison to the PS3, based on everything I've seen and read. I know that I (and, from the XBOx 360 sales number,s millions of others, as well), will wait patiently until the PS3 comes out.
The whole thing is just "blah blah, we don't understand business and refuse to learn". The only thing unsettling is that "programmers" are too stupid to read.
I've never seen as many business illiterate people in one place as I have here on Slashdot. I'm absolutely astounded by the tripe that I read here on a daily basis that demonstrates an absolute and total ignorance about basic business principles.
See, the way I see it, a private company has their reputation to protect, and of course, they have to worry about lawsuits. If a college kid coding OSS decides to put a backdoor in for himself or friends or anybody else, there's really no down side for him to do so. He has no reputation to worry about, and as far as I can tell, odesn't have to worry about lawsuits.
The point is that it is much harder to hide malicious code when the source is available.
My point is that it's beyond unrealistic to think that an average person has any way of auditing code, whether it's going through millions of lines themselves, or hiring an extremely expensive hacker to do the same thing. The end result is the same: it's impossible to know what's in either closed or open source code for 99.999% of the population. So, it comes down to a question of who do you trust: college kids who have nothing at stake, or companies that have everything at stake?
Also, it is worth pointing out that such activity might raise anti-competetitive issues.
Also, it is worth pointing out that OSS itself may raise anti-competitive issues. Giving away a free product is about as bad if not worse than a large company controlling a large part of the market.
Slashdot Mirror
You're right... that WAS a new product, not an upgrade.
Unfortunately for most Slashdotters, I gotta say that MS handles backwards compatibility about a million times better than Apple does. I remember when I had to do Apple tech support, and we had a giant chart of which OS worked on which piece of hardware. Uuugh. That was a nightmare. These days, for 80-90% of all Windows apps, you just have to be running some version of Windows that's at least as new as Windows 95. Personally, I was amazed when I went from Windows 3.1 to NT 3.51 (never used a DOS based OS again). That was a completely different product, but the transition was pretty damn smooth.
They're a hell of a lot smarter than most Slashdotters. I got into a big flame-fest the other day when I said I was still using VB6 and will for the forseeable future because it works great for us. I got a ton of "BUT IT'S NOT SUPPORTED" responses, with no real explination.
Windows 98 to XP wasn't really an upgrade... Going from DOS to the NT kernel really was like getting a whole new product that happened to look the same and was compatible with the old stuff. That's opposed to say, Mac OS.whatever that really are just ARE upgrades (a few new features here and there, but the same basic product) that they sell every few months.
You strike me as the kind of person who thinks doing a 'select' and then formatting the output via PHP makes good software, in my mind, it's barely even software at all.
Actually, from all of my experience working with HEAVY DUTY web apps (10+ TB databases 5 years ago), I've learned that for most web apps, the BEST performance you can get comes from putting as much work at the database as is possible. There's not a "*" in any of my select's, and most of my business logic is in stored procs.
If you're software is any good it should be able to run on more then one DB, at least Postgres.
Actually, I'd say that if your software is any good, it won't be able to run on multiple database. Why? If you can run the same code on multiple databases, then you're not taking advantage of any of the database-specific performance features. Heck, how do you get stored procedures to run across multiple databases? If you're calling "SELECT * FROM TABLENAME" good software because it can run on any database, then you've got some learnin' to do.
Why are you moving cross-country for a job? Is the job THAT great that you're going to destroy any local ties that you have jsut to keep them happy? Call me nuts, but work comes secondary to the rest of my life, and something as critical like where I live isn't decided by work.
That is one of the big problems. the fact that some of my queries will not go cross platform because of stupidities thrown in by Microsoft, MYSQL, and Oracle that cause pain and suffering like this.
I'm just curious... what's the situation where you need the same SQL to talk to multiple kinds of databases? Does your company just have one of everything, or do they gut their infrastructure on a regular basis for fun? When I make a committment to a database, I expect it to be at *least* a 5+ year committment. I don't see the reason for replacing the most important part of most IT infrastuctures unless there's an extemely serious reason to do so.
But for OSS vendors, I think the most important aspect is that the client gets a horse in the race, so to speak. As a paying client, they get a seat at the table, even if by proxy, and have a voice in what will happen with the product. They become Somebody. True, paying a developer to participate is another way - and even more influential - but if your business isn't software in the first place that is just not feasible. Paying a company to, in effect, pay developers by proxy is the next best thing.
That only works with small products. If I'm paying genericJoeSupportGuy for Apache support, that doesn't mean that I get to influence Apache in any way. But you'd also get this with small products with closed source, as well. If you're one of only a few customers, they'll help tailor it to your wants and needs.
Not really.... if the software is good, then little to no service is needed. The "service" model is bogus. It already assumes that broken or difficult to implement software is being sold, and that you're goping to have to pony up to fix it/figure out how to use it later. That doesn't instill very myuch confidence in me. I try to buy software that requires as little support as possible.
Oh, right. Hackers putting together beige-box PC's running Linux and this simple software are going to get to install voting machines. Riiiight. And monkeys might fly out of my butt. Nice idea, but pretty damn silly, nonetheless.
You go ahead and use cheap Linux boxes for routing. I prefer that important devices like routers don't contain parts that fail relatively quickly... like hard drives.
On top of that, you've got all of the complexity of securing every Linux box to deal with. I'd rather buy a box that works when I plug it into the wall, knowing that it was deisgned specifically for that purpose, and is less likely to have security problems, need updating, blah, blah, blah.
There are many compelling reasons to change from VB. "Cost to Market" is the biggest factor. "Time to Market" is what I call the "Bonus Paid Factor". VB? I shake my head an wonder why.
Why? Because if you want to slap a GUI onto a database, along with some business logic, there's nothing faster or easier. What's so difficult to grasp about that? I can make a program, that conforms to all Windows windowing standards (something that most Linux programs still can't do, even with whole teams of developers), and still *does* something, and have it compiled, and functioning in under 60 seconds.
They work now. Nothing guarantees they'll continue working in Windows 2007 or whatever. MS can easily break say, something in ADO, and all the programs using it will be screwed.
Why does everybody feel the need to upgrade every piece of software that comes out? I've never understood that. We use W2K and VB6. It all works. It does it's job. When (IF) we need more functionality out of our software, then I'll look at upgrading as a possiblity. I've been in computers in one way another my whole life, and I always thought that people who jumped on every upgrade were the same kind of suckers that lease their vehicles, and switch vehicles every year, just so they have the newest.
I'm certainly not spending hard earned time and money out of fear, that one day, for no good reason, my software may just stop working. That simply doesn't happen. That's why we have hard drives.
Probably because you'd like to get another job sometime.
.NET already.
Good point. The hamster-wheel career of programming that requires a new skill set, and another $1000 worth of books every 6 months was a real PITA (that was part of the reason I left that career). If I were still a professional developer, I would have learned
his is the risk you take when coding using closed tools. There's no more support for VB6, and no real migration path.
So, what's the problem, *exactly*? I hear OSS zealots saying this all of the time, but all of my VB6 apps still work just fine...
Thanks for the good info. That's some of the first real feedback I've gotten about it. I definitely use all kinds of 3rd party COM objects, and if they're not going to plug in seamlessly, then I'm already losing time/money. And if .NET objects are *easier*, then there's really no need. I've never really run into any problems with the plain ol' vanilla OLE DB stuff except for vendor-supplied drivers (I found one from Oracle where they switched the order of parameters in some important functions that contradicted the documentation).
Thanks for the info. It sounds like there's no real reason to make the jump.
I think that the book should start off by trying to explain to me WHY I should move from VB6 to VB2005. From everything I've read, there's little reason to make the jump. I've got probably a dozen, working, mission-critical VB6 apps right now, and I just don't see the point.
And of course, there are people like me who downloaded it about 10 times at work, but ended up not using any of the copies because the lack of ActiveX makes it pretty much useless for us. The 150M downloaded means just that... it was downloaded 150 million times. There's NO POSSIBLE WAY to get even close to estimating the number of users with that statistic. So there. You're both trolls.
Your input is quite useful, but if you spent $x on a machine and weren't happy with it, why would you spend MORE money buying something else from the same company?
I couldn't agree more. The XBox 360 pales in comparison to the PS3, based on everything I've seen and read. I know that I (and, from the XBOx 360 sales number,s millions of others, as well), will wait patiently until the PS3 comes out.
Unfortunately, The Register's journalistic quality (and integrity) is on par with Slashdot's.
The whole thing is just "blah blah, we don't understand business and refuse to learn". The only thing unsettling is that "programmers" are too stupid to read.
I've never seen as many business illiterate people in one place as I have here on Slashdot. I'm absolutely astounded by the tripe that I read here on a daily basis that demonstrates an absolute and total ignorance about basic business principles.
See, the way I see it, a private company has their reputation to protect, and of course, they have to worry about lawsuits. If a college kid coding OSS decides to put a backdoor in for himself or friends or anybody else, there's really no down side for him to do so. He has no reputation to worry about, and as far as I can tell, odesn't have to worry about lawsuits.
The point is that it is much harder to hide malicious code when the source is available.
My point is that it's beyond unrealistic to think that an average person has any way of auditing code, whether it's going through millions of lines themselves, or hiring an extremely expensive hacker to do the same thing. The end result is the same: it's impossible to know what's in either closed or open source code for 99.999% of the population. So, it comes down to a question of who do you trust: college kids who have nothing at stake, or companies that have everything at stake?
Also, it is worth pointing out that such activity might raise anti-competetitive issues.
Also, it is worth pointing out that OSS itself may raise anti-competitive issues. Giving away a free product is about as bad if not worse than a large company controlling a large part of the market.