Once MS makes an OS that can do all that, I might rejoin the dark side
I'm sure its safe to say MS -wish- they could do it:) They'd just get sued to oblivion by the europeans. MS Office bundled with Windows and forced on the user? Users seeing an MS controled repository of software with everything under the sun? Man, thats their wet dream.
Microsoft had to add all the security stuff. There's a reason DOS was so fast...no checking whatsoever anywhere. So obviously now this would be slower. Unixes did it already, so you can improve on performance there. But Vista -is- indeed the result of extremely poor management decisions, the same kind of constraints that produced the RRODing Xbox360s... and i expect most products of Microsoft that were thought of during that period to have similar issues. Next batch should be better.
That being said, your computer is about the same than the one i run Vista, and its snappy. People saying you need 4 gigs of ram and a dual core for it are in their world.
The amusing thing is that then they will accuse MS of spreading fud when saying things half as innacurate as this:) And I mean, MS does spread lots of fud...but Slashdot has em beat any day of the week, easy =P
It won't be long... I mean, SSIS's script components already use VB.NET (and in the next version can use more languages), so the scaffold is already there.
Of course, dunno about others, but even my Vista install on a not-so-new computer boots almost instantly...ONCE the boot loader kicks in.
Its the RAID, the videocard, the DVD-ROMS, and the bios itself that takes most of the time when im booting (especially the raid). Once the hardware junk is done, and Windows actually -tries- loading, I'm about 2 and a half seconds away from my login screen, and from there another 2-3 seconds from being able to load Internet Explorer (yeah yeah I know).
There's misinformation on both sides. The amount of people who think that your right to a backup also includes making "backups" JUST IN CASE your nearest Blockbuster got broken into and had everything stolen, so you can be the hero and give em back copies of what they once had (/sarcasm) is pretty staggering, too.
I agree with you, though one thing: One has to make the difference between a programmer and a developer. Those are 2 distinct jobs, with a fairly clear cut difference between the two.
A programmer is a "technology specialist", one that knows how to use his hammer to do what he's told. A developer is one that knows what the hell to do with a hammer... When I said the article was good, it was to find a good programmer. Because of the nature of the education system, a typical software development team is made of a lot of people who know how to code algorithms, state machines, and debug a Linux kernel, and 1, maybe 2 people who actually understand the relationship between business processes, requirements, and that code...
It really should be the other way around (Teams would work a lot better if everyone knew how to develop good software and understood the business, and 1-2 people handled the more obscure, scientific or technology intensive part of things, but thats now how universities decided it would be....much easier to train professors to teach things that never change I guess).
The article shows how to detect someone who can take orders and get it done from people who say "I know how to code in Java!" and then are puzzled in front of the javadoc as soon as they see an annotation or a generic. Finding a good -developer- is a whole other beast, as in the current age of software development, its almost impossible. The good ones are already taken.
Dont worrie about it. In general, if the W3C made something, it, by definition, will suck, even if its fully implemented by a bunch of vendors. So what you do is wait for the development tools that abstract it, and use that instead.
W3C: Making over-engineered pieces of trash in an attempt to handle every single darn scenario in existance (instead of using the right tools for the right job) since 19...well, since ever.
Re:Managers have an Oracle bias even in small shop
on
Sun Buys MySQL
·
· Score: 1
To answer my own question, from the look of it, the status is the same as in.NET: the first party drivers are incomplete, so you need third party ones... At least there are a lot of choices I guess.
Re:Managers have an Oracle bias even in small shop
on
Sun Buys MySQL
·
· Score: 1
Well, for small companies and smaller applications, Oracle can also be obtained for free... And while you may not need the features at first (uncommon...standard date handling anyone? was that fixed yet?), when you actually do, its a heck of a lot easier to just "use them" than to have to switch to a more complete system.
If MySQL is enough for your application, you shouldn't even be using Java in the first place. That being said, does MySQL even have a -full- implementation of the latest JDBC specs? If no, (and I'm talking -full- implementation, not partial like it has for ADO.NET, unless you shell out for third party), then its a show stopper right there.
Except that in general, Java is used for backend enterprise solutions on top of the J2EE stack, including business intelligence and such... so you're talking of people who think that even Oracle's features "aren't enough yet". I doubt they'd go for a "lighter" one just because of something like this.
Postgres is already maintained by third party companies and such... if it was to be bought, a -few- forks would probably still live on. Though, MySQL's copyright was assigned to a single company of its own...is Postgres the same? I'd guess it would be different.
I don't know about Sweden, or hell, the US for that matter...but I'd be thinking that if (extreme example incoming!) you were, let say, running a slavery organisation, then move it out of the country, you'd still be liable for what you did BEFORE you moved it out. Unless of course you moved YOURSELF out too, in a country that won't turn you over or doesn't have any kind of treaty along those lines.
My point was that the Open Office install wasn't even worth considering, because the license costs of Office are insignificant.
For the rest, I'll leave it at: new Office versions don't require retrainings (the UI may change, but the stuff you'd actually TRAIN someone on doesn't, as opposed to switching to Open Office: the UI is similar, but the core changes).
And you probably missed the memo, but usually companies don't buy "licenses" of MS products. Its subscription based, which is why its insignificant: you can keep the money invested until its time to shell out, little by little. Doesn't make a big difference when its your spending acount with 500$ in it, but it does when its a large company that invests considerable amount of money. It also comes with full constant support, so YES they'll help yout out with integration and all that jazz, and even send one of their engineers over if you really need help.
Not counting the fact that money spent on software is usually deductable as business spendings. (So is training though, so thats not a point in favor of one or the other, it just shows how little the end cost is in the end).
Now, I'm sounding like I'm advertising for MS: I'm not. I'm just trying to point out how insignificant license fees are compared to the big picture (thats why they actually sell at all...)
For your second point, over time, you're right, no. But on a one shot deal, the cost of training employes is insane. Group seminars are stupidly expensive, plus you have to pay the employes during that time too... that makes enterprise licenses of Office virtually insignificant.
Especially for extremely large corporations, who have "unlimited" license packages... that is, they pay a flat amount for the entire organisation, regardless of how many licenses you use. In those scenarios, you get a new employe, you install Office on the machine, no question asked. If you're using anything else, you need to pay to train em.
I worked for a while for such a large corporation (without naming it, let just say, top 10 in revenue in the world). Now, it wasn't related to Office, but for most of the employes, learning ANYTHING would involve such a seminar, or hiring a third party training to come to the office, or something. Personally I'd think these people should be fired (so its probably a good thing I'm not in charge of HR), especially since this was in IT, and I don't see how you can be a good software developer if you're unwilling to learn new things on your own, but whatever.
Everytime we wanted to introduce something new, it costed hundreds of thousands of dollars. The training budget for a year could pay for every single software license used in the company (including expensive stuff like SAP and various ERP packages...yes, we had a lot of em, with 7-8 digit contracts on each) for half a decade.
Now, if you're in a 10-30 person company and you can hire a trainer for the group for less than 1-2 grands, and you're not in the software undustry (and thus MSDN Premium isn't an efficient option), then definately go with the training, and tell MS to kiss your shiny butt. Currently, I work for a small software firm around that size, and compared to the rest of the cost of doing business, switching from Office to OO.o (and we barely use Office) isn't even worth, money wise, the time it takes to install it.
Don't know about others but me I just use a DateTime field and stick a date object in it, and let the drivers handle the conversation... Now -that- to be seems the obvious best way to do it... Why convert at all, unless someone's using an archaic and incomplete RDBMS.
Office 2003 is supported now. So's 2008. So there's your transition period. Oh, you mean in the same product? Do it like you would any other enterprise application: do a parallel deployement. Once thats done, phase it out.
All serious MS devs knew about this event years ago, no news there.
It is rare that I agree with Slashdot articles on such things, but on this, even the most pro-Microsoft zealot cannot really disagree... everybody wins if these specs are released... They're no more supported, don't compete with Microsoft's newer formats, and would -heavily- show all the entities investigating Microsoft's monopoly that they can "do the right thing".
It would also be a superb PR move (even though they don't deserve the publicity for something they should have done on their own long ago): it would reassure clueless CEOs. "See?? We can use closed source software, because once Microsoft doesn't support it, they'll just open it up!!!". It is far from true, but enough would think that way to make it worth it.
Of course, let say instead of buying from OEM I buy Windows on a CD... I don't have a browser yet. How am I going to go and download one exactly? Windows update? I hardly find that a good solution.
I fully understand your point, but its simply not viable. Programming frameworks and server software not being bundled? They don't sell too many Windows Server 2003 through OEMs... so then you're going to have to install a freagin barebone Windows Server, and go one by one installing all the tools to make a network?
That reminds me of compressed files... having to go and download all the darn softwares if someone dare sending me a rar or ace file.
If nothing else, I'm wishing they'd bundle MORE stuff without having to go through Dell or whatsnot. Want it or not, an operating system, almost by definition in 2008, is a framework of tools to exploit the hardware. If a default install for any OS doesn't include a media player, a browser, networking stacks, as many runtimes as possible, basic text editors, etc, its incomplete, seriously.
I understand that the law doesn't care about that...in which case I feel the law has to be changed or interpreted differently in this case. If MS started bundling Office (themselves, I know Dell and such do) or Exchange, we'd be getting somewhere, yes. But they don't. The only thing i find important is that MS has to make darn sure that OEMs -can- bundle other stuff, and that the APIs are open, and I realise MS hasn't been perfect in this regard, far from it, and they need to improve there. If I type a web address in the explorer, Firefox will pop up if its my default browser. No one even has to know that IE is there. Java programs can run by simply being double clicked on, even though they're not exes. All of the "casual" PC users I've seen install iTunes on any computer they get their hands on (including office computers). FastCGI is up and kicking for IIS, if you don't like ASP.NET.
As long as everything can be replaced and overriden... I don't see the problem. Stripping basic stuff away will just hurt the people who actually LIKE Windows (yes, there are some!).
To your comparison to Apple and such, you'll get a bunch of "When you're a conficted illegal monopolist, the rules change" answers".
That being said, there are just some things that if Microsoft don't buddle, they simply don't have a product anymore. I mean come on, an OS without built in media player or browser? Lol. And the.NET Framework is the main door to interop on Windows, since it allows third parties to more easily build on Windows...
IE sucking balls is the only freagin reason Firefox or Opera have been able to pick up...if it was NOT bundled, but it actually rocked (I know, its hard to imagine), Firefox and Opera would never have been able to gain any Windows market share...
All this garbage is pushing it. Its definately a GOOD thing to keep Microsoft in check, but push it too far and you start hurting the customers. For what? To help other companies, which is just moving the problem around: customers still lose.
Personally, I kind of feel that the contribution to web development is overstated... Rails really doesn't do anything special... its the same paradigms that "serious" developers have known and been using forever. The only big difference is that usually in enterprise development, you'll try and avoid heavily connected "4th Gen" style development, while Rail encourages it, but thats about it...
An MVC Type 2 with a templating language, an OR Mapper, and a messenging system, on top of a 4th Gen RAD environment... There's a couple of things here and there (I recently had to implement an equivalent to ActiveRecord::Migration in C# because all the C# equivalents I could find were not production ready IMO...didn't take long though, since I didn't need platform independence), but overall its stuff thats only really new to PHP developers and hobbyists. Sure, you see literal clones popping up in Java and.NET, but its not like it all didn't exist. The keywords and method names were just different, now they're the same.
As a sidenote, schools are good to teach the science part, but for actual development techniques, they tend to suck. The plural vs singular for table name is a huge debate in the database field, and its far from set in stone. For everything else, chances are, if you learned it in school when it comes to software engineering and development techniques, its wrong. Not always, but its a good general rule to go by.
Slashdot Mirror
Microsoft had to add all the security stuff. There's a reason DOS was so fast...no checking whatsoever anywhere. So obviously now this would be slower. Unixes did it already, so you can improve on performance there. But Vista -is- indeed the result of extremely poor management decisions, the same kind of constraints that produced the RRODing Xbox360s... and i expect most products of Microsoft that were thought of during that period to have similar issues. Next batch should be better.
That being said, your computer is about the same than the one i run Vista, and its snappy. People saying you need 4 gigs of ram and a dual core for it are in their world.
The amusing thing is that then they will accuse MS of spreading fud when saying things half as innacurate as this :) And I mean, MS does spread lots of fud...but Slashdot has em beat any day of the week, easy =P
It won't be long... I mean, SSIS's script components already use VB.NET (and in the next version can use more languages), so the scaffold is already there.
Of course, dunno about others, but even my Vista install on a not-so-new computer boots almost instantly...ONCE the boot loader kicks in.
Its the RAID, the videocard, the DVD-ROMS, and the bios itself that takes most of the time when im booting (especially the raid). Once the hardware junk is done, and Windows actually -tries- loading, I'm about 2 and a half seconds away from my login screen, and from there another 2-3 seconds from being able to load Internet Explorer (yeah yeah I know).
There's misinformation on both sides. The amount of people who think that your right to a backup also includes making "backups" JUST IN CASE your nearest Blockbuster got broken into and had everything stolen, so you can be the hero and give em back copies of what they once had (/sarcasm) is pretty staggering, too.
I agree with you, though one thing: One has to make the difference between a programmer and a developer. Those are 2 distinct jobs, with a fairly clear cut difference between the two.
A programmer is a "technology specialist", one that knows how to use his hammer to do what he's told. A developer is one that knows what the hell to do with a hammer... When I said the article was good, it was to find a good programmer. Because of the nature of the education system, a typical software development team is made of a lot of people who know how to code algorithms, state machines, and debug a Linux kernel, and 1, maybe 2 people who actually understand the relationship between business processes, requirements, and that code...
It really should be the other way around (Teams would work a lot better if everyone knew how to develop good software and understood the business, and 1-2 people handled the more obscure, scientific or technology intensive part of things, but thats now how universities decided it would be....much easier to train professors to teach things that never change I guess).
The article shows how to detect someone who can take orders and get it done from people who say "I know how to code in Java!" and then are puzzled in front of the javadoc as soon as they see an annotation or a generic. Finding a good -developer- is a whole other beast, as in the current age of software development, its almost impossible. The good ones are already taken.
Dont worrie about it. In general, if the W3C made something, it, by definition, will suck, even if its fully implemented by a bunch of vendors. So what you do is wait for the development tools that abstract it, and use that instead.
W3C: Making over-engineered pieces of trash in an attempt to handle every single darn scenario in existance (instead of using the right tools for the right job) since 19...well, since ever.
To answer my own question, from the look of it, the status is the same as in .NET: the first party drivers are incomplete, so you need third party ones... At least there are a lot of choices I guess.
Well, for small companies and smaller applications, Oracle can also be obtained for free... And while you may not need the features at first (uncommon...standard date handling anyone? was that fixed yet?), when you actually do, its a heck of a lot easier to just "use them" than to have to switch to a more complete system.
If MySQL is enough for your application, you shouldn't even be using Java in the first place. That being said, does MySQL even have a -full- implementation of the latest JDBC specs? If no, (and I'm talking -full- implementation, not partial like it has for ADO.NET, unless you shell out for third party), then its a show stopper right there.
The thing is, I don't think Firefox is targeting the same people as Vi. So thats definately not an issue here.
Except that in general, Java is used for backend enterprise solutions on top of the J2EE stack, including business intelligence and such... so you're talking of people who think that even Oracle's features "aren't enough yet". I doubt they'd go for a "lighter" one just because of something like this.
Postgres is already maintained by third party companies and such... if it was to be bought, a -few- forks would probably still live on. Though, MySQL's copyright was assigned to a single company of its own...is Postgres the same? I'd guess it would be different.
I don't know about Sweden, or hell, the US for that matter...but I'd be thinking that if (extreme example incoming!) you were, let say, running a slavery organisation, then move it out of the country, you'd still be liable for what you did BEFORE you moved it out. Unless of course you moved YOURSELF out too, in a country that won't turn you over or doesn't have any kind of treaty along those lines.
My point was that the Open Office install wasn't even worth considering, because the license costs of Office are insignificant.
For the rest, I'll leave it at: new Office versions don't require retrainings (the UI may change, but the stuff you'd actually TRAIN someone on doesn't, as opposed to switching to Open Office: the UI is similar, but the core changes).
And you probably missed the memo, but usually companies don't buy "licenses" of MS products. Its subscription based, which is why its insignificant: you can keep the money invested until its time to shell out, little by little. Doesn't make a big difference when its your spending acount with 500$ in it, but it does when its a large company that invests considerable amount of money. It also comes with full constant support, so YES they'll help yout out with integration and all that jazz, and even send one of their engineers over if you really need help.
Not counting the fact that money spent on software is usually deductable as business spendings. (So is training though, so thats not a point in favor of one or the other, it just shows how little the end cost is in the end).
Now, I'm sounding like I'm advertising for MS: I'm not. I'm just trying to point out how insignificant license fees are compared to the big picture (thats why they actually sell at all...)
Write Once, Piss People Off Everywhere?
For your second point, over time, you're right, no. But on a one shot deal, the cost of training employes is insane. Group seminars are stupidly expensive, plus you have to pay the employes during that time too... that makes enterprise licenses of Office virtually insignificant.
Especially for extremely large corporations, who have "unlimited" license packages... that is, they pay a flat amount for the entire organisation, regardless of how many licenses you use. In those scenarios, you get a new employe, you install Office on the machine, no question asked. If you're using anything else, you need to pay to train em.
I worked for a while for such a large corporation (without naming it, let just say, top 10 in revenue in the world). Now, it wasn't related to Office, but for most of the employes, learning ANYTHING would involve such a seminar, or hiring a third party training to come to the office, or something. Personally I'd think these people should be fired (so its probably a good thing I'm not in charge of HR), especially since this was in IT, and I don't see how you can be a good software developer if you're unwilling to learn new things on your own, but whatever.
Everytime we wanted to introduce something new, it costed hundreds of thousands of dollars. The training budget for a year could pay for every single software license used in the company (including expensive stuff like SAP and various ERP packages...yes, we had a lot of em, with 7-8 digit contracts on each) for half a decade.
Now, if you're in a 10-30 person company and you can hire a trainer for the group for less than 1-2 grands, and you're not in the software undustry (and thus MSDN Premium isn't an efficient option), then definately go with the training, and tell MS to kiss your shiny butt. Currently, I work for a small software firm around that size, and compared to the rest of the cost of doing business, switching from Office to OO.o (and we barely use Office) isn't even worth, money wise, the time it takes to install it.
Don't know about others but me I just use a DateTime field and stick a date object in it, and let the drivers handle the conversation... Now -that- to be seems the obvious best way to do it... Why convert at all, unless someone's using an archaic and incomplete RDBMS.
Office 2003 is supported now. So's 2008. So there's your transition period. Oh, you mean in the same product? Do it like you would any other enterprise application: do a parallel deployement. Once thats done, phase it out.
All serious MS devs knew about this event years ago, no news there.
Well, you see...Microsoft has been BRIBING standard bodies to convince them that their newer formats were easy to remake (not even reverse engineer).
So I doubt that would be a bad thing.
It is rare that I agree with Slashdot articles on such things, but on this, even the most pro-Microsoft zealot cannot really disagree... everybody wins if these specs are released... They're no more supported, don't compete with Microsoft's newer formats, and would -heavily- show all the entities investigating Microsoft's monopoly that they can "do the right thing".
It would also be a superb PR move (even though they don't deserve the publicity for something they should have done on their own long ago): it would reassure clueless CEOs. "See?? We can use closed source software, because once Microsoft doesn't support it, they'll just open it up!!!". It is far from true, but enough would think that way to make it worth it.
So come on MS, do it.
Oups, forgot to switch to plain ol text... hope the wall of text isn't TOO bad.
Of course, let say instead of buying from OEM I buy Windows on a CD... I don't have a browser yet. How am I going to go and download one exactly? Windows update? I hardly find that a good solution. I fully understand your point, but its simply not viable. Programming frameworks and server software not being bundled? They don't sell too many Windows Server 2003 through OEMs... so then you're going to have to install a freagin barebone Windows Server, and go one by one installing all the tools to make a network? That reminds me of compressed files... having to go and download all the darn softwares if someone dare sending me a rar or ace file. If nothing else, I'm wishing they'd bundle MORE stuff without having to go through Dell or whatsnot. Want it or not, an operating system, almost by definition in 2008, is a framework of tools to exploit the hardware. If a default install for any OS doesn't include a media player, a browser, networking stacks, as many runtimes as possible, basic text editors, etc, its incomplete, seriously. I understand that the law doesn't care about that...in which case I feel the law has to be changed or interpreted differently in this case. If MS started bundling Office (themselves, I know Dell and such do) or Exchange, we'd be getting somewhere, yes. But they don't. The only thing i find important is that MS has to make darn sure that OEMs -can- bundle other stuff, and that the APIs are open, and I realise MS hasn't been perfect in this regard, far from it, and they need to improve there. If I type a web address in the explorer, Firefox will pop up if its my default browser. No one even has to know that IE is there. Java programs can run by simply being double clicked on, even though they're not exes. All of the "casual" PC users I've seen install iTunes on any computer they get their hands on (including office computers). FastCGI is up and kicking for IIS, if you don't like ASP.NET. As long as everything can be replaced and overriden... I don't see the problem. Stripping basic stuff away will just hurt the people who actually LIKE Windows (yes, there are some!).
To your comparison to Apple and such, you'll get a bunch of "When you're a conficted illegal monopolist, the rules change" answers".
.NET Framework is the main door to interop on Windows, since it allows third parties to more easily build on Windows...
That being said, there are just some things that if Microsoft don't buddle, they simply don't have a product anymore. I mean come on, an OS without built in media player or browser? Lol. And the
IE sucking balls is the only freagin reason Firefox or Opera have been able to pick up...if it was NOT bundled, but it actually rocked (I know, its hard to imagine), Firefox and Opera would never have been able to gain any Windows market share...
All this garbage is pushing it. Its definately a GOOD thing to keep Microsoft in check, but push it too far and you start hurting the customers. For what? To help other companies, which is just moving the problem around: customers still lose.
Personally, I kind of feel that the contribution to web development is overstated... Rails really doesn't do anything special... its the same paradigms that "serious" developers have known and been using forever. The only big difference is that usually in enterprise development, you'll try and avoid heavily connected "4th Gen" style development, while Rail encourages it, but thats about it...
.NET, but its not like it all didn't exist. The keywords and method names were just different, now they're the same.
An MVC Type 2 with a templating language, an OR Mapper, and a messenging system, on top of a 4th Gen RAD environment... There's a couple of things here and there (I recently had to implement an equivalent to ActiveRecord::Migration in C# because all the C# equivalents I could find were not production ready IMO...didn't take long though, since I didn't need platform independence), but overall its stuff thats only really new to PHP developers and hobbyists. Sure, you see literal clones popping up in Java and
As a sidenote, schools are good to teach the science part, but for actual development techniques, they tend to suck. The plural vs singular for table name is a huge debate in the database field, and its far from set in stone. For everything else, chances are, if you learned it in school when it comes to software engineering and development techniques, its wrong. Not always, but its a good general rule to go by.