You mean something like eclipse and jdeveloper. Eclipse uses native platform widgets and basically behaves like every other ide, JDeveloper 10g uses Swing and renders the UI faster than Mozilla if I run both side by side, it is very snappy.
That recently was arund 40 months ago, when 1.4 was released, 1.4 was the first release which used native rendering on Windows for the 2d parts.
The VM became fast around 1.2 but Swing had lots of performance issues until 1.4.
Actually not really, I used to think that lots of the improvements come from the proc, but then I ran a 1.4 jdk on one of those 1999 machines, and guess what, it was fast.
The MySQL business model works following way. They enforce the GPL and if you want to get out of the GPL you have to pay, just like qt, but MySQL has a problem there.
Qt has its own codebase, MySQL does not, they rely on Berkley DB and on InnoDB, obviously there must be some relicensing contract between them so that people can relicense the InnoDB code non GPL, so what if Oracle refuses this relicensing in the future.
MySQL might have a problem bigger than it seems on their hand.
BDB is not the best repo (ask the SVN guys) and InnoDB is now in the hands of Oracle.
Not that I would be sad to see MySQL going the way of the dodo, but this issue is bigger than it seems.
Actually the windows performance of postgres is pretty up to par to the unix performance
you have to tweak the settings like you have to do it in unix (read the various
postgres performance guides).
The performance used to be bad, but not anymore, since the cygwin layer has been dropped.
Actually Ikea is swedish but their main manufactories are in poland due to cost reasons, and the quality varies, lots of their stuff is due to the wood used not very stable and basically breaks down under heavy load, some of their stuff is good though.
General rule though is, you get what you pay for, in case of Ikea often crap, and it is generally cheap but also very often way too expensive for the quality you get.
The comparison to MySQL is valid in this regard, evn in hype dimensions, you can get better solutions than MySQL (Firebird and Postgresql) really for free, the problem is that MySQL constantly hypes their semi crappy stuff while Postgres and Firebird do not have the hype but the quality on their side.
The IKEA of the database market, way to overpriced and the stability is pretty much the same (aka falls apart under heavy load)
The only thing missing is polish programmers and that the customer has to plug the db together them/theirselfs...
The only downside postgresql had at that time was that it was harder to install and no windows version was available at that time, the windows version problem is solved and besides that Postgres is a fast workhorse.
I have yet to see a hosed prostgres repo...
Ok then send the code to the internet, the MySQL requires you to GPL all the code linked to the DB, which means you have to give the code to the public upon request under GPL...
See that as a semi official request now...
Heck, once you do subselects joins etc... and use transactions heavily, postgres beats mysql by a mile....
Ah yes, MySQL is enforced 500 Dollars upfront or GPL, Postgres is BSD...
Ahh and there still is the issue, that the MySQL devs refused to add transactions for years ditto for subselects because you do not need em:-)
thing of the past, but that attitude alone was enough not to touch this system.
Actually no, Lotus 123 was killed because Lotus could not get the windows interna needed to get the product into the market on day 1, while Microsoft used the windows sourcecode for the excel port left and right...
You should read undocumented windows one day.
Face it Microsoft got where it is today, by screwing partners and competitors left and right.
Well you have that on the small scale as well, my ex boss used to try to use the friendship we had for not paying me some missing payments although the company owners have the money (simply he does not but the co owners do)
After one year of trying to be nice and honest to resolve things, I yesterday had to switch to the tough way...:-(
The problem is and I have been screwed numerous times, that if you try to be nice and honest, people in the business world will screw you left and right. Once you find honest bosses, co workers and customers, stick to them like glue, there are too many crooks around who try to screw you at the first occasion if it fits them. In the end you lose, first you are a total mess mentally, because of all the legal fighting going on after such a mess, secondly you have to dump people mentally you trusted and you really put your heart in.
Playing tough but honest and then over time becoming nicer if you can trust them seems to be the better strategy nowadays with all the crooks floating the business world, unfortunately.
Apple was screwed numerous times by Microsoft, the bailout was because Microsoft had an antitrust case on their hands, Adobe was below the radars until now, Microsoft currently is working actively on shooting adobe with competing products out of the market...
This mentality exists until now... That is usual microsoft tactics, the competition has something, then there is the early press release with, Microsoft soon will have it as well, and then you get a hodj podj implementation where the engineer probably was forced to push it in within a week doing 24 hour shifts.
I can remember several of those things. DCOM in its first incarnations was sold totally broken, at least the java bindings failed back then even on the examples.
Sharepoint was such a hodj podf system because there were several document storage vendors making money (gasp the microsoft management does not even want to seem some mid sized companies making money), the thing was totally broken and Rev 1.0 still does not work properly after three patches.
ActiveX was also lousy in its first marketing incarnations when Microsoft wanted to shoot OpenGL out of the market...
This we have to do something against this evil competition mentatlity drives microsoft until now and does not do the products from them any good.
Features usually are marketing features and once the competition is eliminated those features never get fixed until oblivion. The perfect example was the IE, once Microsoft took over the market they did a dreck, to fix the outstanding numerous and often reported issues, until there was a glimpse of competition.
This company is not driven by wanting to get out good products (the engineers in there are for sure) but by pure greed and not letting anyone else have a piece of the pie, by all means.
The funny thing is I have worked for small projects and big projects
and even for projects within Corporations, never have I really encountered a clean well going dev process..
I think corp projects are the worst in this area, there are so many cooks usually, mostly external consultants (which I was part of) being dragged in and the whole ting lacked consistency entirely.
Ahem OpenStep was a joined effort of Sun and Next Systems...
Sun did not follow the path NeXT did and so does apple, the ultimate component systems currently indeed are OSX/OpenStep/GnuStep and surprisingly KDE, which basically are the first to push such a system decently on C++ (Well Taligent did as well, but they failed for other reasons)
Actually the difference is the binding, to sum it up if you do patchwork coding the binding is the key to success, and that is the reason why it works so well on unix but failes constantly on windows.
Using jakarta and other Apache stuff constantly I can sing a song about that too, but I avoided the pitfalls until now.
In Unix environments, everything is a small self supporting unit, you usually plug things together and dont really care how things work, and they work, because they work as a single entity and work as a bigger entity.
The ultimo of this structure are systems like nextstep and kde, which are composed entirely out of self contained components.
This approach however requires that from time to time, you drop bad design in favor of something better and break old code.
Microsoft tried that in the past with Cairo, and fell flat on its face.
So now we have windows, a hodj podj system of marketing ideas half implemented and often lousy implemented (speaking of OLE one of the main reasons why Cairo never could have been pulled off while NextStep and KDE were able to pull such systems off)
We have legacy Win32 code, we have ole woven around that code, we have components which are interwoven over Win32 and other hodj podj system while they shouldnt, we have anything the marketing departement came along the last 20 years plugged in and we have dead old 20 years code which is triggered in the worst cases when nobody expects it anymore.
This works for unix with its self contained entity approach, in windows everything is tightly coupled there are even old 16 bit subsystems which nobody cares about anymore, no real deprecation marking, you can trigger them.
So giving such a system an overhaul ist a real nightmare and probably what Microsoft had to face. It is less a problem of bad coding or bad project management it is more a problem that microsoft in the past always tried to shove the latest hype into the system 'half assed to sum it up'
and then being half implemented changed its directed and shoved the next marketing hype into the system.
It is short of a miracle that the engineers are even able to keep the system up after 20 years of constantly pushing the next fad of the day into it.
Well the original screenshot maker, did not realize that there is a skin settings option in the configuration dialog of the program, there is an entry for macosx as well.
Slashdot Mirror
You mean something like eclipse and jdeveloper. Eclipse uses native platform widgets and basically behaves like every other ide, JDeveloper 10g uses Swing and renders the UI faster than Mozilla if I run both side by side, it is very snappy.
That recently was arund 40 months ago, when 1.4 was released, 1.4 was the first release which used native rendering on Windows for the 2d parts. The VM became fast around 1.2 but Swing had lots of performance issues until 1.4.
Actually not really, I used to think that lots of the improvements come from the proc, but then I ran a 1.4 jdk on one of those 1999 machines, and guess what, it was fast.
Well they wanted to be the IKEA of db development, the users now have to fix InnoDB themselves ... sounds IKEAish :-)
The MySQL business model works following way. They enforce the GPL and if you want to get out of the GPL you have to pay, just like qt, but MySQL has a problem there. Qt has its own codebase, MySQL does not, they rely on Berkley DB and on InnoDB, obviously there must be some relicensing contract between them so that people can relicense the InnoDB code non GPL, so what if Oracle refuses this relicensing in the future. MySQL might have a problem bigger than it seems on their hand. BDB is not the best repo (ask the SVN guys) and InnoDB is now in the hands of Oracle. Not that I would be sad to see MySQL going the way of the dodo, but this issue is bigger than it seems.
Actually the windows performance of postgres is pretty up to par to the unix performance you have to tweak the settings like you have to do it in unix (read the various postgres performance guides). The performance used to be bad, but not anymore, since the cygwin layer has been dropped.
Actually Ikea is swedish but their main manufactories are in poland due to cost reasons, and the quality varies, lots of their stuff is due to the wood used not very stable and basically breaks down under heavy load, some of their stuff is good though. General rule though is, you get what you pay for, in case of Ikea often crap, and it is generally cheap but also very often way too expensive for the quality you get. The comparison to MySQL is valid in this regard, evn in hype dimensions, you can get better solutions than MySQL (Firebird and Postgresql) really for free, the problem is that MySQL constantly hypes their semi crappy stuff while Postgres and Firebird do not have the hype but the quality on their side.
The IKEA of the database market, way to overpriced and the stability is pretty much the same (aka falls apart under heavy load) The only thing missing is polish programmers and that the customer has to plug the db together them/theirselfs...
The only downside postgresql had at that time was that it was harder to install and no windows version was available at that time, the windows version problem is solved and besides that Postgres is a fast workhorse. I have yet to see a hosed prostgres repo...
Ok then send the code to the internet, the MySQL requires you to GPL all the code linked to the DB, which means you have to give the code to the public upon request under GPL... See that as a semi official request now...
Heck, once you do subselects joins etc... and use transactions heavily, postgres beats mysql by a mile.... Ah yes, MySQL is enforced 500 Dollars upfront or GPL, Postgres is BSD...
Ahh and there still is the issue, that the MySQL devs refused to add transactions for years ditto for subselects because you do not need em :-)
thing of the past, but that attitude alone was enough not to touch this system.
Actually no, Lotus 123 was killed because Lotus could not get the windows interna needed to get the product into the market on day 1, while Microsoft used the windows sourcecode for the excel port left and right...
You should read undocumented windows one day.
Face it Microsoft got where it is today, by screwing partners and competitors left and right.
Sure.. the default settings are set towards mem preservation not speed optimization.
Well you have that on the small scale as well, my ex boss used to try to use the friendship we had for not paying me some missing payments although the company owners have the money (simply he does not but the co owners do) After one year of trying to be nice and honest to resolve things, I yesterday had to switch to the tough way... :-(
The problem is and I have been screwed numerous times, that if you try to be nice and honest, people in the business world will screw you left and right. Once you find honest bosses, co workers and customers, stick to them like glue, there are too many crooks around who try to screw you at the first occasion if it fits them. In the end you lose, first you are a total mess mentally, because of all the legal fighting going on after such a mess, secondly you have to dump people mentally you trusted and you really put your heart in.
Playing tough but honest and then over time becoming nicer if you can trust them seems to be the better strategy nowadays with all the crooks floating the business world, unfortunately.
And please do not use postgres in its default settings but in sane ones...
Not even the managers, as it looks the biggest export in the near future from the US will be laywers and lawsuits...
This guy definitely is, if you read between the lines, he wants to raise the prices and wants to have a share of the ipod revenues... sheesh
Apple was screwed numerous times by Microsoft, the bailout was because Microsoft had an antitrust case on their hands, Adobe was below the radars until now, Microsoft currently is working actively on shooting adobe with competing products out of the market...
This mentality exists until now... That is usual microsoft tactics, the competition has something, then there is the early press release with, Microsoft soon will have it as well, and then you get a hodj podj implementation where the engineer probably was forced to push it in within a week doing 24 hour shifts. I can remember several of those things. DCOM in its first incarnations was sold totally broken, at least the java bindings failed back then even on the examples. Sharepoint was such a hodj podf system because there were several document storage vendors making money (gasp the microsoft management does not even want to seem some mid sized companies making money), the thing was totally broken and Rev 1.0 still does not work properly after three patches. ActiveX was also lousy in its first marketing incarnations when Microsoft wanted to shoot OpenGL out of the market... This we have to do something against this evil competition mentatlity drives microsoft until now and does not do the products from them any good. Features usually are marketing features and once the competition is eliminated those features never get fixed until oblivion. The perfect example was the IE, once Microsoft took over the market they did a dreck, to fix the outstanding numerous and often reported issues, until there was a glimpse of competition. This company is not driven by wanting to get out good products (the engineers in there are for sure) but by pure greed and not letting anyone else have a piece of the pie, by all means.
The funny thing is I have worked for small projects and big projects and even for projects within Corporations, never have I really encountered a clean well going dev process.. I think corp projects are the worst in this area, there are so many cooks usually, mostly external consultants (which I was part of) being dragged in and the whole ting lacked consistency entirely.
Ahem OpenStep was a joined effort of Sun and Next Systems... Sun did not follow the path NeXT did and so does apple, the ultimate component systems currently indeed are OSX/OpenStep/GnuStep and surprisingly KDE, which basically are the first to push such a system decently on C++ (Well Taligent did as well, but they failed for other reasons)
Actually the difference is the binding, to sum it up if you do patchwork coding the binding is the key to success, and that is the reason why it works so well on unix but failes constantly on windows. Using jakarta and other Apache stuff constantly I can sing a song about that too, but I avoided the pitfalls until now. In Unix environments, everything is a small self supporting unit, you usually plug things together and dont really care how things work, and they work, because they work as a single entity and work as a bigger entity. The ultimo of this structure are systems like nextstep and kde, which are composed entirely out of self contained components. This approach however requires that from time to time, you drop bad design in favor of something better and break old code. Microsoft tried that in the past with Cairo, and fell flat on its face. So now we have windows, a hodj podj system of marketing ideas half implemented and often lousy implemented (speaking of OLE one of the main reasons why Cairo never could have been pulled off while NextStep and KDE were able to pull such systems off)
We have legacy Win32 code, we have ole woven around that code, we have components which are interwoven over Win32 and other hodj podj system while they shouldnt, we have anything the marketing departement came along the last 20 years plugged in and we have dead old 20 years code which is triggered in the worst cases when nobody expects it anymore.
This works for unix with its self contained entity approach, in windows everything is tightly coupled there are even old 16 bit subsystems which nobody cares about anymore, no real deprecation marking, you can trigger them.
So giving such a system an overhaul ist a real nightmare and probably what Microsoft had to face. It is less a problem of bad coding or bad project management it is more a problem that microsoft in the past always tried to shove the latest hype into the system 'half assed to sum it up' and then being half implemented changed its directed and shoved the next marketing hype into the system. It is short of a miracle that the engineers are even able to keep the system up after 20 years of constantly pushing the next fad of the day into it.
Good place for US Laywers, Chines Prisons, Russian Gulags... perfect disposal areas for the US waste....
Well the original screenshot maker, did not realize that there is a skin settings option in the configuration dialog of the program, there is an entry for macosx as well.