>
object persistence frameworks [apache.org] are really cool! Get all the benefits of your relational database, but use only [Java] code to access it! SQL is great for some things (generating reports, data mining, etc), but often in an application you just want to map your objects to the DBs
That is the problem. You see data as subordinate to some application. What every company soon discovers, and most power users eventually, is that you want "direct" (meaning, with a powerful, interactive, not limited, declarative data language) access to data not limited by a hierarchical or network data model such as OO.
That you think a persistence library can give you the benefits of a relational database only shows you never understood even the basics of relational theory and practice. Suggestion: read all you can from Date, Darwen, McGoveran and Fabian Pascal.
There is one big problem with your assumptions. You assume Iraq's tyrant was rational and wanted the best. Let me remember you no one, not even Muslim Arabs, trusts Muslim Arab politicians. Words in that culture aren't expected to have the same relationship to reality as in the Western world, even in this age of decadence. The good of the people isn't even a concept for them; fairness is just a cover for agression and self-victimization.
>
All type Pure OODB, ORDB, RDBMS all will have there place.
OO databases don't quite exist... they are more of OO repositories or application-specific DB construction kits. They don't have the flexibility and capabilities of a RDBMS like Alphora Dataphor -- remember, Oracle, DB2 and the such are SQL, not Relational DBMSs. In fact, they are a throwback to 30 years ago when all we had was CODASYL, IMS/DB and the like.
ORDBMSs are ugly -- too complex, too big, and sub-relational at best, being based on SQL. A true RDBMS such as Dataphor can do all OO needs with much less hassle.
>
SDL (Specification and Description Language) as a tool for designing and auto-coding
This is very nice, until you need to deal with data. When you need typing, rules and the such, you need a real programming language with support for the relational model.
Pictorial, as well as natural languages, will sure be interesting -- but then they need to be coded in something, and they are inherently limited.
Why settle for the 3rd? Date just defined a 6th, that is relevant to temporal databases.
>
RDBMS schemas are tightly coupled to the individual database application that uses them
Not at all... SQL yes, but a relational schema provides a really data-independent database to use by any number of different applications. On the contrary, XML and OODBs lack separation of physical and logical layers, thus entailing the coupling you talked about.
>
can't really exist without, well, an RDBMS. Both properties hamper system integration issues.
Not at all... all systems need a RDBMS. That we have only SQL, and usually non-standard at that, does make things difficult in the short term. But if we don't start demanding real RDBMSs from vendors, we will never get them.
Refactoring does not depend on Eclipse: Emacs!
on
Eclipse 2.1 Released
·
· Score: 2, Informative
>
developing under your favorite text editor feels like comparing Eclipse to the dinosaur age - I can't live without refactoring now
>
XML decouples data from platforms and that makes integration easier and saves big bucks.
Yes, XML is nice as a new, enriched Unicode. But it does nothing to make data saner, as the relational model does. This is done by the DTDs, which would be much better done as shared relational data schemas.
The problem is reliability. GNU/Linux clusters are still mostly about performance. Once they get to the level of the VMS clusters, then they will have a reliability solution. Remember, it is not about having three years uptime only; it is about surviving high loads, hardware failures, kernel patching etc.
After we have reliability, there is management. There is a great structure provided by IBM, ISVs and IT old-timers to support these systems. SysAnalysts who really document, security people that really control access, production analysts that really know what is running and what touches which files when to do what.
The fast and loose approach taken by understaffed IT departments supporting GNU/Linux simply will not do. Litmus test: to which of the aforementioned categories do Unix SysAdmins belong? Answer: all of them, as there is not staff enough nor management enough to properly manage systems.
>
If you're working with data that can be meaningfully represented with columns, you're using the wrong damned tool.
All data can be represented in the relational model. Only that it the relational model uses attributes, not columns. Columns are a corruption of the model.
>
XML is for complex structured data, which it does fine.
Yes, it does finely send us back thirty-something years to hierarchical data.
>
It is not for tables.
See above. The relational model is about relations and relvars, not tables. Tables are a corruption. See the pattern? SQL is evil, relational is good.
>
blame the idiot who thought that XML was a good way to do DBs.
Agreed. But not complex structured data should be in relational (not SQL) databases too.
RISC is simple. RISC is a processor architecture, not a type of OS. Modern OSs run on both RISC and CISC processor, or at least are not tied to any particular architecture. X-Box is CISC; that is why it is noisy, big, and uses far too much electrical energy. Some complexity is unavoidable, and an OS can be complex internatlly yet present a simple programming interface: that is the case of the PS/2 but not of the X-Box; the same applies to user interface, in both cases.
>
without accusations from each side that the other is going to hell. Of course, you might still think that
I do not believe doctrinal correctness saves. I do believe over-reliance on doctrine can endanger one, as well as too bad doctrine. But what saves is the Grace of God, in Jesus the Christ, thru faith.
>
what prompter you to read the whole Book of Mormon if you found it so rediculous?
Well, I have to confess to having skipped some parts in the latter half of it...
Anyway, the initial parts sound interesting if read as fiction. Later the moralising, indoctrination and stretching, as well as the pretence to veracity and authority, become tiresome. Also I was strandled in a hotel in Ciudad Guatemala with little else to do; and thought I would read it even if I had already read enough about it. That was when I found that, besides being false, it is also ridiculous, and that is a feeling later Orson Scott Card writing also gives me.
>
pretty rare that you can find a source that dares to examine both sides seriously
What if one side never deserved being taken seriously?
Yes, I read the book of Mormon, and some other LDS materials. But I have to confess I find them too ridiculous, no better than Mohammad and his Koran. Actually a little bit more ridiculous than the Koran, because we are near enough the author in time, culture and space to know what he really was, and where he did get his ideas.
All in all I find it so tiresome now that the only thing that could raise me to the effort would be trying to get someone from entering into, or to escape from, this cult.
Now the Bible... so refreshing it cannot be read too often ever...
Not a book, a speech that made it to the web. Google for you. If you mean the carichatural depictions of Protestants, it is the Ender one in a Brazilian planet -- Orson was a missionary in Brazil.
As I said, looking for a job, will leave homework for you.
Now there are much more than just this OUP books around. One very interesting booklet by http://pfo.org./ is one century old and shows where Joseph Smith picked his themes. Nothing that one Muhammed had not done 1300 years before.
On a related note, what about Orson Scott Card and the Ender books?
He has some really, truly good ideas, but sometimes is just sick -- as in the Protestant planet where the Calvinists are cruel legalists, but the Lutherans are nice, simple people, and Romanists are deeply wise; presumably because Calvinistic idea of grace goes against the grain of Mormon Pelagianism (salvation by works), while Lutheranism in its modern, liberal form, and specially Romanism are more congenial to the Mormon mind.
Worse, he has a speech published in the web defending the Book of Mormon, shamelessly lifted from CS Lewis and his literary argument for the authenticity of Scripture. Just that Orson fails to take in account most of the literary arguments against the Book of Mormon. Moreover, reading the CS Lewis original one sees the argument actually implicates a big contrast between the Bible and the Book of Mormon.
Sorry, too tired, and looking for a job; no time to look for references.
I used NCD WinCenter in conjunction with MS Windows NT 3.51, later Citrix bought if off and renamed it UIS, Unix Integration Services or something the like. Had MS Office displaying nicely in my 21-inches Trinitron Sun Solaris SPARC system. Wonderful!
Perhaps you can find them in some abandonware site?
>
Technically, they are like bastard brothers, usually equally hated by UNIX Lovers.
I am more on the side of Lisp, free software and open standards, than of Unix. But I am not comparing IBM OS/2 to GNU, I am comparing to MS.
>
NT at least had file permissions and multiuser.
So had IBM OS/2 with HFS/386, and using far less system resources. BTW, MS-WNT file permissions are too complex, and its multiuser support (Terminal Server) is still substandard. All in all, OS/2 would have been a better stepping stone for the current and future migration to POSIX than MS-WNT.
>
By the time the big OS/2 push started most of the cool kids were already running Linux.
I didn't have the technical knowledge then necessary.
>
IBM tried to bury the i386 because they were worried about their midrange.
References, please?
>
IBM also heavly marketed OS/2 as part of a Mainframe integration strategy, with it's special "Extended Edition", etc.
What was quite wise.
>
their Unix and Novell machines were second-class citizens in IBM's world -- Which they were.
Not exactly, IBM having their own Unix and even their own, blue-packaged Netware.
>
Windows NT shipped out-of-box with TCP/IP, IPX/SPX, AppleTalk, DECNet, you name it. IBM OS/2 2.0 shipped out of box with jackshit, and then made you pay for NetBEUI and pay some more (a lot more) for TCP/IP.
Yes, price & packaging were horrible at first. Now they are better, but with GNU/Linux around, no one is paying attention anymore, and in a way we are better off.
>
Not to mention that IBM themselves FUDded the idea of OS/2 on non-PS/2 hardware...
Not exactly FUD, since driver availability and quality being so sketchy -- and that, at least in part, MS responsibility too.
>
OS/2's technical marketing was a clusterfuck from day 1
Agreed. But I was arguing not the badness of IBM marketing, which is a given; I was arguing OS/2 technical superiority over MS-Win and MS responsibility on killing it thru betrayal.
>
OS/2 2.1 was horribly crashy on an IBM PS/Valuepoint 486 DX/2 66 with 16mb
I wonder why so many differing experiences? Perhaps I should not be surprised, I knew some people claiming they had unstable installations of MS-WNT and I myself have a crashy installation of a X Window server.
>
Too bad it doesn't have support for modern technologies like AGP, USB, ISA PnP, et cetera.
ISA PnP was never needed, as it do not have the horrible MS-W16 problems with resource allocation. Remember even Linus wanted nothing to do with ISA PnP. But even so you are mistaken, because OS/2 was updated for at least USB. Do your homework check, please.
>
IBM didn't need Microsoft to stab OS/2 in the back: Big Blue did a perfectly good job of that itself.
In part, yes. But MS-OS/2 6.0 probably would have been a better OS than MS-WXP.
>
You never tried running DOS apps in OS/2 1.0, did you.
You mean the DOS torture box? No one expected it to work. The worse it did was preparing people to accept MS-Win.
>
As long as you weren't waiting for applications which ran on your shiny new requires-4MB-of-RAM-costing-to-boot OS/2?
I could not quite make your phrase, but I assume you are talking about performance and resources requirement. I was speaking about the speed at which OS/2 itself was developed. But even so, OS/2 proved to run better and in less memory than MS-WNT.
>
the marketplace didn't want OS/2: even when Microsoft were fully committed to it (the period thru' mid-89) and pushing it as the best thing since sliced bread
That is simply not true. MS attempts at marketing OS/2 were feeble at best. It wrote many drivers for MS-W16, but OS/2 was starving for them. It never wrote a graphical Word for OS/2, and killed all apps for OS/2 as soon as MS-W3.1 was out of the barn. It kept promoting MS-Win as needing less resources than OS/2, but that was only true for the then-current MS-W16 systems.
>
blame the clone makers who had no interest in handing control of the PC marketplace back to IBM
Agreed, but here the clone makers were not at a fault, but IBM pushing MCA as a proprietary standard. Nothing to do with OS/2, but generated lots of ill-will towards IBM, as did bad memories from mainframe times. Not that MS did not became as bad as IBM ever was, and actually more capable of escaping justice.
>
targetting the 286 was a bad idea from the get go, but IBM didn't want to eat into sales of their (expensive) AS/400 line.
If you call the AS/400 expensive or compare it to PCs you never went near one. At any rate, never understood one. The real reason was that i386s were expensive, and IBM ones even more so.
>...as part of SAA - so it would have interoperated with... ooh, let's see... that'd be the IBM AS/400 minicomputer and the IBM S/370 mainframe and... do we all get the idea here?
Are you malicious or just ignorant? Obviously SAA was part of the picture, as it was also for MS which I remember pushing MS interoperability with SNA. But IBM OS/2 also has interoperability with Unix and the Net, and a better POSIX compatibility history than MS-WNT. And that comes as no surprise, since IBM has its own POSIX system, AIX. Or do you think IBM could pull such a compelling GNU/Linux story had not it POSIX experience before?
>
As long as you were prepared to fork over $3000 for said documentation
What is actually quite reasonable for print documentation for an ISV or third-party developer. Remember, this was before CD-ROMs and the Net. Even today, MS documentation is no better, and full of holes.
>
you never used OS/2, did you
Yes, I did: versions 1.3, 2.0, 2.1 and 3. It was rock solid, provided you had enough system resources. Many people tried to run it with the minimal resources, which were enough only for booting. Also, drivers were an issue, but here MS also played MS-Win in violation of its promises to customers and partners.
>
a customer at the time had about 500 PS/2 Model 60s running OS/2 1.3 w/ Presentation Manager
Sad history, but makes me wonder... did they have enough system resources? Were apps native? This were i286 times, remember. Only Xenix ran better for those loads on a i286.
>
Microsoft wanted the application with focus to get most of the CPU and IBM wanted a tight kernel so that all apps got an appropriate amount of CPU time. It took almost 10 years before Microsoft came out with an OS( W2K ) which had decent multitasking for the DESKTOP.
Not quite. MS-W2K is simply OS/2 3.0 NT with MS-W95's GUI & API. The reason MS-WNT was so bad were the drivers and the libraries, not the kernel. MS-WME and antecessors are simply direct descendants of a dumbed-down clone of CP/M.
Slashdot Mirror
Yes, you are.
I singled out Arab Muslim politicians. While one can argue no politicians whatsoever can be trusted, I blamed cultural features to single them out.
That is the problem. You see data as subordinate to some application. What every company soon discovers, and most power users eventually, is that you want "direct" (meaning, with a powerful, interactive, not limited, declarative data language) access to data not limited by a hierarchical or network data model such as OO.
That you think a persistence library can give you the benefits of a relational database only shows you never understood even the basics of relational theory and practice. Suggestion: read all you can from Date, Darwen, McGoveran and Fabian Pascal.
There is one big problem with your assumptions. You assume Iraq's tyrant was rational and wanted the best. Let me remember you no one, not even Muslim Arabs, trusts Muslim Arab politicians. Words in that culture aren't expected to have the same relationship to reality as in the Western world, even in this age of decadence. The good of the people isn't even a concept for them; fairness is just a cover for agression and self-victimization.
OO databases don't quite exist... they are more of OO repositories or application-specific DB construction kits. They don't have the flexibility and capabilities of a RDBMS like Alphora Dataphor -- remember, Oracle, DB2 and the such are SQL, not Relational DBMSs. In fact, they are a throwback to 30 years ago when all we had was CODASYL, IMS/DB and the like.
ORDBMSs are ugly -- too complex, too big, and sub-relational at best, being based on SQL. A true RDBMS such as Dataphor can do all OO needs with much less hassle.
Why settle for the 3rd? Date just defined a 6th, that is relevant to temporal databases.
Not at all... SQL yes, but a relational schema provides a really data-independent database to use by any number of different applications. On the contrary, XML and OODBs lack separation of physical and logical layers, thus entailing the coupling you talked about.
Not at all... all systems need a RDBMS. That we have only SQL, and usually non-standard at that, does make things difficult in the short term. But if we don't start demanding real RDBMSs from vendors, we will never get them.
You can have a better IDE doing refactoring as well.
So sucking money from a struggling, devastated country to a lock-in technology is his work? I'm glad I am unemployed.
Yes, XML is nice as a new, enriched Unicode. But it does nothing to make data saner, as the relational model does. This is done by the DTDs, which would be much better done as shared relational data schemas.
...because the problem is not performance.
The problem is reliability. GNU/Linux clusters are still mostly about performance. Once they get to the level of the VMS clusters, then they will have a reliability solution. Remember, it is not about having three years uptime only; it is about surviving high loads, hardware failures, kernel patching etc.
After we have reliability, there is management. There is a great structure provided by IBM, ISVs and IT old-timers to support these systems. SysAnalysts who really document, security people that really control access, production analysts that really know what is running and what touches which files when to do what.
The fast and loose approach taken by understaffed IT departments supporting GNU/Linux simply will not do. Litmus test: to which of the aforementioned categories do Unix SysAdmins belong? Answer: all of them, as there is not staff enough nor management enough to properly manage systems.
All data can be represented in the relational model. Only that it the relational model uses attributes, not columns. Columns are a corruption of the model.
Yes, it does finely send us back thirty-something years to hierarchical data.
See above. The relational model is about relations and relvars, not tables. Tables are a corruption. See the pattern? SQL is evil, relational is good.
Agreed. But not complex structured data should be in relational (not SQL) databases too.
RISC is simple. RISC is a processor architecture, not a type of OS. Modern OSs run on both RISC and CISC processor, or at least are not tied to any particular architecture. X-Box is CISC; that is why it is noisy, big, and uses far too much electrical energy. Some complexity is unavoidable, and an OS can be complex internatlly yet present a simple programming interface: that is the case of the PS/2 but not of the X-Box; the same applies to user interface, in both cases.
Ergo, you do not know what you are talking about.
I do not believe doctrinal correctness saves. I do believe over-reliance on doctrine can endanger one, as well as too bad doctrine. But what saves is the Grace of God, in Jesus the Christ, thru faith.
Well, I have to confess to having skipped some parts in the latter half of it...
Anyway, the initial parts sound interesting if read as fiction. Later the moralising, indoctrination and stretching, as well as the pretence to veracity and authority, become tiresome. Also I was strandled in a hotel in Ciudad Guatemala with little else to do; and thought I would read it even if I had already read enough about it. That was when I found that, besides being false, it is also ridiculous, and that is a feeling later Orson Scott Card writing also gives me.
What if one side never deserved being taken seriously?
Yes, I read the book of Mormon, and some other LDS materials. But I have to confess I find them too ridiculous, no better than Mohammad and his Koran. Actually a little bit more ridiculous than the Koran, because we are near enough the author in time, culture and space to know what he really was, and where he did get his ideas.
All in all I find it so tiresome now that the only thing that could raise me to the effort would be trying to get someone from entering into, or to escape from, this cult.
Now the Bible... so refreshing it cannot be read too often ever...
Have you read, or do you assume one is biased because he has an opinion to offer?
Me too.
Not a book, a speech that made it to the web. Google for you. If you mean the carichatural depictions of Protestants, it is the Ender one in a Brazilian planet -- Orson was a missionary in Brazil.
As I said, looking for a job, will leave homework for you.
Now there are much more than just this OUP books around. One very interesting booklet by http://pfo.org./ is one century old and shows where Joseph Smith picked his themes. Nothing that one Muhammed had not done 1300 years before.
On a related note, what about Orson Scott Card and the Ender books?
He has some really, truly good ideas, but sometimes is just sick -- as in the Protestant planet where the Calvinists are cruel legalists, but the Lutherans are nice, simple people, and Romanists are deeply wise; presumably because Calvinistic idea of grace goes against the grain of Mormon Pelagianism (salvation by works), while Lutheranism in its modern, liberal form, and specially Romanism are more congenial to the Mormon mind.
Worse, he has a speech published in the web defending the Book of Mormon, shamelessly lifted from CS Lewis and his literary argument for the authenticity of Scripture. Just that Orson fails to take in account most of the literary arguments against the Book of Mormon. Moreover, reading the CS Lewis original one sees the argument actually implicates a big contrast between the Bible and the Book of Mormon.
Sorry, too tired, and looking for a job; no time to look for references.
I used NCD WinCenter in conjunction with MS Windows NT 3.51, later Citrix bought if off and renamed it UIS, Unix Integration Services or something the like. Had MS Office displaying nicely in my 21-inches Trinitron Sun Solaris SPARC system. Wonderful!
Perhaps you can find them in some abandonware site?
I am more on the side of Lisp, free software and open standards, than of Unix. But I am not comparing IBM OS/2 to GNU, I am comparing to MS.
So had IBM OS/2 with HFS/386, and using far less system resources. BTW, MS-WNT file permissions are too complex, and its multiuser support (Terminal Server) is still substandard. All in all, OS/2 would have been a better stepping stone for the current and future migration to POSIX than MS-WNT.
I didn't have the technical knowledge then necessary.
References, please?
What was quite wise.
Not exactly, IBM having their own Unix and even their own, blue-packaged Netware.
Yes, price & packaging were horrible at first. Now they are better, but with GNU/Linux around, no one is paying attention anymore, and in a way we are better off.
Not exactly FUD, since driver availability and quality being so sketchy -- and that, at least in part, MS responsibility too.
Agreed. But I was arguing not the badness of IBM marketing, which is a given; I was arguing OS/2 technical superiority over MS-Win and MS responsibility on killing it thru betrayal.
I wonder why so many differing experiences? Perhaps I should not be surprised, I knew some people claiming they had unstable installations of MS-WNT and I myself have a crashy installation of a X Window server.
ISA PnP was never needed, as it do not have the horrible MS-W16 problems with resource allocation. Remember even Linus wanted nothing to do with ISA PnP. But even so you are mistaken, because OS/2 was updated for at least USB. Do your homework check, please.
In part, yes. But MS-OS/2 6.0 probably would have been a better OS than MS-WXP.
You mean the DOS torture box? No one expected it to work. The worse it did was preparing people to accept MS-Win.
I could not quite make your phrase, but I assume you are talking about performance and resources requirement. I was speaking about the speed at which OS/2 itself was developed. But even so, OS/2 proved to run better and in less memory than MS-WNT.
That is simply not true. MS attempts at marketing OS/2 were feeble at best. It wrote many drivers for MS-W16, but OS/2 was starving for them. It never wrote a graphical Word for OS/2, and killed all apps for OS/2 as soon as MS-W3.1 was out of the barn. It kept promoting MS-Win as needing less resources than OS/2, but that was only true for the then-current MS-W16 systems.
Agreed, but here the clone makers were not at a fault, but IBM pushing MCA as a proprietary standard. Nothing to do with OS/2, but generated lots of ill-will towards IBM, as did bad memories from mainframe times. Not that MS did not became as bad as IBM ever was, and actually more capable of escaping justice.
If you call the AS/400 expensive or compare it to PCs you never went near one. At any rate, never understood one. The real reason was that i386s were expensive, and IBM ones even more so.
Are you malicious or just ignorant? Obviously SAA was part of the picture, as it was also for MS which I remember pushing MS interoperability with SNA. But IBM OS/2 also has interoperability with Unix and the Net, and a better POSIX compatibility history than MS-WNT. And that comes as no surprise, since IBM has its own POSIX system, AIX. Or do you think IBM could pull such a compelling GNU/Linux story had not it POSIX experience before?
What is actually quite reasonable for print documentation for an ISV or third-party developer. Remember, this was before CD-ROMs and the Net. Even today, MS documentation is no better, and full of holes.
Yes, I did: versions 1.3, 2.0, 2.1 and 3. It was rock solid, provided you had enough system resources. Many people tried to run it with the minimal resources, which were enough only for booting. Also, drivers were an issue, but here MS also played MS-Win in violation of its promises to customers and partners.
Sad history, but makes me wonder... did they have enough system resources? Were apps native? This were i286 times, remember. Only Xenix ran better for those loads on a i286.
Yes.
No. OS/2 is still sold and supported both by IBM and third parties. It just is not promoted.
Beware hearsay.
Not quite. MS-W2K is simply OS/2 3.0 NT with MS-W95's GUI & API. The reason MS-WNT was so bad were the drivers and the libraries, not the kernel. MS-WME and antecessors are simply direct descendants of a dumbed-down clone of CP/M.