When Codd created the relational model, there wasn't the current Unix filesystem idea... the relational model was always intended to store data, and files are data.
System R, SQL and DB2 prototype, was intended to be the basis for IBM FS.
IBM realised this in OS/400, which being proprietary hasn't the influence it deserves.
MS also wanted Jet to be the building block of its OSs since its inception, that is, sometime before MS Access release.
>
newdocms announced on Slashdot in January 2003
Sorry, but NewDocMS is based on SQLite, which is typeless and but a library... simply not good enough to be attractive. Storage is based on PostgreSQL, the real thing, and aims high.
Before that. MS Access was intended to be a development platform for Jet, and MS Windows would eventually gain a Jet-based filesystem. Only Jet was considered harmful, so MSDE was substituted and the whole thing delayed a decade or so.
This adds nothing, it substitutes the hierarchical store while presenting a compatible interface. It has been proven SQL if correctly modeled gives usually comparable performance to application-specific databases such as a hierarchical filesystem; it also opens the way to a relational system that could significantly better SQL performance while being more powerful and simpler.
That said, the question is how sane will be the database model... for such a fundamental stuff you'd want 6NF.
>
The only that you need is a good and fast btree system
As for implementation, a bTree is OK. But here the goal is semantics: the hierarchical semantics is fundamentally broken, because the world can't be nicely organised that way. Instead, even is SQL is corrupted from the relational model, it allows a much more flexible, powerful, logical interface.
>
creating something like Pro/ENGINEER becomes a multi-billion dollar task.
Not at all... Pro/Engineer may be a multi-billion affair nowadays, but it doesn't need to start like that. If one wanted to compete, it could start supporting GNU/Linux with the latest, more efficient stuff available -- Lisp or Mono at your choice, latest standards and all that. The simple lack of a legacy to support slashes costs enormously, as does the bare-bones startup culture. When users' needs grow to include ports to other platforms, the choice of Lisp or Mono, OpenGL and POSIX and DPS and all that would have made it much easier. MS Windows? OK, as long as it has the POSIX extension installed.
>
platform independent, as long as wise architectural choices are made (no DirectX, no MFC, no.NET, no C#, and highly abstracted platform dependencies).
Actually DirectX is unneeded, as DPS and OpenGL suffice; MFC too, POSIX being already better; and.Net and C# are already implemented in Mono.
Actually there is only one x86-64 manufacturer, AMD; IPF is quite another game, and they are not compatible.
If alternative sources for these architectures will appear is quite another story; I'd say IPF doesn't stand a chance at creating alternatives, being too complex, but it might be that neither PPC, nor x86-64 nor IPF are enough documented and 'IP'-free to be independently implemented. I don't know.
SPARC, on the other hand, is supposed to be an open standard, but currently Sun lacks IBM's enthusiasm for Linux.
Currently I'm going PPC, it being more efficient, portables and desktops being available, and it being a clear enough stand against the Wintel duopoly and mass behaviour. But I'd go SPARC if Sun, or some alternative SPARC supplier, took an initiative to match IBM's POP. Hopefully this would lead to SPARC desktops and notebooks, not only workstations and servers.
Is it? I didn't know. Do you have supporting facts or references? If true, it could explain China's choice.
I know for sure that SPARC is the open architecture, including being the birthplace of OF and having a backing consortium etc, but Sun's focus on Solaris is stealing its momentum. I certainly would welcome a Leo-based GNU/Linux desktop!
>
MIPS has even fewer desktop choices with regard to software than PPC
Agreed, but SGI is shifting to IPF and China isn't still there... if China really pushes it, it could yet be a containder again.
>
I've checked out a 1.6 GHz G5 PowerMac at my campus computer store, and I must say that I was amazed at how quiet it is. (And this is coming from a Cube owner/phanatic.)
I believe you. Thing is, I tend to work during the night, and I live by the Swiss fields. You can hardly imagine anything more silent. During the wee hours, you can hear your breath. My iBook is too loud for me, I'm looking forward to either a Cube or a Genesi Pegasos.
>
only IBM produces the G5, motorola only produces
32 bit variants
This is a transitory situation for a specific generation. The fundamental concept here is the architecture; the PowerPC one is shared and quite open (OpenFirmware, GNU/Linux the primary OS supported by vendors), while x86-64 and IPF aren't: proprietary boot schemas, preference for MS Windows...
This is not relevant. Betamax was proprietary to Sony, while the PowerPC has quite a momentum and community around it.
>
Just because on paper it might be arguably better
This has nothing to do with paper. The Apple G5s are there, IBM's are just around the corner, and Genesi and EyeTech are already preparing to fit their current G4 systems with G5. I write from a 366MHz G3 that is quite enough for me, and silent.
True enough, for the moment. But the future perspective is based more on fundations than on the current situation, and here the PowerPC is more open.
Now it can come to happen that either the IPF or the x86-64 become more open, but for the moment both have single sources of supply, closed architectures and not much you can run on them.
Actually I want to have a cool, silent, energy-efficient machine. While current Apples aren't as silent and energy-efficient as I'd like, I can get a silent, energy-efficient EyeTech or Genesi.
AFAH/WG, the PowerPC is even more open than x86-64 or IPF, having two chip vendors (IBM and Moto) while the latter have one each (AMD and Intel) only.
AFAS/WG, the same holds true: the PowerPC runs GNU/Linux and BSD just fine, besides Apple's semi-proprietary Mac OS X instead of MS's completely proprietary Windows; the Hurd is being ported, there is Amiga-derived MorphOS and the AmigaOS itself.
Some PCI stuff is more expensive due to the presence of x86 assembly code in firmware of dirty cheap adaptors, but this should be fixed by AMD and Intel adopting OpenFirmware's Forth dialect. So all in all, it is a more open platform.
And BTW, if someday we get to use processors made from GNU GPL'd designs, RISC in general is a better candidate than x86-64 or IPF, being much simpler and more efficient. See that China is trying to go MIPS... MIPS in the East and PowerPC in the West would make for a much more open environment than, say, IPF in the US, x86-64 in Europe, MIPS in China and PowerPC in the high-end...
>
the G5 is ok, a little better, but you've gotta go all apple to get it.
For now you're right, but IBM will sell them too later. IBM prices tend to be high, but also quality; for the low end, EyeTech in the UK and Genesi in Luxembourg are now selling G4 systems. They should have G5 systems just a little down the road after IBM.
>
I really dont want to be locked into a whole platform.
So go G5. There are two vendors of compatible processors, IBM and Motorola, while the only vendor of x86-64 is AMD, and the only vendor of IPF is Intel... not only that, the PowerPC is more efficient and has a technically brighter future.
Actually, if people get used to lend and borrow each other's books, and to deliver them intact afterwards, this could help foster the idea of sharing and free software... nice!
Actually you are half a century late... there were lots of Arabs in Eretz Israel already in 1948. Besides the already settled population before Zionism -- admitedly small --, the huge influx of immigrants to work for Jewish settlers came during British mandate, from 1917 to 1948, and a little while before during the initial Zionist settlements in the last years of the Ottoman Empire, from the 1890s.
>
they smack a mosque dome on top of the Jewish Temple Mount
Actually this was done for political reasons, but not against Zionism. In the VIII century, lots of Arab Muslim conquerors were converting to the Eastern Church due to the splendour and meaningfulness of the rites and the church buildings where these rites were carried, not only in the Temple Mount church but also in the Calvary Church and others. Then the Muslims commandeered the Temple Mount church for a mosque -- the Al Aqsa -- and built the Dome to outmatch the remaining churches, thus giving the Muslims something more visible to cling to, in almost direct contradiction to the Koran's injunctions.
Admitedly, this was a local phenomenon only until Zionism. Even so, half of the Arab population in the Mediterranean was culturally Christian until the Jerusalem mufti ordered the assassination of their leaders and pan-Arabism started using first secularism, then Islamism as tools against the West, Zionism and freedom. Today only 20% of the Arab population in Palestine is Christian; the rest has migrated to France, the US, Brazil and other assorted countries.
Just to complicate matters further, part of the Muslim populational predominance is also due to higher birth rates, their ignorance keeping them poor; and lately Muslims have began migrating to the West too, due to the general insecurity: the Arab infighting and despotism is so bad even some of them prefer to live elsewhere.
When? Actually I am not running Mac OS X because I feel deceived by Apple's broken promises, but I understand 10.2 is the first serious versions. 10.0 was alpha, 10.1 beta quality.
Slashdot Mirror
Indexing is already a standard feature. POSIX systems have locate, Gnome has Medusa, the Mac OS indexing is ancestral...
Using a quasi-relational system is about enabling more powerful interaction with the data, not only indexes.
How can a query be useful if it is not accurate?
How SQL is worse than anything else, save a relational system?
No!
When Codd created the relational model, there wasn't the current Unix filesystem idea... the relational model was always intended to store data, and files are data.
System R, SQL and DB2 prototype, was intended to be the basis for IBM FS.
IBM realised this in OS/400, which being proprietary hasn't the influence it deserves.
MS also wanted Jet to be the building block of its OSs since its inception, that is, sometime before MS Access release.
Sorry, but NewDocMS is based on SQLite, which is typeless and but a library... simply not good enough to be attractive. Storage is based on PostgreSQL, the real thing, and aims high.
Before that. MS Access was intended to be a development platform for Jet, and MS Windows would eventually gain a Jet-based filesystem. Only Jet was considered harmful, so MSDE was substituted and the whole thing delayed a decade or so.
This adds nothing, it substitutes the hierarchical store while presenting a compatible interface. It has been proven SQL if correctly modeled gives usually comparable performance to application-specific databases such as a hierarchical filesystem; it also opens the way to a relational system that could significantly better SQL performance while being more powerful and simpler.
That said, the question is how sane will be the database model... for such a fundamental stuff you'd want 6NF.
As for implementation, a bTree is OK. But here the goal is semantics: the hierarchical semantics is fundamentally broken, because the world can't be nicely organised that way. Instead, even is SQL is corrupted from the relational model, it allows a much more flexible, powerful, logical interface.
Letter to the Patent Office From Professor Donald Knuth
Not at all... Pro/Engineer may be a multi-billion affair nowadays, but it doesn't need to start like that. If one wanted to compete, it could start supporting GNU/Linux with the latest, more efficient stuff available -- Lisp or Mono at your choice, latest standards and all that. The simple lack of a legacy to support slashes costs enormously, as does the bare-bones startup culture. When users' needs grow to include ports to other platforms, the choice of Lisp or Mono, OpenGL and POSIX and DPS and all that would have made it much easier. MS Windows? OK, as long as it has the POSIX extension installed.
Actually DirectX is unneeded, as DPS and OpenGL suffice; MFC too, POSIX being already better; and .Net and C# are already implemented in Mono.
Actually there is only one x86-64 manufacturer, AMD; IPF is quite another game, and they are not compatible.
If alternative sources for these architectures will appear is quite another story; I'd say IPF doesn't stand a chance at creating alternatives, being too complex, but it might be that neither PPC, nor x86-64 nor IPF are enough documented and 'IP'-free to be independently implemented. I don't know.
SPARC, on the other hand, is supposed to be an open standard, but currently Sun lacks IBM's enthusiasm for Linux.
Currently I'm going PPC, it being more efficient, portables and desktops being available, and it being a clear enough stand against the Wintel duopoly and mass behaviour. But I'd go SPARC if Sun, or some alternative SPARC supplier, took an initiative to match IBM's POP. Hopefully this would lead to SPARC desktops and notebooks, not only workstations and servers.
The article mentions solid skills as a sure job getter.
Sorry, but no. Employers want certifications and 'perfect matches', not real knowledge.
Is it? I didn't know. Do you have supporting facts or references? If true, it could explain China's choice.
I know for sure that SPARC is the open architecture, including being the birthplace of OF and having a backing consortium etc, but Sun's focus on Solaris is stealing its momentum. I certainly would welcome a Leo-based GNU/Linux desktop!
Agreed, but SGI is shifting to IPF and China isn't still there... if China really pushes it, it could yet be a containder again.
Apple uses OpenFirmware. You can't get more open than that AFA boot goes.
But the x86 BIOS causes PCI cards, for one, to be exclusively x86. While Mac PCI cards, working with OF, are cross-platform.
G5s are quite competitive, and G[34]s are silent and energy-efficient.
China has a 'Dragon' or something processor about to launch to run GNU/Linux, and it is MIPS.
Today's high end is embedded five years hence.
I believe you. Thing is, I tend to work during the night, and I live by the Swiss fields. You can hardly imagine anything more silent. During the wee hours, you can hear your breath. My iBook is too loud for me, I'm looking forward to either a Cube or a Genesi Pegasos.
This is a transitory situation for a specific generation. The fundamental concept here is the architecture; the PowerPC one is shared and quite open (OpenFirmware, GNU/Linux the primary OS supported by vendors), while x86-64 and IPF aren't: proprietary boot schemas, preference for MS Windows...
This is not relevant. Betamax was proprietary to Sony, while the PowerPC has quite a momentum and community around it.
This has nothing to do with paper. The Apple G5s are there, IBM's are just around the corner, and Genesi and EyeTech are already preparing to fit their current G4 systems with G5. I write from a 366MHz G3 that is quite enough for me, and silent.
True enough, for the moment. But the future perspective is based more on fundations than on the current situation, and here the PowerPC is more open.
Now it can come to happen that either the IPF or the x86-64 become more open, but for the moment both have single sources of supply, closed architectures and not much you can run on them.
I am not a kid, I have a kid.
Actually I want to have a cool, silent, energy-efficient machine. While current Apples aren't as silent and energy-efficient as I'd like, I can get a silent, energy-efficient EyeTech or Genesi.
I should think before answering... third time.
What does this mean?
AFAH/WG, the PowerPC is even more open than x86-64 or IPF, having two chip vendors (IBM and Moto) while the latter have one each (AMD and Intel) only.
AFAS/WG, the same holds true: the PowerPC runs GNU/Linux and BSD just fine, besides Apple's semi-proprietary Mac OS X instead of MS's completely proprietary Windows; the Hurd is being ported, there is Amiga-derived MorphOS and the AmigaOS itself.
Some PCI stuff is more expensive due to the presence of x86 assembly code in firmware of dirty cheap adaptors, but this should be fixed by AMD and Intel adopting OpenFirmware's Forth dialect. So all in all, it is a more open platform.
And BTW, if someday we get to use processors made from GNU GPL'd designs, RISC in general is a better candidate than x86-64 or IPF, being much simpler and more efficient. See that China is trying to go MIPS... MIPS in the East and PowerPC in the West would make for a much more open environment than, say, IPF in the US, x86-64 in Europe, MIPS in China and PowerPC in the high-end...
For now you're right, but IBM will sell them too later. IBM prices tend to be high, but also quality; for the low end, EyeTech in the UK and Genesi in Luxembourg are now selling G4 systems. They should have G5 systems just a little down the road after IBM.
So go G5. There are two vendors of compatible processors, IBM and Motorola, while the only vendor of x86-64 is AMD, and the only vendor of IPF is Intel... not only that, the PowerPC is more efficient and has a technically brighter future.
Actually, if people get used to lend and borrow each other's books, and to deliver them intact afterwards, this could help foster the idea of sharing and free software... nice!
Are you sure your users are among the 5% that can't live with Gimp instead of Adobe Photoshop?
Actually you are half a century late... there were lots of Arabs in Eretz Israel already in 1948. Besides the already settled population before Zionism -- admitedly small --, the huge influx of immigrants to work for Jewish settlers came during British mandate, from 1917 to 1948, and a little while before during the initial Zionist settlements in the last years of the Ottoman Empire, from the 1890s.
Actually this was done for political reasons, but not against Zionism. In the VIII century, lots of Arab Muslim conquerors were converting to the Eastern Church due to the splendour and meaningfulness of the rites and the church buildings where these rites were carried, not only in the Temple Mount church but also in the Calvary Church and others. Then the Muslims commandeered the Temple Mount church for a mosque -- the Al Aqsa -- and built the Dome to outmatch the remaining churches, thus giving the Muslims something more visible to cling to, in almost direct contradiction to the Koran's injunctions.
Admitedly, this was a local phenomenon only until Zionism. Even so, half of the Arab population in the Mediterranean was culturally Christian until the Jerusalem mufti ordered the assassination of their leaders and pan-Arabism started using first secularism, then Islamism as tools against the West, Zionism and freedom. Today only 20% of the Arab population in Palestine is Christian; the rest has migrated to France, the US, Brazil and other assorted countries.
Just to complicate matters further, part of the Muslim populational predominance is also due to higher birth rates, their ignorance keeping them poor; and lately Muslims have began migrating to the West too, due to the general insecurity: the Arab infighting and despotism is so bad even some of them prefer to live elsewhere.
When? Actually I am not running Mac OS X because I feel deceived by Apple's broken promises, but I understand 10.2 is the first serious versions. 10.0 was alpha, 10.1 beta quality.
Already tried M-x recover-file? In my experience GNU Emacs saves to a temporary file each few characters.