Long ago, hardware made much more of a difference than it does today and was one reason relational databases "won" out.
Hardware makes just as big of a difference today, which is why distributed key/value stores are gaining currency at the moment. The hardware-related difference that was a big win for relational databases was their efficient use of disk space when normalized; the hardware-related difference that is a big win for distributed key/value stores now is their efficient scalability by distribution across multiple nodes.
I am going to tear my eyes out if I see "yet another tuple store or graph db." Welcome to the last century, please try again.
The big thing isn't "tuple stores or graph dbs" its distributed tuple stores, and, even better, distributed transactional tuple stores. Not a whole of them from the last century.
If you are willing to get rid of ACID like the other solutions, there are no limitations.
The other solutions (see below) do not, in all cases, "get rid of ACID".
Please site one example, just one, where a simple key/pair data system is the "better" solution for a high volume site than a more powerful database like PostgreSQL wouldn't do a better job.
Scalaris, a distributed transactional key/value store that does not get rid of ACID, is one of the "other solutions" (and one that has been demonstrated, by replicating Wikipedia on a distributed cluster, to scale better, at least, than Wikipedia's existing MySQL platform).
Yes I read most the article and perhaps reading the rest would answer this, but how is a key/value database different from a MySQL databse running MyISAM where you store a bunch of different objects as a string, maybe json_encoded or whatever in the row?
Purpose-designed distributed key/value stores (which are what is interesting, not just "key/value stores") that aren't pretending to be RDBMS's are generally easier to scale out, since scalability is what they are designed for.
The problem with any of it is that relational databases rule the enterprise space. You cannot get away from them, and they are far from dead, because you will always have business people wanting to do ad hoc reports, and those are best done against denormalized models where object stores tend to get super normalized which is just bad for reporting because cross table joins are the most expensive thing you can do in any database.)
Huh? Object databases tend to be extremely denormalized in the relational sense, the whole point of them is to have a closer mapping to the application domain than a normalized relational representation would provide. Furthermore, regular, anticipated reports are often best done against denormalized models, since you know exactly how the report is going to be structured and can optimized the data storate to support it; truly ad hoc reports (i.e., where you've got a pile of data and the questions being asked about it aren't anticipated in advanced) gain the least, performance-wise, from denormalization, and are often made much more difficult (in terms of actually putting a query together to get what you want) by denormalization.
So we can establish that a SQL relational database can do *everything* a simpler system can do.
In terms of expressive power, sure, but no one is arguing that distributed key/value stores are going to gain against RDBMS's because they have superior expressive power. What is being argued is that they will do so because they have superior scalability and distribution properties, and that in many real-world applications those are more important than the having the full expressive power of relational algebra. Particularly as you get ones that can provide ACID guarantees, that becomes a compelling selling point in many applications where RDBMS's would otherwise be used simply because they are the only available tool, but where distributed key/value stores are a better tool.
The big dumb thing about key store values is that they are actually just a subset of relational algebra in theory and are thus readily implementable in a relational database in fact.
Yeah, sure you get implement a key/value store in any RDBMS, but then, to make it into a distributed database, you'd have all the problems of doing that with an RDBMS, whereas purpose-built key/value stores and similar distribute much better than traditional RDBMS's.
If you really wanted to have a database just do key / store values, you could quite easily do that in any rdms.
OTOH, if you wanted to do seamless scaling, you can't do that easily in most RDBMS's. But you can do that with a lot of the purpose-built key-value stores. OTOH, most distribtued key/value stores don't provide ACID guarantees, though Scalaris seems to.
Since it has a native GCC toolchain, just about anything you'd care to recompile will run on it. Firefox runs on it, for example, as the story summary states.
The summary doesn't state that, though, it states that the recent addition of a GCC toolchain "clears the way" for a port of Firefox. Having the GCC toolchain is a start, but not the whole shebang.
That's what you get if you discourage the use of primary sources in favor of secondary sources.
No, its what you get when you fail to distinguish secondary sources from tertiary sources. Any source which relies on any encyclopedia directly for information, rather than using it as a research tool, is at best a tertiary source (and arguably worse than that), and should not be given the weight of a reliable secondary source in assembling an encyclopedia or similar tertiary source.
Der Spiegel superficially seems to meet the criteria of a secondary source, but the fact that they are using material from Wikipedia without any independent verification means they are, in fact, no better than a tertiary source, and therefore not the kind of source a Wikipedia article should rely on (even if the tertiary source they were using on wasn't Wikipedia, but was instead, e.g., Britannica.)
It seems like, if you are going to do an encyclopedia article, the source for someone's full name would not be a news article (which is a 'secondary' source - that is, the journalist had to get that info from somewhere else), but you should instead source facts from 'primary' sources - like a birth certificate (or other similar 'official' document - I'm not sure if Germany has the concept of a birth certificate like we do in the USA).
Wikipedia, as an encyclopedia and, itself, a tertiary source, generally prefers secondary sources to primary sources. (see WP:NOR)
People writing in media that are considered secondary sources (including news articles) should prefer primary sources, not tertiary sources.
On spiegel.de an article cites Mr. von und zu Guttenberg using his 'full name'; however, while the quote might have been real, the full name seems to have been looked up on Wikipedia while the false edit was in place. So the circle was closed: Wikipedia states a false fact, a reputable media outlet copies the false fact, and this outlet is then used as the source to prove the false fact to Wikipedia.
Any outlet which uses Wikipedia as a source, rather than a research tool to find other sources, should, ipso facto, not be considered a "reputable media outlet" for any purpose, especially as a valid source of citations in Wikipedia.
A reasonable ruling would be that the market is "computer operating systems". There are plenty of alternatives and in many spaces MS isn't even dominant.
The fact that there are many definable "spaces" within that market that are not equivalent, and that dominance or the lack of it in one is entirely separate from dominance in others demonstrates that this is not a "reasonable ruling" as to what the relevant market is.
A more reasonable ruling would be that antitrust law should only be applied to physically limited resources of vast public importance
That would even more emphatically not be a "reasonable ruling". It might be a reasonable law (I don't think it would, but that's a separate debate), but it clearly violates both the letter and the spirit of the laws on the books (both in the US and in most other places that have anti-trust/competition laws), and as such would be an unreasonable ruling.
You are (by simple fact) forced to admit there are actually many alternative OS's, but none of them are Microsoft Windows compatible.
Legally, a monopoly doesn't generally mean that there are no alternatives, but that a particular person (including a corporation) dominates the market; the existence of many alternatives that are insignificant in market share does not refute the existence of a monopoly. So, no, the relevant issue isn't that Microsoft has an absolute monopoly on Windows-compatible operating systems, but that it has a legal monopoly on desktop computer operating systems.
Why do we keep using a term that means, essentially, monarch or "supreme ruler"?
The whole "czar" thing in US government was coined to represent an official with overall coordinating responsibility over functions which are within the domains of several different executive branch departments (and usually, in each, within the domain of some obscure office within the department); it makes sense when you consider the common description of the problem "czars" are intended to address that the various offices with related responsibilities tend to operate as independent "fiefdoms" with no overall direction in addressing the larger policy problem. Within a particular narrow policy area, the "czar" is over all of the "fiefdoms".
Why is it that using a code's API makes something derivative work, but using a program's CLI is non-derivative work, and even allowed to be non-GPL?
There's a pretty good argument to be made that using a code's API or dynamically linking to a library is not generally a "derivative work", although this is pretty clearly not the FSF's view.
i understand that - but being the co-founder there were no stipulations of how MySQL were to be maintained in the future?
Again, $1 billion buys a lot. Even a very small fraction of that (don't know how much Monty got) means you can easily quit and do your own thing (as Monty is) if you don't like what your new employer is doing; its pretty easy to see why taking the money without stipulations rather than imposing stipulations which would have scuttled the deal or reduced the value to Sun could easily make sense.
This is all good until real money starts pouring in and someone wants/needs more money or claims they are more deserving than another or something along those lines. Then the "fun" arrangement becomes less fun and more bitter.
And then Monty sells the company and IP to some big firm that likes the big money that's pouring in, and after a suitable transition period quits and goes on starts some other venture with like-minded folks for fun.
And I doubt Monty is going to be hurting for money, himself, if he keeps succesfully doing that. Where the fun ends is if too much time and money are poured into a venture but money never starts pouring in.
IANAL, but patents are for protection from copying.
No, copyrights are for protection from copying. Patents protect methods whether or not the the other person trying to use them is copying them or has indepedently developed them without any knowledge of your work.
The problem with Windows is its backup software is Veritas. Its disk defragmenter is... I forget who, I think it may be Disk Keeper. Most of the internal tools are licensed from companies that Microsoft doesn't own; you can buy a much better Veritas backup system or a full Disk Keeper license and get network control and everything. They can't open source this, and they can't give it away for free because they have to pay it back somehow
Microsoft could probably just buy the companies the technology is licensed from outright, if open sourcing Windows was critical to their business model. Or reimplement those features.
Congress is the one doing this, not the president.
Since the President has requested the action and, presumably, will sign the bill rather than vetoing it and having Congress attempt to override the veto, it is the President and Congress doing it; it is not "Congress , not the President".
Also, why would Microsoft open-sourcing things be good for Microsoft? Either people shift to Linux because they drink the RMS kool aid (that'd include me), or because it's the better product for them (I then found out this also included me).
If they shift because they drink the RMS kool aid, then we can assume that they prefer a completely free OS (including application stack), which MS won't give out (according to the article, at least).
If they shift because Linux is the better product (technically, that is), Windows being open source(d) won't change the fact that Linux is the better product.
Arguably, being open source contributes to something being a better product, particularly if its big enough and interesting enough that an active developer and user community exists for it. I don't think open source Windows, were it to happen (which it won't, but let's pretend) would have any problem getting an active developer and user community (MS might manage it in a way that hurt participation in the "official" Windows project, even if it was open source, but someone else would fork it and run it in a way that was community-friendly if that happened.)
So, I'd be inclined to think that at least one fork of open source Windows would be a far better product than commercial Windows.
I have heard (but beware of echo chamber effects) that it's nigh impossible to write two implementations of the OOXML spec that renders identical outputs.
Last I heard, there wasn't any compliant implementation of the OOXML spec and even the actual Microsoft Office's OOXML doesn't comply with the spec. (Note that I have neither read the whole spec nor attempted to verify any programs compliance with it.)
He suggests that the money to be made from the things MS builds on top of Windows (Office, Server, SQL Server, Exchange, Sharepoint, etc) is so much greater than what can be made from Windows itself that MS will have to give up the revenue stream from Windows in order to maintain these other, more valuable, revenue streams.
Opening Windows does nothing to protect those other revenue streams. If anything, it would weaken them, because then there would be Windows distributions available that included competitors to those packages out-of-the-box, and were supported top-to-bottom by the distribution makers (who may or may not also be the vendor for the competing product.) Sure, Joe's Windows distro with PostgreSQL and no professional support may not be a big threat to Microsoft in the enterprise, but Oracle Unbreakable Windows might be.
You are, to some extent, talking about creating two different things: public domain Vs. GPLed.
OP seems pretty clear that they are looking to release the protocol to the public domain, while the software they have so far released that implements the protocol is released under the GPL. Those are, quite clearly, two different things, but they are not at all in conflict.
Slashdot Mirror
Hardware makes just as big of a difference today, which is why distributed key/value stores are gaining currency at the moment. The hardware-related difference that was a big win for relational databases was their efficient use of disk space when normalized; the hardware-related difference that is a big win for distributed key/value stores now is their efficient scalability by distribution across multiple nodes.
The big thing isn't "tuple stores or graph dbs" its distributed tuple stores, and, even better, distributed transactional tuple stores. Not a whole of them from the last century.
The other solutions (see below) do not, in all cases, "get rid of ACID".
Scalaris, a distributed transactional key/value store that does not get rid of ACID, is one of the "other solutions" (and one that has been demonstrated, by replicating Wikipedia on a distributed cluster, to scale better, at least, than Wikipedia's existing MySQL platform).
Purpose-designed distributed key/value stores (which are what is interesting, not just "key/value stores") that aren't pretending to be RDBMS's are generally easier to scale out, since scalability is what they are designed for.
Huh? Object databases tend to be extremely denormalized in the relational sense, the whole point of them is to have a closer mapping to the application domain than a normalized relational representation would provide. Furthermore, regular, anticipated reports are often best done against denormalized models, since you know exactly how the report is going to be structured and can optimized the data storate to support it; truly ad hoc reports (i.e., where you've got a pile of data and the questions being asked about it aren't anticipated in advanced) gain the least, performance-wise, from denormalization, and are often made much more difficult (in terms of actually putting a query together to get what you want) by denormalization.
In terms of expressive power, sure, but no one is arguing that distributed key/value stores are going to gain against RDBMS's because they have superior expressive power. What is being argued is that they will do so because they have superior scalability and distribution properties, and that in many real-world applications those are more important than the having the full expressive power of relational algebra. Particularly as you get ones that can provide ACID guarantees, that becomes a compelling selling point in many applications where RDBMS's would otherwise be used simply because they are the only available tool, but where distributed key/value stores are a better tool.
Yeah, sure you get implement a key/value store in any RDBMS, but then, to make it into a distributed database, you'd have all the problems of doing that with an RDBMS, whereas purpose-built key/value stores and similar distribute much better than traditional RDBMS's.
OTOH, if you wanted to do seamless scaling, you can't do that easily in most RDBMS's. But you can do that with a lot of the purpose-built key-value stores. OTOH, most distribtued key/value stores don't provide ACID guarantees, though Scalaris seems to.
The summary doesn't state that, though, it states that the recent addition of a GCC toolchain "clears the way" for a port of Firefox. Having the GCC toolchain is a start, but not the whole shebang.
That's what you get if you discourage the use of primary sources in favor of secondary sources.
No, its what you get when you fail to distinguish secondary sources from tertiary sources. Any source which relies on any encyclopedia directly for information, rather than using it as a research tool, is at best a tertiary source (and arguably worse than that), and should not be given the weight of a reliable secondary source in assembling an encyclopedia or similar tertiary source.
Der Spiegel superficially seems to meet the criteria of a secondary source, but the fact that they are using material from Wikipedia without any independent verification means they are, in fact, no better than a tertiary source, and therefore not the kind of source a Wikipedia article should rely on (even if the tertiary source they were using on wasn't Wikipedia, but was instead, e.g., Britannica.)
Wikipedia, as an encyclopedia and, itself, a tertiary source, generally prefers secondary sources to primary sources. (see WP:NOR)
People writing in media that are considered secondary sources (including news articles) should prefer primary sources, not tertiary sources.
Any outlet which uses Wikipedia as a source, rather than a research tool to find other sources, should, ipso facto, not be considered a "reputable media outlet" for any purpose, especially as a valid source of citations in Wikipedia.
The fact that there are many definable "spaces" within that market that are not equivalent, and that dominance or the lack of it in one is entirely separate from dominance in others demonstrates that this is not a "reasonable ruling" as to what the relevant market is.
That would even more emphatically not be a "reasonable ruling". It might be a reasonable law (I don't think it would, but that's a separate debate), but it clearly violates both the letter and the spirit of the laws on the books (both in the US and in most other places that have anti-trust/competition laws), and as such would be an unreasonable ruling.
Legally, a monopoly doesn't generally mean that there are no alternatives, but that a particular person (including a corporation) dominates the market; the existence of many alternatives that are insignificant in market share does not refute the existence of a monopoly. So, no, the relevant issue isn't that Microsoft has an absolute monopoly on Windows-compatible operating systems, but that it has a legal monopoly on desktop computer operating systems.
The whole "czar" thing in US government was coined to represent an official with overall coordinating responsibility over functions which are within the domains of several different executive branch departments (and usually, in each, within the domain of some obscure office within the department); it makes sense when you consider the common description of the problem "czars" are intended to address that the various offices with related responsibilities tend to operate as independent "fiefdoms" with no overall direction in addressing the larger policy problem. Within a particular narrow policy area, the "czar" is over all of the "fiefdoms".
Perhaps, but what about PostgreSQL?
There's a pretty good argument to be made that using a code's API or dynamically linking to a library is not generally a "derivative work", although this is pretty clearly not the FSF's view.
Again, $1 billion buys a lot. Even a very small fraction of that (don't know how much Monty got) means you can easily quit and do your own thing (as Monty is) if you don't like what your new employer is doing; its pretty easy to see why taking the money without stipulations rather than imposing stipulations which would have scuttled the deal or reduced the value to Sun could easily make sense.
And then Monty sells the company and IP to some big firm that likes the big money that's pouring in, and after a suitable transition period quits and goes on starts some other venture with like-minded folks for fun.
And I doubt Monty is going to be hurting for money, himself, if he keeps succesfully doing that. Where the fun ends is if too much time and money are poured into a venture but money never starts pouring in.
No, copyrights are for protection from copying. Patents protect methods whether or not the the other person trying to use them is copying them or has indepedently developed them without any knowledge of your work.
As counterevidence, I present SQLite.
Microsoft could probably just buy the companies the technology is licensed from outright, if open sourcing Windows was critical to their business model. Or reimplement those features.
Since the President has requested the action and, presumably, will sign the bill rather than vetoing it and having Congress attempt to override the veto, it is the President and Congress doing it; it is not "Congress , not the President".
Arguably, being open source contributes to something being a better product, particularly if its big enough and interesting enough that an active developer and user community exists for it. I don't think open source Windows, were it to happen (which it won't, but let's pretend) would have any problem getting an active developer and user community (MS might manage it in a way that hurt participation in the "official" Windows project, even if it was open source, but someone else would fork it and run it in a way that was community-friendly if that happened.)
So, I'd be inclined to think that at least one fork of open source Windows would be a far better product than commercial Windows.
Last I heard, there wasn't any compliant implementation of the OOXML spec and even the actual Microsoft Office's OOXML doesn't comply with the spec. (Note that I have neither read the whole spec nor attempted to verify any programs compliance with it.)
Opening Windows does nothing to protect those other revenue streams. If anything, it would weaken them, because then there would be Windows distributions available that included competitors to those packages out-of-the-box, and were supported top-to-bottom by the distribution makers (who may or may not also be the vendor for the competing product.) Sure, Joe's Windows distro with PostgreSQL and no professional support may not be a big threat to Microsoft in the enterprise, but Oracle Unbreakable Windows might be.
OP seems pretty clear that they are looking to release the protocol to the public domain, while the software they have so far released that implements the protocol is released under the GPL. Those are, quite clearly, two different things, but they are not at all in conflict.