Oh, I dunno, but it sure looks like the AC in question has not read the Byte August 1981 issue. And especially the article Design Principles Behind Smalltalk by Dan Ingalls.
Where he defines Factoring: Each independent component in a system should appear in only one place.
Well, actually, if that notion, however silly it might be, gets to be law then, at least to my simple programmer mind, that in itself would mean that Disney would indeed owe royalties to the heirs of Grimm.
It could even be the case that such a law could be effectuated retro-actively.
In which case it could even bankrupt Disney!
One has to wonder if the powers that be have really thought things through;).
The whole point is that you don't care. Or rather, that you learn to live with it. That's why I've marked some of software 'All Rights Relinquished' which gets that sentiment across. In both ways.
It's also very Taoistic of nature. Very Lao Tse's Tao Te Ching, which btw was written well over 2,500 years ago!
Just tried 'm both and noticed this whilst sitting behind a FreeBSD NAT box, that BT has an upload and download rate of 40 KB/sec whereas WinMX claims no upload capabilities are present.
OK, so my ADSL hookup is pretty much swamped, still I oughta have some KB/secs spare...
Roelof -- the red nosed hacker -- Osinga
PS this is not criticism per se, just something I noticed, is all
Hm. Guess we were created in Its own image after all;). Hack first, debug later.
For it is The motto of Gods!
Roelof
Re:AI is going wherever it wants
on
AI Going Nowhere?
·
· Score: 2, Interesting
Sure, but the whole point of AI was that we were supposed to be able to ask it where it thought it was going... and that not only it would know, but would give a well thought out answer too!
Me too. I also agree to a point with what you're saying.
However, easing off on the registration requirements does have two real advantages.
In the first place it brings ease of mind. Not to mention that it saves time. And stress. And thus prolongues live.
More important however is that it might help you leverage off of Metcalf's -- I believe -- law which states that the Value of a thing like this is e ^ Nusers, where ^ denotes to the power of.
It will also allow you to tap into the business market more easily. Due to brand recognition.
It is, in other words, not a clear cut situation.
For example, the very fact... Sorry, I got derailed by a beautiful being whom I spotted outside my window. Signing became talking, became drinking thee, more talking, etc. Couple of hours later I find that I lost the thread of my thoughts.
Too bad too, 'coz it was a darned good point;). Still, the being known as Maja provided a very welcome experience indeed. So can't say I'm sorry that I can not finish that point I was building.
But I do imagine you'll be getting the gist of it. That a relaxed attitude will still allow you, even assist you, build a profitable thingum.
Remember Phil Katz? Destroyed himself quite comfortably on the proceeds of pkzip and maybe some pkarc too. Don't quite remember. Long time ago.
Roelof - as in (t)rue love - Osinga
PS Let us not forget that these days the Internet is of the utmost importance. Would you like to serve your customers from a Win32 base? I think not. So in that way the picture, even on the desktop, has changed irrevocably.
That I personally am not alltogether happy with Java, comes from my background. Though I do confess that the latest on the Java front looks good, what with the inlusion of polymorphic types.
What I am trying to bring across is that the adding of an extra layer of abstraction, or two, brings new opportunities for business. Brings new degrees of freedom to application development.
I have to disagree. And agree as well at the same time, of course. But the latter mostly because of the abundant choice in OSS, the majority of which are free.
The former, however, is the most important. At least, let us hope it is.
As I said in this topic earlier, most animals recognize good and respond in kind. As it happens, humans are basically animals. So humans validate good deeds appropriately.
One counter example to your case sprong to mind reading your post. And that was the case of OpenBSD. Theo de Raadt gets by one the sale of the CDs, T shirts and what nots.
Even though it is completely free for the taking.
And I do mean free as in free beer! [Hick]
I my self bought several T-shirts of, whatever, don't remember, and stickers and stuff just because a part of the sale went to sponsor the project.
I mean, why not?
As others said before. It makes me feel good.
Roelof -- pronounced [t]rue love -- Osinga
PS as to the market being limited, two things. First, yes, but growing! Second, sure so target where it is in large use. Like servers.
I agree. The registration is icing on the cake. If you have fun writing the daft thinglet than honest people will want to give you your just reward. Most people fortunately are honest.
Heck, my first attempt at shareware was even distributed in source code, I kid you not! Did not stop even companies from ordering it and sending checks. Received orders from universities up to Xerox and Intel.
This was a Smalltalk 'goodie' I had named LifeSaver and is probably still out there. Even though that was a decade ago.
In answer to my sibling poster, I failed miserably even though I succeeded beyond my wildest dreams!
Turned out it cost me almost as much to clear the checks I received as the checks were for. Making it a money loosing proposition I alas had to quit.
The second attempt at shareware was also not quite the money making venture I had hoped. This is the JPEG Sheriff. A file, especially JPEG files, verifier and collection completenes checker.
Here too necessity -- if one can call it that -- was the mother of invention. You see, I was a beta tester of Borland. Delphi 2.0 one of them. This had support for threads!
WOW!
That begged for thorough testing as I am sure you can imagine. But what to test it with? It had to make sense.
So I made the JPEG Sheriff a multi-threaded app. Just for the heck of it. Well, OK, and because I was a serieus bèta tester. Of course. Goes without saying.
This I put there as just freeware. Not in source form. Mostly because the code lacked a certain elegance from an OO (purists) view.
Especially the first three or four years I got all kinds of requests -- believe it or not, almost no bug reports -- which I gladly honored. If they made sense to me.
Also had people telling me, why don't you ask a bit for it? We'll be glad to pay.
Thus, when PayPal became available overseas I did just that.
Am still waiting for the first donation. This while the JPEG Sheriff has been downloaded several thousand times.
To be fair I must confess that I fell ill 'in my prime'. In 2000 I got carted into a hospital for MS (multiple sclerosis). Took me out of the loop for quite a while. Don't know what would've happened, could've happened, otherwise. Had I been able to keep up the good work.
So there you have it. It depends. With a bit of luck, with the right timing, you can build momentum. If you can do that...
But as to registration and stuff. Who needs it? If they don't want to pay. They don't pay. If they want to copy the source or even need the source as with Smalltalk, here it is. Well, if it's presentable, you understand.
That having been said however. Must confess I am playing with this nice idea about cryptographic hashes. Last work I did on the JPEG Sheriff, though unreleased due to my bout in the hospital, was with better check figures. Added the MDs (4&5) and SHA-1.
Which got me thinking about a registration system based on those hashes. You know, to get your name in the verification lists you make or request lists you send out.
I am still recuperating -- don't ever, EVER, try to get high on prednisone; it works, but boy will you regret it (1000 mg twice intravenously) -- but when I find my wits...
That's what they used to say about minicomputers and later microcomputers. Sure, it has way less market share. But hardly less funcionality. In some regards it can probably still be considered to be more advanced (e.g. versioning engine) than the others.
Mini's, supermini's, microcomputers and LANs did eat into mainframe territory. There is a decided possibility Interbase can now eat harder into the market share kept by those vendors. Why not? Look at what MS did with their bought SQL Server.
But I do realize mainframes are still around and doing very well. Likewise will the mainframes in the RDBMS market.
SQL has nothing to do with the relational model as defined by Codd. Other than that it is one of the many query languages defined to work with implementations of the relational model. It turned into the most popular supplanting Sequel, Quel, QBE, etc. But popularity has nothing to do with essentiality.
In fact you could make a case that there are no true RDBMSs at all. Most, if not all, lack some features or have added shortcuts to the pure relational model. E.g. the addition of TOP to the SELECT statement in MS SQL. You can only have TOP if you supported relational sequences. As far as I am aware sequences are not part of the relational model. Besides which MS does not provide a rigorous definition of TOP in terms of the relational model.
TOP can only make sense given an ordering such that no duplicates can occur. From that sequence you can take the first N tuples. Since SELECT is defined without ordering or distinctness requirements how can TOP be explained? Within the relational model that know nothing about, not recorgnizes, orderings implied by an implementation?
Another thing are relations. I believe Codd included the notion of domains in his model. That would indicate valid relations/joins can only be made on keys of the same domain. In practice you can darned well join anything with whatever you fancy whenever you want. Though it can be defended I don't think it is what was intended.
I'm not sure it has already been mentioned in this still growing thread. But what got me interested in functional languages was the 1985 Byte August issue.
Back then, and maybe still , Byte had a tradition of devoting its August issue to a language. So after Smalltalk, Logo, etc. the 85 issue was about functional programming.
B.t.w. one reason this particular implementation could be a little on the slow side is because it has a reference counting GC. I needed that to have quick finalization.
I find functional languages extremely interesting. In fact I've spent well over three years implementing a kind of Hope derivative called very originally TLC (Typed Lambda Calculus).
What made me do it was that at the time (early nineties) I knew of no implementation that defined a FL metacircularly. Or rather, a strict FL. So there was a challenge and one I did succeed to master.
So, TLC (or Tender Loving Care as Hal Hildebrand would have it) is a strict FL implemented in itself. Albeit just the interpreter and type inference engine.
The compiler which currently compiles TLC to plain C with gets linked to the also C VM uses the FPM compilation scheme.
Though it doesn't use optimazation like memoization or abstract interpretation it is a strict language and as such supposed to be faster than a lazy implementation.
Alas, I have found even Java to be way faster. And I do not consider Java to be anywhere near fast.
Unfortunately it is not available for download. You see, I wrote it in Smalltalk/V-PM. Even the running C interpreter is currently useless since it's for BCOS2 and relies on a OS/2 database library. Been meaning to regenerate an interpreter that does not rely on any OS/2 quirks but somehow haven't come around to doing it.
Back in '87 I was running MicroSoft Xenix. Which was basically a System III clone or version which was more or less Release 7.
When I came to Linux when it was one, or nearly 1.0 anyway, via the SLR distrib I stumbled upon Slack. The second Slack cought.
What I liked about Slack was that its resource configuration system was so very much akin to what I was used to!
Now as already has been described, back in the olden days things were a bleedin' mess. We went to UNIX because in those days UNIX was hyped as the new OS that would kill all other OS's. Something we've since seen when NT came along as well as when Linux broke through. (Let's face it, 10 percent of the Fortune-500 is a big deal).
From that mess two streams emerged. System V on the one hand and BSD on the other. BSD's rc system, like Slack's I still grock, SysV's OTOH can not capture my imagination.
In that sense Slack's BSD feel is anything but an illusion. That's it's not BSD is a given, but qua feel it's much more BSDish then SVish!
Please note that I come from the UNIX corner, not the BSD one:).
The thing I missed in this story is its ISO
code. It all hinges on its sovereignty. Which
in turn depends on more then just one positive
verdict in an English court.
Like, what's Sealand's ISO code? Can we surf
to http://havenco.sd/ or something? If not
then international recognition is still a
way off.
Slashdot Mirror
vi will let you pipe your text through ispell easily enough :)
Heck, no. Not with the way the, ehm, civilized world is switching en masse to OSS it isn't ;).
No, no. It's just that their users are in veritable IT heaven!
It is not quite the same you know.
Roelof
Oh, I dunno, but it sure looks like the AC in question has not read the Byte August 1981 issue. And especially the article Design Principles Behind Smalltalk by Dan Ingalls.
Where he defines Factoring: Each independent component in a system should appear in only one place.
Roelof Osinga
Well, actually, if that notion, however silly it might be, gets to be law then, at least to my simple programmer mind, that in itself would mean that Disney would indeed owe royalties to the heirs of Grimm.
;).
It could even be the case that such a law could be effectuated retro-actively.
In which case it could even bankrupt Disney!
One has to wonder if the powers that be have really thought things through
Roelof - the red nose 'grammer - Osinga
Thus mathematicians aren't scientists.
The whole point is that you don't care. Or rather, that you learn to live with it. That's why I've marked some of software 'All Rights Relinquished' which gets that sentiment across. In both ways.
It's also very Taoistic of nature. Very Lao Tse's Tao Te Ching, which btw was written well over 2,500 years ago!
Rulof
Ventriloquist Cheater
Just tried 'm both and noticed this whilst sitting behind a FreeBSD NAT box, that BT has an upload and download rate of 40 KB/sec whereas WinMX claims no upload capabilities are present.
OK, so my ADSL hookup is pretty much swamped, still I oughta have some KB/secs spare...
Roelof -- the red nosed hacker -- Osinga
PS this is not criticism per se, just something I noticed, is all
Well, if that were indeed the case then even I would have no problem with such a patent.
:).
However, I do not think the patent is about an actual working AI that can speculate and even think!
It is about extrapolating based on too few data points
Roelof
Hm. Guess we were created in Its own image after all ;). Hack first, debug later.
For it is The motto of Gods!
Roelof
Sure, but the whole point of AI was that we were supposed to be able to ask it where it thought it was going... and that not only it would know, but would give a well thought out answer too!
Roelof
Me too. I also agree to a point with what you're saying.
...
;). Still, the being known as Maja provided a very welcome experience indeed. So can't say I'm sorry that I can not finish that point I was building.
However, easing off on the registration requirements does have two real advantages.
In the first place it brings ease of mind. Not to mention that it saves time. And stress. And thus prolongues live.
More important however is that it might help you leverage off of Metcalf's -- I believe -- law which states that the Value of a thing like this is e ^ Nusers, where ^ denotes to the power of.
It will also allow you to tap into the business market more easily. Due to brand recognition.
It is, in other words, not a clear cut situation.
For example, the very fact
Sorry, I got derailed by a beautiful being whom I spotted outside my window. Signing became talking, became drinking thee, more talking, etc. Couple of hours later I find that I lost the thread of my thoughts.
Too bad too, 'coz it was a darned good point
But I do imagine you'll be getting the gist of it. That a relaxed attitude will still allow you, even assist you, build a profitable thingum.
Remember Phil Katz? Destroyed himself quite comfortably on the proceeds of pkzip and maybe some pkarc too. Don't quite remember. Long time ago.
Roelof - as in (t)rue love - Osinga
PS Let us not forget that these days the Internet is of the utmost importance. Would you like to serve your customers from a Win32 base? I think not. So in that way the picture, even on the desktop, has changed irrevocably.
That I personally am not alltogether happy with Java, comes from my background. Though I do confess that the latest on the Java front looks good, what with the inlusion of polymorphic types.
What I am trying to bring across is that the adding of an extra layer of abstraction, or two, brings new opportunities for business. Brings new degrees of freedom to application development.
In short, it is all very interesting!
I have to disagree. And agree as well at the same time, of course. But the latter mostly because of the abundant choice in OSS, the majority of which are free.
The former, however, is the most important. At least, let us hope it is.
As I said in this topic earlier, most animals recognize good and respond in kind. As it happens, humans are basically animals. So humans validate good deeds appropriately.
One counter example to your case sprong to mind reading your post. And that was the case of OpenBSD. Theo de Raadt gets by one the sale of the CDs, T shirts and what nots.
Even though it is completely free for the taking.
And I do mean free as in free beer! [Hick]
I my self bought several T-shirts of, whatever, don't remember, and stickers and stuff just because a part of the sale went to sponsor the project.
I mean, why not?
As others said before. It makes me feel good.
Roelof -- pronounced [t]rue love -- Osinga
PS as to the market being limited, two things. First, yes, but growing! Second, sure so target where it is in large use. Like servers.
I agree. The registration is icing on the cake. If you have fun writing the daft thinglet than honest people will want to give you your just reward. Most people fortunately are honest.
Heck, my first attempt at shareware was even distributed in source code, I kid you not! Did not stop even companies from ordering it and sending checks. Received orders from universities up to Xerox and Intel.
This was a Smalltalk 'goodie' I had named LifeSaver and is probably still out there. Even though that was a decade ago.
In answer to my sibling poster, I failed miserably even though I succeeded beyond my wildest dreams!
Turned out it cost me almost as much to clear the checks I received as the checks were for. Making it a money loosing proposition I alas had to quit.
The second attempt at shareware was also not quite the money making venture I had hoped. This is the JPEG Sheriff. A file, especially JPEG files, verifier and collection completenes checker.
Here too necessity -- if one can call it that -- was the mother of invention. You see, I was a beta tester of Borland. Delphi 2.0 one of them. This had support for threads!
WOW!
That begged for thorough testing as I am sure you can imagine. But what to test it with? It had to make sense.
So I made the JPEG Sheriff a multi-threaded app. Just for the heck of it. Well, OK, and because I was a serieus bèta tester. Of course. Goes without saying.
This I put there as just freeware. Not in source form. Mostly because the code lacked a certain elegance from an OO (purists) view.
Especially the first three or four years I got all kinds of requests -- believe it or not, almost no bug reports -- which I gladly honored. If they made sense to me.
Also had people telling me, why don't you ask a bit for it? We'll be glad to pay.
Thus, when PayPal became available overseas I did just that.
Am still waiting for the first donation. This while the JPEG Sheriff has been downloaded several thousand times.
To be fair I must confess that I fell ill 'in my prime'. In 2000 I got carted into a hospital for MS (multiple sclerosis). Took me out of the loop for quite a while. Don't know what would've happened, could've happened, otherwise. Had I been able to keep up the good work.
So there you have it. It depends. With a bit of luck, with the right timing, you can build momentum. If you can do that...
But as to registration and stuff. Who needs it? If they don't want to pay. They don't pay. If they want to copy the source or even need the source as with Smalltalk, here it is. Well, if it's presentable, you understand.
That having been said however. Must confess I am playing with this nice idea about cryptographic hashes. Last work I did on the JPEG Sheriff, though unreleased due to my bout in the hospital, was with better check figures. Added the MDs (4&5) and SHA-1.
Which got me thinking about a registration system based on those hashes. You know, to get your name in the verification lists you make or request lists you send out.
I am still recuperating -- don't ever, EVER, try to get high on prednisone; it works, but boy will you regret it (1000 mg twice intravenously) -- but when I find my wits...
Does the name "Micro Soft" ring a bell?
Roelof
Sure, but the README clearly states that it is definitely alpha-test code and NOT yet ready for production use. (CVSup a week ago)
;)
At the end it also gave an URL about snapshots and stuff, http://www.mckusick.com/softdep/
Anyway it is a two step approach. First you make a snapshot in the filesystem being snapshotted, then you can use dump to safeguard it.
Roelof (Rulof almost as in the reindeer
Nah.
;).
It's not that the EE's are slipping... it's more like they're being diluted
Roelof - EE
That's what they used to say about minicomputers and later microcomputers. Sure, it has way less market share. But hardly less funcionality. In some regards it can probably still be considered to be more advanced (e.g. versioning engine) than the others.
Mini's, supermini's, microcomputers and LANs did eat into mainframe territory. There is a decided possibility Interbase can now eat harder into the market share kept by those vendors. Why not? Look at what MS did with their bought SQL Server.
But I do realize mainframes are still around and doing very well. Likewise will the mainframes in the RDBMS market.
SQL has nothing to do with the relational model as defined by Codd. Other than that it is one of the many query languages defined to work with implementations of the relational model. It turned into the most popular supplanting Sequel, Quel, QBE, etc. But popularity has nothing to do with essentiality.
:)
In fact you could make a case that there are no true RDBMSs at all. Most, if not all, lack some features or have added shortcuts to the pure relational model. E.g. the addition of TOP to the SELECT statement in MS SQL. You can only have TOP if you supported relational sequences. As far as I am aware sequences are not part of the relational model. Besides which MS does not provide a rigorous definition of TOP in terms of the relational model.
TOP can only make sense given an ordering such that no duplicates can occur. From that sequence you can take the first N tuples. Since SELECT is defined without ordering or distinctness requirements how can TOP be explained? Within the relational model that know nothing about, not recorgnizes, orderings implied by an implementation?
Another thing are relations. I believe Codd included the notion of domains in his model. That would indicate valid relations/joins can only be made on keys of the same domain. In practice you can darned well join anything with whatever you fancy whenever you want. Though it can be defended I don't think it is what was intended.
Ah well, whatever
I'm not sure it has already been mentioned in this still growing thread. But what got me interested in functional languages was the 1985 Byte August issue.
Back then, and maybe still , Byte had a tradition of devoting its August issue to a language. So after Smalltalk, Logo, etc. the 85 issue was about functional programming.
Scheme was featured in that issue.
B.t.w. one reason this particular implementation
could be a little on the slow side is because it has a reference counting GC. I needed that to have quick finalization.
I find functional languages extremely interesting. In fact I've spent well over three years implementing a kind of Hope derivative called very originally TLC (Typed Lambda Calculus).
What made me do it was that at the time (early nineties) I knew of no implementation that defined a FL metacircularly. Or rather, a strict FL. So there was a challenge and one I did succeed to master.
So, TLC (or Tender Loving Care as Hal Hildebrand would have it) is a strict FL implemented in itself. Albeit just the interpreter and type inference engine.
The compiler which currently compiles TLC to plain C with gets linked to the also C VM uses the FPM compilation scheme.
Though it doesn't use optimazation like memoization or abstract interpretation it is a strict language and as such supposed to be faster than a lazy implementation.
Alas, I have found even Java to be way faster. And I do not consider Java to be anywhere near fast.
Unfortunately it is not available for download. You see, I wrote it in Smalltalk/V-PM. Even the running C interpreter is currently useless since it's for BCOS2 and relies on a OS/2 database library. Been meaning to regenerate an interpreter that does not rely on any OS/2 quirks but somehow haven't come around to doing it.
My 2 c.
Roelof
Back in '87 I was running MicroSoft Xenix. Which was basically a System III clone or version which was more or less Release 7.
:).
When I came to Linux when it was one, or nearly 1.0 anyway, via the SLR distrib I stumbled upon Slack. The second Slack cought.
What I liked about Slack was that its resource configuration system was so very much akin to what I was used to!
Now as already has been described, back in the olden days things were a bleedin' mess. We went to UNIX because in those days UNIX was hyped as the new OS that would kill all other OS's. Something we've since seen when NT came along as well as when Linux broke through. (Let's face it, 10 percent of the Fortune-500 is a big deal).
From that mess two streams emerged. System V on the one hand and BSD on the other. BSD's rc system, like Slack's I still grock, SysV's OTOH can not capture my imagination.
In that sense Slack's BSD feel is anything but an illusion. That's it's not BSD is a given, but qua feel it's much more BSDish then SVish!
Please note that I come from the UNIX corner, not the BSD one
The thing I missed in this story is its ISO
code. It all hinges on its sovereignty. Which
in turn depends on more then just one positive
verdict in an English court.
Like, what's Sealand's ISO code? Can we surf
to http://havenco.sd/ or something? If not
then international recognition is still a
way off.
Roelof