one thing Google can do is look at their user base's interaction with a given query
Now that's an answer:-)
You must be right that they're experimenting with this -- I noticed that result pages sometimes give redirect links that go through Google's servers. Things like http://google.com/somescript?http://www.aol.com, which btw, gave me an eerie feeling of being tracked...
(Not sure what was determining this -- e.g. I'd get such links with Mozilla, but not with Safari at right the same time.)
PageRank was a good idea and clearly rooted in the technology; but people have caught up on it. Now how many others like that can Google really hope to have so that it stays useful?
One, Silverstein acknowledges that AI problems are basically hopeless (gonna take "about 200 to 300 years").
Two, when asked if PageRank is dead and what they are doing to fight false popularity, he says they are "tweaking it in new ways".
Three, when asked how ("do you have algorithms?) he answers,
Well, there are certainly other techniques that we are using. Talking
about it is the trickier part. In broad terms, techniques we use fall
into, like, two or three categories, and one is we try to understand
and leverage human intelligence. We look for signals that people put in
to indicate intelligence, like deciding to link from one page to
another or annotating text with the description of what the text is
about.
OK, they are looking at the anchor text. Then what? As long as HTML is the language, I'm afraid there aren't that many more things they can do.
By 1989, I had already started numbering Apollo
objects using gaulish gods. One which I had not used was Toutatis since I thought
it was an invention of
Goscinny and Uderzo,
authors of the well known comic book series "Les aventures d'Asterix". There are
several dozens sites about this comic book series, you may want to look at few
of them :
One of their constant saying is "By Toutatis", another one is that their only
fear is that the sky may fall onto their heads.
I discovered my ignorance of gaulish culture when I learned that Toutatis was
( or had been ) a real God. I also learned that the citation in Asterix was
not a joke, but that it had been reported by some historians of Alexander the
great who had met some gaulish warriors ( who had once invaded Italy and Great
Britain ).
One of the first thing we learned about Toutatis was its record low inclination.
This meant that it is indeed ( in a remote future ) a good candidate to fall
onto our heads. The name stuck almost immediately at the telescope when I proposed
it. Toutatis, also sometimes spelled "teutates" is a totemic deity, to which
human sacrifices were made.
Don't be misled, very few french persons do know about the cruel god Toutatis,
but most will talk to you about Asterix and his friends if you come to swear
" By Toutatis ! ", provided you get the right (i.e. french) accent...
Why is Barrett complaining about teachers being incompetent when the education system is paying them wages that could only attract people who are incompetent
Hmm, I think you misunderstand... and contradict yourself. Barrett is precisely suggesting that we as a nation devote more tax money to research and education (thus making it attractive to better teachers), so that we can better compete, rather than unproductively waste money on subsidies (agricultural, steel, tax cuts, etc.) that only temporarily hide the effects of competition.
More spending on education sounds like exactly what you want (based on your sentence above).
How's that for a zigzag strategy? They used to have the inventor of IMAP, actively working on Mozilla Mail. No longer it seems? his bugzilla address isn't netscape.com or aol.net anymore.
The
core of OSX's Network Utility application is a little command line
program (...)stroke.
Stroke isn't "the" core of Network Utility... the app drives eleven command line programs, of which stroke is just one. (The others are in [/usr]/sbin.)
The sort of apps you're talking about here (there are tons more, e.g. SimpleWget) are very limited though: basically they just exec CLI apps, and all they can be is some sort of form-based interface to enter CLI arguments. Which invariably turns out to be crippling -- soon enough we're back to the command line and manpage, if we want to dig -x or wget less than the entire internet.
Much more interesting (and relevant to the TUI question, IMO) are apps that expose libraries: ReSTedit, TestXSLT, TeXShop,...
Recently I got a (used) Forester, and within a week the "check engine" light came up. To even diagnose what it means requires an electronic gizmo that Subaru 1) charges an hour for just plugging in 2) will only sell at outrageous prices. This effectively locked out my trusty local mechanic in favor of the Subaru dealership. It's a trend with all car makers, he said.
(Happy ending, we found another mechanic who could bypass that, and thanks to good Connecticut consumer laws, the dealer paid for the repair. Still, that was a sobering experience -- especially for him, who got the car from this very same dealership...)
Seeing as Vasters' site is still slow as hell, people might be interested in the "reply to the replies" posted there. (I've left his mangled links ["guid"?!; the site is a complete mess in my browser] as is, I you can more or less figure out where the external ones go.)
>
Of course my letter to Aiden is prompting some opposition.It may be worth noting that a very large proportionof the codethat I writeendsup being publicand there's more stuff brewing as we speak. There islittle need to educate me about giving. I am an educator. Sharing insight and therefore sharing manifestations of that insight in form of source code is my mission and part of my business. But this is not the businessmy clients are in and neitheris it the business ofmost of the thousands of developers I am honored to speakfor at conferenceseach year. Their business is about being paid for writing software. If they weren't paid, I wouldn't be paid. My job description is to figure out fundamental stuff anduse my natural"understand very complex things thoroughly and rapidly" skill that I was luckily blessed with, so that I can explain those things to them and they can focus on solving customer problems. My free stuff helps my customers and is also playing a marketing role for me an my company. Our free stuff is a calculated investment. We can and do attach a number to it. dasBlog is a freebie for others but represents a significantinvestment that's worth several tens of thousands of Euros. It's not free, at all.
We support a project that brings us some indirect value.However, we do not in any wayforce any code republishing requirementsupon the folks who'd like to reuse our code(we have a strict "no GPL" policy; our code is BSD licensed). We don't depend on a community of volunteers toturn dasBlog intoa dominant blogging tool that we can benefit from by commerically supporting it.We believe that if we wanted to benefit from the software directly, we would have to rearchitect and rebuild it (or at least restrict ourselves to newtelligence contributions) and then sell it as a fully supported commercial product. My personal sense of respect and fairness tells me that I will not and should not exploit the others guys that have contributed to the free version of dasBlog. It's their hobby and their work is their work. I think a company like Red Hat, whichis a public company (whichdid yielda significant "going public benefit" to their founders)and is profitingfrom the work of countless unpaid volunteersand enthusiasts, is a very clever, but deeply unethical entity.
Ido believe in giving andI do believe that there is value for the community at large in sharing insight through source code. But we don't share theview that software is free or should be free. Someone pays for it. We have an investment in software that is free for others to use, MySQL has, HP has, IBM has, Sun has and - believe it or not - even Microsoft has. We do that as part of a well thought out and well understood business strategy.
I thought, I'll call up Apple's CS people (...) when he discovered that the connection to the internet was through a linuux box, he demanded that I reboot it. I balked at that, and let him know in no uncertain terms that this was utterly unacceptable.
Sorry but Tier-1 support addresses the needs of clueless users, and you were just that. Obviously, having you turn off the DHCP server you had set up in conflict with your Airport's was the right thing to do.
One thing I learned from the linux tools was that the airport was running a DHCP server.
Pray tell, which of those tools couldn't you have used on the Mac? If anything it has more, including an obvious GUI to configure Airport, which will tell you all about the DHCP in it.
Generally speaking the problem is that anything bleeding edge is better supported under Linux. Of course that's only natural, but it's (imho) reason enough to install Linux too.
The relevant passage from the article you cited is: (...)
Well, that and (my emphasis):
"Gecko does not fit in well with
Objective C and Cocoa (...) Gecko is a thick cross-platform codebase that
doesn't reuse
components that have already been implemented by the OS."
So yeah, it looks to me like they did go for maximal reuse. It's just that, with all cross-platform requirements gone, the choice was not between Gecko and KHTML but between Gecko and KHTML + Cocoa...
I will agree with you that the great-granparent overreached with that "virtual standstill" comment. But still I think you're stretching those examples quite a bit. Maybe you can find better ones:-)
As you admit OS X makes significant (I'd say huge) reuse of (Mach,) BSD and NeXTstep. Didn't previous attempts to do it from scratch almost kill Apple? (I forget the projects' names... Taligent? Copland?)
As to KHTML vs. Gecko, they seem to simply not have found the latter so "superior"...
"It seems unlikely this is going to create a material, significant security problem," said Rob Enderle, a technology expert and principal analyst with the Enderle Group.
you exaggerate the effects of not being able to reuse code
Case in point: Microsoft started nearly from scratch (licensed a simpler browser, IIRC) with IE, at around the same time Netscape decided it was unable to maintain its aging source code. IE overtook Netscape
You might want to read Eric Sink on how this happened:
What was interesting was the day we learned that Netscape didn't have the funding to keep up with Microsoft. (...) At one of those meetings we sat down for a talk which was a major turning point for me and for Spyglass. Scott told me that the IE team had over 1,000 people.
I was stunned. That was 50 times the size of the Spyglass browser team. It was almost as many people as Netscape had in their whole company. I could have written the rest of the history of web browsers on that day -- no other outcomes were possible.
Apple then did the same years later, starting with KHTML (generally considered inferior to Gecko), and within a pretty short time has a really polished Safari browser.
Well you're making the other guy's point, since KHTML was, precisely, (open source and) being reused.
Slashdot Mirror
Now that's an answer :-)
You must be right that they're experimenting with this -- I noticed that result pages sometimes give redirect links that go through Google's servers. Things like http://google.com/somescript?http://www.aol.com, which btw, gave me an eerie feeling of being tracked...
(Not sure what was determining this -- e.g. I'd get such links with Mozilla, but not with Safari at right the same time.)
One, Silverstein acknowledges that AI problems are basically hopeless (gonna take "about 200 to 300 years").
Two, when asked if PageRank is dead and what they are doing to fight false popularity, he says they are "tweaking it in new ways".
Three, when asked how ("do you have algorithms?) he answers,
OK, they are looking at the anchor text. Then what? As long as HTML is the language, I'm afraid there aren't that many more things they can do.Time to cash in, perhaps?
By 1989, I had already started numbering Apollo objects using gaulish gods. One which I had not used was Toutatis since I thought it was an invention of Goscinny and Uderzo, authors of the well known comic book series "Les aventures d'Asterix". There are several dozens sites about this comic book series, you may want to look at few of them :
One of their constant saying is "By Toutatis", another one is that their only fear is that the sky may fall onto their heads.
I discovered my ignorance of gaulish culture when I learned that Toutatis was ( or had been ) a real God. I also learned that the citation in Asterix was not a joke, but that it had been reported by some historians of Alexander the great who had met some gaulish warriors ( who had once invaded Italy and Great Britain ).
One of the first thing we learned about Toutatis was its record low inclination. This meant that it is indeed ( in a remote future ) a good candidate to fall onto our heads. The name stuck almost immediately at the telescope when I proposed it. Toutatis, also sometimes spelled "teutates" is a totemic deity, to which human sacrifices were made.
Don't be misled, very few french persons do know about the cruel god Toutatis, but most will talk to you about Asterix and his friends if you come to swear " By Toutatis ! ", provided you get the right (i.e. french) accent...
Hmm, I think you misunderstand... and contradict yourself. Barrett is precisely suggesting that we as a nation devote more tax money to research and education (thus making it attractive to better teachers), so that we can better compete, rather than unproductively waste money on subsidies (agricultural, steel, tax cuts, etc.) that only temporarily hide the effects of competition.
More spending on education sounds like exactly what you want (based on your sentence above).
Windows?
How's that for a zigzag strategy? They used to have the inventor of IMAP, actively working on Mozilla Mail. No longer it seems? his bugzilla address isn't netscape.com or aol.net anymore.
Stroke isn't "the" core of Network Utility... the app drives eleven command line programs, of which stroke is just one. (The others are in [/usr]/sbin.)
The sort of apps you're talking about here (there are tons more, e.g. SimpleWget) are very limited though: basically they just exec CLI apps, and all they can be is some sort of form-based interface to enter CLI arguments. Which invariably turns out to be crippling -- soon enough we're back to the command line and manpage, if we want to dig -x or wget less than the entire internet.
Much more interesting (and relevant to the TUI question, IMO) are apps that expose libraries: ReSTedit, TestXSLT, TeXShop,...
/.'s search seems down at the moment, so: here.
Well, what's for sure is that they still won't run on powerpc (12" powerbooks, low-end G5s,...). Please wake up nVidia!
So, talking about benchmarks, how does that stack up against gcc?
Ever heard of Applescript? This lets you not only program UI actions in code, but even automatically *record* them into code...
(Happy ending, we found another mechanic who could bypass that, and thanks to good Connecticut consumer laws, the dealer paid for the repair. Still, that was a sobering experience -- especially for him, who got the car from this very same dealership...)
Oh my, what /. is turning into.
Sorry but Tier-1 support addresses the needs of clueless users, and you were just that. Obviously, having you turn off the DHCP server you had set up in conflict with your Airport's was the right thing to do.
One thing I learned from the linux tools was that the airport was running a DHCP server.
Pray tell, which of those tools couldn't you have used on the Mac? If anything it has more, including an obvious GUI to configure Airport, which will tell you all about the DHCP in it.
http://thread.gmane.org/gmane.comp.python.apple/21 91
5 56
1 77
http://thread.gmane.org/gmane.comp.python.apple/2
http://thread.gmane.org/gmane.comp.python.apple/3
Generally speaking the problem is that anything bleeding edge is better supported under Linux. Of course that's only natural, but it's (imho) reason enough to install Linux too.
Perhaps because it uses a video card which nVidia can't be bothered to support on PPC?
Cf.
- (lack of) linux support
- requests to nVidia
- (lack of) response from nVidia
Now if only we had MOL running under Darwin/OS X, that would make for a great Linux-on-Mac solution.Last I checked, many (most) Linux IDEs weren't available under OS X:
Google is your friend, the first six hits (after which I stopped checking) all send you to the right place.
Someone failed to teach you the difference between a constant and a variable, I'm afraid.
Pudge seems to believe that researching and crediting sources is retarded behavior. May be right, may be wrong...
Well, that and (my emphasis):
So yeah, it looks to me like they did go for maximal reuse. It's just that, with all cross-platform requirements gone, the choice was not between Gecko and KHTML but between Gecko and KHTML + Cocoa...As you admit OS X makes significant (I'd say huge) reuse of (Mach,) BSD and NeXTstep. Didn't previous attempts to do it from scratch almost kill Apple? (I forget the projects' names... Taligent? Copland?)
As to KHTML vs. Gecko, they seem to simply not have found the latter so "superior"...
Case in point: Microsoft started nearly from scratch (licensed a simpler browser, IIRC) with IE, at around the same time Netscape decided it was unable to maintain its aging source code. IE overtook Netscape
You might want to read Eric Sink on how this happened:
Apple then did the same years later, starting with KHTML (generally considered inferior to Gecko), and within a pretty short time has a really polished Safari browser.Well you're making the other guy's point, since KHTML was, precisely, (open source and) being reused.