Actually, I'd argue that the genericity of lisp's syntax is as much a hindrance as it is a help, emacs parenthesis-matching aside.
The above quoted lisp sexp could mean almost anything or nothing, depending on the context in which that sexp occurs. (Is it data? Is it code? In what evaluation context will it be processed?). Java, C, and most other languages at least give you more distinct contextual tokens to guide you in your understanding.
I'm not that impressed with the NASA study linked. Anyone who has done any Lisp programming in school will know that Lisp is a great language for algorithmic problem solving, and the problem used in the study really plays to that strength in Lisp. Perl would also likely kick C++ and Java's ass in this task. That doesn't mean that Perl or Lisp would necessarily be as good a choice as C++ or Java in a wide range of problem domains.
Lisp is a great language, but it traditionally has not interfaced as well with the Real World as a number of other languages that were more practically oriented.
I know that people absolutely rave about the pleasures of programming on all-Lisp machines, but most of us don't have those, and without either having a Java-style complete set of API's for doing GUI and system interfacing, or having C's ability to directly call system API's, one is left without the tools to directly influence the environment. And, yes, I'm sure there are implementations that have those things there, but without them being absolutely standardized, you can wind up locked in a ghetto of your own making.
As a side note, Kent, don't you think it a little churlish to denigrate Java as merely 'an abstract, high level assembly language'?;-)
Re:So what is new if some hates God?
on
God's Debris
·
· Score: 1
For all of you out there who think God is not watching, you are in the dark, and are in for a big unpleasant suprise.
Hm, sure you're not thinking of Santa Claus? The way I remember it, he was the guy in charge of the whole naughty/nice thing. Or was that L. Ron Hubbard? Joseph Smith? Muhammed? David Koresh? Jim Jones? Bob?
When you leave the land of evidence and logic for the land of Word From On High, you've really got to trust the person who tells you what that Word happens to be.
Counting college students making MP3 players for free, Linux wins, but capitalism hasn't seemed to recognize the vast Linux desktop market, if in fact it exists.
The Linux market is hard to make money on because all of the damn Linux users keep insisting on writing their own software and sharing it for free. This will change eventually, as the user base expands beyond the hard core tech set. Things like StarOffice and Evolution will help drive that by expanding the user base, but anyone proposing to sell software on Linux will have to have products that are good enough that they won't be easily duplicated by free software authors.
I myself have bought a dozen commercial linux packages, mostly RedHat distros and Loki games. If the non-technical Linux user population was to grow to Macintosh levels, to say nothing of Windows levels, you'd see plenty of people willing to buy Linux software.
Full agreement on the value of getting the GNOME and GTK apps going on Windows. If Evolution is to be a truly meaningful alternative to Outlook, it has to go where the Outlook users are.
That said, I'd love to know whether Evolution would turn into a COM shell on Windows or whether all of Bonobo would have to be ported? Is Bonobo similar enough to COM these days that the various Evolution modules could be rebuilt as COM objects easily?
Our laboratory started off heavily invested in Macintoshes, but over the last decade the numbers of Macs have greatly declined. Most of that decline is in favor of desktops running some flavor of Windows, of course, but looking at our database, it looks like Macs and Linux systems in the lab are just about at a dead heat. A number of those Linux systems are actually YellowDog Linux running on Apple hardware, even.
Walking around the lab, I see lots and lots of Gnome or KDE or WindowMaker desktops. Seems like all the k00l k1dz coming in want their own Linux desktops.
I would even go so far as to claim that it is more likely that a new machine coming into the laboratory will be a Linux system than that it would be a Macintosh, now. I know my division hasn't bought a Mac in the last 3 years, but we have several people with Linux desktops.
The lab certainly has far more people running Linux desktops than we do running MacOS X desktops today.
</anecdotal>
The ideal situation, as far as our lab is concerned for something like Evolution, would be if we could get it on Windows and Mac as well as Linux/UNIX.. being able to standardize on something other than Outlook would be a blessing from heaven for us. From what I understand of the ways of GTK, I imagine that it's more likely that we might see a Win32 port of Evolution than a MacOS [8,9] version. We do have enough Macs around that that's a real consideration, but the Mac users are used to having to run oddball software anyway.
Fun single player.. think "Norweigan Tomb Raider"
on
Rune for Linux Review
·
· Score: 2
It's a fun enough game to play single player, I think. The levels are interesting to look at, and kind of fun to play through. Combat's not all that, but getting to see the next level and cutscenes have proven to be enough of a motivator to keep me playing. The game is very well done from a thematic perspective, and it's big.. lots and lots of levels to satisfy your idle 'hm, i should be gaming right now' urges. Easy to pop up the game and work through a level or two before going back to whatever you were doing.
Exactly. Microsoft will be able to cover every wire protocol they have with this loophole. And the codicil that they get to verify the business model of the supplicant, and that the supplicant may not share that data makes the whole open protocol section a farce. Oh, you're a Samba hacker? Go suck eggs. Not to mention that Microsoft is not under any obligation if they are licensing the IP from a third party.. all Bill would have to do would be to start a new company, transfer Microsoft's secrets there, and license them back. Bam, problem solved.
Man, I could spend all day imagining ways for Microsoft to exploit these loopholes.
On the matter of the present loophole, however.. I wonder if we could impose upon Bruce Schneier to file comments during the published commentary period upon the real (lack of) necessity and utility of sooper sekret protocols for a real security system?
Take a look at section III.J, which shields Microsoft from having to disclose any authorization or authentication wire protocols. This loophole would allow Microsoft to maintain the secrecy of their BDC and PDC protocols, thereby locking out Samba. Ditto the III.J terms which require the company to have a 'verifiable business plan', on Microsoft's terms, in order to get release of this information. Likewise, Microsoft is able to force vendors who get access to the API's and protocols not to release them to the public.
All of these terms put together will shield Microsoft from revealing any interoperability information if the protocol in question includes authorization or authentication (which all protocols of significance will do), and will shield Microsoft in any case from having to let the grubby open sourcers get their hands on the info.
If you read the decision, you'll see that the judges are not establishing that the First Amendment always or typically trumps Copyright, just that the First Amendment trumps prior restraint in the form of preliminary injunctions in a trade secret case of this kind. The appeals court could still conceivaly come back and approve a final judgement against the distribution of DeCSS, and the court will surely uphold actions against individuals distributing copyrighted DVD materials through benefit of DeCSS.
Which puts matters back into the interesting realm of practical enforcement of copyright on an open Internet.
Of course, one of the "ideas" of XML is that you can just strip out all of the tags and have a document you can sort of read. That would be anathema to a Lisp person, and for good reason. Lisp is all about simple, minimalistic expression and manipulation of hierarchical data. XML is about an underspecified hodgepodge of structure and free form data.
Which is not to say that it's not useful, regardless.
But OpenSSL allows you to create a CA cert as well. Just preload your clients' browsers with your CA's public key cert and then you are every bit as safe against man in the middle attacks as you would be if Verisign or Thawte had signed your server's cert, SO LONG AS you keep your private CA key safe on your servers.
Which you have to do in the presence of a Verisign or Thawte-signed server cert anyway. All that having Verisign or Thawte sign your cert gets you is convenience (so you don't have to distribute your public cert keys to your browsers) and the ability to provide data to a wider audience.
The big hole that I wonder about is that most browsers have cert keys for a whole lot of CA's.. do they cross-check with each other to ensure that more than one Cert Authority among them are issuing keys for a given host or domain? If my web servers' keys are signed with Verisign and someone else can get a cert for the same domain signed by Joe Fourth Party CA, then new visitors to the fake site would be none the wiser.
Second this recommendation. Kohan feels a whole lot like a cross between Heroes of Might and Magic III, Civilization, and something like WarCraft. The underlying game is sort of tile based, where each company has a certain area of influence. This lets you think strategically, but the game's has the real-time pace and action.
is that Quake3 already has the ability to let the user turn down the texture details, an option that makes Quake3 look very much like it does with these cheating Radeon drivers.
All ATI is doing is saying, "no high quality mode for you, we want those high benchmark numbers", which is pathetic. If they want to go that route, they should put on their web page a notice that says, "our Radeon card sucks too hard to play Quake3 with high texture quality enabled. For the best playing experience, please play at low quality, or just go out and buy an Nvidia card."
MSN.com renders *perfectly* with my Mozilla nightly build. I changed the reported user agent as per an above post, and got to see everything on msn.com in all its glory.
Mozilla is far more compatible with all versions of IE than Netscape 4 ever was, and is indeed at least as W3C standards compliant as any version of Internet Explorer. Mozilla is the best tool for the job.
It's just that it happens to compete with the Microsoft hegemony, so I and other Mozilla users can go eat bark.
Nonsense. If that was what their users wanted, their users could have selected something other than 'High Quality' in the Quake3 video configuration screen.
ATI did this because it was what ATI wanted, a high framerate number on benchmarks.
Hm.. direct hardware acceleration.. you mean, like with OpenGL/GLX?
I don't get people militating to ditch X11 for mainstream desktops. The performance is fine, the technology is extensible, and you can run software written 15 years ago without any hassles.
GNU Emacs 21? I've been using xemacs for 5 years
on
GNU Emacs 21
·
· Score: 3, Informative
Seriously, XEmacs has been leading the FSF's GNU Emacs for a whole lotta years now, in terms of the object model, the GUI, and the packaging. What's new in GNU Emacs 21 to make it the new leader? And how long will it be before the XEmacs folks adopt the worthwhile new features?
The XEmacs/FSF Emacs split was the big project fork, for those of you who don't track Emacsen.
I called it a tangent because even if those projects achieve perfect, complete success (and patents and the like may well make such impossible), Microsoft's control of the clients will still allow them to push the client away from the (newly freed) protocols and standards towards something which they can and do control. Or at least towards something that will act as a network effect driver to keep their software relevant long enough while we chase them down.
I posted the dcerpc.net link because I do think that it's fantastic work.
I wish Luke Howard and his band of merry men every success in their Active Directory cloning work, Project XAD. Microsoft is genuinely pushing the state of the art of directory services and management forward (at least relative to the UNIX world, I haven't been much exposed to Novell's stuff), it would just be nice if they had the confidence in their stuff to document their protocols the way that Sun does.
Of course, Sun sells big boxes with electronics stuff in them, and Microsoft doesn't sell anything but bits. (Sooper SECRET bits).
If we Linux folks give up on the desktop, we will eventually have to give up on the server, unless the states and the DOJ get really wise about remedies.
As it stands now, the biggest single factor, by far, driving Microsoft server technology into the enterprise is the fact that Microsoft desktops want to talk to Microsoft servers. Jeremy Allison made this point on the LinuxToday talkbacks for this article, that the reason Exchange gets pulled into companies is because Outlook (part of office, and so bundled everywhere) has to talk to Exchange to do calendaring and scheduling. Exchange 2000, at least, needs to talk to ActiveDirectory. ActiveDirectory and Windows 2000 really, really want to absorb the DNS function (or else you're stuck with either a lot of manual overhead to manage the SRV records, or else you have to enable Dynamic DNS updates with a total lack of security because Microsoft doesn't support any open DDNS standards, they simply use the ActiveDirectory ACL's for security..)
See how that works? It's like dominoes, and Microsoft is supremely willing to set them up and knock them down.
Even though we spent 5+ years developing Ganymede, we're getting massive pressure on us to adopt ActiveDirectory because that's what Microsoft says Windows 2000 really needs, and because the protocols that Windows 2000 uses to talk to its directory services are proprietary and non-documented.
Microsoft is like a cuckoo bird, that lays its eggs in the nests of other birds. The eggs hatch, and out pop the baby cuckoos, who then proceed to shove all the other eggs out of the nest.
Slashdot Mirror
Only arguably truthful. Python is a far more dynamic language than Java, to be sure, but I've never seen any evidence that it is more portable.
Sounds wonderful. I still have my prized Dylan manual that I ordered from Apple's Cambridge labs back in the day.
Is ScriptX still a going concern in any way?
Please. I completely understand that about Lisp. I did quite a lot of Lisp in school, actually.
My point was that that aspect of Lisp can work against its readability, which was what this particular sub-thread was about, nothing more.
Read my post again, please.
Actually, I'd argue that the genericity of lisp's syntax is as much a hindrance as it is a help, emacs parenthesis-matching aside.
The above quoted lisp sexp could mean almost anything or nothing, depending on the context in which that sexp occurs. (Is it data? Is it code? In what evaluation context will it be processed?). Java, C, and most other languages at least give you more distinct contextual tokens to guide you in your understanding.
I'm not that impressed with the NASA study linked. Anyone who has done any Lisp programming in school will know that Lisp is a great language for algorithmic problem solving, and the problem used in the study really plays to that strength in Lisp. Perl would also likely kick C++ and Java's ass in this task. That doesn't mean that Perl or Lisp would necessarily be as good a choice as C++ or Java in a wide range of problem domains.
Lisp is a great language, but it traditionally has not interfaced as well with the Real World as a number of other languages that were more practically oriented.
I know that people absolutely rave about the pleasures of programming on all-Lisp machines, but most of us don't have those, and without either having a Java-style complete set of API's for doing GUI and system interfacing, or having C's ability to directly call system API's, one is left without the tools to directly influence the environment. And, yes, I'm sure there are implementations that have those things there, but without them being absolutely standardized, you can wind up locked in a ghetto of your own making.
As a side note, Kent, don't you think it a little churlish to denigrate Java as merely 'an abstract, high level assembly language'? ;-)
For all of you out there who think God is not watching, you are in the dark, and are in for a big unpleasant suprise.
Hm, sure you're not thinking of Santa Claus? The way I remember it, he was the guy in charge of the whole naughty/nice thing. Or was that L. Ron Hubbard? Joseph Smith? Muhammed? David Koresh? Jim Jones? Bob?
When you leave the land of evidence and logic for the land of Word From On High, you've really got to trust the person who tells you what that Word happens to be.
Why should we trust you?
Counting college students making MP3 players for free, Linux wins, but capitalism hasn't seemed to recognize the vast Linux desktop market, if in fact it exists.
The Linux market is hard to make money on because all of the damn Linux users keep insisting on writing their own software and sharing it for free. This will change eventually, as the user base expands beyond the hard core tech set. Things like StarOffice and Evolution will help drive that by expanding the user base, but anyone proposing to sell software on Linux will have to have products that are good enough that they won't be easily duplicated by free software authors.
I myself have bought a dozen commercial linux packages, mostly RedHat distros and Loki games. If the non-technical Linux user population was to grow to Macintosh levels, to say nothing of Windows levels, you'd see plenty of people willing to buy Linux software.
Sure would be nice if we had an LSB, though.
Full agreement on the value of getting the GNOME and GTK apps going on Windows. If Evolution is to be a truly meaningful alternative to Outlook, it has to go where the Outlook users are.
That said, I'd love to know whether Evolution would turn into a COM shell on Windows or whether all of Bonobo would have to be ported? Is Bonobo similar enough to COM these days that the various Evolution modules could be rebuilt as COM objects easily?
<anecdotal>
Our laboratory started off heavily invested in Macintoshes, but over the last decade the numbers of Macs have greatly declined. Most of that decline is in favor of desktops running some flavor of Windows, of course, but looking at our database, it looks like Macs and Linux systems in the lab are just about at a dead heat. A number of those Linux systems are actually YellowDog Linux running on Apple hardware, even.
Walking around the lab, I see lots and lots of Gnome or KDE or WindowMaker desktops. Seems like all the k00l k1dz coming in want their own Linux desktops.
I would even go so far as to claim that it is more likely that a new machine coming into the laboratory will be a Linux system than that it would be a Macintosh, now. I know my division hasn't bought a Mac in the last 3 years, but we have several people with Linux desktops.
The lab certainly has far more people running Linux desktops than we do running MacOS X desktops today.
</anecdotal>
The ideal situation, as far as our lab is concerned for something like Evolution, would be if we could get it on Windows and Mac as well as Linux/UNIX.. being able to standardize on something other than Outlook would be a blessing from heaven for us. From what I understand of the ways of GTK, I imagine that it's more likely that we might see a Win32 port of Evolution than a MacOS [8,9] version. We do have enough Macs around that that's a real consideration, but the Mac users are used to having to run oddball software anyway.
It's a fun enough game to play single player, I think. The levels are interesting to look at, and kind of fun to play through. Combat's not all that, but getting to see the next level and cutscenes have proven to be enough of a motivator to keep me playing. The game is very well done from a thematic perspective, and it's big.. lots and lots of levels to satisfy your idle 'hm, i should be gaming right now' urges. Easy to pop up the game and work through a level or two before going back to whatever you were doing.
Exactly. Microsoft will be able to cover every wire protocol they have with this loophole. And the codicil that they get to verify the business model of the supplicant, and that the supplicant may not share that data makes the whole open protocol section a farce. Oh, you're a Samba hacker? Go suck eggs. Not to mention that Microsoft is not under any obligation if they are licensing the IP from a third party.. all Bill would have to do would be to start a new company, transfer Microsoft's secrets there, and license them back. Bam, problem solved.
Man, I could spend all day imagining ways for Microsoft to exploit these loopholes.
On the matter of the present loophole, however.. I wonder if we could impose upon Bruce Schneier to file comments during the published commentary period upon the real (lack of) necessity and utility of sooper sekret protocols for a real security system?
Take a look at section III.J, which shields Microsoft from having to disclose any authorization or authentication wire protocols. This loophole would allow Microsoft to maintain the secrecy of their BDC and PDC protocols, thereby locking out Samba. Ditto the III.J terms which require the company to have a 'verifiable business plan', on Microsoft's terms, in order to get release of this information. Likewise, Microsoft is able to force vendors who get access to the API's and protocols not to release them to the public.
All of these terms put together will shield Microsoft from revealing any interoperability information if the protocol in question includes authorization or authentication (which all protocols of significance will do), and will shield Microsoft in any case from having to let the grubby open sourcers get their hands on the info.
Yuck. Microsoft has very good lawyers, indeed.
If you read the decision, you'll see that the judges are not establishing that the First Amendment always or typically trumps Copyright, just that the First Amendment trumps prior restraint in the form of preliminary injunctions in a trade secret case of this kind. The appeals court could still conceivaly come back and approve a final judgement against the distribution of DeCSS, and the court will surely uphold actions against individuals distributing copyrighted DVD materials through benefit of DeCSS.
Which puts matters back into the interesting realm of practical enforcement of copyright on an open Internet.
Of course, one of the "ideas" of XML is that you can just strip out all of the tags and have a document you can sort of read. That would be anathema to a Lisp person, and for good reason. Lisp is all about simple, minimalistic expression and manipulation of hierarchical data. XML is about an underspecified hodgepodge of structure and free form data.
Which is not to say that it's not useful, regardless.
But OpenSSL allows you to create a CA cert as well. Just preload your clients' browsers with your CA's public key cert and then you are every bit as safe against man in the middle attacks as you would be if Verisign or Thawte had signed your server's cert, SO LONG AS you keep your private CA key safe on your servers.
Which you have to do in the presence of a Verisign or Thawte-signed server cert anyway. All that having Verisign or Thawte sign your cert gets you is convenience (so you don't have to distribute your public cert keys to your browsers) and the ability to provide data to a wider audience.
The big hole that I wonder about is that most browsers have cert keys for a whole lot of CA's.. do they cross-check with each other to ensure that more than one Cert Authority among them are issuing keys for a given host or domain? If my web servers' keys are signed with Verisign and someone else can get a cert for the same domain signed by Joe Fourth Party CA, then new visitors to the fake site would be none the wiser.
Second this recommendation. Kohan feels a whole lot like a cross between Heroes of Might and Magic III, Civilization, and something like WarCraft. The underlying game is sort of tile based, where each company has a certain area of influence. This lets you think strategically, but the game's has the real-time pace and action.
And it kicks ass for multiplayer.
is that Quake3 already has the ability to let the user turn down the texture details, an option that makes Quake3 look very much like it does with these cheating Radeon drivers.
All ATI is doing is saying, "no high quality mode for you, we want those high benchmark numbers", which is pathetic. If they want to go that route, they should put on their web page a notice that says, "our Radeon card sucks too hard to play Quake3 with high texture quality enabled. For the best playing experience, please play at low quality, or just go out and buy an Nvidia card."
MSN.com renders *perfectly* with my Mozilla nightly build. I changed the reported user agent as per an above post, and got to see everything on msn.com in all its glory.
Mozilla is far more compatible with all versions of IE than Netscape 4 ever was, and is indeed at least as W3C standards compliant as any version of Internet Explorer. Mozilla is the best tool for the job.
It's just that it happens to compete with the Microsoft hegemony, so I and other Mozilla users can go eat bark.
Microsoft's arrogance never ceases to astonish.
Nonsense. If that was what their users wanted, their users could have selected something other than 'High Quality' in the Quake3 video configuration screen.
ATI did this because it was what ATI wanted, a high framerate number on benchmarks.
Hm.. direct hardware acceleration.. you mean, like with OpenGL/GLX?
I don't get people militating to ditch X11 for mainstream desktops. The performance is fine, the technology is extensible, and you can run software written 15 years ago without any hassles.
Seriously, XEmacs has been leading the FSF's GNU Emacs for a whole lotta years now, in terms of the object model, the GUI, and the packaging. What's new in GNU Emacs 21 to make it the new leader? And how long will it be before the XEmacs folks adopt the worthwhile new features?
The XEmacs/FSF Emacs split was the big project fork, for those of you who don't track Emacsen.
I called it a tangent because even if those projects achieve perfect, complete success (and patents and the like may well make such impossible), Microsoft's control of the clients will still allow them to push the client away from the (newly freed) protocols and standards towards something which they can and do control. Or at least towards something that will act as a network effect driver to keep their software relevant long enough while we chase them down.
I posted the dcerpc.net link because I do think that it's fantastic work.
Thanks, that cheers me up considerably. ;-)
<tangent>I wish Luke Howard and his band of merry men every success in their Active Directory cloning work, Project XAD. Microsoft is genuinely pushing the state of the art of directory services and management forward (at least relative to the UNIX world, I haven't been much exposed to Novell's stuff), it would just be nice if they had the confidence in their stuff to document their protocols the way that Sun does.
Of course, Sun sells big boxes with electronics stuff in them, and Microsoft doesn't sell anything but bits. (Sooper SECRET bits).
</tangent>If we Linux folks give up on the desktop, we will eventually have to give up on the server, unless the states and the DOJ get really wise about remedies.
As it stands now, the biggest single factor, by far, driving Microsoft server technology into the enterprise is the fact that Microsoft desktops want to talk to Microsoft servers. Jeremy Allison made this point on the LinuxToday talkbacks for this article, that the reason Exchange gets pulled into companies is because Outlook (part of office, and so bundled everywhere) has to talk to Exchange to do calendaring and scheduling. Exchange 2000, at least, needs to talk to ActiveDirectory. ActiveDirectory and Windows 2000 really, really want to absorb the DNS function (or else you're stuck with either a lot of manual overhead to manage the SRV records, or else you have to enable Dynamic DNS updates with a total lack of security because Microsoft doesn't support any open DDNS standards, they simply use the ActiveDirectory ACL's for security..)
See how that works? It's like dominoes, and Microsoft is supremely willing to set them up and knock them down.
Even though we spent 5+ years developing Ganymede, we're getting massive pressure on us to adopt ActiveDirectory because that's what Microsoft says Windows 2000 really needs, and because the protocols that Windows 2000 uses to talk to its directory services are proprietary and non-documented.
Microsoft is like a cuckoo bird, that lays its eggs in the nests of other birds. The eggs hatch, and out pop the baby cuckoos, who then proceed to shove all the other eggs out of the nest.
I hope that folks remember this sort of flamage when the KDE Kids next komplain about Gnome people dissing their work.
Can't we all grow up?